##import data

pc3_ctrl_217 <- read.table(“217_pc3_ctrl_pc3_ctrl_seg_nobad.txt”, header = TRUE)

#subset data to keep columns: chr, start, end, and seg.mean

pc3_ctrl_217 <- pc3_ctrl_217[, c(“chr”, “start”, “end”, “seg.mean”, “abspos”)]

#1: cutting the genome into losses, gains, and neutral regions # Using seg.mean column

Cut offs:

File_neutral: 0.8 – 1.2

File_gain: > 1.2

File_loss: < 0.8

pc3_ctrl_217_NEUTRAL <- pc3_ctrl_217[pc3_ctrl_217$seg.mean >= 0.8 & pc3_ctrl_217$seg.mean <= 1.2, ]

pc3_ctrl_217_GAIN <- pc3_ctrl_217[pc3_ctrl_217$seg.mean > 1.2, ]

pc3_ctrl_217_LOSS <- pc3_ctrl_217[pc3_ctrl_217$seg.mean < 0.8, ]

#load gene_overlaps_CNA.txt into R

gene_overlaps <- read.table(“gene_overlaps_CNA.txt”, header = TRUE)

Combine rows w/ (1) same seg/means & (2) sequential end/start

library(dplyr)

#pc3_ctrl_217_GAIN <- pc3_ctrl_217_GAIN %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end), abspos = first(abspos)) %>% # ungroup()

#write.table(pc3_ctrl_217_GAIN, file = “pc3_ctrl_217_GAIN.txt”)

#pc3_ctrl_217_LOSS <- pc3_ctrl_217_LOSS %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end), abspos = first(abspos)) %>% # ungroup()

#write.table(pc3_ctrl_217_LOSS, file = “pc3_ctrl_217_LOSS.txt”)

#pc3_ctrl_217_NEUTRAL <- pc3_ctrl_217_NEUTRAL %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end), abspos = first(abspos)) %>% # ungroup()

#write.table(pc3_ctrl_217_NEUTRAL, file = “pc3_ctrl_217_NEUTRAL.txt”)

###Mike’s Lines #files A - Loss/Gain/Neutral #files B - Gene overlap

#want to merge gene names w/ segmented DNA

overlap <- read.table(“gene_overlaps_CNA.txt”) head(overlap)

unique_rows <- !duplicated(overlap$gene) unique_df <- overlap[unique_rows, ] head(unique_df)

gain<- read.table(“pc3_ctrl_217_GAIN.txt”) head(gain)

neutral <- read.table(“pc3_ctrl_217_NEUTRAL.txt”) head(neutral)

loss <- read.table(“pc3_ctrl_217_LOSS.txt”) head(loss)

###data A is overlap file data_A = overlap write.table(data_A, “data_A.txt”, sep = “, quote = F)

###data B is segmented copy number file # do gain, neutral, and loss data_B = gain write.table(data_B, “data_B.txt”, sep = “, quote = F) unique_data_B <- unique(data_B)

#merge.py outputs output_file.csv, but then I changed name to #respective sample & duplication level (GAIN, NEUTRAL, LOSS)

merged_217_GAIN <- read.csv(“output_217_GAIN.csv”)

merged_217_NEUTRAL <- read.csv(“output_217_NEUTRAL.csv”)

merged_217_LOSS <- read.csv(“output_217_LOSS.csv”)

###Now filter for genes that occur in multiple files #ex: #GAIN - A, B, C, D #LOSS - A, E, F, G #NEUTRAL - B, C, H, I

#Gets rid of A & B

#OG code: #filtered_data <- fileA %>% #anti_join(fileB, by = “gene”) %>% # anti_join(fileC, by = “gene”)

unique_merged_217_GAIN <- merged_217_GAIN %>% anti_join(merged_217_NEUTRAL, by = “gene”) %>% anti_join(merged_217_LOSS, by = “gene”)

unique_merged_217_NEUTRAL <- merged_217_NEUTRAL %>% anti_join(merged_217_GAIN, by = “gene”) %>% anti_join(merged_217_LOSS, by = “gene”)

unique_merged_217_LOSS <- merged_217_LOSS %>% anti_join(merged_217_GAIN, by = “gene”) %>% anti_join(merged_217_NEUTRAL, by = “gene”)

write.csv(unique_merged_217_GAIN, “unique_output_217_GAIN.csv”) write.csv(unique_merged_217_NEUTRAL, “unique_output_217_NEUTRAL.csv”) write.csv(unique_merged_217_LOSS, “unique_output_217_LOSS.csv”)

#####################Below: Cleaned Up Version of Above

pc3_ctrl_224 <- read.table(“224_pc3_ctrl_pc3_ctrl_seg_nobad.txt”, header = TRUE)

#change working directory back to monica_genes

pc3_ctrl_224 <- pc3_ctrl_224[, c(“chr”, “start”, “end”, “seg.mean”)]

pc3_ctrl_224_GAIN <- pc3_ctrl_224[pc3_ctrl_224$seg.mean > 1.2, ]

pc3_ctrl_224_NEUTRAL <- pc3_ctrl_224[pc3_ctrl_224$seg.mean >= 0.8 & pc3_ctrl_224$seg.mean <= 1.2, ]

pc3_ctrl_224_LOSS <- pc3_ctrl_224[pc3_ctrl_224$seg.mean < 0.8, ]

#pc3_ctrl_224_GAIN <- pc3_ctrl_224_GAIN %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end)) %>% # ungroup()

#pc3_ctrl_224_NEUTRAL <- pc3_ctrl_224_NEUTRAL %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end)) %>% # ungroup()

#pc3_ctrl_224_LOSS <- pc3_ctrl_224_LOSS %>% # group_by(chr, seg.mean) %>% # summarize(start = min(start), end = max(end)) %>% # ungroup()

gain <- pc3_ctrl_224_GAIN

neutral <- pc3_ctrl_224_NEUTRAL

loss <- pc3_ctrl_224_LOSS

data_B = gain write.table(data_B, “data_B.txt”, sep = “, quote = F) unique_data_B <- unique(data_B)

#NOW DO PYTHON #**change name out output_file.txt

data_B = neutral write.table(data_B, “data_B.txt”, sep = “, quote = F) unique_data_B <- unique(data_B)

#NOW DO PYTHON #**change name out output_file.txt

data_B = loss write.table(data_B, “data_B.txt”, sep = “, quote = F) unique_data_B <- unique(data_B)

#NOW DO PYTHON #**change name out output_file.txt

output_224_GAIN <- read.csv(“output_224_GAIN.csv”)

output_224_NEUTRAL <- read.csv(“output_224_NEUTRAL.csv”)

output_224_LOSS <- read.csv(“output_224_LOSS.csv”)

unique_output_224_GAIN <- output_224_GAIN %>% anti_join(output_224_NEUTRAL, by = “gene”) %>% anti_join(output_224_LOSS, by = “gene”)

unique_output_224_NEUTRAL <- output_224_NEUTRAL %>% anti_join(output_224_GAIN, by = “gene”) %>% anti_join(output_224_LOSS, by = “gene”)

unique_output_224_LOSS <- output_224_LOSS %>% anti_join(output_224_GAIN, by = “gene”) %>% anti_join(output_224_NEUTRAL, by = “gene”)

write.csv(unique_output_224_GAIN, “unique_output_224_GAIN.csv”) write.csv(unique_output_224_NEUTRAL, “unique_output_224_NEUTRAL.csv”) write.csv(unique_output_224_LOSS, “unique_output_224_LOSS.csv”)

########NEXT STEP: Merge unique outputs w/ counts

#Want: gene name, seg.mean, type of protein, & counts columns

pc3_224_counts <- read.table(“224_pc3_ctrlpc3_ctrl.count.txt”, header = FALSE, sep = “)

names(pc3_224_counts)[names(pc3_224_counts) == “V1”] <- “gene_ID” names(pc3_224_counts)[names(pc3_224_counts) == “V2”] <- “gene” names(pc3_224_counts)[names(pc3_224_counts) == “V3”] <- “type” names(pc3_224_counts)[names(pc3_224_counts) == “V4”] <- “counts”

write.csv(pc3_224_counts, “pc3_224_counts.csv”)

Load the dplyr package

library(dplyr)

Merge the data frames based on the ‘gene’ column

pc3_217_GAIN <- inner_join(unique_output_217_GAIN, pc3_217_counts, by = ‘gene’) pc3_217_NEUTRAL <- inner_join(unique_output_217_NEUTRAL, pc3_217_counts, by = ‘gene’) pc3_217_LOSS <- inner_join(unique_output_217_LOSS, pc3_217_counts, by = ‘gene’)

Select the desired columns

selected_columns <- c(‘chr’, ‘gene’, ‘type’, ‘seg.mean’, ‘counts’) pc3_217_GAIN <- pc3_217_GAIN[selected_columns] pc3_217_NEUTRAL <- pc3_217_NEUTRAL[selected_columns] pc3_217_LOSS <- pc3_217_LOSS[selected_columns]

Save the final merged data frame to a new file

write.csv(pc3_217_GAIN, ‘final_pc3_217_GAIN.csv’) write.csv(pc3_217_NEUTRAL, ‘final_pc3_217_NEUTRAL.csv’) write.csv(pc3_217_LOSS, ‘final_pc3_217_LOSS.csv’)

MERGE AGAIN

#09/25/2023 #need to merge unique files w/ seg.mean files again to get gene ID #then merge with TPM files to get (1) TPM and (2) length

unique_output_217_GAIN <- read.csv(“unique_output_217_GAIN.csv”) unique_output_217_NEUTRAL <- read.csv(“unique_output_217_NEUTRAL.csv”) unique_output_217_LOSS <- read.csv(“unique_output_217_LOSS.csv”)

pc3_217_gene_counts <- read.csv(“pc3_217_counts.csv”, header = TRUE)

Merge the data frames based on the ‘gene’ column

merged_pc3_217_GAIN <- inner_join(unique_output_217_GAIN, pc3_217_gene_counts, by = ‘gene’) merged_pc3_217_NEUTRAL <- inner_join(unique_output_217_NEUTRAL, pc3_217_gene_counts, by = ‘gene’) merged_pc3_217_LOSS <- inner_join(unique_output_217_LOSS, pc3_217_gene_counts, by = ‘gene’)

Import TPM files

pc3_217_TPM <- read.table(“217_pc3_ctrlpc3_ctrl.genes.results.txt”, sep = “, header = TRUE) #rename column names(pc3_217_TPM)[names(pc3_217_TPM) == ”gene_id”] <-”gene_ID”

Merge files w/ TPM to add TPM and length columns

merged_pc3_217_GAIN <- inner_join(merged_pc3_217_GAIN, pc3_217_TPM, by = ‘gene_ID’) merged_pc3_217_NEUTRAL <- inner_join(merged_pc3_217_NEUTRAL, pc3_217_TPM, by = ‘gene_ID’) merged_pc3_217_LOSS <- inner_join(merged_pc3_217_LOSS, pc3_217_TPM, by = ‘gene_ID’)

Select the UPDATED desired columns

selected_columns <- c(‘chr’, ‘gene_ID’, ‘gene’, ‘type’, ‘seg.mean’, ‘abspos’, ‘counts’, ‘TPM’, ‘length’) merged_pc3_217_GAIN <- merged_pc3_217_GAIN[selected_columns] merged_pc3_217_NEUTRAL <- merged_pc3_217_NEUTRAL[selected_columns] merged_pc3_217_LOSS <- merged_pc3_217_LOSS[selected_columns]

Save as a csv

write.csv(merged_pc3_217_GAIN, ‘final_merged_pc3_217_GAIN.csv’) write.csv(merged_pc3_217_NEUTRAL, ‘final_merged_pc3_217_NEUTRAL.csv’) write.csv(merged_pc3_217_LOSS, ‘final_merged_pc3_217_LOSS.csv’)

#go to unique folder unique_output_224_GAIN <- read.csv(“unique_output_224_GAIN.csv”) unique_output_224_NEUTRAL <- read.csv(“unique_output_224_NEUTRAL.csv”) unique_output_224_LOSS <- read.csv(“unique_output_224_LOSS.csv”)

#go to pc3_counts folder pc3_224_gene_counts <- read.csv(“pc3_224_counts.csv”, header = TRUE)

Merge the data frames based on the ‘gene’ column

merged_pc3_224_GAIN <- inner_join(unique_output_224_GAIN, pc3_224_gene_counts, by = ‘gene’) merged_pc3_224_NEUTRAL <- inner_join(unique_output_224_NEUTRAL, pc3_224_gene_counts, by = ‘gene’) merged_pc3_224_LOSS <- inner_join(unique_output_224_LOSS, pc3_224_gene_counts, by = ‘gene’)

Import TPM files

pc3_224_TPM <- read.table(“224_pc3_ctrlpc3_ctrl.genes.results.txt”, sep = “, header = TRUE) #rename column names(pc3_224_TPM)[names(pc3_224_TPM) == ”gene_id”] <-”gene_ID”

Merge files w/ TPM to add TPM and length columns

merged_pc3_224_GAIN <- inner_join(merged_pc3_224_GAIN, pc3_224_TPM, by = ‘gene_ID’) merged_pc3_224_NEUTRAL <- inner_join(merged_pc3_224_NEUTRAL, pc3_224_TPM, by = ‘gene_ID’) merged_pc3_224_LOSS <- inner_join(merged_pc3_224_LOSS, pc3_224_TPM, by = ‘gene_ID’)

Select the UPDATED desired columns

selected_columns <- c(‘chr’, ‘gene_ID’, ‘gene’, ‘type’, ‘seg.mean’, ‘counts’, ‘TPM’, ‘length’) merged_pc3_224_GAIN <- merged_pc3_224_GAIN[selected_columns] merged_pc3_224_NEUTRAL <- merged_pc3_224_NEUTRAL[selected_columns] merged_pc3_224_LOSS <- merged_pc3_224_LOSS[selected_columns]

#change directory to final_merge_TPM # Save as a csv write.csv(merged_pc3_224_GAIN, ‘final_merged_pc3_224_GAIN.csv’) write.csv(merged_pc3_224_NEUTRAL, ‘final_merged_pc3_224_NEUTRAL.csv’) write.csv(merged_pc3_224_LOSS, ‘final_merged_pc3_224_LOSS.csv’)

###################BOXPLOT TIMEEEE library(ggplot2)

pc3_217_GAIN_boxplot <- ggplot(pc3_217_GAIN, aes(x = counts, y = seg.mean)) + geom_boxplot() + labs(x = “counts”, y = “seg.mean”, title = “pc3_217_GAIN”)

print(pc3_217_GAIN_boxplot) ggsave(“pc3_217_GAIN_boxplot.png”, plot = pc3_217_GAIN_boxplot, width = 8, height = 6, units = “in”)

###########I lied lmfao…

################VOLCANO PLOT TIMEEEE

Load the necessary packages

#library(ggplot2)

Create separate layers for each data frame

#layer_pc3_217_GAIN <- geom_point(data = pc3_217_GAIN, aes(x = seg.mean, y = counts), color = “green”, size = 1) #layer_pc3_217_NEUTRAL <- geom_point(data = pc3_217_NEUTRAL, aes(x = seg.mean, y = counts), color = “blue”, size = 1) #layer_pc3_217_LOSS <- geom_point(data = pc3_217_LOSS, aes(x = seg.mean, y = counts), color = “red”, size = 1)

Create a combined volcano plot by adding the layers together

#pc3_217_volcano <- ggplot() + # layer_pc3_217_GAIN + layer_pc3_217_NEUTRAL + layer_pc3_217_LOSS + #labs(x = “seg.mean”, y = “counts”, title = “pc3_ctrl_217 Volcano”) + # theme_minimal()

Save the volcano plot as a PNG file

#ggsave(“pc3_ctrl_217_Volcano.png”, plot = pc3_217_volcano, width = 8, height = 6, units = “in”)

Print the combined volcano plot

#print(pc3_217_volcano)

Create separate layers for each data frame

#layer_pc3_224_GAIN <- geom_point(data = pc3_224_GAIN, aes(x = seg.mean, y = counts), color = “green”, size = 1) #layer_pc3_224_NEUTRAL <- geom_point(data = pc3_224_NEUTRAL, aes(x = seg.mean, y = counts), color = “blue”, size = 1) #layer_pc3_224_LOSS <- geom_point(data = pc3_224_LOSS, aes(x = seg.mean, y = counts), color = “red”, size = 1)

Create a combined volcano plot by adding the layers together

#pc3_224_volcano <- ggplot() + #layer_pc3_224_GAIN + layer_pc3_224_NEUTRAL + layer_pc3_224_LOSS + #labs(x = “seg.mean”, y = “counts”, title = “pc3_ctrl_223 Volcano”) + #theme_minimal() + #theme(plot.background = element_rect(fill = “white”))

Print the combined volcano plot

#print(pc3_224_volcano)

########Violin Plot

#funny story, R crashed so I have to re upload certain files WHICH IS WHY YOU #SAVE EVERY NEW FILE AS A CSV

#so gotta upload the merged files (seg.mean and counts) FOR EVERY SAMPLE’S GAIN, #NEUTRAL, AND LOSS

pc3_217_GAIN <- read.csv(“final_merged_pc3_217_GAIN.csv”) pc3_217_NEUTRAL <- read.csv(“final_merged_pc3_217_NEUTRAL.csv”) pc3_217_LOSS <- read.csv(“final_merged_pc3_217_LOSS.csv”)

#Add column with whicg group they came from to color code & separate on plot pc3_217_GAIN$source <- "gain" pc3_217_NEUTRAL$source <- “neutral” pc3_217_LOSS$source <- “loss”

#NOTE: they haven’t been subsetted to throw away rows w/o 0 #SO they all contain genes with counts of 0

#FIRST bind them into another file w/ 0 @ end of name #want violin plot w/ 0 AND w/o 0 combined_2170 <- rbind(pc3_217_GAIN, pc3_217_NEUTRAL, pc3_217_LOSS)

#THEN subset for a file w/o 0 #name of file will NOT have 0 @ end combined_217 <- combined_2170[combined_2170$counts!=0,]

#note of 0’s thrown away for each (loss, gain, etc.) #note of # of genes in each (loss, gain, etc.)

#individual violin plots of each (loss, gain, etc.) #^w/ and w/o 0’s

#for all samples

#cow plot (for all samples)

Create the violin plots

combined_217$source <- factor(combined_217$source , levels=c(“gain”, “neutral”, “loss”))

pc3_217_violin_plot <- ggplot(combined_217, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(x = “source”, y = “counts”, title = “pc3_ctrl_217 - Violin Plot”) + scale_fill_manual(values = c(“loss” = “#8c5c47”, “neutral” = “#f8ede5”, “gain” = “#4a8cb0”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”))

print(pc3_217_violin_plot)

Save the violin plot as a PNG file

pc3_217_violin_plot.png

combined_2170$source <- factor(combined_2170$source , levels=c(“gain”, “neutral”, “loss”))

pc3_2170_violin_plot <- ggplot(combined_2170, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(x = “source”, y = “counts”, title = “pc3_ctrl_2170 - Violin Plot”) + scale_fill_manual(values = c(“loss” = “#8c5c47”, “neutral” = “#f8ede5”, “gain” = “#4a8cb0”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”))

print(pc3_2170_violin_plot)

Save the violin plot as a PNG file

pc3_2170_violin_plot.png

######individual plots for EACH gain, neutral, loss pc3_2170_GAIN <- pc3_217_GAIN pc3_217_GAIN <- subset(pc3_217_GAIN, counts != 0)

pc3_217_GAIN_violin_plot <- ggplot(pc3_217_GAIN, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_217_GAIN”) + scale_fill_manual(values = c(“#4a8cb0”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_217_GAIN_violin_plot) ggsave(“pc3_217_GAIN_violin.png”, plot = pc3_217_GAIN_violin_plot, width = 8, height = 6, units = “in”)

pc3_2170_GAIN_violin_plot <- ggplot(pc3_2170_GAIN, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_2170_GAIN”) + scale_fill_manual(values = c(“#4a8cb0”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_2170_GAIN_violin_plot) ggsave(“pc3_2170_GAIN_violin.png”, plot = pc3_2170_GAIN_violin_plot, width = 8, height = 6, units = “in”)

pc3_2170_NEUTRAL <- pc3_217_NEUTRAL pc3_217_NEUTRAL <- subset(pc3_217_NEUTRAL, counts != 0)

pc3_217_NEUTRAL_violin_plot <- ggplot(pc3_217_NEUTRAL, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_217_NEUTRAL”) + scale_fill_manual(values = c(“#f8ede5”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_217_NEUTRAL_violin_plot) ggsave(“pc3_217_NEUTRAL_violin.png”, plot = pc3_217_NEUTRAL_violin_plot, width = 8, height = 6, units = “in”)

pc3_2170_NEUTRAL_violin_plot <- ggplot(pc3_2170_NEUTRAL, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_2170_NEUTRAL”) + scale_fill_manual(values = c(“#f8ede5”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_2170_NEUTRAL_violin_plot) ggsave(“pc3_2170_NEUTRAL_violin.png”, plot = pc3_2170_NEUTRAL_violin_plot, width = 8, height = 6, units = “in”)

pc3_2170_LOSS <- pc3_217_LOSS pc3_217_LOSS <- subset(pc3_217_LOSS, counts != 0)

pc3_217_LOSS_violin_plot <- ggplot(pc3_217_LOSS, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_217_LOSS”) + scale_fill_manual(values = c(“#8c5c47”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_217_LOSS_violin_plot) ggsave(“pc3_217_LOSS_violin.png”, plot = pc3_217_LOSS_violin_plot, width = 8, height = 6, units = “in”)

pc3_2170_LOSS_violin_plot <- ggplot(pc3_2170_LOSS, aes(x = source, y = counts, fill = source)) + geom_violin() + geom_boxplot(width = 0.1, color = “black”, position = position_nudge(x = 0.2)) + labs(y = “counts”, x = “gain”, title = “pc3_ctrl_2170_LOSS”) + scale_fill_manual(values = c(“#8c5c47”)) + coord_cartesian(ylim = c(0, 1000)) + # Set y-axis limit to 0-100 theme_minimal() + theme(plot.background = element_rect(fill = “white”), axis.title.x = element_blank())

print(pc3_2170_LOSS_violin_plot) ggsave(“pc3_2170_LOSS_violin.png”, plot = pc3_2170_LOSS_violin_plot, width = 8, height = 6, units = “in”)

#install.packages(“cowplot”) #library(cowplot)

plot_grid(pc3_217_GAIN_violin_plot, pc3_217_NEUTRAL_violin_plot, pc3_217_LOSS_violin_plot, ncol=3)

plot_grid(pc3_2170_GAIN_violin_plot, pc3_2170_NEUTRAL_violin_plot, pc3_2170_LOSS_violin_plot, ncol=3)

sum(pc3_2170_GAIN[[“counts”]] == 0) sum(pc3_2170_GAIN[[“counts”]] != 0) summary(pc3_2170_GAIN)

sum(pc3_2170_NEUTRAL[[“counts”]] == 0) sum(pc3_2170_NEUTRAL[[“counts”]] != 0) summary(pc3_2170_NEUTRAL)

sum(pc3_2170_LOSS[[“counts”]] == 0) sum(pc3_2170_LOSS[[“counts”]] != 0) summary(pc3_2170_LOSS)

##########see which top 15 genes are consistent in gain, neutral, loss #expand to top 100 #intersect to see common genes btwn 2 columns

pc3_ctrl_gain100 <- read.csv(“pc3_ctrl_gain100.csv”)

pc3_ctrl_neutral100 <- read.csv(“pc3_ctrl_neutral100.csv”)

pc3_ctrl_loss100 <- read.csv(“pc3_ctrl_loss100.csv”)

#common_gain <- intersect(pc3_ctrl_gain100$pc3_ctrl_217, pc3_ctrl_gain100$pc3_ctrl_218)

n <- nrow(pc3_ctrl_gain100) # Get the number of rows in pc3_ctrl_gain100 common_gain100 <- data.frame(matrix(NA, nrow = n, ncol = 0)) # Initialize common_gain with the correct number of rows

for (i in 1:7) { for (j in (i + 1):8) { col_name <- paste(i, “vs”, j, sep = ““) common_values <- intersect(pc3_ctrl_gain100[, i], pc3_ctrl_gain100[, j])

# Create a vector with the same number of rows as pc3_ctrl_gain100
common_values_with_missing <- rep(NA, n)
common_values_with_missing[1:length(common_values)] <- common_values

common_gain100[[col_name]] <- common_values_with_missing

} }

#rename columns #kinda pointless to name ngl colnames(common_gain100)[colnames(common_gain100) == “7vs8”] <- “223vs224”

##consistent is at least 70%

gene_names <- unique(unlist(pc3_ctrl_gain100))

gene_count_df <- data.frame(Gene = character(0), Count = integer(0))

for (gene in gene_names) { gene_count_gain100 <- sum(sapply(pc3_ctrl_gain100, function(col) sum(col == gene))) gene_count_df <- bind_rows(gene_count_df, data.frame(Gene = gene, Count = gene_count_gain100)) }

occurences_gain100 <- as.data.frame(gene_count_df)

write.csv(occurences_gain100, “occurences_gain100.csv”)

occurences70_gain100 <- subset(occurences_gain100, Count >= 8)

subset_gene_count_df <- subset(occurences70_gain100, select = -Count)

######bar plots of counts = 0 & > 0

pc3_ctrl_counts <- read.csv(“pc3_ctrl_counts.csv”)

colnames(pc3_ctrl_counts)[2] =“G/N/L” colnames(pc3_ctrl_counts)[3] =“counts_0” colnames(pc3_ctrl_counts)[4] =“counts_greater_0”

pc3_ctrl_217_counts <- subset(pc3_ctrl_counts, sample == “pc3_ctrl_217”) library(tidyr) pc3_ctrl_217_counts <- gather(pc3_ctrl_217_counts, key = “count_type”, value = “count”, counts_0, counts_greater_0, total)

custom_colors <- c(“Counts = 0” = “#e6ecee”, “Counts > 0” = “#aac2ce”, “Total” = “#d2b08c”)

ggplot(pc3_ctrl_217_counts, aes(x = count_type, y = count, fill = count_type)) + geom_bar(stat = “identity”) + labs(title = “Counts for PC3_ctrl_217_GAIN”, x = “Count Type”, y = “Counts”) + scale_fill_manual(name = “Count Type”, values = custom_colors) + # Set legend title and colors scale_x_discrete(labels = c(“Counts = 0”, “Counts > 0”, “Total”)) + # Set custom x-axis breaks theme_minimal()

Scatterplot: TPM vs abspos

Should look like CNV Plot

Y Axis: TPM

X Axis: Abspos

color code

2 different plots to make

- all gains, neutrals, and losses (use data WITH 0’s)

- individual plots for each gain, neutral, and loss (use data WITH 0’s)

Self Note:

Either reupload OG csv and assign to variable

OR

Make sure to use variable that WASN’T subsetted w/o 0’s

ex: pc3_224_GAIN was subsetted w/o 0’s and pc3_2240_GAIN contains all data even / 0’s

pc3_224_TPM_abspos <- ggplot() + geom_point(data = pc3_2240_GAIN, aes(x = abspos, y = TPM), color = “#4a8cb0”) + geom_point(data = pc3_2240_NEUTRAL, aes(x = abspos, y = TPM), color = “#e2b390”) + geom_point(data = pc3_2240_LOSS, aes(x = abspos, y = TPM), color = “#8c5c47”) + labs( title = “pc3_ctrl_224: TPM vs. abspos”, x = “absolute position”, y = “TPM” ) + theme_minimal()

print(pc3_224_TPM_abspos)

sample2 <- ggplot() + geom_point(data = pc3_2170_GAIN, aes(x = abspos, y = TPM, color = “Gain”), size = 3) + geom_point(data = pc3_2170_NEUTRAL, aes(x = abspos, y = TPM, color = “Neutral”), size = 3) + geom_point(data = pc3_2170_LOSS, aes(x = abspos, y = TPM, color = “Loss”), size = 3) + labs( title = “pc3_ctrl_217: TPM vs. Absolute Position”, x = “Absolute Position”, y = “TPM” ) + scale_color_manual(values = c(“Gain” = “#4a8cb0”, “Neutral” = “#e2b390”, “Loss” = “#8c5c47”)) + theme_minimal() + theme(legend.title = element_blank()) + guides(color = guide_legend(title = “Legend Title”))

print(sample2)

Scatterplot: TPM vs seg.mean

TPM - Y Axis

Seg.mean - X Axis

2 plots

all data (gain, neutral, loss)

individuals (gain, neutral, loss)

w/ or w/o 0’s??

pc3_224_TPM_segmean <- ggplot() + geom_point(data = pc3_2240_GAIN, aes(x = seg.mean, y = TPM), color = “#4a8cb0”) + geom_point(data = pc3_2240_NEUTRAL, aes(x = seg.mean, y = TPM), color = “#e2b390”) + geom_point(data = pc3_2240_LOSS, aes(x = seg.mean, y = TPM), color = “#8c5c47”) + labs( title = “pc3_ctrl_224: TPM vs. Segmented Mean”, x = “segmented mean”, y = “TPM” ) + theme_minimal()

print(pc3_224_TPM_segmean)

sample <- ggplot() + geom_point(data = pc3_2170_GAIN, aes(x = seg.mean, y = TPM, color = “Gain”), size = 3) + geom_point(data = pc3_2170_NEUTRAL, aes(x = seg.mean, y = TPM, color = “Neutral”), size = 3) + geom_point(data = pc3_2170_LOSS, aes(x = seg.mean, y = TPM, color = “Loss”), size = 3) + labs( title = “pc3_ctrl_217: TPM vs. Segmented Mean”, x = “Segmented Mean”, y = “TPM” ) + scale_color_manual(values = c(“Gain” = “#4a8cb0”, “Neutral” = “#e2b390”, “Loss” = “#8c5c47”)) + theme_minimal() + theme(legend.title = element_blank()) + guides(color = guide_legend(title = “Legend Title”))

print(sample)

TPM vs abspos

count bins first

greater than or equal to 1000 (1000 ≥)

mildly: between 10 and 1000 (10 ≤ x < 1000)

lowly: between 0.5 and 10 (0.5 ≤ x < 10)

not expressed: lower than 0.5 (x < 0.5)

greater than or equal to 1000 (1000 ≥)

nrow(pc3_2170_GAIN[pc3_2170_GAIN$TPM >= 1000, ])

mildly: between 10 and 1000 (10 ≤ x < 1000)

nrow(pc3_2170_GAIN[pc3_2170_GAIN$TPM >= 10 & pc3_2170_GAIN$TPM < 1000, ])

lowly: between 0.5 and 10 (0.5 ≤ x < 10)

nrow(pc3_2170_GAIN[pc3_2170_GAIN$TPM >= 0.5 & pc3_2170_GAIN$TPM < 10, ])

not expressed: lower than 0.5 (x < 0.5)

nrow(pc3_2170_GAIN[pc3_2170_GAIN$TPM <= 0.5, ])

pc3_2240_GAIN pc3_2240_NEUTRAL pc3_2240_LOSS

nrow(pc3_2240_LOSS[pc3_2240_LOSS$TPM >= 1000, ]) nrow(pc3_2240_LOSS[pc3_2240_LOSS$TPM >= 10 & pc3_2240_LOSS$TPM < 1000, ]) nrow(pc3_2240_LOSS[pc3_2240_LOSS$TPM >= 0.5 & pc3_2240_LOSS$TPM < 10, ]) nrow(pc3_2240_LOSS[pc3_2240_LOSS$TPM <= 0.5, ])

adjust TPM: everything over 1000 is set to 1000

make a copy of data frame

adjusted_pc3_2240_GAIN <- pc3_2240_GAIN

adjusted_pc3_2240_GAIN <- adjusted_pc3_2240_GAIN %>% mutate(TPM = pmin(TPM, 1000))

adjusted_pc3_2240_NEUTRAL <- pc3_2240_NEUTRAL

adjusted_pc3_2240_NEUTRAL <- adjusted_pc3_2240_NEUTRAL %>% mutate(TPM = pmin(TPM, 1000))

adjusted_pc3_2240_LOSS <- pc3_2240_LOSS

adjusted_pc3_2240_LOSS <- adjusted_pc3_2240_LOSS %>% mutate(TPM = pmin(TPM, 1000))

#gain, neutral, loss together adjusted_pc3_224_TPM_abspos <- ggplot() + geom_point(data = adjusted_pc3_2240_GAIN, aes(x = abspos, y = TPM), color = “#4a8cb0”) + geom_point(data = adjusted_pc3_2240_NEUTRAL, aes(x = abspos, y = TPM), color = “#e2b390”) + geom_point(data = adjusted_pc3_2240_LOSS, aes(x = abspos, y = TPM), color = “#8c5c47”) + labs( title = “adjusted_pc3_ctrl_224: TPM vs. abspos”, x = “absolute position”, y = “TPM” ) + theme_minimal()

print(adjusted_pc3_224_TPM_abspos)

sample3 <- ggplot() + geom_point(data = adjusted_pc3_2170_GAIN, aes(x = abspos, y = TPM, color = “Gain”), size = 3) + geom_point(data = adjusted_pc3_2170_NEUTRAL, aes(x = abspos, y = TPM, color = “Neutral”), size = 3) + geom_point(data = adjusted_pc3_2170_LOSS, aes(x = abspos, y = TPM, color = “Loss”), size = 3) + labs( title = “adjusted_pc3_ctrl_217: TPM vs. Absolute Position”, x = “Absolute Position”, y = “TPM” ) + scale_color_manual(values = c(“Gain” = “#4a8cb0”, “Neutral” = “#e2b390”, “Loss” = “#8c5c47”)) + theme_minimal() + theme(legend.title = element_blank()) + guides(color = guide_legend(title = “Legend Title”))

print(sample3)

###########gain, neutral, loss individual adjusted_pc3_217_GAIN_TPM_abspos <- ggplot() + geom_point(data = adjusted_pc3_2170_GAIN, aes(x = abspos, y = TPM), color = “#4a8cb0”) + labs( title = “adjusted_pc3_ctrl_217_GAIN: TPM vs. abspos”, x = “absolute position”, y = “TPM” ) + theme_minimal()

print(adjusted_pc3_217_GAIN_TPM_abspos)

adjusted_pc3_217_NEUTRAL_TPM_abspos <- ggplot() + geom_point(data = adjusted_pc3_2170_NEUTRAL, aes(x = abspos, y = TPM), color = “#e2b390”) + labs( title = “adjusted_pc3_ctrl_217_NEUTRAL: TPM vs. abspos”, x = “absolute position”, y = “TPM” ) + theme_minimal()

print(adjusted_pc3_217_NEUTRAL_TPM_abspos)

adjusted_pc3_217_LOSS_TPM_abspos <- ggplot() + geom_point(data = adjusted_pc3_2170_LOSS, aes(x = abspos, y = TPM), color = “#8c5c47”) + labs( title = “adjusted_pc3_ctrl_217_LOSS: TPM vs. abspos”, x = “absolute position”, y = “TPM” ) + theme_minimal()

print(adjusted_pc3_217_LOSS_TPM_abspos)

bar plots

bins <- read.csv(“bins_pc3.csv”, header = TRUE)

pc3_ctrl_224_bins <- bins[bins$sample == ‘pc3_ctrl_224’,]

pc3_ctrl_224_bins$expression <- factor(pc3_ctrl_224_bins$expression, levels = c(“greater_1000”, “mildly”, “lowly”, “not_expressed”)) pc3_ctrl_224_bins$CNV_status <- factor(pc3_ctrl_224_bins$CNV_status, levels = c(“GAIN”, “NEUTRAL”, “LOSS”))

grouped bar plot

pc3_ctrl_224_bins_grouped <- ggplot(pc3_ctrl_224_bins, aes(x = CNV_status, y = expression_value, fill = expression)) + geom_bar(stat = “identity”, position = position_dodge(width = 0.9)) + labs(title = “pc3_ctrl_224: TPM Bins”, x = “CNV Status”, y = “Nunber of Genes”, fill = “Expression Type”) + scale_fill_manual(values = c(“greater_1000” = “#18465a”, “mildly” = “#9ec1d7”, “lowly” = “#65403a”, “not_expressed” = “#1b1c27”), labels = c(“greater_1000” = “greater than 1000”, “mildly” = “mildly”, “lowly” = “lowly”, “not_expressed” = “not expressed”)) + theme_minimal() + geom_text(aes(label = expression_value), position = position_dodge(width = 0.9), vjust = -0.5, size = 2) + theme(plot.background = element_rect(fill = “white”))

print(pc3_ctrl_224_bins_grouped)

ggsave(“pc3_ctrl_224_bins_grouped.png”, plot = pc3_ctrl_224_bins_grouped, width = 8, height = 6, units = “in”)

stacked bar plot

pc3_ctrl_217_bins_stacked <- ggplot(pc3_ctrl_217_bins, aes(x = CNV_status, y = expression_value, fill = expression)) + geom_bar(stat = “identity”, position = “stack”) + labs(title = “pc3_ctrl_217: TPM Bins”, x = “CNV Status”, y = “Nunber of Genes”, fill = “Expression Type”) + scale_fill_manual(values = c(“greater_1000” = “#18465a”, “mildly” = “#9ec1d7”, “lowly” = “#65403a”, “not_expressed” = “#1b1c27”), labels = c(“greater_1000” = “greater than 1000”, “mildly” = “mildly”, “lowly” = “lowly”, “not_expressed” = “not expressed”)) + theme_minimal() + theme(plot.background = element_rect(fill = “white”))

print(pc3_ctrl_217_bins_stacked)

ggsave(“pc3_ctrl_217_bins_stacked.png”, plot = pc3_ctrl_217_bins_stacked, width = 8, height = 6, units = “in”)

genes w/ TPMs 1000 ≥ & 70% occurence

pc3_2170_GAIN_TPM1000 <- pc3_2170_GAIN[pc3_2170_GAIN$TPM >= 1000, ] pc3_2170_GAIN_TPM1000 <- pc3_2170_GAIN_TPM1000[“gene”] colnames(pc3_2170_GAIN_TPM1000) <- “pc3_217_GAIN”

pc3_2240_NEUTRAL_TPM1000 <- pc3_2240_NEUTRAL[pc3_2240_NEUTRAL$TPM >= 1000, ] pc3_2240_NEUTRAL_TPM1000 <- pc3_2240_NEUTRAL_TPM1000[“gene”] colnames(pc3_2240_NEUTRAL_TPM1000) <- “pc3_224_NEUTRAL”

pc3_2240_LOSS_TPM1000 <- pc3_2240_LOSS[pc3_2240_LOSS$TPM >= 1000, ] pc3_2240_LOSS_TPM1000 <- pc3_2240_LOSS_TPM1000[“gene”] colnames(pc3_2240_LOSS_TPM1000) <- “pc3_224_LOSS”

^ done with samples 217-224

list_df <- list(pc3_2170_GAIN_TPM1000, pc3_2180_GAIN_TPM1000, pc3_2190_GAIN_TPM1000, pc3_2200_GAIN_TPM1000, pc3_2210_GAIN_TPM1000, pc3_2220_GAIN_TPM1000, pc3_2230_GAIN_TPM1000, pc3_2240_GAIN_TPM1000) n_r <- seq_len(max(sapply(list_df, nrow))) pc3_ctrl_GAIN_TPM1000 <- do.call(cbind, lapply(list_df, [, n_r, )) pc3_ctrl_GAIN_TPM1000

list_df <- list(pc3_2170_NEUTRAL_TPM1000, pc3_2180_NEUTRAL_TPM1000, pc3_2190_NEUTRAL_TPM1000, pc3_2200_NEUTRAL_TPM1000, pc3_2210_NEUTRAL_TPM1000, pc3_2220_NEUTRAL_TPM1000, pc3_2230_NEUTRAL_TPM1000, pc3_2240_NEUTRAL_TPM1000) n_r <- seq_len(max(sapply(list_df, nrow))) pc3_ctrl_NEUTRAL_TPM1000 <- do.call(cbind, lapply(list_df, [, n_r, )) pc3_ctrl_NEUTRAL_TPM1000

list_df <- list(pc3_2170_LOSS_TPM1000, pc3_2180_LOSS_TPM1000, pc3_2190_LOSS_TPM1000, pc3_2200_LOSS_TPM1000, pc3_2210_LOSS_TPM1000, pc3_2220_LOSS_TPM1000, pc3_2230_LOSS_TPM1000, pc3_2240_LOSS_TPM1000) n_r <- seq_len(max(sapply(list_df, nrow))) pc3_ctrl_LOSS_TPM1000 <- do.call(cbind, lapply(list_df, [, n_r, )) pc3_ctrl_LOSS_TPM1000

#rename columns

new_column_names <- c(“pc3_217_GAIN”, “pc3_218_GAIN”, “pc3_219_GAIN”, “pc3_220_GAIN”, “pc3_221_GAIN”, “pc3_222_GAIN”, “pc3_223_GAIN”, “pc3_224_GAIN”) colnames(pc3_ctrl_GAIN_TPM1000) <- new_column_names

new_column_names <- c(“pc3_217_NEUTRAL”, “pc3_218_NEUTRAL”, “pc3_219_NEUTRAL”, “pc3_220_NEUTRAL”, “pc3_221_NEUTRAL”, “pc3_222_NEUTRAL”, “pc3_223_NEUTRAL”, “pc3_224_NEUTRAL”) colnames(pc3_ctrl_NEUTRAL_TPM1000) <- new_column_names

new_column_names <- c(“pc3_217_LOSS”, “pc3_218_LOSS”, “pc3_219_LOSS”, “pc3_220_LOSS”, “pc3_221_LOSS”, “pc3_222_LOSS”, “pc3_223_LOSS”, “pc3_224_LOSS”) colnames(pc3_ctrl_LOSS_TPM1000) <- new_column_names

write.csv(pc3_ctrl_GAIN_TPM1000, “pc3_ctrl_GAIN_TPM1000.csv”) write.csv(pc3_ctrl_NEUTRAL_TPM1000, “pc3_ctrl_NEUTRAL_TPM1000.csv”) write.csv(pc3_ctrl_LOSS_TPM1000, “pc3_ctrl_LOSS_TPM1000.csv”)

to count occurences

gene_vector <- unlist(pc3_ctrl_LOSS_TPM1000)

Step 2: Use the table() function to count the occurrences of each gene

gene_counts <- table(gene_vector)

Step 3: Create a new dataframe from the result

occurences_pc3_ctrl_LOSS_TPM1000 <- data.frame( Gene = names(gene_counts), Count = as.numeric(gene_counts) )

write.csv(occurences_pc3_ctrl_LOSS_TPM1000, “occurences_pc3_ctrl_LOSS_TPM1000.CSV”)

consistent_occurences_pc3_ctrl_LOSS_TPM1000 <- subset(occurences_pc3_ctrl_LOSS_TPM1000, Count >= 6)

write.csv(consistent_occurences_pc3_ctrl_LOSS_TPM1000, “consistent_occurences_pc3_ctrl_LOSS_TPM1000.CSV”)

compare TPMs in chromosomes

Chromosome 1 (beginning)

pc3_ctrl_217 = baseline

^neutral

download the neutral/gain/loss file for the sample

pc3_217_NEUTRAL <- read.csv(“final_merged_pc3_217_NEUTRAL.csv”) pc3_218_NEUTRAL <- read.csv(“final_merged_pc3_218_NEUTRAL.csv”) pc3_219_NEUTRAL <- read.csv(“final_merged_pc3_219_NEUTRAL.csv”) pc3_220_NEUTRAL <- read.csv(“final_merged_pc3_220_NEUTRAL.csv”) pc3_222_NEUTRAL <- read.csv(“final_merged_pc3_222_NEUTRAL.csv”) pc3_223_NEUTRAL <- read.csv(“final_merged_pc3_223_NEUTRAL.csv”) pc3_224_NEUTRAL <- read.csv(“final_merged_pc3_224_NEUTRAL.csv”) pc3_219_GAIN <- read.csv(“final_merged_pc3_219_GAIN.csv”) pc3_219_LOSS <- read.csv(“final_merged_pc3_219_LOSS.csv”) pc3_220_LOSS <- read.csv(“final_merged_pc3_220_LOSS.csv”) pc3_221_LOSS <- read.csv(“final_merged_pc3_221_LOSS.csv”) pc3_222_LOSS <- read.csv(“final_merged_pc3_222_LOSS.csv”) pc3_223_LOSS <- read.csv(“final_merged_pc3_223_LOSS.csv”) pc3_224_LOSS <- read.csv(“final_merged_pc3_224_LOSS.csv”) pc3_221_GAIN <- read.csv(“final_merged_pc3_221_GAIN.csv”) pc3_222_GAIN <- read.csv(“final_merged_pc3_222_GAIN.csv”) pc3_223_GAIN <- read.csv(“final_merged_pc3_223_GAIN.csv”) pc3_224_GAIN <- read.csv(“final_merged_pc3_224_GAIN.csv”)

subset for genes on the specific chromosome

library(dplyr)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”) pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

once you have your comparisons (from excel sheet), only keep genes in BOTH data sets

make list of common names

common_genes_217vs219_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_219_chr1_GAIN$gene)

subset data using list of common names

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs219_chr1, ] pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_217vs219_chr1, ]

plot TPM_neutral (217_neutral on y axis vs 219_gain on x axis)

library(ggplot2)

#ggplot() + # geom_point(aes(x = pc3_219_chr1_GAIN$TPM, y = pc3_217_chr1_NEUTRAL$TPM)) + # labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_217_chr1_NEUTRAL (TPM)”, # title = “pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)”)

pc3_217vs219_chr1 <- merge(pc3_219_chr1_GAIN, pc3_217_chr1_NEUTRAL, by = “gene”)

create new column for fold change

pc3_217vs219_chr1$fold_change <- pc3_217vs219_chr1$TPM.x / pc3_217vs219_chr1$TPM.y

pc3_217vs219_chr1_genes_06 <- pc3_217vs219_chr1$gene[pc3_217vs219_chr1$fold_change < 0.6] pc3_217vs219_chr1_genes_15 <- pc3_217vs219_chr1$gene[pc3_217vs219_chr1$fold_change > 1.5]

Create a scatterplot

#ggplot(pc3_217vs219_chr1, aes(x = TPM.x, y = TPM.y)) + # geom_point() + #labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_217_chr1_NEUTRAL (TPM)”, # title = “pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)”)

ggplot(pc3_217vs219_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( gene %in% pc3_217vs219_chr1_genes_06 ~ “< 0.6”, gene %in% pc3_217vs219_chr1_genes_15 ~ “> 1.5”, TRUE ~ “Other” ) ))) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + geom_text(aes(label = ifelse(gene %in% c(pc3_217vs219_chr1_genes_06, pc3_217vs219_chr1_genes_15), gene, ““)), size = 2, nudge_x = 0.5, nudge_y = 0.5) + labs(x =”pc3_219_chr1_GAIN (TPM)“, y =”pc3_217_chr1_NEUTRAL (TPM)“, title =”pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)“, color =”Fold Change”) + theme(legend.position = “right”)

JUST COLOR CODE

ggplot(pc3_217vs219_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_217_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”)

##with ggreppel ggplot(pc3_217vs219_chr1, aes(x = TPM.x, y = TPM.y, label = gene)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs219_chr1, fold_change < 0.6 | fold_change > 1.5), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_217_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

ONLY keep dots that meet fold change criteria & add names

Filter data for fold change criteria

pc3_217vs219_chr1_filtered_data <- pc3_217vs219_chr1[pc3_217vs219_chr1$fold_change < 0.6 | pc3_217vs219_chr1$fold_change > 1.5, ]

Create a scatter plot for filtered data with colored points based on fold change criteria

ggplot(pc3_217vs219_chr1_filtered_data, aes(x = TPM.x, y = TPM.y, label = gene)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5” ) ))) + geom_text(size = 2, nudge_x = 0.5, nudge_y = 0.5, check_overlap = FALSE) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_217_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 217 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”)

library(ggrepel)

220 LOSS vs 223 GAIN

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

once you have your comparisons (from excel sheet), only keep genes in BOTH data sets

make list of common names

common_genes_220vs223_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_223_chr1_GAIN$gene)

subset data using list of common names

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_220vs223_chr1, ]

plot TPM_neutral (217_neutral on y axis vs 219_gain on x axis)

library(ggplot2)

pc3_220vs223_chr1 <- merge(pc3_220_chr1_LOSS, pc3_223_chr1_GAIN, by = “gene”)

create new column for fold change

pc3_220vs223_chr1$fold_change <- pc3_220vs223_chr1$TPM.x / pc3_220vs223_chr1$TPM.y

pc3_220vs223_chr1_genes_06 <- pc3_220vs223_chr1$gene[pc3_220vs223_chr1$fold_change < 0.6] pc3_220vs223_chr1_genes_15 <- pc3_220vs223_chr1$gene[pc3_220vs223_chr1$fold_change > 1.5]

library(dplyr)

ggplot(pc3_220vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( gene %in% pc3_220vs223_chr1_genes_06 ~ “< 0.6”, gene %in% pc3_220vs223_chr1_genes_15 ~ “> 1.5”, TRUE ~ “Other” ) ))) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + geom_text(aes(label = ifelse(gene %in% c(pc3_220vs223_chr1_genes_06, pc3_220vs223_chr1_genes_15), gene, ““)), size = 2, nudge_x = 0.5, nudge_y = 0.5) + labs(x =”pc3_220_chr1_LOSS (TPM)“, y =”pc3_223_chr1_GAIN (TPM)“, title =”pc3_chr1: 220 (LOSS) vs 223 (GAIN)“, color =”Fold Change”) + theme(legend.position = “right”)

JUST COLOR CODE

ggplot(pc3_220vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “Prostate Cancer Cell Line - Chromosome 1: Sample 220 (LOSS) vs Sample 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

ONLY keep dots that meet fold change criteria & add names

Filter data for fold change criteria

pc3_217vs219_chr1_filtered_data <- pc3_217vs219_chr1[pc3_217vs219_chr1$fold_change < 0.6 | pc3_217vs219_chr1$fold_change > 1.5, ]

Create a scatter plot for filtered data with colored points based on fold change criteria

library(ggrepel)

every comparison, need to reset the subset for specific chromosome in N/G/L

^bc won’t have all genes, just genes in common btwn previous comparison

217 (neutral) vs 219 (gain)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”)

common_genes_217vs219_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_219_chr1_GAIN$gene)

pc3_217vs219_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_219_chr1_GAIN, by = “gene”)

pc3_217vs219_chr1$fold_change <- pc3_217vs219_chr1$TPM.x / pc3_217vs219_chr1$TPM.y

pc3_217vs219_chr1_genes_06 <- pc3_217vs219_chr1$gene[pc3_217vs219_chr1$fold_change < 0.6] pc3_217vs219_chr1_genes_15 <- pc3_217vs219_chr1$gene[pc3_217vs219_chr1$fold_change > 1.5]

ggplot(pc3_217vs219_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs219_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_219_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 219 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs219_chr1, “pc3_217(NEUTRAL)vs219(GAIN)_chr1.csv”, row.names = FALSE)

pc3_217Nvs219G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs219_chr1_genes_06)) write.csv(pc3_217vs219_chr1_genes_06, “pc3_217(NEUTRAL)vs219(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs219G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs219_chr1_genes_15)) write.csv(pc3_217vs219_chr1_genes_15, “pc3_217(NEUTRAL)vs219(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 219 (gain)

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”)

common_genes_218vs219_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_219_chr1_GAIN$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs219_chr1, ] pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_218vs219_chr1, ]

pc3_218vs219_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_219_chr1_GAIN, by = “gene”)

pc3_218vs219_chr1$fold_change <- pc3_218vs219_chr1$TPM.x / pc3_218vs219_chr1$TPM.y

pc3_218vs219_chr1_genes_06 <- pc3_218vs219_chr1$gene[pc3_218vs219_chr1$fold_change < 0.6] pc3_218vs219_chr1_genes_15 <- pc3_218vs219_chr1$gene[pc3_218vs219_chr1$fold_change > 1.5]

ggplot(pc3_218vs219_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs219_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_219_chr1_GAIN (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 219 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs219_chr1, “pc3_218(NEUTRAL)vs219(GAIN)_chr1.csv”, row.names = FALSE)

pc3_218Nvs219G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs219_chr1_genes_06)) write.csv(pc3_218Nvs219G_chr1_genes_06, “pc3_218(NEUTRAL)vs219(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs219G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_218vs219_chr1_genes_15)) write.csv(pc3_218Nvs219G_chr1_genes_15, “pc3_218(NEUTRAL)vs219(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

######compare 220 with others

218 (neutral) vs 220 (loss)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”)

common_genes_220vs218_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_218_chr1_NEUTRAL$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs218_chr1, ] pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_220vs218_chr1, ]

pc3_220vs218_chr1 <- merge(pc3_220_chr1_LOSS, pc3_218_chr1_NEUTRAL, by = “gene”)

pc3_220vs218_chr1$fold_change <- pc3_220vs218_chr1$TPM.x / pc3_220vs218_chr1$TPM.y

pc3_220vs218_chr1_genes_06 <- pc3_220vs218_chr1$gene[pc3_220vs218_chr1$fold_change < 0.6] pc3_220vs218_chr1_genes_15 <- pc3_220vs218_chr1$gene[pc3_220vs218_chr1$fold_change > 1.5]

ggplot(pc3_220vs218_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr17, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_218_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 218 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

other way

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”)

common_genes_218vs220_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_220_chr1_LOSS$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs220_chr1, ] pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_218vs220_chr1, ]

pc3_218vs220_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_220_chr1_LOSS, by = “gene”)

pc3_218vs220_chr1$fold_change <- pc3_218vs220_chr1$TPM.x / pc3_218vs220_chr1$TPM.y

pc3_218vs220_chr1_genes_06 <- pc3_218vs220_chr1$gene[pc3_218vs220_chr1$fold_change < 0.6] pc3_218vs220_chr1_genes_15 <- pc3_218vs220_chr1$gene[pc3_218vs220_chr1$fold_change > 1.5]

ggplot(pc3_218vs220_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs220_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_220_chr1_LOSS (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs220_chr1, “pc3_218(NEUTRAL)vs220(LOSS)_chr1.csv”, row.names = FALSE)

pc3_218Nvs220L_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs220_chr1_genes_06)) write.csv(pc3_218Nvs220L_chr1_genes_06, “pc3_218(NEUTRAL)vs220(LOSS)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs220L_chr1_genes_15 <- data.frame(gene = na.omit(pc3_218vs220_chr1_genes_15)) write.csv(pc3_218Nvs220L_chr1_genes_15, “pc3_218(NEUTRAL)vs220(LOSS)_chr1_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 220 (loss)

pc3_219_NEUTRAL <- read.csv(“final_merged_pc3_219_NEUTRAL.csv”)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_219_chr1_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr1”)

common_genes_220vs219_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_219_chr1_NEUTRAL$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs219_chr1, ] pc3_219_chr1_NEUTRAL <- pc3_219_chr1_NEUTRAL[pc3_219_chr1_NEUTRAL$gene %in% common_genes_220vs219_chr1, ]

pc3_220vs219_chr1 <- merge(pc3_220_chr1_LOSS, pc3_219_chr1_NEUTRAL, by = “gene”)

pc3_220vs219_chr1$fold_change <- pc3_220vs219_chr1$TPM.x / pc3_220vs219_chr1$TPM.y

pc3_220vs219_chr1_genes_06 <- pc3_220vs219_chr1$gene[pc3_220vs219_chr1$fold_change < 0.6] pc3_220vs219_chr1_genes_15 <- pc3_220vs219_chr1$gene[pc3_220vs219_chr1$fold_change > 1.5]

ggplot(pc3_220vs219_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_219_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 219 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

other way

pc3_219_NEUTRAL <- read.csv(“final_merged_pc3_219_NEUTRAL.csv”)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_219_chr1_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr1”)

common_genes_219vs220_chr1 <- intersect(pc3_219_chr1_NEUTRAL$gene, pc3_220_chr1_LOSS$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_219vs220_chr1, ] pc3_219_chr1_NEUTRAL <- pc3_219_chr1_NEUTRAL[pc3_219_chr1_NEUTRAL$gene %in% common_genes_219vs220_chr1, ]

pc3_219vs220_chr1 <- merge(pc3_219_chr1_NEUTRAL, pc3_220_chr1_LOSS, by = “gene”)

pc3_219vs220_chr1$fold_change <- pc3_219vs220_chr1$TPM.x / pc3_219vs220_chr1$TPM.y

pc3_219vs220_chr1_genes_06 <- pc3_219vs220_chr1$gene[pc3_219vs220_chr1$fold_change < 0.6] pc3_219vs220_chr1_genes_15 <- pc3_219vs220_chr1$gene[pc3_219vs220_chr1$fold_change > 1.5]

ggplot(pc3_219vs220_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs220_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_NEUTRAL (TPM)”, y = “pc3_220_chr1_LOSS (TPM)”, title = “pc3_chr1: 219 (NEUTRAL) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs220_chr1, “pc3_219(NEUTRAL)vs220(LOSS)_chr1.csv”, row.names = FALSE)

pc3_219Nvs220L_chr1 <- pc3_219vs220_chr1

pc3_219Nvs220L_chr1_genes_06 <- data.frame(gene = na.omit(pc3_219vs220_chr1_genes_06)) write.csv(pc3_219Nvs220L_chr1_genes_06, “pc3_219(NEUTRAL)vs220(LOSS)_chr1_genes_06.csv”, row.names = FALSE)

pc3_219Nvs220L_chr1_genes_15 <- data.frame(gene = na.omit(pc3_219vs220_chr1_genes_15)) write.csv(pc3_219Nvs220L_chr1_genes_15, “pc3_219(NEUTRAL)vs220(LOSS)_chr1_genes_15.csv”, row.names = FALSE)

221 (gain) vs 220 (loss)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”)

common_genes_220vs221_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_221_chr1_GAIN$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs221_chr1, ] pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_220vs221_chr1, ]

pc3_220vs221_chr1 <- merge(pc3_220_chr1_LOSS, pc3_221_chr1_GAIN, by = “gene”)

pc3_220vs221_chr1$fold_change <- pc3_220vs221_chr1$TPM.x / pc3_220vs221_chr1$TPM.y

pc3_220vs221_chr1_genes_06 <- pc3_220vs221_chr1$gene[pc3_220vs221_chr1$fold_change < 0.6] pc3_220vs221_chr1_genes_15 <- pc3_220vs221_chr1$gene[pc3_220vs221_chr1$fold_change > 1.5]

ggplot(pc3_220vs221_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs221_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_221_chr1_GAIN (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs221_chr1, “pc3_220(LOSS)vs221(GAIN)_chr1.csv”, row.names = FALSE)

pc3_220Lvs221G_chr1 <- pc3_220vs221_chr1

pc3_220Lvs221G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_220vs221_chr1_genes_06)) write.csv(pc3_220Lvs221G_chr1_genes_06, “pc3_220(LOSS)vs221(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_220Lvs221G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_220vs221_chr1_genes_15)) write.csv(pc3_220Lvs221G_chr1_genes_15, “pc3_220(LOSS)vs221(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

222 (gain) vs 220 (loss)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”)

common_genes_220vs222_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_222_chr1_GAIN$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs222_chr1, ] pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_220vs222_chr1, ]

pc3_220vs222_chr1 <- merge(pc3_220_chr1_LOSS, pc3_222_chr1_GAIN, by = “gene”)

pc3_220vs222_chr1$fold_change <- pc3_220vs222_chr1$TPM.x / pc3_220vs222_chr1$TPM.y

pc3_220vs222_chr1_genes_06 <- pc3_220vs222_chr1$gene[pc3_220vs222_chr1$fold_change < 0.6] pc3_220vs222_chr1_genes_15 <- pc3_220vs222_chr1$gene[pc3_220vs222_chr1$fold_change > 1.5]

ggplot(pc3_220vs222_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs222_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_222_chr1_GAIN (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 222 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs222_chr1, “pc3_220(LOSS)vs222(GAIN)_chr1.csv”, row.names = FALSE)

pc3_220Lvs222G_chr1 <- pc3_220vs222_chr1

pc3_220Lvs222G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_220vs222_chr1_genes_06)) write.csv(pc3_220Lvs222G_chr1_genes_06, “pc3_220(LOSS)vs222(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_220Lvs222G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_220vs222_chr1_genes_15)) write.csv(pc3_220Lvs222G_chr1_genes_15, “pc3_220(LOSS)vs222(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

223 (gain) vs 220 (loss)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_220vs223_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_223_chr1_GAIN$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_220vs223_chr1, ]

pc3_220vs223_chr1 <- merge(pc3_220_chr1_LOSS, pc3_223_chr1_GAIN, by = “gene”)

pc3_220vs223_chr1$fold_change <- pc3_220vs223_chr1$TPM.x / pc3_220vs223_chr1$TPM.y

pc3_220vs223_chr1_genes_06 <- pc3_220vs223_chr1$gene[pc3_220vs223_chr1$fold_change < 0.6] pc3_220vs223_chr1_genes_15 <- pc3_220vs223_chr1$gene[pc3_220vs223_chr1$fold_change > 1.5]

library(ggplot2) library(ggrepel) library(dplyr)

ggplot(pc3_220vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs223_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + geom_text(data = subset(pc3_220vs223_chr1, gene %in% c(“ERRFI1”)), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs223_chr1, “pc3_220(LOSS)vs223(GAIN)_chr1.csv”, row.names = FALSE)

pc3_220Lvs223G_chr1 <- pc3_220vs223_chr1

pc3_220Lvs223G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_220vs223_chr1_genes_06)) write.csv(pc3_220Lvs223G_chr1_genes_06, “pc3_220(LOSS)vs223(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_220Lvs223G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_220vs223_chr1_genes_15)) write.csv(pc3_220Lvs223G_chr1_genes_15, “pc3_220(LOSS)vs223(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

224 (neutral) vs 220 (loss)

pc3_224_NEUTRAL <- read.csv(“final_merged_pc3_224_NEUTRAL.csv”)

pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”) pc3_224_chr1_NEUTRAL <- subset(pc3_224_NEUTRAL, chr == “chr1”)

common_genes_220vs224_chr1 <- intersect(pc3_220_chr1_LOSS$gene, pc3_224_chr1_NEUTRAL$gene)

pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_220vs224_chr1, ] pc3_224_chr1_NEUTRAL <- pc3_224_chr1_NEUTRAL[pc3_224_chr1_NEUTRAL$gene %in% common_genes_220vs224_chr1, ]

pc3_220vs224_chr1 <- merge(pc3_220_chr1_LOSS, pc3_224_chr1_NEUTRAL, by = “gene”)

pc3_220vs224_chr1$fold_change <- pc3_220vs224_chr1$TPM.x / pc3_220vs224_chr1$TPM.y

pc3_220vs224_chr1_genes_06 <- pc3_220vs224_chr1$gene[pc3_220vs224_chr1$fold_change < 0.6] pc3_220vs224_chr1_genes_15 <- pc3_220vs224_chr1$gene[pc3_220vs224_chr1$fold_change > 1.5]

ggplot(pc3_220vs224_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs224_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr1_LOSS (TPM)”, y = “pc3_224_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 220 (LOSS) vs 224 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs224_chr1, “pc3_220(LOSS)vs224(GAIN)_chr1.csv”, row.names = FALSE)

pc3_220Lvs224G_chr1 <- pc3_220vs224_chr1

pc3_220Lvs224G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_220vs224_chr1_genes_06)) write.csv(pc3_220Lvs224G_chr1_genes_06, “pc3_220(LOSS)vs224(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_220Lvs224G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_220vs224_chr1_genes_15)) write.csv(pc3_220Lvs224G_chr1_genes_15, “pc3_220(LOSS)vs224(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 220 (loss)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_220_chr1_LOSS <- subset(pc3_220_LOSS, chr == “chr1”)

common_genes_217vs220_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_220_chr1_LOSS$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs220_chr1, ] pc3_220_chr1_LOSS <- pc3_220_chr1_LOSS[pc3_220_chr1_LOSS$gene %in% common_genes_217vs220_chr1, ]

pc3_217vs220_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_220_chr1_LOSS, by = “gene”)

pc3_217vs220_chr1$fold_change <- pc3_217vs220_chr1$TPM.x / pc3_217vs220_chr1$TPM.y

pc3_217vs220_chr1_genes_06 <- pc3_217vs220_chr1$gene[pc3_217vs220_chr1$fold_change < 0.6] pc3_217vs220_chr1_genes_15 <- pc3_217vs220_chr1$gene[pc3_217vs220_chr1$fold_change > 1.5]

ggplot(pc3_217vs220_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 15) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_220_chr1_LOSS (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs220_chr1, “pc3_217(NEUTRAL)vs220(LOSS)_chr1.csv”, row.names = FALSE)

pc3_217Nvs220L_chr1 <- pc3_217vs220_chr1

pc3_217Nvs220L_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr1_genes_06)) write.csv(pc3_217Nvs220L_chr1_genes_06, “pc3_217(NEUTRAL)vs220(LOSS)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs220L_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr1_genes_15)) write.csv(pc3_217Nvs220L_chr1_genes_15, “pc3_217(NEUTRAL)vs220(LOSS)_chr1_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 221 (gain)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”)

common_genes_217vs221_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_221_chr1_GAIN$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs221_chr1, ] pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_217vs221_chr1, ]

pc3_217vs221_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_221_chr1_GAIN, by = “gene”)

pc3_217vs221_chr1$fold_change <- pc3_217vs221_chr1$TPM.x / pc3_217vs221_chr1$TPM.y

pc3_217vs221_chr1_genes_06 <- pc3_217vs221_chr1$gene[pc3_217vs221_chr1$fold_change < 0.6] pc3_217vs221_chr1_genes_15 <- pc3_217vs221_chr1$gene[pc3_217vs221_chr1$fold_change > 1.5]

ggplot(pc3_217vs221_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs221_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_221_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs221_chr1, “pc3_217(NEUTRAL)vs221(GAIN)_chr1.csv”, row.names = FALSE)

pc3_217Nvs221G_chr1 <- pc3_217vs221_chr1

pc3_217Nvs221G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs221_chr1_genes_06)) write.csv(pc3_217Nvs221G_chr1_genes_06, “pc3_217(NEUTRAL)vs221(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs221G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs221_chr1_genes_15)) write.csv(pc3_217Nvs221G_chr1_genes_15, “pc3_217(NEUTRAL)vs221(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 222 (gain)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”)

common_genes_217vs222_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_222_chr1_GAIN$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs222_chr1, ] pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_217vs222_chr1, ]

pc3_217vs222_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_222_chr1_GAIN, by = “gene”)

pc3_217vs222_chr1$fold_change <- pc3_217vs222_chr1$TPM.x / pc3_217vs222_chr1$TPM.y

pc3_217vs222_chr1_genes_06 <- pc3_217vs222_chr1$gene[pc3_217vs222_chr1$fold_change < 0.6] pc3_217vs222_chr1_genes_15 <- pc3_217vs222_chr1$gene[pc3_217vs222_chr1$fold_change > 1.5]

ggplot(pc3_217vs222_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs222_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_222_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 222 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs222_chr1, “pc3_217(NEUTRAL)vs222(GAIN)_chr1.csv”, row.names = FALSE)

pc3_217Nvs222G_chr1 <- pc3_217vs222_chr1

pc3_217Nvs222G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs222_chr1_genes_06)) write.csv(pc3_217Nvs222G_chr1_genes_06, “pc3_217(NEUTRAL)vs222(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs222G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs222_chr1_genes_15)) write.csv(pc3_217Nvs222G_chr1_genes_15, “pc3_217(NEUTRAL)vs222(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 223 (gain)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_217vs223_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_223_chr1_GAIN$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_217vs223_chr1, ]

pc3_217vs223_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_223_chr1_GAIN, by = “gene”)

pc3_217vs223_chr1$fold_change <- pc3_217vs223_chr1$TPM.x / pc3_217vs223_chr1$TPM.y

pc3_217vs223_chr1_genes_06 <- pc3_217vs223_chr1$gene[pc3_217vs223_chr1$fold_change < 0.6] pc3_217vs223_chr1_genes_15 <- pc3_217vs223_chr1$gene[pc3_217vs223_chr1$fold_change > 1.5]

ggplot(pc3_217vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs223_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 2, nudge_y = 2) + geom_text(data = subset(pc3_217vs223_chr1, gene %in% c(“ATAD3A”, “DVL1”, “MTCO1P12”, “NOL9”, “NPHP4”)), aes(label = gene), size = 3, nudge_x = 10, nudge_y = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

##for cure symposum library(ggplot2) library(ggrepel)

ggplot(pc3_217vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs223_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, box.padding = 2) + geom_text_repel(data = subset(pc3_217vs223_chr1, gene %in% c(“ATAD3A”, “DVL1”, “MTCO1P12”, “NOL9”, “NPHP4”)), aes(label = gene), size = 3, box.padding = 1.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs223_chr1, “pc3_217(NEUTRAL)vs223(GAIN)_chr1.csv”, row.names = FALSE)

pc3_217Nvs223G_chr1 <- pc3_217vs223_chr1

pc3_217Nvs223G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs223_chr1_genes_06)) write.csv(pc3_217Nvs223G_chr1_genes_06, “pc3_217(NEUTRAL)vs223(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs223G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs223_chr1_genes_15)) write.csv(pc3_217Nvs223G_chr1_genes_15, “pc3_217(NEUTRAL)vs223(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 224 (gain)

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_217vs224_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_224_chr1_GAIN$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_217vs224_chr1, ]

pc3_217vs224_chr1 <- merge(pc3_217_chr1_NEUTRAL, pc3_224_chr1_GAIN, by = “gene”)

pc3_217vs224_chr1$fold_change <- pc3_217vs224_chr1$TPM.x / pc3_217vs224_chr1$TPM.y

pc3_217vs224_chr1_genes_06 <- pc3_217vs224_chr1$gene[pc3_217vs224_chr1$fold_change < 0.6] pc3_217vs224_chr1_genes_15 <- pc3_217vs224_chr1$gene[pc3_217vs224_chr1$fold_change > 1.5]

ggplot(pc3_217vs224_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs224_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs224_chr1, “pc3_217(NEUTRAL)vs224(GAIN)_chr1.csv”, row.names = FALSE)

pc3_217Nvs224G_chr1 <- pc3_217vs224_chr1

pc3_217Nvs224G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_217vs224_chr1_genes_06)) write.csv(pc3_217Nvs224G_chr1_genes_06, “pc3_217(NEUTRAL)vs224(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_217Nvs224G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs224_chr1_genes_15)) write.csv(pc3_217Nvs224G_chr1_genes_15, “pc3_217(NEUTRAL)vs224(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

218 NEUTRAL

218 (neutral) vs 221 (gain)

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”)

common_genes_218vs221_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_221_chr1_GAIN$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs221_chr1, ] pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_218vs221_chr1, ]

pc3_218vs221_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_221_chr1_GAIN, by = “gene”)

pc3_218vs221_chr1$fold_change <- pc3_218vs221_chr1$TPM.x / pc3_218vs221_chr1$TPM.y

pc3_218vs221_chr1_genes_06 <- pc3_218vs221_chr1$gene[pc3_218vs221_chr1$fold_change < 0.6] pc3_218vs221_chr1_genes_15 <- pc3_218vs221_chr1$gene[pc3_218vs221_chr1$fold_change > 1.5]

ggplot(pc3_218vs221_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs221_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_221_chr1_GAIN (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs221_chr1, “pc3_218(NEUTRAL)vs221(GAIN)_chr1.csv”, row.names = FALSE)

pc3_218Nvs221G_chr1 <- pc3_218vs221_chr1

pc3_218Nvs221G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs221_chr1_genes_06)) write.csv(pc3_218Nvs221G_chr1_genes_06, “pc3_218(NEUTRAL)vs221(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs221G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_218vs221_chr1_genes_15)) write.csv(pc3_218Nvs221G_chr1_genes_15, “pc3_218(NEUTRAL)vs221(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 222 (gain)

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”)

common_genes_218vs222_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_222_chr1_GAIN$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs222_chr1, ] pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_218vs222_chr1, ]

pc3_218vs222_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_222_chr1_GAIN, by = “gene”)

pc3_218vs222_chr1$fold_change <- pc3_218vs222_chr1$TPM.x / pc3_218vs222_chr1$TPM.y

pc3_218vs222_chr1_genes_06 <- pc3_218vs222_chr1$gene[pc3_218vs222_chr1$fold_change < 0.6] pc3_218vs222_chr1_genes_15 <- pc3_218vs222_chr1$gene[pc3_218vs222_chr1$fold_change > 1.5]

ggplot(pc3_218vs222_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs222_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_222_chr1_GAIN (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 222 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs222_chr1, “pc3_218(NEUTRAL)vs222(GAIN)_chr1.csv”, row.names = FALSE)

pc3_218Nvs222G_chr1 <- pc3_218vs222_chr1

pc3_218Nvs222G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs222_chr1_genes_06)) write.csv(pc3_218Nvs222G_chr1_genes_06, “pc3_218(NEUTRAL)vs222(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs222G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_218vs222_chr1_genes_15)) write.csv(pc3_218Nvs222G_chr1_genes_15, “pc3_218(NEUTRAL)vs222(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 223 (gain)

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_218vs223_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_223_chr1_GAIN$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_218vs223_chr1, ]

pc3_218vs223_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_223_chr1_GAIN, by = “gene”)

pc3_218vs223_chr1$fold_change <- pc3_218vs223_chr1$TPM.x / pc3_218vs223_chr1$TPM.y

pc3_218vs223_chr1_genes_06 <- pc3_218vs223_chr1$gene[pc3_218vs223_chr1$fold_change < 0.6] pc3_218vs223_chr1_genes_15 <- pc3_218vs223_chr1$gene[pc3_218vs223_chr1$fold_change > 1.5]

ggplot(pc3_218vs223_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs223_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs223_chr1, “pc3_218(NEUTRAL)vs223(GAIN)_chr1.csv”, row.names = FALSE)

pc3_218Nvs223G_chr1 <- pc3_218vs223_chr1

pc3_218Nvs223G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs223_chr1_genes_06)) write.csv(pc3_218Nvs223G_chr1_genes_06, “pc3_218(NEUTRAL)vs223(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs223G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_218vs223_chr1_genes_15)) write.csv(pc3_218Nvs223G_chr1_genes_15, “pc3_218(NEUTRAL)vs223(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 224 (gain)

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_218vs224_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_224_chr1_GAIN$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_218vs224_chr1, ]

pc3_218vs224_chr1 <- merge(pc3_218_chr1_NEUTRAL, pc3_224_chr1_GAIN, by = “gene”)

pc3_218vs224_chr1$fold_change <- pc3_218vs224_chr1$TPM.x / pc3_218vs224_chr1$TPM.y

pc3_218vs224_chr1_genes_06 <- pc3_218vs224_chr1$gene[pc3_218vs224_chr1$fold_change < 0.6] pc3_218vs224_chr1_genes_15 <- pc3_218vs224_chr1$gene[pc3_218vs224_chr1$fold_change > 1.5]

ggplot(pc3_218vs224_chr1, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs224_chr1, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs224_chr1, “pc3_218(NEUTRAL)vs224(GAIN)_chr1.csv”, row.names = FALSE)

pc3_218Nvs224G_chr1 <- pc3_218vs224_chr1

pc3_218Nvs224G_chr1_genes_06 <- data.frame(gene = na.omit(pc3_218vs224_chr1_genes_06)) write.csv(pc3_218Nvs224G_chr1_genes_06, “pc3_218(NEUTRAL)vs224(GAIN)_chr1_genes_06.csv”, row.names = FALSE)

pc3_218Nvs224G_chr1_genes_15 <- data.frame(gene = na.omit(pc3_217vs224_chr1_genes_15)) write.csv(pc3_218Nvs224G_chr1_genes_15, “pc3_218(NEUTRAL)vs224(GAIN)_chr1_genes_15.csv”, row.names = FALSE)

CHR 1 - FC_Genes

neutrals vs neutrals

217 vs 218

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”)

common_genes_217vs218_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_218_chr1_NEUTRAL$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs218_chr1, ] pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_217vs218_chr1, ]

pc3_217vs218_chr1_neutral <- merge(pc3_217_chr1_NEUTRAL, pc3_218_chr1_NEUTRAL, by = “gene”)

pc3_217vs218_chr1_neutral$fold_change <- pc3_217vs218_chr1_neutral$TPM.x / pc3_217vs218_chr1_neutral$TPM.y

pc3_217vs218_chr1_neutral_genes_06 <- pc3_217vs218_chr1_neutral$gene[pc3_217vs218_chr1_neutral$fold_change < 0.6] pc3_217vs218_chr1_neutral_genes_15 <- pc3_217vs218_chr1_neutral$gene[pc3_217vs218_chr1_neutral$fold_change > 1.5]

ggplot(pc3_217vs218_chr1_neutral, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs218_chr1_neutral, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_218_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 218 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs218_chr1_neutral, “pc3_217vs218_NEUTRAL_chr1.csv”, row.names = FALSE)

pc3_217vs218_chr1_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_217vs218_chr1_neutral_genes_06)) write.csv(pc3_217vs218_chr1_NEUTRAL_genes_06, “pc3_217vs218_NEUTRAL_chr1_genes_06.csv”, row.names = FALSE)

pc3_217vs218_chr1_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_217vs218_chr1_neutral_genes_15)) write.csv(pc3_217vs218_chr1_NEUTRAL_genes_15, “pc3_217vs218_NEUTRAL_chr1_genes_15.csv”, row.names = FALSE)

217 vs 219

pc3_217_chr1_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr1”) pc3_219_chr1_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr1”)

common_genes_217vs219_chr1 <- intersect(pc3_217_chr1_NEUTRAL$gene, pc3_219_chr1_NEUTRAL$gene)

pc3_217_chr1_NEUTRAL <- pc3_217_chr1_NEUTRAL[pc3_217_chr1_NEUTRAL$gene %in% common_genes_217vs219_chr1, ] pc3_219_chr1_NEUTRAL <- pc3_219_chr1_NEUTRAL[pc3_219_chr1_NEUTRAL$gene %in% common_genes_217vs219_chr1, ]

pc3_217vs219_chr1_neutral <- merge(pc3_217_chr1_NEUTRAL, pc3_219_chr1_NEUTRAL, by = “gene”)

pc3_217vs219_chr1_neutral$fold_change <- pc3_217vs219_chr1_neutral$TPM.x / pc3_217vs219_chr1_neutral$TPM.y

pc3_217vs219_chr1_neutral_genes_06 <- pc3_217vs219_chr1_neutral$gene[pc3_217vs219_chr1_neutral$fold_change < 0.6] pc3_217vs219_chr1_neutral_genes_15 <- pc3_217vs219_chr1_neutral$gene[pc3_217vs219_chr1_neutral$fold_change > 1.5]

ggplot(pc3_217vs219_chr1_neutral, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs219_chr1_neutral, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr1_NEUTRAL (TPM)”, y = “pc3_219_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 217 (NEUTRAL) vs 219 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs219_chr1_neutral, “pc3_217vs219_NEUTRAL_chr1.csv”, row.names = FALSE)

pc3_217vs219_chr1_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_217vs219_chr1_neutral_genes_06)) write.csv(pc3_217vs219_chr1_NEUTRAL_genes_06, “pc3_217vs219_NEUTRAL_chr1_genes_06.csv”, row.names = FALSE)

pc3_217vs219_chr1_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_217vs219_chr1_neutral_genes_15)) write.csv(pc3_217vs219_chr1_NEUTRAL_genes_15, “pc3_217vs219_NEUTRAL_chr1_genes_15.csv”, row.names = FALSE)

218 vs 219

pc3_218_chr1_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr1”) pc3_219_chr1_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr1”)

common_genes_218vs219_chr1 <- intersect(pc3_218_chr1_NEUTRAL$gene, pc3_219_chr1_NEUTRAL$gene)

pc3_218_chr1_NEUTRAL <- pc3_218_chr1_NEUTRAL[pc3_218_chr1_NEUTRAL$gene %in% common_genes_218vs219_chr1, ] pc3_219_chr1_NEUTRAL <- pc3_219_chr1_NEUTRAL[pc3_219_chr1_NEUTRAL$gene %in% common_genes_218vs219_chr1, ]

pc3_218vs219_chr1_NEUTRAL <- merge(pc3_218_chr1_NEUTRAL, pc3_219_chr1_NEUTRAL, by = “gene”)

pc3_218vs219_chr1_NEUTRAL$fold_change <- pc3_218vs219_chr1_NEUTRAL$TPM.x / pc3_218vs219_chr1_NEUTRAL$TPM.y

pc3_218vs219_chr1_NEUTRAL_genes_06 <- pc3_218vs219_chr1_NEUTRAL$gene[pc3_218vs219_chr1_NEUTRAL$fold_change < 0.6] pc3_218vs219_chr1_NEUTRAL_genes_15 <- pc3_218vs219_chr1_NEUTRAL$gene[pc3_218vs219_chr1_NEUTRAL$fold_change > 1.5]

ggplot(pc3_218vs219_chr1_NEUTRAL, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs219_chr1_NEUTRAL, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr1_NEUTRAL (TPM)”, y = “pc3_219_chr1_NEUTRAL (TPM)”, title = “pc3_chr1: 218 (NEUTRAL) vs 219 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs219_chr1_NEUTRAL, “pc3_218vs219_NEUTRAL_chr1.csv”, row.names = FALSE)

pc3_218vs219_chr1_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_218vs219_chr1_NEUTRAL_genes_06)) write.csv(pc3_218vs219_chr1_NEUTRAL_genes_06, “pc3_218vs219_NEUTRAL_chr1_genes_06.csv”, row.names = FALSE)

pc3_218vs219_chr1_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_218vs219_chr1_NEUTRAL_genes_15)) write.csv(pc3_218vs219_chr1_NEUTRAL_genes_15, “pc3_218vs219_NEUTRAL_chr1_genes_15.csv”, row.names = FALSE)

gains vs gains

219 vs 221 (gain)

pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”) pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”)

common_genes_219vs221_chr1 <- intersect(pc3_219_chr1_GAIN$gene, pc3_221_chr1_GAIN$gene)

pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_219vs221_chr1, ] pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_219vs221_chr1, ]

pc3_219vs221_chr1_GAIN <- merge(pc3_219_chr1_GAIN, pc3_221_chr1_GAIN, by = “gene”)

pc3_219vs221_chr1_GAIN$fold_change <- pc3_219vs221_chr1_GAIN$TPM.x / pc3_219vs221_chr1_GAIN$TPM.y

pc3_219vs221_chr1_GAIN_genes_06 <- pc3_219vs221_chr1_GAIN$gene[pc3_219vs221_chr1_GAIN$fold_change < 0.6] pc3_219vs221_chr1_GAIN_genes_15 <- pc3_219vs221_chr1_GAIN$gene[pc3_219vs221_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_219vs221_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs221_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_221_chr1_GAIN (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs221_chr1_GAIN, “pc3_219vs221_GAIN_chr1.csv”, row.names = FALSE)

pc3_219vs221_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_219vs221_chr1_GAIN_genes_06)) write.csv(pc3_219vs221_chr1_GAIN_genes_06, “pc3_219vs221_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_219vs221_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_219vs221_chr1_GAIN_genes_15)) write.csv(pc3_219vs221_chr1_GAIN_genes_15, “pc3_219vs221_GAIN_chr1_genes_15.csv”, row.names = FALSE)

219 vs 222

pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”)

common_genes_219vs222_chr1 <- intersect(pc3_219_chr1_GAIN$gene, pc3_222_chr1_GAIN$gene)

pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_219vs222_chr1, ] pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_219vs222_chr1, ]

pc3_219vs222_chr1_GAIN <- merge(pc3_219_chr1_GAIN, pc3_222_chr1_GAIN, by = “gene”)

pc3_219vs222_chr1_GAIN$fold_change <- pc3_219vs222_chr1_GAIN$TPM.x / pc3_219vs222_chr1_GAIN$TPM.y

pc3_219vs222_chr1_GAIN_genes_06 <- pc3_219vs222_chr1_GAIN$gene[pc3_219vs222_chr1_GAIN$fold_change < 0.6] pc3_219vs222_chr1_GAIN_genes_15 <- pc3_219vs222_chr1_GAIN$gene[pc3_219vs222_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_219vs222_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs222_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_222_chr1_GAIN (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 222 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs222_chr1_GAIN, “pc3_219vs222_GAIN_chr1.csv”, row.names = FALSE)

pc3_219vs222_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_219vs222_chr1_GAIN_genes_06)) write.csv(pc3_219vs222_chr1_GAIN_genes_06, “pc3_219vs222_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_219vs222_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_219vs222_chr1_GAIN_genes_15)) write.csv(pc3_219vs222_chr1_GAIN_genes_15, “pc3_219vs222_GAIN_chr1_genes_15.csv”, row.names = FALSE)

219 vs 223 (gain)

pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_219vs223_chr1 <- intersect(pc3_219_chr1_GAIN$gene, pc3_223_chr1_GAIN$gene)

pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_219vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_219vs223_chr1, ]

pc3_219vs223_chr1_GAIN <- merge(pc3_219_chr1_GAIN, pc3_223_chr1_GAIN, by = “gene”)

pc3_219vs223_chr1_GAIN$fold_change <- pc3_219vs223_chr1_GAIN$TPM.x / pc3_219vs223_chr1_GAIN$TPM.y

pc3_219vs223_chr1_GAIN_genes_06 <- pc3_219vs223_chr1_GAIN$gene[pc3_219vs223_chr1_GAIN$fold_change < 0.6] pc3_219vs223_chr1_GAIN_genes_15 <- pc3_219vs223_chr1_GAIN$gene[pc3_219vs223_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_219vs223_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs223_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs223_chr1_GAIN, “pc3_219vs223_GAIN_chr1.csv”, row.names = FALSE)

pc3_219vs223_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_219vs223_chr1_GAIN_genes_06)) write.csv(pc3_219vs223_chr1_GAIN_genes_06, “pc3_219vs223_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_219vs223_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_219vs223_chr1_GAIN_genes_15)) write.csv(pc3_219vs223_chr1_GAIN_genes_15, “pc3_219vs223_GAIN_chr1_genes_15.csv”, row.names = FALSE)

219 vs 224 (gain)

pc3_219_chr1_GAIN <- subset(pc3_219_GAIN, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_219vs224_chr1 <- intersect(pc3_219_chr1_GAIN$gene, pc3_224_chr1_GAIN$gene)

pc3_219_chr1_GAIN <- pc3_219_chr1_GAIN[pc3_219_chr1_GAIN$gene %in% common_genes_219vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_219vs224_chr1, ]

pc3_219vs224_chr1_GAIN <- merge(pc3_219_chr1_GAIN, pc3_224_chr1_GAIN, by = “gene”)

pc3_219vs224_chr1_GAIN$fold_change <- pc3_219vs224_chr1_GAIN$TPM.x / pc3_219vs224_chr1_GAIN$TPM.y

pc3_219vs224_chr1_GAIN_genes_06 <- pc3_219vs224_chr1_GAIN$gene[pc3_219vs224_chr1_GAIN$fold_change < 0.6] pc3_219vs224_chr1_GAIN_genes_15 <- pc3_219vs224_chr1_GAIN$gene[pc3_219vs224_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_219vs224_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs224_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr1_GAIN (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 219 (GAIN) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs224_chr1_GAIN, “pc3_219vs224_GAIN_chr1.csv”, row.names = FALSE)

pc3_219vs224_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_219vs224_chr1_GAIN_genes_06)) write.csv(pc3_219vs224_chr1_GAIN_genes_06, “pc3_219vs224_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_219vs224_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_219vs224_chr1_GAIN_genes_15)) write.csv(pc3_219vs224_chr1_GAIN_genes_15, “pc3_219vs224_GAIN_chr1_genes_15.csv”, row.names = FALSE)

221 vs 222

pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”) pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”)

common_genes_221vs222_chr1 <- intersect(pc3_221_chr1_GAIN$gene, pc3_222_chr1_GAIN$gene)

pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_221vs222_chr1, ] pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_221vs222_chr1, ]

pc3_221vs222_chr1_GAIN <- merge(pc3_221_chr1_GAIN, pc3_222_chr1_GAIN, by = “gene”)

pc3_221vs222_chr1_GAIN$fold_change <- pc3_221vs222_chr1_GAIN$TPM.x / pc3_221vs222_chr1_GAIN$TPM.y

pc3_221vs222_chr1_GAIN_genes_06 <- pc3_221vs222_chr1_GAIN$gene[pc3_221vs222_chr1_GAIN$fold_change < 0.6] pc3_221vs222_chr1_GAIN_genes_15 <- pc3_221vs222_chr1_GAIN$gene[pc3_221vs222_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_221vs222_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs222_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr1_GAIN (TPM)”, y = “pc3_222_chr1_GAIN (TPM)”, title = “pc3_chr1: 221 (GAIN) vs 222 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs222_chr1_GAIN, “pc3_221vs222_GAIN_chr1.csv”, row.names = FALSE)

pc3_221vs222_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_221vs222_chr1_GAIN_genes_06)) write.csv(pc3_221vs222_chr1_GAIN_genes_06, “pc3_221vs222_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_221vs222_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_221vs222_chr1_GAIN_genes_15)) write.csv(pc3_221vs222_chr1_GAIN_genes_15, “pc3_221vs222_GAIN_chr1_genes_15.csv”, row.names = FALSE)

221 vs 223 (gain)

pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_221vs223_chr1 <- intersect(pc3_221_chr1_GAIN$gene, pc3_223_chr1_GAIN$gene)

pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_221vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_221vs223_chr1, ]

pc3_221vs223_chr1_GAIN <- merge(pc3_221_chr1_GAIN, pc3_223_chr1_GAIN, by = “gene”)

pc3_221vs223_chr1_GAIN$fold_change <- pc3_221vs223_chr1_GAIN$TPM.x / pc3_221vs223_chr1_GAIN$TPM.y

pc3_221vs223_chr1_GAIN_genes_06 <- pc3_221vs223_chr1_GAIN$gene[pc3_221vs223_chr1_GAIN$fold_change < 0.6] pc3_221vs223_chr1_GAIN_genes_15 <- pc3_221vs223_chr1_GAIN$gene[pc3_221vs223_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_221vs223_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs223_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr1_GAIN (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 221 (GAIN) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs223_chr1_GAIN, “pc3_221vs223_GAIN_chr1.csv”, row.names = FALSE)

pc3_221vs223_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_221vs223_chr1_GAIN_genes_06)) write.csv(pc3_221vs223_chr1_GAIN_genes_06, “pc3_221vs223_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_221vs223_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_221vs223_chr1_GAIN_genes_15)) write.csv(pc3_221vs223_chr1_GAIN_genes_15, “pc3_221vs223_GAIN_chr1_genes_15.csv”, row.names = FALSE)

221 vs 224 (gain)

pc3_221_chr1_GAIN <- subset(pc3_221_GAIN, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_221vs224_chr1 <- intersect(pc3_221_chr1_GAIN$gene, pc3_224_chr1_GAIN$gene)

pc3_221_chr1_GAIN <- pc3_221_chr1_GAIN[pc3_221_chr1_GAIN$gene %in% common_genes_221vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_221vs224_chr1, ]

pc3_221vs224_chr1_GAIN <- merge(pc3_221_chr1_GAIN, pc3_224_chr1_GAIN, by = “gene”)

pc3_221vs224_chr1_GAIN$fold_change <- pc3_221vs224_chr1_GAIN$TPM.x / pc3_221vs224_chr1_GAIN$TPM.y

pc3_221vs224_chr1_GAIN_genes_06 <- pc3_221vs224_chr1_GAIN$gene[pc3_221vs224_chr1_GAIN$fold_change < 0.6] pc3_221vs224_chr1_GAIN_genes_15 <- pc3_221vs224_chr1_GAIN$gene[pc3_221vs224_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_221vs224_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs224_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr1_GAIN (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 221 (GAIN) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs224_chr1_GAIN, “pc3_221vs224_GAIN_chr1.csv”, row.names = FALSE)

pc3_221vs224_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_221vs224_chr1_GAIN_genes_06)) write.csv(pc3_221vs224_chr1_GAIN_genes_06, “pc3_221vs224_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_221vs224_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_221vs224_chr1_GAIN_genes_15)) write.csv(pc3_221vs224_chr1_GAIN_genes_15, “pc3_221vs224_GAIN_chr1_genes_15.csv”, row.names = FALSE)

222 vs 223 (gain)

pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”) pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”)

common_genes_222vs223_chr1 <- intersect(pc3_222_chr1_GAIN$gene, pc3_223_chr1_GAIN$gene)

pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_222vs223_chr1, ] pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_222vs223_chr1, ]

pc3_222vs223_chr1_GAIN <- merge(pc3_222_chr1_GAIN, pc3_223_chr1_GAIN, by = “gene”)

pc3_222vs223_chr1_GAIN$fold_change <- pc3_222vs223_chr1_GAIN$TPM.x / pc3_222vs223_chr1_GAIN$TPM.y

pc3_222vs223_chr1_GAIN_genes_06 <- pc3_222vs223_chr1_GAIN$gene[pc3_222vs223_chr1_GAIN$fold_change < 0.6] pc3_222vs223_chr1_GAIN_genes_15 <- pc3_222vs223_chr1_GAIN$gene[pc3_222vs223_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_222vs223_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs223_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr1_GAIN (TPM)”, y = “pc3_223_chr1_GAIN (TPM)”, title = “pc3_chr1: 222 (GAIN) vs 223 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs223_chr1_GAIN, “pc3_222vs223_GAIN_chr1.csv”, row.names = FALSE)

pc3_222vs223_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_222vs223_chr1_GAIN_genes_06)) write.csv(pc3_222vs223_chr1_GAIN_genes_06, “pc3_222vs223_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_222vs223_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_222vs223_chr1_GAIN_genes_15)) write.csv(pc3_222vs223_chr1_GAIN_genes_15, “pc3_222vs223_GAIN_chr1_genes_15.csv”, row.names = FALSE)

222 vs 224 (gain)

pc3_222_chr1_GAIN <- subset(pc3_222_GAIN, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_222vs224_chr1 <- intersect(pc3_222_chr1_GAIN$gene, pc3_224_chr1_GAIN$gene)

pc3_222_chr1_GAIN <- pc3_222_chr1_GAIN[pc3_222_chr1_GAIN$gene %in% common_genes_222vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_222vs224_chr1, ]

pc3_222vs224_chr1_GAIN <- merge(pc3_222_chr1_GAIN, pc3_224_chr1_GAIN, by = “gene”)

pc3_222vs224_chr1_GAIN$fold_change <- pc3_222vs224_chr1_GAIN$TPM.x / pc3_222vs224_chr1_GAIN$TPM.y

pc3_222vs224_chr1_GAIN_genes_06 <- pc3_222vs224_chr1_GAIN$gene[pc3_222vs224_chr1_GAIN$fold_change < 0.6] pc3_222vs224_chr1_GAIN_genes_15 <- pc3_222vs224_chr1_GAIN$gene[pc3_222vs224_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_222vs224_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs224_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr1_GAIN (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 222 (GAIN) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs224_chr1_GAIN, “pc3_222vs224_GAIN_chr1.csv”, row.names = FALSE)

pc3_222vs224_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_222vs224_chr1_GAIN_genes_06)) write.csv(pc3_222vs224_chr1_GAIN_genes_06, “pc3_222vs224_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_222vs224_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_222vs224_chr1_GAIN_genes_15)) write.csv(pc3_222vs224_chr1_GAIN_genes_15, “pc3_222vs224_GAIN_chr1_genes_15.csv”, row.names = FALSE)

223 vs 224 (gain)

pc3_223_chr1_GAIN <- subset(pc3_223_GAIN, chr == “chr1”) pc3_224_chr1_GAIN <- subset(pc3_224_GAIN, chr == “chr1”)

common_genes_223vs224_chr1 <- intersect(pc3_223_chr1_GAIN$gene, pc3_224_chr1_GAIN$gene)

pc3_223_chr1_GAIN <- pc3_223_chr1_GAIN[pc3_223_chr1_GAIN$gene %in% common_genes_223vs224_chr1, ] pc3_224_chr1_GAIN <- pc3_224_chr1_GAIN[pc3_224_chr1_GAIN$gene %in% common_genes_223vs224_chr1, ]

pc3_223vs224_chr1_GAIN <- merge(pc3_223_chr1_GAIN, pc3_224_chr1_GAIN, by = “gene”)

pc3_223vs224_chr1_GAIN$fold_change <- pc3_223vs224_chr1_GAIN$TPM.x / pc3_223vs224_chr1_GAIN$TPM.y

pc3_223vs224_chr1_GAIN_genes_06 <- pc3_223vs224_chr1_GAIN$gene[pc3_223vs224_chr1_GAIN$fold_change < 0.6] pc3_223vs224_chr1_GAIN_genes_15 <- pc3_223vs224_chr1_GAIN$gene[pc3_223vs224_chr1_GAIN$fold_change > 1.5]

ggplot(pc3_223vs224_chr1_GAIN, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_223vs224_chr1_GAIN, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_223_chr1_GAIN (TPM)”, y = “pc3_224_chr1_GAIN (TPM)”, title = “pc3_chr1: 223 (GAIN) vs 224 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_223vs224_chr1_GAIN, “pc3_223vs224_GAIN_chr1.csv”, row.names = FALSE)

pc3_223vs224_chr1_GAIN_genes_06 <- data.frame(gene = na.omit(pc3_223vs224_chr1_GAIN_genes_06)) write.csv(pc3_223vs224_chr1_GAIN_genes_06, “pc3_223vs224_GAIN_chr1_genes_06.csv”, row.names = FALSE)

pc3_223vs224_chr1_GAIN_genes_15 <- data.frame(gene = na.omit(pc3_223vs224_chr1_GAIN_genes_15)) write.csv(pc3_223vs224_chr1_GAIN_genes_15, “pc3_223vs224_GAIN_chr1_genes_15.csv”, row.names = FALSE)

CHROMOSOME 8

217 (neutral) vs 221 (gain)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_217vs221_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_217vs221_chr8, ]

pc3_217vs221_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_217vs221_chr8$fold_change <- pc3_217vs221_chr8$TPM.x / pc3_217vs221_chr8$TPM.y

pc3_217vs221_chr8_genes_06 <- pc3_217vs221_chr8$gene[pc3_217vs221_chr8$fold_change < 0.6] pc3_217vs221_chr8_genes_15 <- pc3_217vs221_chr8$gene[pc3_217vs221_chr8$fold_change > 1.5]

ggplot(pc3_217vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs221_chr8, “pc3_217(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_217Nvs221G_chr8 <- pc3_217vs221_chr8

pc3_217Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs221_chr8_genes_06)) write.csv(pc3_217Nvs221G_chr8_genes_06, “pc3_217(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs221_chr8_genes_15)) write.csv(pc3_217Nvs221G_chr8_genes_15, “pc3_217(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 221 (gain)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_218vs221_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_218vs221_chr8, ]

pc3_218vs221_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_218vs221_chr8$fold_change <- pc3_218vs221_chr8$TPM.x / pc3_218vs221_chr8$TPM.y

pc3_218vs221_chr8_genes_06 <- pc3_218vs221_chr8$gene[pc3_218vs221_chr8$fold_change < 0.6] pc3_218vs221_chr8_genes_15 <- pc3_218vs221_chr8$gene[pc3_218vs221_chr8$fold_change > 1.5]

ggplot(pc3_218vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs221_chr8, “pc3_218(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_218Nvs221G_chr8 <- pc3_218vs221_chr8

pc3_218Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs221_chr8_genes_06)) write.csv(pc3_218Nvs221G_chr8_genes_06, “pc3_218(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs221_chr8_genes_15)) write.csv(pc3_218Nvs221G_chr8_genes_15, “pc3_218(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 221 (gain)

pc3_219_chr8_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_219vs221_chr8 <- intersect(pc3_219_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_219_chr8_NEUTRAL <- pc3_219_chr8_NEUTRAL[pc3_219_chr8_NEUTRAL$gene %in% common_genes_219vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_219vs221_chr8, ]

pc3_219vs221_chr8 <- merge(pc3_219_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_219vs221_chr8$fold_change <- pc3_219vs221_chr8$TPM.x / pc3_219vs221_chr8$TPM.y

pc3_219vs221_chr8_genes_06 <- pc3_219vs221_chr8$gene[pc3_219vs221_chr8$fold_change < 0.6] pc3_219vs221_chr8_genes_15 <- pc3_219vs221_chr8$gene[pc3_219vs221_chr8$fold_change > 1.5]

ggplot(pc3_219vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 25) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 219 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs221_chr8, “pc3_219(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_219Nvs221G_chr8 <- pc3_219vs221_chr8

pc3_219Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_219vs221_chr8_genes_06)) write.csv(pc3_219Nvs221G_chr8_genes_06, “pc3_219(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_219Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_219vs221_chr8_genes_15)) write.csv(pc3_219Nvs221G_chr8_genes_15, “pc3_219(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

220 (neutral) vs 221 (gain)

pc3_220_chr8_NEUTRAL <- subset(pc3_220_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_220vs221_chr8 <- intersect(pc3_220_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_220_chr8_NEUTRAL <- pc3_220_chr8_NEUTRAL[pc3_220_chr8_NEUTRAL$gene %in% common_genes_220vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_220vs221_chr8, ]

pc3_220vs221_chr8 <- merge(pc3_220_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_220vs221_chr8$fold_change <- pc3_220vs221_chr8$TPM.x / pc3_220vs221_chr8$TPM.y

pc3_220vs221_chr8_genes_06 <- pc3_220vs221_chr8$gene[pc3_220vs221_chr8$fold_change < 0.6] pc3_220vs221_chr8_genes_15 <- pc3_220vs221_chr8$gene[pc3_220vs221_chr8$fold_change > 1.5]

ggplot(pc3_220vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 220 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs221_chr8, “pc3_220(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_220Nvs221G_chr8 <- pc3_220vs221_chr8

pc3_220Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_220vs221_chr8_genes_06)) write.csv(pc3_220Nvs221G_chr8_genes_06, “pc3_220(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_220Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_220vs221_chr8_genes_15)) write.csv(pc3_220Nvs221G_chr8_genes_15, “pc3_220(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

222 (neutral) vs 221 (gain)

pc3_222_chr8_NEUTRAL <- subset(pc3_222_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_222vs221_chr8 <- intersect(pc3_222_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_222_chr8_NEUTRAL <- pc3_222_chr8_NEUTRAL[pc3_222_chr8_NEUTRAL$gene %in% common_genes_222vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_222vs221_chr8, ]

pc3_222vs221_chr8 <- merge(pc3_222_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_222vs221_chr8$fold_change <- pc3_222vs221_chr8$TPM.x / pc3_222vs221_chr8$TPM.y

pc3_222vs221_chr8_genes_06 <- pc3_222vs221_chr8$gene[pc3_222vs221_chr8$fold_change < 0.6] pc3_222vs221_chr8_genes_15 <- pc3_222vs221_chr8$gene[pc3_222vs221_chr8$fold_change > 1.5]

ggplot(pc3_222vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 65) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 222 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs221_chr8, “pc3_222(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_222Nvs221G_chr8 <- pc3_222vs221_chr8

pc3_222Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_222vs221_chr8_genes_06)) write.csv(pc3_222Nvs221G_chr8_genes_06, “pc3_222(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_222Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_222vs221_chr8_genes_15)) write.csv(pc3_222Nvs221G_chr8_genes_15, “pc3_222(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

223 (neutral) vs 221 (gain)

pc3_223_chr8_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_223vs221_chr8 <- intersect(pc3_223_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_223_chr8_NEUTRAL <- pc3_223_chr8_NEUTRAL[pc3_223_chr8_NEUTRAL$gene %in% common_genes_223vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_223vs221_chr8, ]

pc3_223vs221_chr8 <- merge(pc3_223_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_223vs221_chr8$fold_change <- pc3_223vs221_chr8$TPM.x / pc3_223vs221_chr8$TPM.y

pc3_223vs221_chr8_genes_06 <- pc3_223vs221_chr8$gene[pc3_223vs221_chr8$fold_change < 0.6] pc3_223vs221_chr8_genes_15 <- pc3_223vs221_chr8$gene[pc3_223vs221_chr8$fold_change > 1.5]

ggplot(pc3_223vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_223vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_223_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 223 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_223vs221_chr8, “pc3_223(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_223Nvs221G_chr8 <- pc3_223vs221_chr8

pc3_223Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_223vs221_chr8_genes_06)) write.csv(pc3_223Nvs221G_chr8_genes_06, “pc3_223(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_223Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_223vs221_chr8_genes_15)) write.csv(pc3_223Nvs221G_chr8_genes_15, “pc3_223(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

224 (neutral) vs 221 (gain)

pc3_224_chr8_NEUTRAL <- subset(pc3_224_NEUTRAL, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_224vs221_chr8 <- intersect(pc3_224_chr8_NEUTRAL$gene, pc3_221_chr8_GAIN$gene)

pc3_224_chr8_NEUTRAL <- pc3_224_chr8_NEUTRAL[pc3_224_chr8_NEUTRAL$gene %in% common_genes_224vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_224vs221_chr8, ]

pc3_224vs221_chr8 <- merge(pc3_224_chr8_NEUTRAL, pc3_221_chr8_GAIN, by = “gene”)

pc3_224vs221_chr8$fold_change <- pc3_224vs221_chr8$TPM.x / pc3_224vs221_chr8$TPM.y

pc3_224vs221_chr8_genes_06 <- pc3_224vs221_chr8$gene[pc3_224vs221_chr8$fold_change < 0.6] pc3_224vs221_chr8_genes_15 <- pc3_224vs221_chr8$gene[pc3_224vs221_chr8$fold_change > 1.5]

ggplot(pc3_224vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_224vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 35) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_224_chr8_NEUTRAL (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 224 (NEUTRAL) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_224vs221_chr8, “pc3_224(NEUTRAL)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_224Nvs221G_chr8 <- pc3_224vs221_chr8

pc3_224Nvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_224vs221_chr8_genes_06)) write.csv(pc3_224Nvs221G_chr8_genes_06, “pc3_224(NEUTRAL)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_224Nvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_224vs221_chr8_genes_15)) write.csv(pc3_224Nvs221G_chr8_genes_15, “pc3_224(NEUTRAL)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 219 (loss)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”)

common_genes_217vs219_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_219_chr8_LOSS$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs219_chr8, ] pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_217vs219_chr8, ]

pc3_217vs219_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_219_chr8_LOSS, by = “gene”)

pc3_217vs219_chr8$fold_change <- pc3_217vs219_chr8$TPM.x / pc3_217vs219_chr8$TPM.y

pc3_217vs219_chr8_genes_06 <- pc3_217vs219_chr8$gene[pc3_217vs219_chr8$fold_change < 0.6] pc3_217vs219_chr8_genes_15 <- pc3_217vs219_chr8$gene[pc3_217vs219_chr8$fold_change > 1.5]

ggplot(pc3_217vs219_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs219_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_219_chr8_LOSS (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 219 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs219_chr8, “pc3_217(NEUTRAL)vs219(LOSS)_chr8.csv”, row.names = FALSE)

pc3_217Nvs219L_chr8 <- pc3_217vs219_chr8

pc3_217Nvs219L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs219_chr8_genes_06)) write.csv(pc3_217Nvs219L_chr8_genes_06, “pc3_217(NEUTRAL)vs219(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs219L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs219_chr8_genes_15)) write.csv(pc3_217Nvs219L_chr8_genes_15, “pc3_217(NEUTRAL)vs219(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 220 (loss)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”)

common_genes_217vs220_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_220_chr8_LOSS$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs220_chr8, ] pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_217vs220_chr8, ]

pc3_217vs220_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_220_chr8_LOSS, by = “gene”)

pc3_217vs220_chr8$fold_change <- pc3_217vs220_chr8$TPM.x / pc3_217vs220_chr8$TPM.y

pc3_217vs220_chr8_genes_06 <- pc3_217vs220_chr8$gene[pc3_217vs220_chr8$fold_change < 0.6] pc3_217vs220_chr8_genes_15 <- pc3_217vs220_chr8$gene[pc3_217vs220_chr8$fold_change > 1.5]

ggplot(pc3_217vs220_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_220_chr8_LOSS (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs220_chr8, “pc3_217(NEUTRAL)vs220(LOSS)_chr8.csv”, row.names = FALSE)

pc3_217Nvs220L_chr8 <- pc3_217vs220_chr8

pc3_217Nvs220L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr8_genes_06)) write.csv(pc3_217Nvs220L_chr8_genes_06, “pc3_217(NEUTRAL)vs220(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs220L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr8_genes_15)) write.csv(pc3_217Nvs220L_chr8_genes_15, “pc3_217(NEUTRAL)vs220(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 222 (loss)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”)

common_genes_217vs222_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_222_chr8_LOSS$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs222_chr8, ] pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_217vs222_chr8, ]

pc3_217vs222_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_222_chr8_LOSS, by = “gene”)

pc3_217vs222_chr8$fold_change <- pc3_217vs222_chr8$TPM.x / pc3_217vs222_chr8$TPM.y

pc3_217vs222_chr8_genes_06 <- pc3_217vs222_chr8$gene[pc3_217vs222_chr8$fold_change < 0.6] pc3_217vs222_chr8_genes_15 <- pc3_217vs222_chr8$gene[pc3_217vs222_chr8$fold_change > 1.5]

ggplot(pc3_217vs222_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs222_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_222_chr8_LOSS (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs222_chr8, “pc3_217(NEUTRAL)vs222(LOSS)_chr8.csv”, row.names = FALSE)

pc3_217Nvs222L_chr8 <- pc3_217vs222_chr8

pc3_217Nvs222L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs222_chr8_genes_06)) write.csv(pc3_217Nvs222L_chr8_genes_06, “pc3_217(NEUTRAL)vs222(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs222L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs222_chr8_genes_15)) write.csv(pc3_217Nvs222L_chr8_genes_15, “pc3_217(NEUTRAL)vs222(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 223 (loss)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”)

common_genes_217vs223_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_223_chr8_LOSS$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs223_chr8, ] pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_217vs223_chr8, ]

pc3_217vs223_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_223_chr8_LOSS, by = “gene”)

pc3_217vs223_chr8$fold_change <- pc3_217vs223_chr8$TPM.x / pc3_217vs223_chr8$TPM.y

pc3_217vs223_chr8_genes_06 <- pc3_217vs223_chr8$gene[pc3_217vs223_chr8$fold_change < 0.6] pc3_217vs223_chr8_genes_15 <- pc3_217vs223_chr8$gene[pc3_217vs223_chr8$fold_change > 1.5]

ggplot(pc3_217vs223_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs223_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_223_chr8_LOSS (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 223 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs223_chr8, “pc3_217(NEUTRAL)vs223(LOSS)_chr8.csv”, row.names = FALSE)

pc3_217Nvs223L_chr8 <- pc3_217vs223_chr8

pc3_217Nvs223L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs223_chr8_genes_06)) write.csv(pc3_217Nvs223L_chr8_genes_06, “pc3_217(NEUTRAL)vs223(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs223L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs223_chr8_genes_15)) write.csv(pc3_217Nvs223L_chr8_genes_15, “pc3_217(NEUTRAL)vs223(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 224 (loss)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”)

common_genes_217vs224_chr8 <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_224_chr8_LOSS$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs224_chr8, ] pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_217vs224_chr8, ]

pc3_217vs224_chr8 <- merge(pc3_217_chr8_NEUTRAL, pc3_224_chr8_LOSS, by = “gene”)

pc3_217vs224_chr8$fold_change <- pc3_217vs224_chr8$TPM.x / pc3_217vs224_chr8$TPM.y

pc3_217vs224_chr8_genes_06 <- pc3_217vs224_chr8$gene[pc3_217vs224_chr8$fold_change < 0.6] pc3_217vs224_chr8_genes_15 <- pc3_217vs224_chr8$gene[pc3_217vs224_chr8$fold_change > 1.5]

ggplot(pc3_217vs224_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs224_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_224_chr8_LOSS (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs224_chr8, “pc3_217(NEUTRAL)vs224(LOSS)_chr8.csv”, row.names = FALSE)

pc3_217Nvs224L_chr8 <- pc3_217vs224_chr8

pc3_217Nvs224L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_217vs224_chr8_genes_06)) write.csv(pc3_217Nvs224L_chr8_genes_06, “pc3_217(NEUTRAL)vs224(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_217Nvs224L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_217vs224_chr8_genes_15)) write.csv(pc3_217Nvs224L_chr8_genes_15, “pc3_217(NEUTRAL)vs224(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 219 (loss)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”)

common_genes_218vs219_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_219_chr8_LOSS$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs219_chr8, ] pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_218vs219_chr8, ]

pc3_218vs219_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_219_chr8_LOSS, by = “gene”)

pc3_218vs219_chr8$fold_change <- pc3_218vs219_chr8$TPM.x / pc3_218vs219_chr8$TPM.y

pc3_218vs219_chr8_genes_06 <- pc3_218vs219_chr8$gene[pc3_218vs219_chr8$fold_change < 0.6] pc3_218vs219_chr8_genes_15 <- pc3_218vs219_chr8$gene[pc3_218vs219_chr8$fold_change > 1.5]

ggplot(pc3_218vs219_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs219_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_219_chr8_LOSS (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 219 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs219_chr8, “pc3_218(NEUTRAL)vs219(LOSS)_chr8.csv”, row.names = FALSE)

pc3_218Nvs219L_chr8 <- pc3_218vs219_chr8

pc3_218Nvs219L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs219_chr8_genes_06)) write.csv(pc3_218Nvs219L_chr8_genes_06, “pc3_218(NEUTRAL)vs219(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs219L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs219_chr8_genes_15)) write.csv(pc3_218Nvs219L_chr8_genes_15, “pc3_218(NEUTRAL)vs219(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 220 (loss)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”)

common_genes_218vs220_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_220_chr8_LOSS$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs220_chr8, ] pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_218vs220_chr8, ]

pc3_218vs220_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_220_chr8_LOSS, by = “gene”)

pc3_218vs220_chr8$fold_change <- pc3_218vs220_chr8$TPM.x / pc3_218vs220_chr8$TPM.y

pc3_218vs220_chr8_genes_06 <- pc3_218vs220_chr8$gene[pc3_218vs220_chr8$fold_change < 0.6] pc3_218vs220_chr8_genes_15 <- pc3_218vs220_chr8$gene[pc3_218vs220_chr8$fold_change > 1.5]

ggplot(pc3_218vs220_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs220_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_220_chr8_LOSS (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs220_chr8, “pc3_218(NEUTRAL)vs220(LOSS)_chr8.csv”, row.names = FALSE)

pc3_218Nvs220L_chr8 <- pc3_218vs220_chr8

pc3_218Nvs220L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs220_chr8_genes_06)) write.csv(pc3_218Nvs220L_chr8_genes_06, “pc3_218(NEUTRAL)vs220(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs220L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs220_chr8_genes_15)) write.csv(pc3_218Nvs220L_chr8_genes_15, “pc3_218(NEUTRAL)vs220(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 222 (loss)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”)

common_genes_218vs222_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_222_chr8_LOSS$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs222_chr8, ] pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_218vs222_chr8, ]

pc3_218vs222_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_222_chr8_LOSS, by = “gene”)

pc3_218vs222_chr8$fold_change <- pc3_218vs222_chr8$TPM.x / pc3_218vs222_chr8$TPM.y

pc3_218vs222_chr8_genes_06 <- pc3_218vs222_chr8$gene[pc3_218vs222_chr8$fold_change < 0.6] pc3_218vs222_chr8_genes_15 <- pc3_218vs222_chr8$gene[pc3_218vs222_chr8$fold_change > 1.5]

ggplot(pc3_218vs222_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs222_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_222_chr8_LOSS (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs222_chr8, “pc3_218(NEUTRAL)vs222(LOSS)_chr8.csv”, row.names = FALSE)

pc3_218Nvs222L_chr8 <- pc3_218vs222_chr8

pc3_218Nvs222L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs222_chr8_genes_06)) write.csv(pc3_218Nvs222L_chr8_genes_06, “pc3_218(NEUTRAL)vs222(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs222L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs222_chr8_genes_15)) write.csv(pc3_218Nvs222L_chr8_genes_15, “pc3_218(NEUTRAL)vs222(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 223 (loss)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”)

common_genes_218vs223_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_223_chr8_LOSS$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs223_chr8, ] pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_218vs223_chr8, ]

pc3_218vs223_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_223_chr8_LOSS, by = “gene”)

pc3_218vs223_chr8$fold_change <- pc3_218vs223_chr8$TPM.x / pc3_218vs223_chr8$TPM.y

pc3_218vs223_chr8_genes_06 <- pc3_218vs223_chr8$gene[pc3_218vs223_chr8$fold_change < 0.6] pc3_218vs223_chr8_genes_15 <- pc3_218vs223_chr8$gene[pc3_218vs223_chr8$fold_change > 1.5]

ggplot(pc3_218vs223_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs223_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_223_chr8_LOSS (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 223 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs223_chr8, “pc3_218(NEUTRAL)vs223(LOSS)_chr8.csv”, row.names = FALSE)

pc3_218Nvs223L_chr8 <- pc3_218vs223_chr8

pc3_218Nvs223L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs223_chr8_genes_06)) write.csv(pc3_218Nvs223L_chr8_genes_06, “pc3_218(NEUTRAL)vs223(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs223L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs223_chr8_genes_15)) write.csv(pc3_218Nvs223L_chr8_genes_15, “pc3_218(NEUTRAL)vs223(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

218 (neutral) vs 224 (loss)

pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”) pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”)

common_genes_218vs224_chr8 <- intersect(pc3_218_chr8_NEUTRAL$gene, pc3_224_chr8_LOSS$gene)

pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_218vs224_chr8, ] pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_218vs224_chr8, ]

pc3_218vs224_chr8 <- merge(pc3_218_chr8_NEUTRAL, pc3_224_chr8_LOSS, by = “gene”)

pc3_218vs224_chr8$fold_change <- pc3_218vs224_chr8$TPM.x / pc3_218vs224_chr8$TPM.y

pc3_218vs224_chr8_genes_06 <- pc3_218vs224_chr8$gene[pc3_218vs224_chr8$fold_change < 0.6] pc3_218vs224_chr8_genes_15 <- pc3_218vs224_chr8$gene[pc3_218vs224_chr8$fold_change > 1.5]

ggplot(pc3_218vs224_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs224_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr8_NEUTRAL (TPM)”, y = “pc3_224_chr8_LOSS (TPM)”, title = “pc3_chr8: 218 (NEUTRAL) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs224_chr8, “pc3_218(NEUTRAL)vs224(LOSS)_chr8.csv”, row.names = FALSE)

pc3_218Nvs224L_chr8 <- pc3_218vs224_chr8

pc3_218Nvs224L_chr8_genes_06 <- data.frame(gene = na.omit(pc3_218vs224_chr8_genes_06)) write.csv(pc3_218Nvs224L_chr8_genes_06, “pc3_218(NEUTRAL)vs224(LOSS)_chr8_genes_06.csv”, row.names = FALSE)

pc3_218Nvs224L_chr8_genes_15 <- data.frame(gene = na.omit(pc3_218vs224_chr8_genes_15)) write.csv(pc3_218Nvs224L_chr8_genes_15, “pc3_218(NEUTRAL)vs224(LOSS)_chr8_genes_15.csv”, row.names = FALSE)

219 (LOSS) vs 221 (gain)

pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_219vs221_chr8 <- intersect(pc3_219_chr8_LOSS$gene, pc3_221_chr8_GAIN$gene)

pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_219vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_219vs221_chr8, ]

pc3_219vs221_chr8 <- merge(pc3_219_chr8_LOSS, pc3_221_chr8_GAIN, by = “gene”)

pc3_219vs221_chr8$fold_change <- pc3_219vs221_chr8$TPM.x / pc3_219vs221_chr8$TPM.y

pc3_219vs221_chr8_genes_06 <- pc3_219vs221_chr8$gene[pc3_219vs221_chr8$fold_change < 0.6] pc3_219vs221_chr8_genes_15 <- pc3_219vs221_chr8$gene[pc3_219vs221_chr8$fold_change > 1.5]

ggplot(pc3_219vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 25) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_LOSS (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 219 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs221_chr8, “pc3_219(LOSS)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_219Lvs221G_chr8 <- pc3_219vs221_chr8

pc3_219Lvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_219vs221_chr8_genes_06)) write.csv(pc3_219Lvs221G_chr8_genes_06, “pc3_219(LOSS)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_219Lvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_219vs221_chr8_genes_15)) write.csv(pc3_219Lvs221G_chr8_genes_15, “pc3_219(LOSS)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

220 (LOSS) vs 221 (gain)

pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_220vs221_chr8 <- intersect(pc3_220_chr8_LOSS$gene, pc3_221_chr8_GAIN$gene)

pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_220vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_220vs221_chr8, ]

pc3_220vs221_chr8 <- merge(pc3_220_chr8_LOSS, pc3_221_chr8_GAIN, by = “gene”)

pc3_220vs221_chr8$fold_change <- pc3_220vs221_chr8$TPM.x / pc3_220vs221_chr8$TPM.y

pc3_220vs221_chr8_genes_06 <- pc3_220vs221_chr8$gene[pc3_220vs221_chr8$fold_change < 0.6] pc3_220vs221_chr8_genes_15 <- pc3_220vs221_chr8$gene[pc3_220vs221_chr8$fold_change > 1.5]

ggplot(pc3_220vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 25) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr8_LOSS (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 220 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs221_chr8, “pc3_220(LOSS)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_220Lvs221G_chr8 <- pc3_220vs221_chr8

pc3_220Lvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_220vs221_chr8_genes_06)) write.csv(pc3_220Lvs221G_chr8_genes_06, “pc3_220(LOSS)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_220Lvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_220vs221_chr8_genes_15)) write.csv(pc3_220Lvs221G_chr8_genes_15, “pc3_220(LOSS)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

222 (LOSS) vs 221 (gain)

pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_222vs221_chr8 <- intersect(pc3_222_chr8_LOSS$gene, pc3_221_chr8_GAIN$gene)

pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_222vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_222vs221_chr8, ]

pc3_222vs221_chr8 <- merge(pc3_222_chr8_LOSS, pc3_221_chr8_GAIN, by = “gene”)

pc3_222vs221_chr8$fold_change <- pc3_222vs221_chr8$TPM.x / pc3_222vs221_chr8$TPM.y

pc3_222vs221_chr8_genes_06 <- pc3_222vs221_chr8$gene[pc3_222vs221_chr8$fold_change < 0.6] pc3_222vs221_chr8_genes_15 <- pc3_222vs221_chr8$gene[pc3_222vs221_chr8$fold_change > 1.5]

ggplot(pc3_222vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 25) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr8_LOSS (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 222 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs221_chr8, “pc3_222(LOSS)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_222Lvs221G_chr8 <- pc3_222vs221_chr8

pc3_222Lvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_222vs221_chr8_genes_06)) write.csv(pc3_222Lvs221G_chr8_genes_06, “pc3_222(LOSS)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_222Lvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_222vs221_chr8_genes_15)) write.csv(pc3_222Lvs221G_chr8_genes_15, “pc3_222(LOSS)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

223 (LOSS) vs 221 (gain)

pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_223vs221_chr8 <- intersect(pc3_223_chr8_LOSS$gene, pc3_221_chr8_GAIN$gene)

pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_223vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_223vs221_chr8, ]

pc3_223vs221_chr8 <- merge(pc3_223_chr8_LOSS, pc3_221_chr8_GAIN, by = “gene”)

pc3_223vs221_chr8$fold_change <- pc3_223vs221_chr8$TPM.x / pc3_223vs221_chr8$TPM.y

pc3_223vs221_chr8_genes_06 <- pc3_223vs221_chr8$gene[pc3_223vs221_chr8$fold_change < 0.6] pc3_223vs221_chr8_genes_15 <- pc3_223vs221_chr8$gene[pc3_223vs221_chr8$fold_change > 1.5]

ggplot(pc3_223vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_223vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 25) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_223_chr8_LOSS (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 223 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_223vs221_chr8, “pc3_223(LOSS)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_223Lvs221G_chr8 <- pc3_223vs221_chr8

pc3_223Lvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_223vs221_chr8_genes_06)) write.csv(pc3_223Lvs221G_chr8_genes_06, “pc3_223(LOSS)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_223Lvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_223vs221_chr8_genes_15)) write.csv(pc3_223Lvs221G_chr8_genes_15, “pc3_223(LOSS)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

224 (LOSS) vs 221 (gain)

pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”) pc3_221_chr8_GAIN <- subset(pc3_221_GAIN, chr == “chr8”)

common_genes_224vs221_chr8 <- intersect(pc3_224_chr8_LOSS$gene, pc3_221_chr8_GAIN$gene)

pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_224vs221_chr8, ] pc3_221_chr8_GAIN <- pc3_221_chr8_GAIN[pc3_221_chr8_GAIN$gene %in% common_genes_224vs221_chr8, ]

pc3_224vs221_chr8 <- merge(pc3_224_chr8_LOSS, pc3_221_chr8_GAIN, by = “gene”)

pc3_224vs221_chr8$fold_change <- pc3_224vs221_chr8$TPM.x / pc3_224vs221_chr8$TPM.y

pc3_224vs221_chr8_genes_06 <- pc3_224vs221_chr8$gene[pc3_224vs221_chr8$fold_change < 0.6] pc3_224vs221_chr8_genes_15 <- pc3_224vs221_chr8$gene[pc3_224vs221_chr8$fold_change > 1.5]

ggplot(pc3_224vs221_chr8, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_224vs221_chr8, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 35) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_224_chr8_LOSS (TPM)”, y = “pc3_221_chr8_GAIN (TPM)”, title = “pc3_chr8: 224 (LOSS) vs 221 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_224vs221_chr8, “pc3_224(LOSS)vs221(GAIN)_chr8.csv”, row.names = FALSE)

pc3_224Lvs221G_chr8 <- pc3_224vs221_chr8

pc3_224Lvs221G_chr8_genes_06 <- data.frame(gene = na.omit(pc3_224vs221_chr8_genes_06)) write.csv(pc3_224Lvs221G_chr8_genes_06, “pc3_224(LOSS)vs221(GAIN)_chr8_genes_06.csv”, row.names = FALSE)

pc3_224Lvs221G_chr8_genes_15 <- data.frame(gene = na.omit(pc3_224vs221_chr8_genes_15)) write.csv(pc3_224Lvs221G_chr8_genes_15, “pc3_224(LOSS)vs221(GAIN)_chr8_genes_15.csv”, row.names = FALSE)

NEUTRALS

217 (neutral) vs 220 (neutral)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_220_chr8_NEUTRAL<- subset(pc3_220_NEUTRAL, chr == “chr8”)

common_genes_217vs220_chr8_NEUTRAL <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_220_chr8_NEUTRAL$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs220_chr8_NEUTRAL, ] pc3_220_chr8_NEUTRAL <- pc3_220_chr8_NEUTRAL[pc3_220_chr8_NEUTRAL$gene %in% common_genes_217vs220_chr8_NEUTRAL, ]

pc3_217vs220_chr8_NEUTRAL <- merge(pc3_217_chr8_NEUTRAL, pc3_220_chr8_NEUTRAL, by = “gene”)

pc3_217vs220_chr8_NEUTRAL$fold_change <- pc3_217vs220_chr8_NEUTRAL$TPM.x / pc3_217vs220_chr8_NEUTRAL$TPM.y

pc3_217vs220_chr8_NEUTRAL_genes_06 <- pc3_217vs220_chr8_NEUTRAL$gene[pc3_217vs220_chr8_NEUTRAL$fold_change < 0.6] pc3_217vs220_chr8_NEUTRAL_genes_15 <- pc3_217vs220_chr8_NEUTRAL$gene[pc3_217vs220_chr8_NEUTRAL$fold_change > 1.5]

ggplot(pc3_217vs220_chr8_NEUTRAL, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr8_NEUTRAL, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_220_chr8_NEUTRAL (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 220 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs220_chr8_NEUTRAL, “pc3_217vs220_NEUTRAL_chr8.csv”, row.names = FALSE)

pc3_217vs220_chr8_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr8_NEUTRAL_genes_06)) write.csv(pc3_217vs220_chr8_NEUTRAL_genes_06, “pc3_217vs220_NEUTRAL_chr8_genes_06.csv”, row.names = FALSE)

pc3_217vs220_chr8_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr8_NEUTRAL_genes_15)) write.csv(pc3_217vs220_chr8_NEUTRAL_genes_15, “pc3_217vs220_NEUTRAL_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 218 (neutral)

pc3_217_chr8_NEUTRAL <- subset(pc3_217_NEUTRAL, chr == “chr8”) pc3_218_chr8_NEUTRAL <- subset(pc3_218_NEUTRAL, chr == “chr8”)

common_genes_217vs218_chr8_NEUTRAL <- intersect(pc3_217_chr8_NEUTRAL$gene, pc3_218_chr8_NEUTRAL$gene)

pc3_217_chr8_NEUTRAL <- pc3_217_chr8_NEUTRAL[pc3_217_chr8_NEUTRAL$gene %in% common_genes_217vs218_chr8_NEUTRAL, ] pc3_218_chr8_NEUTRAL <- pc3_218_chr8_NEUTRAL[pc3_218_chr8_NEUTRAL$gene %in% common_genes_217vs218_chr8_NEUTRAL, ]

pc3_217vs218_chr8_NEUTRAL <- merge(pc3_217_chr8_NEUTRAL, pc3_218_chr8_NEUTRAL, by = “gene”)

pc3_217vs218_chr8_NEUTRAL$fold_change <- pc3_217vs218_chr8_NEUTRAL$TPM.x / pc3_217vs218_chr8_NEUTRAL$TPM.y

pc3_217vs218_chr8_NEUTRAL_genes_06 <- pc3_217vs218_chr8_NEUTRAL$gene[pc3_217vs218_chr8_NEUTRAL$fold_change < 0.6] pc3_217vs218_chr8_NEUTRAL_genes_15 <- pc3_217vs218_chr8_NEUTRAL$gene[pc3_217vs218_chr8_NEUTRAL$fold_change > 1.5]

ggplot(pc3_217vs218_chr8_NEUTRAL, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs218_chr8_NEUTRAL, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr8_NEUTRAL (TPM)”, y = “pc3_218_chr8_NEUTRAL (TPM)”, title = “pc3_chr8: 217 (NEUTRAL) vs 218 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs218_chr8_NEUTRAL, “pc3_217vs218_NEUTRAL_chr8.csv”, row.names = FALSE)

pc3_217vs218_chr8_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_217vs218_chr8_NEUTRAL_genes_06)) write.csv(pc3_217vs218_chr8_NEUTRAL_genes_06, “pc3_217vs218_NEUTRAL_chr8_genes_06.csv”, row.names = FALSE)

pc3_217vs218_chr8_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_217vs218_chr8_NEUTRAL_genes_15)) write.csv(pc3_217vs218_chr8_NEUTRAL_genes_15, “pc3_217vs218_NEUTRAL_chr8_genes_15.csv”, row.names = FALSE)

217 (neutral) vs 219 (neutral) - nothing

217 (neutral) vs 220 (neutral) - nothing

217 (neutral) vs 222 (neutral) - nothing

217 (neutral) vs 223 (neutral) - nothing

217 (neutral) vs 224 (neutral) - nothing

219 (loss) vs 220 (loss)

pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”) pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”)

common_genes_219vs220_chr8_LOSS <- intersect(pc3_219_chr8_LOSS$gene, pc3_220_chr8_LOSS$gene)

pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_219vs220_chr8_LOSS, ] pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_219vs220_chr8_LOSS, ]

pc3_219vs220_chr8_LOSS <- merge(pc3_219_chr8_LOSS, pc3_220_chr8_LOSS, by = “gene”)

pc3_219vs220_chr8_LOSS$fold_change <- pc3_219vs220_chr8_LOSS$TPM.x / pc3_219vs220_chr8_LOSS$TPM.y

pc3_219vs220_chr8_LOSS_genes_06 <- pc3_219vs220_chr8_LOSS$gene[pc3_219vs220_chr8_LOSS$fold_change < 0.6] pc3_219vs220_chr8_LOSS_genes_15 <- pc3_219vs220_chr8_LOSS$gene[pc3_219vs220_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_219vs220_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs220_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_LOSS (TPM)”, y = “pc3_220_chr8_LOSS (TPM)”, title = “pc3_chr8: 219 (LOSS) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs220_chr8_LOSS, “pc3_219vs220_LOSS_chr8.csv”, row.names = FALSE)

pc3_219vs220_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_219vs220_chr8_LOSS_genes_06)) write.csv(pc3_219vs220_chr8_LOSS_genes_06, “pc3_219vs220_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_219vs220_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_219vs220_chr8_LOSS_genes_15)) write.csv(pc3_219vs220_chr8_LOSS_genes_15, “pc3_219vs220_LOSS_chr8_genes_15.csv”, row.names = FALSE)

219 (LOSS) vs 222 (LOSS)

pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”) pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”)

common_genes_219vs222_chr8_LOSS <- intersect(pc3_219_chr8_LOSS$gene, pc3_222_chr8_LOSS$gene)

pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_219vs222_chr8_LOSS, ] pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_219vs222_chr8_LOSS, ]

pc3_219vs222_chr8_LOSS <- merge(pc3_219_chr8_LOSS, pc3_222_chr8_LOSS, by = “gene”)

pc3_219vs222_chr8_LOSS$fold_change <- pc3_219vs222_chr8_LOSS$TPM.x / pc3_219vs222_chr8_LOSS$TPM.y

pc3_219vs222_chr8_LOSS_genes_06 <- pc3_219vs222_chr8_LOSS$gene[pc3_219vs222_chr8_LOSS$fold_change < 0.6] pc3_219vs222_chr8_LOSS_genes_15 <- pc3_219vs222_chr8_LOSS$gene[pc3_219vs222_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_219vs222_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs222_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_LOSS (TPM)”, y = “pc3_222_chr8_LOSS (TPM)”, title = “pc3_chr8: 219 (LOSS) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs222_chr8_LOSS, “pc3_219vs222_LOSS_chr8.csv”, row.names = FALSE)

pc3_219vs222_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_219vs222_chr8_LOSS_genes_06)) write.csv(pc3_219vs222_chr8_LOSS_genes_06, “pc3_219vs222_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_219vs222_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_219vs222_chr8_LOSS_genes_15)) write.csv(pc3_219vs222_chr8_LOSS_genes_15, “pc3_219vs222_LOSS_chr8_genes_15.csv”, row.names = FALSE)

219 (LOSS) vs 223 (LOSS)

pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”) pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”)

common_genes_219vs223_chr8_LOSS <- intersect(pc3_219_chr8_LOSS$gene, pc3_223_chr8_LOSS$gene)

pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_219vs223_chr8_LOSS, ] pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_219vs223_chr8_LOSS, ]

pc3_219vs223_chr8_LOSS <- merge(pc3_219_chr8_LOSS, pc3_223_chr8_LOSS, by = “gene”)

pc3_219vs223_chr8_LOSS$fold_change <- pc3_219vs223_chr8_LOSS$TPM.x / pc3_219vs223_chr8_LOSS$TPM.y

pc3_219vs223_chr8_LOSS_genes_06 <- pc3_219vs223_chr8_LOSS$gene[pc3_219vs223_chr8_LOSS$fold_change < 0.6] pc3_219vs223_chr8_LOSS_genes_15 <- pc3_219vs223_chr8_LOSS$gene[pc3_219vs223_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_219vs223_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs223_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_LOSS (TPM)”, y = “pc3_223_chr8_LOSS (TPM)”, title = “pc3_chr8: 219 (LOSS) vs 223 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs223_chr8_LOSS, “pc3_219vs223_LOSS_chr8.csv”, row.names = FALSE)

pc3_219vs223_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_219vs223_chr8_LOSS_genes_06)) write.csv(pc3_219vs223_chr8_LOSS_genes_06, “pc3_219vs223_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_219vs223_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_219vs223_chr8_LOSS_genes_15)) write.csv(pc3_219vs223_chr8_LOSS_genes_15, “pc3_219vs223_LOSS_chr8_genes_15.csv”, row.names = FALSE)

219 (LOSS) vs 224 (LOSS)

pc3_219_chr8_LOSS <- subset(pc3_219_LOSS, chr == “chr8”) pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”)

common_genes_219vs224_chr8_LOSS <- intersect(pc3_219_chr8_LOSS$gene, pc3_224_chr8_LOSS$gene)

pc3_219_chr8_LOSS <- pc3_219_chr8_LOSS[pc3_219_chr8_LOSS$gene %in% common_genes_219vs224_chr8_LOSS, ] pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_219vs224_chr8_LOSS, ]

pc3_219vs224_chr8_LOSS <- merge(pc3_219_chr8_LOSS, pc3_224_chr8_LOSS, by = “gene”)

pc3_219vs224_chr8_LOSS$fold_change <- pc3_219vs224_chr8_LOSS$TPM.x / pc3_219vs224_chr8_LOSS$TPM.y

pc3_219vs224_chr8_LOSS_genes_06 <- pc3_219vs224_chr8_LOSS$gene[pc3_219vs224_chr8_LOSS$fold_change < 0.6] pc3_219vs224_chr8_LOSS_genes_15 <- pc3_219vs224_chr8_LOSS$gene[pc3_219vs224_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_219vs224_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs224_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr8_LOSS (TPM)”, y = “pc3_224_chr8_LOSS (TPM)”, title = “pc3_chr8: 219 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs224_chr8_LOSS, “pc3_219vs224_LOSS_chr8.csv”, row.names = FALSE)

pc3_219vs224_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_219vs224_chr8_LOSS_genes_06)) write.csv(pc3_219vs224_chr8_LOSS_genes_06, “pc3_219vs224_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_219vs224_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_219vs224_chr8_LOSS_genes_15)) write.csv(pc3_219vs224_chr8_LOSS_genes_15, “pc3_219vs224_LOSS_chr8_genes_15.csv”, row.names = FALSE)

220 (LOSS) vs 222 (LOSS)

pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”) pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”)

common_genes_220vs222_chr8_LOSS <- intersect(pc3_220_chr8_LOSS$gene, pc3_222_chr8_LOSS$gene)

pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_220vs222_chr8_LOSS, ] pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_220vs222_chr8_LOSS, ]

pc3_220vs222_chr8_LOSS <- merge(pc3_220_chr8_LOSS, pc3_222_chr8_LOSS, by = “gene”)

pc3_220vs222_chr8_LOSS$fold_change <- pc3_220vs222_chr8_LOSS$TPM.x / pc3_220vs222_chr8_LOSS$TPM.y

pc3_220vs222_chr8_LOSS_genes_06 <- pc3_220vs222_chr8_LOSS$gene[pc3_220vs222_chr8_LOSS$fold_change < 0.6] pc3_220vs222_chr8_LOSS_genes_15 <- pc3_220vs222_chr8_LOSS$gene[pc3_220vs222_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_220vs222_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs222_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr8_LOSS (TPM)”, y = “pc3_222_chr8_LOSS (TPM)”, title = “pc3_chr8: 220 (LOSS) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs222_chr8_LOSS, “pc3_220vs222_LOSS_chr8.csv”, row.names = FALSE)

pc3_220vs222_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_220vs222_chr8_LOSS_genes_06)) write.csv(pc3_220vs222_chr8_LOSS_genes_06, “pc3_220vs222_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_220vs222_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_220vs222_chr8_LOSS_genes_15)) write.csv(pc3_220vs222_chr8_LOSS_genes_15, “pc3_220vs222_LOSS_chr8_genes_15.csv”, row.names = FALSE)

220 (LOSS) vs 223 (LOSS)

pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”) pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”)

common_genes_220vs223_chr8_LOSS <- intersect(pc3_220_chr8_LOSS$gene, pc3_223_chr8_LOSS$gene)

pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_220vs223_chr8_LOSS, ] pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_220vs223_chr8_LOSS, ]

pc3_220vs223_chr8_LOSS <- merge(pc3_220_chr8_LOSS, pc3_223_chr8_LOSS, by = “gene”)

pc3_220vs223_chr8_LOSS$fold_change <- pc3_220vs223_chr8_LOSS$TPM.x / pc3_220vs223_chr8_LOSS$TPM.y

pc3_220vs223_chr8_LOSS_genes_06 <- pc3_220vs223_chr8_LOSS$gene[pc3_220vs223_chr8_LOSS$fold_change < 0.6] pc3_220vs223_chr8_LOSS_genes_15 <- pc3_220vs223_chr8_LOSS$gene[pc3_220vs223_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_220vs223_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs223_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr8_LOSS (TPM)”, y = “pc3_223_chr8_LOSS (TPM)”, title = “pc3_chr8: 220 (LOSS) vs 223 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs223_chr8_LOSS, “pc3_220vs223_LOSS_chr8.csv”, row.names = FALSE)

pc3_220vs223_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_220vs223_chr8_LOSS_genes_06)) write.csv(pc3_220vs223_chr8_LOSS_genes_06, “pc3_220vs223_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_220vs223_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_220vs223_chr8_LOSS_genes_15)) write.csv(pc3_220vs223_chr8_LOSS_genes_15, “pc3_220vs223_LOSS_chr8_genes_15.csv”, row.names = FALSE)

220 (LOSS) vs 224 (LOSS)

pc3_220_chr8_LOSS <- subset(pc3_220_LOSS, chr == “chr8”) pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”)

common_genes_220vs224_chr8_LOSS <- intersect(pc3_220_chr8_LOSS$gene, pc3_224_chr8_LOSS$gene)

pc3_220_chr8_LOSS <- pc3_220_chr8_LOSS[pc3_220_chr8_LOSS$gene %in% common_genes_220vs224_chr8_LOSS, ] pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_220vs224_chr8_LOSS, ]

pc3_220vs224_chr8_LOSS <- merge(pc3_220_chr8_LOSS, pc3_224_chr8_LOSS, by = “gene”)

pc3_220vs224_chr8_LOSS$fold_change <- pc3_220vs224_chr8_LOSS$TPM.x / pc3_220vs224_chr8_LOSS$TPM.y

pc3_220vs224_chr8_LOSS_genes_06 <- pc3_220vs224_chr8_LOSS$gene[pc3_220vs224_chr8_LOSS$fold_change < 0.6] pc3_220vs224_chr8_LOSS_genes_15 <- pc3_220vs224_chr8_LOSS$gene[pc3_220vs224_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_220vs224_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs224_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr8_LOSS (TPM)”, y = “pc3_224_chr8_LOSS (TPM)”, title = “pc3_chr8: 220 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs224_chr8_LOSS, “pc3_220vs224_LOSS_chr8.csv”, row.names = FALSE)

pc3_220vs224_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_220vs224_chr8_LOSS_genes_06)) write.csv(pc3_220vs224_chr8_LOSS_genes_06, “pc3_220vs224_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_220vs224_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_220vs224_chr8_LOSS_genes_15)) write.csv(pc3_220vs224_chr8_LOSS_genes_15, “pc3_220vs224_LOSS_chr8_genes_15.csv”, row.names = FALSE)

222 (LOSS) vs 223 (LOSS)

pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”) pc3_223_chr8_LOSS <- subset(pc3_223_LOSS, chr == “chr8”)

common_genes_222vs223_chr8_LOSS <- intersect(pc3_222_chr8_LOSS$gene, pc3_223_chr8_LOSS$gene)

pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_222vs223_chr8_LOSS, ] pc3_223_chr8_LOSS <- pc3_223_chr8_LOSS[pc3_223_chr8_LOSS$gene %in% common_genes_222vs223_chr8_LOSS, ]

pc3_222vs223_chr8_LOSS <- merge(pc3_222_chr8_LOSS, pc3_223_chr8_LOSS, by = “gene”)

pc3_222vs223_chr8_LOSS$fold_change <- pc3_222vs223_chr8_LOSS$TPM.x / pc3_222vs223_chr8_LOSS$TPM.y

pc3_222vs223_chr8_LOSS_genes_06 <- pc3_222vs223_chr8_LOSS$gene[pc3_222vs223_chr8_LOSS$fold_change < 0.6] pc3_222vs223_chr8_LOSS_genes_15 <- pc3_222vs223_chr8_LOSS$gene[pc3_222vs223_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_222vs223_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs223_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr8_LOSS (TPM)”, y = “pc3_223_chr8_LOSS (TPM)”, title = “pc3_chr8: 222 (LOSS) vs 223 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs223_chr8_LOSS, “pc3_222vs223_LOSS_chr8.csv”, row.names = FALSE)

pc3_222vs223_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_222vs223_chr8_LOSS_genes_06)) write.csv(pc3_222vs223_chr8_LOSS_genes_06, “pc3_222vs223_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_222vs223_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_222vs223_chr8_LOSS_genes_15)) write.csv(pc3_222vs223_chr8_LOSS_genes_15, “pc3_222vs223_LOSS_chr8_genes_15.csv”, row.names = FALSE)

222 (LOSS) vs 224 (LOSS)

pc3_222_chr8_LOSS <- subset(pc3_222_LOSS, chr == “chr8”) pc3_224_chr8_LOSS <- subset(pc3_224_LOSS, chr == “chr8”)

common_genes_222vs224_chr8_LOSS <- intersect(pc3_222_chr8_LOSS$gene, pc3_224_chr8_LOSS$gene)

pc3_222_chr8_LOSS <- pc3_222_chr8_LOSS[pc3_222_chr8_LOSS$gene %in% common_genes_222vs224_chr8_LOSS, ] pc3_224_chr8_LOSS <- pc3_224_chr8_LOSS[pc3_224_chr8_LOSS$gene %in% common_genes_222vs224_chr8_LOSS, ]

pc3_222vs224_chr8_LOSS <- merge(pc3_222_chr8_LOSS, pc3_224_chr8_LOSS, by = “gene”)

pc3_222vs224_chr8_LOSS$fold_change <- pc3_222vs224_chr8_LOSS$TPM.x / pc3_222vs224_chr8_LOSS$TPM.y

pc3_222vs224_chr8_LOSS_genes_06 <- pc3_222vs224_chr8_LOSS$gene[pc3_222vs224_chr8_LOSS$fold_change < 0.6] pc3_222vs224_chr8_LOSS_genes_15 <- pc3_222vs224_chr8_LOSS$gene[pc3_222vs224_chr8_LOSS$fold_change > 1.5]

ggplot(pc3_222vs224_chr8_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs224_chr8_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr8_LOSS (TPM)”, y = “pc3_224_chr8_LOSS (TPM)”, title = “pc3_chr8: 222 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs224_chr8_LOSS, “pc3_222vs224_LOSS_chr8.csv”, row.names = FALSE)

pc3_222vs224_chr8_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_222vs224_chr8_LOSS_genes_06)) write.csv(pc3_222vs224_chr8_LOSS_genes_06, “pc3_222vs224_LOSS_chr8_genes_06.csv”, row.names = FALSE)

pc3_222vs224_chr8_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_222vs224_chr8_LOSS_genes_15)) write.csv(pc3_222vs224_chr8_LOSS_genes_15, “pc3_222vs224_LOSS_chr8_genes_15.csv”, row.names = FALSE)

CHROMOSOME 16

pc3_217_LOSS <- read.csv(“final_merged_pc3_217_LOSS.csv”) pc3_218_LOSS <- read.csv(“final_merged_pc3_218_LOSS.csv”) pc3_219_NEUTRAL <- read.csv(“final_merged_pc3_219_NEUTRAL.csv”) pc3_220_GAIN <- read.csv(“final_merged_pc3_220_GAIN.csv”) pc3_221_LOSS <- read.csv(“final_merged_pc3_221_LOSS.csv”) pc3_222_LOSS <- read.csv(“final_merged_pc3_222_LOSS.csv”) pc3_223_NEUTRAL <- read.csv(“final_merged_pc3_223_NEUTRAL.csv”) pc3_224_LOSS <- read.csv(“final_merged_pc3_224_LOSS.csv”)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”) pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

217 (loss) vs 219 (neutral)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”)

common_genes_217vs219_chr16 <- intersect(pc3_217_chr16_LOSS$gene, pc3_219_chr16_NEUTRAL$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs219_chr16, ] pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_217vs219_chr16, ]

pc3_217vs219_chr16 <- merge(pc3_217_chr16_LOSS, pc3_219_chr16_NEUTRAL, by = “gene”)

pc3_217vs219_chr16$fold_change <- pc3_217vs219_chr16$TPM.x / pc3_217vs219_chr16$TPM.y

pc3_217vs219_chr16_genes_06 <- pc3_217vs219_chr16$gene[pc3_217vs219_chr16$fold_change < 0.6] pc3_217vs219_chr16_genes_15 <- pc3_217vs219_chr16$gene[pc3_217vs219_chr16$fold_change > 1.5]

ggplot(pc3_217vs219_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs219_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 200) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_219_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 219 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs219_chr16, “pc3_217(LOSS)vs219(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_217Lvs219N_chr16 <- pc3_217vs219_chr16

pc3_217Lvs219N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_217vs219_chr16_genes_06)) write.csv(pc3_217Lvs219N_chr16_genes_06, “pc3_217(LOSS)vs219(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_217Lvs219N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_217vs219_chr16_genes_15)) write.csv(pc3_217Lvs219N_chr16_genes_15, “pc3_217(LOSS)vs219(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

217 (loss) vs 220 (gain)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”)

common_genes_217vs220_chr16 <- intersect(pc3_217_chr16_LOSS$gene, pc3_220_chr16_GAIN$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs220_chr16, ] pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_217vs220_chr16, ]

pc3_217vs220_chr16 <- merge(pc3_217_chr16_LOSS, pc3_220_chr16_GAIN, by = “gene”)

pc3_217vs220_chr16$fold_change <- pc3_217vs220_chr16$TPM.x / pc3_217vs220_chr16$TPM.y

pc3_217vs220_chr16_genes_06 <- pc3_217vs220_chr16$gene[pc3_217vs220_chr16$fold_change < 0.6] pc3_217vs220_chr16_genes_15 <- pc3_217vs220_chr16$gene[pc3_217vs220_chr16$fold_change > 1.5]

ggplot(pc3_217vs220_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_220_chr16_GAIN (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 220 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs220_chr16, “pc3_217(LOSS)vs220(GAIN)_chr16.csv”, row.names = FALSE)

pc3_217Lvs220G_chr16 <- pc3_217vs220_chr16

pc3_217Lvs220G_chr16_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr16_genes_06)) write.csv(pc3_217Lvs220G_chr16_genes_06, “pc3_217(LOSS)vs220(GAIN)_chr16_genes_06.csv”, row.names = FALSE)

pc3_217Lvs220G_chr16_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr16_genes_15)) write.csv(pc3_217Lvs220G_chr16_genes_15, “pc3_217(LOSS)vs220(GAIN)_chr16_genes_15.csv”, row.names = FALSE)

217 (loss) vs 223 (neutral)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_217vs223_chr16 <- intersect(pc3_217_chr16_LOSS$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs223_chr16, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_217vs223_chr16, ]

pc3_217vs223_chr16 <- merge(pc3_217_chr16_LOSS, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_217vs223_chr16$fold_change <- pc3_217vs223_chr16$TPM.x / pc3_217vs223_chr16$TPM.y

pc3_217vs223_chr16_genes_06 <- pc3_217vs223_chr16$gene[pc3_217vs223_chr16$fold_change < 0.6] pc3_217vs223_chr16_genes_15 <- pc3_217vs223_chr16$gene[pc3_217vs223_chr16$fold_change > 1.5]

ggplot(pc3_217vs223_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs223_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 15) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs223_chr16, “pc3_217(LOSS)vs223(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_217Lvs223N_chr16 <- pc3_217vs223_chr16

pc3_217Lvs223N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_217vs223_chr16_genes_06)) write.csv(pc3_217Lvs223N_chr16_genes_06, “pc3_217(LOSS)vs223(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_217Lvs223N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_217vs223_chr16_genes_15)) write.csv(pc3_217Lvs223N_chr16_genes_15, “pc3_217(LOSS)vs223(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

218 (loss) vs 219 (neutral)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”)

common_genes_218vs219_chr16 <- intersect(pc3_218_chr16_LOSS$gene, pc3_219_chr16_NEUTRAL$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs219_chr16, ] pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_218vs219_chr16, ]

pc3_218vs219_chr16 <- merge(pc3_218_chr16_LOSS, pc3_219_chr16_NEUTRAL, by = “gene”)

pc3_218vs219_chr16$fold_change <- pc3_218vs219_chr16$TPM.x / pc3_218vs219_chr16$TPM.y

pc3_218vs219_chr16_genes_06 <- pc3_218vs219_chr16$gene[pc3_218vs219_chr16$fold_change < 0.6] pc3_218vs219_chr16_genes_15 <- pc3_218vs219_chr16$gene[pc3_218vs219_chr16$fold_change > 1.5]

ggplot(pc3_218vs219_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs219_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 200) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_219_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 219 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs219_chr16, “pc3_218(LOSS)vs219(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_218Lvs219N_chr16 <- pc3_218vs219_chr16

pc3_218Lvs219N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_218vs219_chr16_genes_06)) write.csv(pc3_218Lvs219N_chr16_genes_06, “pc3_218(LOSS)vs219(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_218Lvs219N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_218vs219_chr16_genes_15)) write.csv(pc3_218Lvs219N_chr16_genes_15, “pc3_218(LOSS)vs219(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

218 (loss) vs 220 (gain)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”)

common_genes_218vs220_chr16 <- intersect(pc3_218_chr16_LOSS$gene, pc3_220_chr16_GAIN$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs220_chr16, ] pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_218vs220_chr16, ]

pc3_218vs220_chr16 <- merge(pc3_218_chr16_LOSS, pc3_220_chr16_GAIN, by = “gene”)

pc3_218vs220_chr16$fold_change <- pc3_218vs220_chr16$TPM.x / pc3_218vs220_chr16$TPM.y

pc3_218vs220_chr16_genes_06 <- pc3_218vs220_chr16$gene[pc3_218vs220_chr16$fold_change < 0.6] pc3_218vs220_chr16_genes_15 <- pc3_218vs220_chr16$gene[pc3_218vs220_chr16$fold_change > 1.5]

ggplot(pc3_218vs220_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs220_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_220_chr16_GAIN (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 220 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs220_chr16, “pc3_218(LOSS)vs220(GAIN)_chr16.csv”, row.names = FALSE)

pc3_218Lvs220G_chr16 <- pc3_218vs220_chr16

pc3_218Lvs220G_chr16_genes_06 <- data.frame(gene = na.omit(pc3_218vs220_chr16_genes_06)) write.csv(pc3_218Lvs220G_chr16_genes_06, “pc3_218(LOSS)vs220(GAIN)_chr16_genes_06.csv”, row.names = FALSE)

pc3_218Lvs220G_chr16_genes_15 <- data.frame(gene = na.omit(pc3_218vs220_chr16_genes_15)) write.csv(pc3_218Lvs220G_chr16_genes_15, “pc3_218(LOSS)vs220(GAIN)_chr16_genes_15.csv”, row.names = FALSE)

218 (loss) vs 223 (neutral)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_218vs223_chr16 <- intersect(pc3_218_chr16_LOSS$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs223_chr16, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_218vs223_chr16, ]

pc3_218vs223_chr16 <- merge(pc3_218_chr16_LOSS, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_218vs223_chr16$fold_change <- pc3_218vs223_chr16$TPM.x / pc3_218vs223_chr16$TPM.y

pc3_218vs223_chr16_genes_06 <- pc3_218vs223_chr16$gene[pc3_218vs223_chr16$fold_change < 0.6] pc3_218vs223_chr16_genes_15 <- pc3_218vs223_chr16$gene[pc3_218vs223_chr16$fold_change > 1.5]

ggplot(pc3_218vs223_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs223_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs223_chr16, “pc3_218(LOSS)vs223(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_218Lvs223N_chr16 <- pc3_218vs223_chr16

pc3_218Lvs223N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_218vs223_chr16_genes_06)) write.csv(pc3_218Lvs223N_chr16_genes_06, “pc3_218(LOSS)vs223(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_218Lvs223N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_218vs223_chr16_genes_15)) write.csv(pc3_218Lvs223N_chr16_genes_15, “pc3_218(LOSS)vs223(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 220 (gain)

pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”)

common_genes_219vs220_chr16 <- intersect(pc3_219_chr16_NEUTRAL$gene, pc3_220_chr16_GAIN$gene)

pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_219vs220_chr16, ] pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_219vs220_chr16, ]

pc3_219vs220_chr16 <- merge(pc3_219_chr16_NEUTRAL, pc3_220_chr16_GAIN, by = “gene”)

pc3_219vs220_chr16$fold_change <- pc3_219vs220_chr16$TPM.x / pc3_219vs220_chr16$TPM.y

pc3_219vs220_chr16_genes_06 <- pc3_219vs220_chr16$gene[pc3_219vs220_chr16$fold_change < 0.6] pc3_219vs220_chr16_genes_15 <- pc3_219vs220_chr16$gene[pc3_219vs220_chr16$fold_change > 1.5]

ggplot(pc3_219vs220_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs220_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 200) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr16_NEUTRAL (TPM)”, y = “pc3_220_chr16_GAIN (TPM)”, title = “pc3_chr16: 219 (NEUTRAL) vs 220 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs220_chr16, “pc3_219(NEUTRAL)vs220(GAIN)_chr16.csv”, row.names = FALSE)

pc3_219Nvs220G_chr16 <- pc3_219vs220_chr16

pc3_219Nvs220G_chr16_genes_06 <- data.frame(gene = na.omit(pc3_219vs220_chr16_genes_06)) write.csv(pc3_219Nvs220G_chr16_genes_06, “pc3_219(NEUTRAL)vs220(GAIN)_chr16_genes_06.csv”, row.names = FALSE)

pc3_219Nvs220G_chr16_genes_15 <- data.frame(gene = na.omit(pc3_219vs220_chr16_genes_15)) write.csv(pc3_219Nvs220G_chr16_genes_15, “pc3_219(NEUTRAL)vs220(GAIN)_chr16_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 221 (loss)

pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”)

common_genes_219vs221_chr16 <- intersect(pc3_219_chr16_NEUTRAL$gene, pc3_221_chr16_LOSS$gene)

pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_219vs221_chr16, ] pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_219vs221_chr16, ]

pc3_219vs221_chr16 <- merge(pc3_219_chr16_NEUTRAL, pc3_221_chr16_LOSS, by = “gene”)

pc3_219vs221_chr16$fold_change <- pc3_219vs221_chr16$TPM.x / pc3_219vs221_chr16$TPM.y

pc3_219vs221_chr16_genes_06 <- pc3_219vs221_chr16$gene[pc3_219vs221_chr16$fold_change < 0.6] pc3_219vs221_chr16_genes_15 <- pc3_219vs221_chr16$gene[pc3_219vs221_chr16$fold_change > 1.5]

ggplot(pc3_219vs221_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs221_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 20) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr16_NEUTRAL (TPM)”, y = “pc3_221_chr16_LOSS (TPM)”, title = “pc3_chr16: 219 (NEUTRAL) vs 221 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs221_chr16, “pc3_219(NEUTRAL)vs221(LOSS)_chr16.csv”, row.names = FALSE)

pc3_219Nvs221L_chr16 <- pc3_219vs221_chr16

pc3_219Nvs221L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_219vs221_chr16_genes_06)) write.csv(pc3_219Nvs221L_chr16_genes_06, “pc3_219(NEUTRAL)vs221(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_219Nvs221L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_219vs221_chr16_genes_15)) write.csv(pc3_219Nvs221L_chr16_genes_15, “pc3_219(NEUTRAL)vs221(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 222 (loss)

pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”)

common_genes_219vs222_chr16 <- intersect(pc3_219_chr16_NEUTRAL$gene, pc3_222_chr16_LOSS$gene)

pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_219vs222_chr16, ] pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_219vs222_chr16, ]

pc3_219vs222_chr16 <- merge(pc3_219_chr16_NEUTRAL, pc3_222_chr16_LOSS, by = “gene”)

pc3_219vs222_chr16$fold_change <- pc3_219vs222_chr16$TPM.x / pc3_219vs222_chr16$TPM.y

pc3_219vs222_chr16_genes_06 <- pc3_219vs222_chr16$gene[pc3_219vs222_chr16$fold_change < 0.6] pc3_219vs222_chr16_genes_15 <- pc3_219vs222_chr16$gene[pc3_219vs222_chr16$fold_change > 1.5]

ggplot(pc3_219vs222_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs222_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 40) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr16_NEUTRAL (TPM)”, y = “pc3_222_chr16_LOSS (TPM)”, title = “pc3_chr16: 219 (NEUTRAL) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs222_chr16, “pc3_219(NEUTRAL)vs222(LOSS)_chr16.csv”, row.names = FALSE)

pc3_219Nvs222L_chr16 <- pc3_219vs222_chr16

pc3_219Nvs222L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_219vs222_chr16_genes_06)) write.csv(pc3_219Nvs222L_chr16_genes_06, “pc3_219(NEUTRAL)vs222(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_219Nvs222L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_219vs222_chr16_genes_15)) write.csv(pc3_219Nvs222L_chr16_genes_15, “pc3_219(NEUTRAL)vs222(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

219 (neutral) vs 224 (loss)

pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_219vs224_chr16 <- intersect(pc3_219_chr16_NEUTRAL$gene, pc3_224_chr16_LOSS$gene)

pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_219vs224_chr16, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_219vs224_chr16, ]

pc3_219vs224_chr16 <- merge(pc3_219_chr16_NEUTRAL, pc3_224_chr16_LOSS, by = “gene”)

pc3_219vs224_chr16$fold_change <- pc3_219vs224_chr16$TPM.x / pc3_219vs224_chr16$TPM.y

pc3_219vs224_chr16_genes_06 <- pc3_219vs224_chr16$gene[pc3_219vs224_chr16$fold_change < 0.6] pc3_219vs224_chr16_genes_15 <- pc3_219vs224_chr16$gene[pc3_219vs224_chr16$fold_change > 1.5]

ggplot(pc3_219vs224_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs224_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr16_NEUTRAL (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 219 (NEUTRAL) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs224_chr16, “pc3_219(NEUTRAL)vs224(LOSS)_chr16.csv”, row.names = FALSE)

pc3_219Nvs224L_chr16 <- pc3_219vs224_chr16

pc3_219Nvs224L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_219vs224_chr16_genes_06)) write.csv(pc3_219Nvs224L_chr16_genes_06, “pc3_219(NEUTRAL)vs224(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_219Nvs224L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_219vs224_chr16_genes_15)) write.csv(pc3_219Nvs224L_chr16_genes_15, “pc3_219(NEUTRAL)vs224(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

220 (gain) vs 221 (loss)

pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”) pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”)

common_genes_220vs221_chr16 <- intersect(pc3_220_chr16_GAIN$gene, pc3_221_chr16_LOSS$gene)

pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_220vs221_chr16, ] pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_220vs221_chr16, ]

pc3_220vs221_chr16 <- merge(pc3_220_chr16_GAIN, pc3_221_chr16_LOSS, by = “gene”)

pc3_220vs221_chr16$fold_change <- pc3_220vs221_chr16$TPM.x / pc3_220vs221_chr16$TPM.y

pc3_220vs221_chr16_genes_06 <- pc3_220vs221_chr16$gene[pc3_220vs221_chr16$fold_change < 0.6] pc3_220vs221_chr16_genes_15 <- pc3_220vs221_chr16$gene[pc3_220vs221_chr16$fold_change > 1.5]

ggplot(pc3_220vs221_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs221_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr16_GAIN (TPM)”, y = “pc3_221_chr16_LOSS (TPM)”, title = “pc3_chr16: 220 (GAIN) vs 221 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs221_chr16, “pc3_220(GAIN)vs221(LOSS)_chr16.csv”, row.names = FALSE)

pc3_220Gvs221L_chr16 <- pc3_220vs221_chr16

pc3_220Gvs221L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_220vs221_chr16_genes_06)) write.csv(pc3_220Gvs221L_chr16_genes_06, “pc3_220(GAIN)vs221(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_220Gvs221L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_220vs221_chr16_genes_15)) write.csv(pc3_220Gvs221L_chr16_genes_15, “pc3_220(GAIN)vs221(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

220 (gain) vs 222 (loss)

pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”)

common_genes_220vs222_chr16 <- intersect(pc3_220_chr16_GAIN$gene, pc3_222_chr16_LOSS$gene)

pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_220vs222_chr16, ] pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_220vs222_chr16, ]

pc3_220vs222_chr16 <- merge(pc3_220_chr16_GAIN, pc3_222_chr16_LOSS, by = “gene”)

pc3_220vs222_chr16$fold_change <- pc3_220vs222_chr16$TPM.x / pc3_220vs222_chr16$TPM.y

pc3_220vs222_chr16_genes_06 <- pc3_220vs222_chr16$gene[pc3_220vs222_chr16$fold_change < 0.6] pc3_220vs222_chr16_genes_15 <- pc3_220vs222_chr16$gene[pc3_220vs222_chr16$fold_change > 1.5]

ggplot(pc3_220vs222_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs222_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr16_GAIN (TPM)”, y = “pc3_222_chr16_LOSS (TPM)”, title = “pc3_chr16: 220 (GAIN) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs222_chr16, “pc3_220(GAIN)vs222(LOSS)_chr16.csv”, row.names = FALSE)

pc3_220Gvs222L_chr16 <- pc3_220vs222_chr16

pc3_220Gvs222L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_220vs222_chr16_genes_06)) write.csv(pc3_220Gvs222L_chr16_genes_06, “pc3_220(GAIN)vs222(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_220Gvs222L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_220vs222_chr16_genes_15)) write.csv(pc3_220Gvs222L_chr16_genes_15, “pc3_220(GAIN)vs222(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

220 (gain) vs 223 (neutral)

pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_220vs223_chr16 <- intersect(pc3_220_chr16_GAIN$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_220vs223_chr16, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_220vs223_chr16, ]

pc3_220vs223_chr16 <- merge(pc3_220_chr16_GAIN, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_220vs223_chr16$fold_change <- pc3_220vs223_chr16$TPM.x / pc3_220vs223_chr16$TPM.y

pc3_220vs223_chr16_genes_06 <- pc3_220vs223_chr16$gene[pc3_220vs223_chr16$fold_change < 0.6] pc3_220vs223_chr16_genes_15 <- pc3_220vs223_chr16$gene[pc3_220vs223_chr16$fold_change > 1.5]

ggplot(pc3_220vs223_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs223_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr16_GAIN (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 220 (GAIN) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs223_chr16, “pc3_220(GAIN)vs223(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_220Gvs223N_chr16 <- pc3_220vs223_chr16

pc3_220Gvs223N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_220vs223_chr16_genes_06)) write.csv(pc3_220Gvs223N_chr16_genes_06, “pc3_220(GAIN)vs223(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_220Gvs223N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_220vs223_chr16_genes_15)) write.csv(pc3_220Gvs223N_chr16_genes_15, “pc3_220(GAIN)vs223(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

220 (gain) vs 224 (loss)

pc3_220_chr16_GAIN <- subset(pc3_220_GAIN, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_220vs224_chr16 <- intersect(pc3_220_chr16_GAIN$gene, pc3_224_chr16_LOSS$gene)

pc3_220_chr16_GAIN <- pc3_220_chr16_GAIN[pc3_220_chr16_GAIN$gene %in% common_genes_220vs224_chr16, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_220vs224_chr16, ]

pc3_220vs224_chr16 <- merge(pc3_220_chr16_GAIN, pc3_224_chr16_LOSS, by = “gene”)

pc3_220vs224_chr16$fold_change <- pc3_220vs224_chr16$TPM.x / pc3_220vs224_chr16$TPM.y

pc3_220vs224_chr16_genes_06 <- pc3_220vs224_chr16$gene[pc3_220vs224_chr16$fold_change < 0.6] pc3_220vs224_chr16_genes_15 <- pc3_220vs224_chr16$gene[pc3_220vs224_chr16$fold_change > 1.5]

ggplot(pc3_220vs224_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_220vs224_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_220_chr16_GAIN (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 220 (GAIN) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_220vs224_chr16, “pc3_220(GAIN)vs224(LOSS)_chr16.csv”, row.names = FALSE)

pc3_220Gvs224L_chr16 <- pc3_220vs224_chr16

pc3_220Gvs224L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_220vs224_chr16_genes_06)) write.csv(pc3_220Gvs224L_chr16_genes_06, “pc3_220(GAIN)vs224(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_220Gvs224L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_220vs224_chr16_genes_15)) write.csv(pc3_220Gvs224L_chr16_genes_15, “pc3_220(GAIN)vs224(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

221 (loss) vs 223 (neutral)

pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_221vs223_chr16 <- intersect(pc3_221_chr16_LOSS$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_221vs223_chr16, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_221vs223_chr16, ]

pc3_221vs223_chr16 <- merge(pc3_221_chr16_LOSS, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_221vs223_chr16$fold_change <- pc3_221vs223_chr16$TPM.x / pc3_221vs223_chr16$TPM.y

pc3_221vs223_chr16_genes_06 <- pc3_221vs223_chr16$gene[pc3_221vs223_chr16$fold_change < 0.6] pc3_221vs223_chr16_genes_15 <- pc3_221vs223_chr16$gene[pc3_221vs223_chr16$fold_change > 1.5]

ggplot(pc3_221vs223_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs223_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 10) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr16_LOSS (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 221 (LOSS) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs223_chr16, “pc3_221(LOSS)vs223(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_221Lvs223N_chr16 <- pc3_221vs223_chr16

pc3_221Lvs223N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_221vs223_chr16_genes_06)) write.csv(pc3_221Lvs223N_chr16_genes_06, “pc3_221(LOSS)vs223(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_221Lvs223N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_221vs223_chr16_genes_15)) write.csv(pc3_221Lvs223N_chr16_genes_15, “pc3_221(LOSS)vs223(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

222 (loss) vs 223 (neutral)

pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_222vs223_chr16 <- intersect(pc3_222_chr16_LOSS$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_222vs223_chr16, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_222vs223_chr16, ]

pc3_222vs223_chr16 <- merge(pc3_222_chr16_LOSS, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_222vs223_chr16$fold_change <- pc3_222vs223_chr16$TPM.x / pc3_222vs223_chr16$TPM.y

pc3_222vs223_chr16_genes_06 <- pc3_222vs223_chr16$gene[pc3_222vs223_chr16$fold_change < 0.6] pc3_222vs223_chr16_genes_15 <- pc3_222vs223_chr16$gene[pc3_222vs223_chr16$fold_change > 1.5]

ggplot(pc3_222vs223_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs223_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 15) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr16_LOSS (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 222 (LOSS) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs223_chr16, “pc3_222(LOSS)vs223(NEUTRAL)_chr16.csv”, row.names = FALSE)

pc3_222Lvs223N_chr16 <- pc3_222vs223_chr16

pc3_222Lvs223N_chr16_genes_06 <- data.frame(gene = na.omit(pc3_222vs223_chr16_genes_06)) write.csv(pc3_222Lvs223N_chr16_genes_06, “pc3_222(LOSS)vs223(NEUTRAL)_chr16_genes_06.csv”, row.names = FALSE)

pc3_222Lvs223N_chr16_genes_15 <- data.frame(gene = na.omit(pc3_222vs223_chr16_genes_15)) write.csv(pc3_222Lvs223N_chr16_genes_15, “pc3_222(LOSS)vs223(NEUTRAL)_chr16_genes_15.csv”, row.names = FALSE)

223 (neutral) vs 224 (loss)

pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_223vs224_chr16 <- intersect(pc3_223_chr16_NEUTRAL$gene, pc3_224_chr16_LOSS$gene)

pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_223vs224_chr16, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_223vs224_chr16, ]

pc3_223vs224_chr16 <- merge(pc3_223_chr16_NEUTRAL, pc3_224_chr16_LOSS, by = “gene”)

pc3_223vs224_chr16$fold_change <- pc3_223vs224_chr16$TPM.x / pc3_223vs224_chr16$TPM.y

pc3_223vs224_chr16_genes_06 <- pc3_223vs224_chr16$gene[pc3_223vs224_chr16$fold_change < 0.6] pc3_223vs224_chr16_genes_15 <- pc3_223vs224_chr16$gene[pc3_223vs224_chr16$fold_change > 1.5]

ggplot(pc3_223vs224_chr16, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_223vs224_chr16, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5, max.overlaps = 15) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_223_chr16_NEUTRAL (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 223 (NEUTRAL) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_223vs224_chr16, “pc3_223(NEUTRAL)vs224(LOSS)_chr16.csv”, row.names = FALSE)

pc3_223Nvs224L_chr16 <- pc3_223vs224_chr16

pc3_223Nvs224L_chr16_genes_06 <- data.frame(gene = na.omit(pc3_223vs224_chr16_genes_06)) write.csv(pc3_223Nvs224L_chr16_genes_06, “pc3_223(NEUTRAL)vs224(LOSS)_chr16_genes_06.csv”, row.names = FALSE)

pc3_223Nvs224L_chr16_genes_15 <- data.frame(gene = na.omit(pc3_223vs224_chr16_genes_15)) write.csv(pc3_223Nvs224L_chr16_genes_15, “pc3_223(NEUTRAL)vs224(LOSS)_chr16_genes_15.csv”, row.names = FALSE)

FOLD CHANGE GENES

LOSS

217 (LOSS) vs 218 (LOSS)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”)

common_genes_217vs218_chr16_LOSS <- intersect(pc3_217_chr16_LOSS$gene, pc3_218_chr16_LOSS$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs218_chr16_LOSS, ] pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_217vs218_chr16_LOSS, ]

pc3_217vs218_chr16_LOSS <- merge(pc3_217_chr16_LOSS, pc3_218_chr16_LOSS, by = “gene”)

pc3_217vs218_chr16_LOSS$fold_change <- pc3_217vs218_chr16_LOSS$TPM.x / pc3_217vs218_chr16_LOSS$TPM.y

pc3_217vs218_chr16_LOSS_genes_06 <- pc3_217vs218_chr16_LOSS$gene[pc3_217vs218_chr16_LOSS$fold_change < 0.6] pc3_217vs218_chr16_LOSS_genes_15 <- pc3_217vs218_chr16_LOSS$gene[pc3_217vs218_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_217vs218_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs218_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_218_chr16_LOSS (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 218 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs218_chr16_LOSS, “pc3_217vs218_LOSS_chr16.csv”, row.names = FALSE)

pc3_217vs218_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_217vs218_chr16_LOSS_genes_06)) write.csv(pc3_217vs218_chr16_LOSS_genes_06, “pc3_217vs218_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_217vs218_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_217vs218_chr16_LOSS_genes_15)) write.csv(pc3_217vs218_chr16_LOSS_genes_15, “pc3_217vs218_LOSS_chr16_genes_15.csv”, row.names = FALSE)

217 (LOSS) vs 221 (LOSS)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”)

common_genes_217vs221_chr16_LOSS <- intersect(pc3_217_chr16_LOSS$gene, pc3_221_chr16_LOSS$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs221_chr16_LOSS, ] pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_217vs221_chr16_LOSS, ]

pc3_217vs221_chr16_LOSS <- merge(pc3_217_chr16_LOSS, pc3_221_chr16_LOSS, by = “gene”)

pc3_217vs221_chr16_LOSS$fold_change <- pc3_217vs221_chr16_LOSS$TPM.x / pc3_217vs221_chr16_LOSS$TPM.y

pc3_217vs221_chr16_LOSS_genes_06 <- pc3_217vs221_chr16_LOSS$gene[pc3_217vs221_chr16_LOSS$fold_change < 0.6] pc3_217vs221_chr16_LOSS_genes_15 <- pc3_217vs221_chr16_LOSS$gene[pc3_217vs221_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_217vs221_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs221_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_221_chr16_LOSS (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 221 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs221_chr16_LOSS, “pc3_217vs221_LOSS_chr16.csv”, row.names = FALSE)

pc3_217vs221_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_217vs221_chr16_LOSS_genes_06)) write.csv(pc3_217vs221_chr16_LOSS_genes_06, “pc3_217vs221_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_217vs221_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_217vs221_chr16_LOSS_genes_15)) write.csv(pc3_217vs221_chr16_LOSS_genes_15, “pc3_217vs221_LOSS_chr16_genes_15.csv”, row.names = FALSE)

217 (LOSS) vs 222 (LOSS)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”)

common_genes_217vs222_chr16_LOSS <- intersect(pc3_217_chr16_LOSS$gene, pc3_222_chr16_LOSS$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs222_chr16_LOSS, ] pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_217vs222_chr16_LOSS, ]

pc3_217vs222_chr16_LOSS <- merge(pc3_217_chr16_LOSS, pc3_222_chr16_LOSS, by = “gene”)

pc3_217vs222_chr16_LOSS$fold_change <- pc3_217vs222_chr16_LOSS$TPM.x / pc3_217vs222_chr16_LOSS$TPM.y

pc3_217vs222_chr16_LOSS_genes_06 <- pc3_217vs222_chr16_LOSS$gene[pc3_217vs222_chr16_LOSS$fold_change < 0.6] pc3_217vs222_chr16_LOSS_genes_15 <- pc3_217vs222_chr16_LOSS$gene[pc3_217vs222_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_217vs222_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs222_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_222_chr16_LOSS (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs222_chr16_LOSS, “pc3_217vs222_LOSS_chr16.csv”, row.names = FALSE)

pc3_217vs222_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_217vs222_chr16_LOSS_genes_06)) write.csv(pc3_217vs222_chr16_LOSS_genes_06, “pc3_217vs222_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_217vs222_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_217vs222_chr16_LOSS_genes_15)) write.csv(pc3_217vs222_chr16_LOSS_genes_15, “pc3_217vs222_LOSS_chr16_genes_15.csv”, row.names = FALSE)

217 (LOSS) vs 224 (LOSS)

pc3_217_chr16_LOSS <- subset(pc3_217_LOSS, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_217vs224_chr16_LOSS <- intersect(pc3_217_chr16_LOSS$gene, pc3_224_chr16_LOSS$gene)

pc3_217_chr16_LOSS <- pc3_217_chr16_LOSS[pc3_217_chr16_LOSS$gene %in% common_genes_217vs224_chr16_LOSS, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_217vs224_chr16_LOSS, ]

pc3_217vs224_chr16_LOSS <- merge(pc3_217_chr16_LOSS, pc3_224_chr16_LOSS, by = “gene”)

pc3_217vs224_chr16_LOSS$fold_change <- pc3_217vs224_chr16_LOSS$TPM.x / pc3_217vs224_chr16_LOSS$TPM.y

pc3_217vs224_chr16_LOSS_genes_06 <- pc3_217vs224_chr16_LOSS$gene[pc3_217vs224_chr16_LOSS$fold_change < 0.6] pc3_217vs224_chr16_LOSS_genes_15 <- pc3_217vs224_chr16_LOSS$gene[pc3_217vs224_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_217vs224_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs224_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr16_LOSS (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 217 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_217vs224_chr16_LOSS, “pc3_217vs224_LOSS_chr16.csv”, row.names = FALSE)

pc3_217vs224_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_217vs224_chr16_LOSS_genes_06)) write.csv(pc3_217vs224_chr16_LOSS_genes_06, “pc3_217vs224_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_217vs224_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_217vs224_chr16_LOSS_genes_15)) write.csv(pc3_217vs224_chr16_LOSS_genes_15, “pc3_217vs224_LOSS_chr16_genes_15.csv”, row.names = FALSE)

218 (LOSS) vs 221 (LOSS)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”)

common_genes_218vs221_chr16_LOSS <- intersect(pc3_218_chr16_LOSS$gene, pc3_221_chr16_LOSS$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs221_chr16_LOSS, ] pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_218vs221_chr16_LOSS, ]

pc3_218vs221_chr16_LOSS <- merge(pc3_218_chr16_LOSS, pc3_221_chr16_LOSS, by = “gene”)

pc3_218vs221_chr16_LOSS$fold_change <- pc3_218vs221_chr16_LOSS$TPM.x / pc3_218vs221_chr16_LOSS$TPM.y

pc3_218vs221_chr16_LOSS_genes_06 <- pc3_218vs221_chr16_LOSS$gene[pc3_218vs221_chr16_LOSS$fold_change < 0.6] pc3_218vs221_chr16_LOSS_genes_15 <- pc3_218vs221_chr16_LOSS$gene[pc3_218vs221_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_218vs221_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs221_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_221_chr16_LOSS (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 221 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs221_chr16_LOSS, “pc3_218vs221_LOSS_chr16.csv”, row.names = FALSE)

pc3_218vs221_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_218vs221_chr16_LOSS_genes_06)) write.csv(pc3_218vs221_chr16_LOSS_genes_06, “pc3_218vs221_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_218vs221_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_218vs221_chr16_LOSS_genes_15)) write.csv(pc3_218vs221_chr16_LOSS_genes_15, “pc3_218vs221_LOSS_chr16_genes_15.csv”, row.names = FALSE)

218 (LOSS) vs 222 (LOSS)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”)

common_genes_218vs222_chr16_LOSS <- intersect(pc3_218_chr16_LOSS$gene, pc3_222_chr16_LOSS$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs222_chr16_LOSS, ] pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_218vs222_chr16_LOSS, ]

pc3_218vs222_chr16_LOSS <- merge(pc3_218_chr16_LOSS, pc3_222_chr16_LOSS, by = “gene”)

pc3_218vs222_chr16_LOSS$fold_change <- pc3_218vs222_chr16_LOSS$TPM.x / pc3_218vs222_chr16_LOSS$TPM.y

pc3_218vs222_chr16_LOSS_genes_06 <- pc3_218vs222_chr16_LOSS$gene[pc3_218vs222_chr16_LOSS$fold_change < 0.6] pc3_218vs222_chr16_LOSS_genes_15 <- pc3_218vs222_chr16_LOSS$gene[pc3_218vs222_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_218vs222_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs222_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_222_chr16_LOSS (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs222_chr16_LOSS, “pc3_218vs222_LOSS_chr16.csv”, row.names = FALSE)

pc3_218vs222_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_218vs222_chr16_LOSS_genes_06)) write.csv(pc3_218vs222_chr16_LOSS_genes_06, “pc3_218vs222_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_218vs222_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_218vs222_chr16_LOSS_genes_15)) write.csv(pc3_218vs222_chr16_LOSS_genes_15, “pc3_218vs222_LOSS_chr16_genes_15.csv”, row.names = FALSE)

218 (LOSS) vs 224 (LOSS)

pc3_218_chr16_LOSS <- subset(pc3_218_LOSS, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_218vs224_chr16_LOSS <- intersect(pc3_218_chr16_LOSS$gene, pc3_224_chr16_LOSS$gene)

pc3_218_chr16_LOSS <- pc3_218_chr16_LOSS[pc3_218_chr16_LOSS$gene %in% common_genes_218vs224_chr16_LOSS, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_218vs224_chr16_LOSS, ]

pc3_218vs224_chr16_LOSS <- merge(pc3_218_chr16_LOSS, pc3_224_chr16_LOSS, by = “gene”)

pc3_218vs224_chr16_LOSS$fold_change <- pc3_218vs224_chr16_LOSS$TPM.x / pc3_218vs224_chr16_LOSS$TPM.y

pc3_218vs224_chr16_LOSS_genes_06 <- pc3_218vs224_chr16_LOSS$gene[pc3_218vs224_chr16_LOSS$fold_change < 0.6] pc3_218vs224_chr16_LOSS_genes_15 <- pc3_218vs224_chr16_LOSS$gene[pc3_218vs224_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_218vs224_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_218vs224_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_218_chr16_LOSS (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 218 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_218vs224_chr16_LOSS, “pc3_218vs224_LOSS_chr16.csv”, row.names = FALSE)

pc3_218vs224_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_218vs224_chr16_LOSS_genes_06)) write.csv(pc3_218vs224_chr16_LOSS_genes_06, “pc3_218vs224_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_218vs224_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_218vs224_chr16_LOSS_genes_15)) write.csv(pc3_218vs224_chr16_LOSS_genes_15, “pc3_218vs224_LOSS_chr16_genes_15.csv”, row.names = FALSE)

221 (LOSS) vs 222 (LOSS)

pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”) pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”)

common_genes_221vs222_chr16_LOSS <- intersect(pc3_221_chr16_LOSS$gene, pc3_222_chr16_LOSS$gene)

pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_221vs222_chr16_LOSS, ] pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_221vs222_chr16_LOSS, ]

pc3_221vs222_chr16_LOSS <- merge(pc3_221_chr16_LOSS, pc3_222_chr16_LOSS, by = “gene”)

pc3_221vs222_chr16_LOSS$fold_change <- pc3_221vs222_chr16_LOSS$TPM.x / pc3_221vs222_chr16_LOSS$TPM.y

pc3_221vs222_chr16_LOSS_genes_06 <- pc3_221vs222_chr16_LOSS$gene[pc3_221vs222_chr16_LOSS$fold_change < 0.6] pc3_221vs222_chr16_LOSS_genes_15 <- pc3_221vs222_chr16_LOSS$gene[pc3_221vs222_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_221vs222_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs222_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr16_LOSS (TPM)”, y = “pc3_222_chr16_LOSS (TPM)”, title = “pc3_chr16: 221 (LOSS) vs 222 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs222_chr16_LOSS, “pc3_221vs222_LOSS_chr16.csv”, row.names = FALSE)

pc3_221vs222_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_221vs222_chr16_LOSS_genes_06)) write.csv(pc3_221vs222_chr16_LOSS_genes_06, “pc3_221vs222_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_221vs222_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_221vs222_chr16_LOSS_genes_15)) write.csv(pc3_221vs222_chr16_LOSS_genes_15, “pc3_221vs222_LOSS_chr16_genes_15.csv”, row.names = FALSE)

221 (LOSS) vs 224 (LOSS)

pc3_221_chr16_LOSS <- subset(pc3_221_LOSS, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_221vs224_chr16_LOSS <- intersect(pc3_221_chr16_LOSS$gene, pc3_224_chr16_LOSS$gene)

pc3_221_chr16_LOSS <- pc3_221_chr16_LOSS[pc3_221_chr16_LOSS$gene %in% common_genes_221vs224_chr16_LOSS, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_221vs224_chr16_LOSS, ]

pc3_221vs224_chr16_LOSS <- merge(pc3_221_chr16_LOSS, pc3_224_chr16_LOSS, by = “gene”)

pc3_221vs224_chr16_LOSS$fold_change <- pc3_221vs224_chr16_LOSS$TPM.x / pc3_221vs224_chr16_LOSS$TPM.y

pc3_221vs224_chr16_LOSS_genes_06 <- pc3_221vs224_chr16_LOSS$gene[pc3_221vs224_chr16_LOSS$fold_change < 0.6] pc3_221vs224_chr16_LOSS_genes_15 <- pc3_221vs224_chr16_LOSS$gene[pc3_221vs224_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_221vs224_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_221vs224_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_221_chr16_LOSS (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 221 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_221vs224_chr16_LOSS, “pc3_221vs224_LOSS_chr16.csv”, row.names = FALSE)

pc3_221vs224_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_221vs224_chr16_LOSS_genes_06)) write.csv(pc3_221vs224_chr16_LOSS_genes_06, “pc3_221vs224_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_221vs224_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_221vs224_chr16_LOSS_genes_15)) write.csv(pc3_221vs224_chr16_LOSS_genes_15, “pc3_221vs224_LOSS_chr16_genes_15.csv”, row.names = FALSE)

222 (LOSS) vs 224 (LOSS)

pc3_222_chr16_LOSS <- subset(pc3_222_LOSS, chr == “chr16”) pc3_224_chr16_LOSS <- subset(pc3_224_LOSS, chr == “chr16”)

common_genes_222vs224_chr16_LOSS <- intersect(pc3_222_chr16_LOSS$gene, pc3_224_chr16_LOSS$gene)

pc3_222_chr16_LOSS <- pc3_222_chr16_LOSS[pc3_222_chr16_LOSS$gene %in% common_genes_222vs224_chr16_LOSS, ] pc3_224_chr16_LOSS <- pc3_224_chr16_LOSS[pc3_224_chr16_LOSS$gene %in% common_genes_222vs224_chr16_LOSS, ]

pc3_222vs224_chr16_LOSS <- merge(pc3_222_chr16_LOSS, pc3_224_chr16_LOSS, by = “gene”)

pc3_222vs224_chr16_LOSS$fold_change <- pc3_222vs224_chr16_LOSS$TPM.x / pc3_222vs224_chr16_LOSS$TPM.y

pc3_222vs224_chr16_LOSS_genes_06 <- pc3_222vs224_chr16_LOSS$gene[pc3_222vs224_chr16_LOSS$fold_change < 0.6] pc3_222vs224_chr16_LOSS_genes_15 <- pc3_222vs224_chr16_LOSS$gene[pc3_222vs224_chr16_LOSS$fold_change > 1.5]

ggplot(pc3_222vs224_chr16_LOSS, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_222vs224_chr16_LOSS, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_222_chr16_LOSS (TPM)”, y = “pc3_224_chr16_LOSS (TPM)”, title = “pc3_chr16: 222 (LOSS) vs 224 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_222vs224_chr16_LOSS, “pc3_222vs224_LOSS_chr16.csv”, row.names = FALSE)

pc3_222vs224_chr16_LOSS_genes_06 <- data.frame(gene = na.omit(pc3_222vs224_chr16_LOSS_genes_06)) write.csv(pc3_222vs224_chr16_LOSS_genes_06, “pc3_222vs224_LOSS_chr16_genes_06.csv”, row.names = FALSE)

pc3_222vs224_chr16_LOSS_genes_15 <- data.frame(gene = na.omit(pc3_222vs224_chr16_LOSS_genes_15)) write.csv(pc3_222vs224_chr16_LOSS_genes_15, “pc3_222vs224_LOSS_chr16_genes_15.csv”, row.names = FALSE)

NEUTRALS

219 (NEUTRAL) vs 223 (NEUTRAL)

pc3_219_chr16_NEUTRAL <- subset(pc3_219_NEUTRAL, chr == “chr16”) pc3_223_chr16_NEUTRAL <- subset(pc3_223_NEUTRAL, chr == “chr16”)

common_genes_219vs223_chr16_NEUTRAL <- intersect(pc3_219_chr16_NEUTRAL$gene, pc3_223_chr16_NEUTRAL$gene)

pc3_219_chr16_NEUTRAL <- pc3_219_chr16_NEUTRAL[pc3_219_chr16_NEUTRAL$gene %in% common_genes_219vs223_chr16_NEUTRAL, ] pc3_223_chr16_NEUTRAL <- pc3_223_chr16_NEUTRAL[pc3_223_chr16_NEUTRAL$gene %in% common_genes_219vs223_chr16_NEUTRAL, ]

pc3_219vs223_chr16_NEUTRAL <- merge(pc3_219_chr16_NEUTRAL, pc3_223_chr16_NEUTRAL, by = “gene”)

pc3_219vs223_chr16_NEUTRAL$fold_change <- pc3_219vs223_chr16_NEUTRAL$TPM.x / pc3_219vs223_chr16_NEUTRAL$TPM.y

pc3_219vs223_chr16_NEUTRAL_genes_06 <- pc3_219vs223_chr16_NEUTRAL$gene[pc3_219vs223_chr16_NEUTRAL$fold_change < 0.6] pc3_219vs223_chr16_NEUTRAL_genes_15 <- pc3_219vs223_chr16_NEUTRAL$gene[pc3_219vs223_chr16_NEUTRAL$fold_change > 1.5]

ggplot(pc3_219vs223_chr16_NEUTRAL, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_219vs223_chr16_NEUTRAL, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_219_chr16_NEUTRAL (TPM)”, y = “pc3_223_chr16_NEUTRAL (TPM)”, title = “pc3_chr16: 219 (NEUTRAL) vs 223 (NEUTRAL)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

write.csv(pc3_219vs223_chr16_NEUTRAL, “pc3_219vs223_NEUTRAL_chr16.csv”, row.names = FALSE)

pc3_219vs223_chr16_NEUTRAL_genes_06 <- data.frame(gene = na.omit(pc3_219vs223_chr16_NEUTRAL_genes_06)) write.csv(pc3_219vs223_chr16_NEUTRAL_genes_06, “pc3_219vs223_NEUTRAL_chr16_genes_06.csv”, row.names = FALSE)

pc3_219vs223_chr16_NEUTRAL_genes_15 <- data.frame(gene = na.omit(pc3_219vs223_chr16_NEUTRAL_genes_15)) write.csv(pc3_219vs223_chr16_NEUTRAL_genes_15, “pc3_219vs223_NEUTRAL_chr16_genes_15.csv”, row.names = FALSE)

chromosome 17

pc3_217_LOSS <- read.csv(“final_merged_pc3_217_LOSS.csv”) pc3_217_GAIN <- read.csv(“final_merged_pc3_217_GAIN.csv”) pc3_220_LOSS <- read.csv(“final_merged_pc3_220_LOSS.csv”) pc3_220_GAIN <- read.csv(“final_merged_pc3_220_GAIN.csv”)

217 (gain) vs 220 (gain)

pc3_217_chr17_GAIN <- subset(pc3_217_GAIN, chr == “chr17”) pc3_220_chr17_GAIN<- subset(pc3_220_GAIN, chr == “chr17”)

common_genes_217vs220_chr17 <- intersect(pc3_217_chr17_GAIN$gene, pc3_220_chr17_GAIN$gene)

pc3_217_chr17_GAIN <- pc3_217_chr17_GAIN[pc3_217_chr17_GAIN$gene %in% common_genes_217vs220_chr17, ] pc3_220_chr17_GAIN <- pc3_220_chr17_GAIN[pc3_220_chr17_GAIN$gene %in% common_genes_217vs220_chr17, ]

pc3_217vs220_chr17 <- merge(pc3_217_chr17_GAIN, pc3_220_chr17_GAIN, by = “gene”)

pc3_217vs220_chr17$fold_change <- pc3_217vs220_chr17$TPM.x / pc3_217vs220_chr17$TPM.y

pc3_217vs220_chr17_genes_06 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change < 0.6] pc3_217vs220_chr17_genes_15 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change > 1.5]

ggplot(pc3_217vs220_chr17, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr17, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr17_GAIN (TPM)”, y = “pc3_220_chr17_GAIN (TPM)”, title = “pc3_chr17: 217 (GAIN) vs 220 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

pc3_217vs220_chr17_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr17_genes_06))

pc3_217vs220_chr17_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr17_genes_15))

217 (gain) vs 220 (loss) -> nothing

pc3_217_chr17_GAIN <- subset(pc3_217_GAIN, chr == “chr17”) pc3_220_chr17_LOSS<- subset(pc3_220_LOSS, chr == “chr17”)

common_genes_217vs220_chr17 <- intersect(pc3_217_chr17_GAIN$gene, pc3_220_chr17_LOSS$gene)

pc3_217_chr17_GAIN <- pc3_217_chr17_GAIN[pc3_217_chr17_GAIN$gene %in% common_genes_217vs220_chr17, ] pc3_220_chr17_LOSS <- pc3_220_chr17_LOSS[pc3_220_chr17_LOSS$gene %in% common_genes_217vs220_chr17, ]

pc3_217vs220_chr17 <- merge(pc3_217_chr17_GAIN, pc3_220_chr17_LOSS, by = “gene”)

pc3_217vs220_chr17$fold_change <- pc3_217vs220_chr17$TPM.x / pc3_217vs220_chr17$TPM.y

pc3_217vs220_chr17_genes_06 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change < 0.6] pc3_217vs220_chr17_genes_15 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change > 1.5]

ggplot(pc3_217vs220_chr17, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr17, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr17_GAIN (TPM)”, y = “pc3_220_chr17_LOSS (TPM)”, title = “pc3_chr17: 217 (GAIN) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

217 (loss) vs 220 (gain) -> nothing

pc3_217_chr17_LOSS <- subset(pc3_217_LOSS, chr == “chr17”) pc3_220_chr17_GAIN<- subset(pc3_220_GAIN, chr == “chr17”)

common_genes_217vs220_chr17 <- intersect(pc3_217_chr17_LOSS$gene, pc3_220_chr17_GAIN$gene)

pc3_217_chr17_LOSS <- pc3_217_chr17_LOSS[pc3_217_chr17_LOSS$gene %in% common_genes_217vs220_chr17, ] pc3_220_chr17_GAIN <- pc3_220_chr17_GAIN[pc3_220_chr17_GAIN$gene %in% common_genes_217vs220_chr17, ]

pc3_217vs220_chr17 <- merge(pc3_217_chr17_LOSS, pc3_220_chr17_GAIN, by = “gene”)

pc3_217vs220_chr17$fold_change <- pc3_217vs220_chr17$TPM.x / pc3_217vs220_chr17$TPM.y

pc3_217vs220_chr17_genes_06 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change < 0.6] pc3_217vs220_chr17_genes_15 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change > 1.5]

ggplot(pc3_217vs220_chr17, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr17, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr17_LOSS (TPM)”, y = “pc3_220_chr17_GAIN (TPM)”, title = “pc3_chr17: 217 (LOSS) vs 220 (GAIN)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

217 (loss) vs 220 (loss)

pc3_217_chr17_LOSS <- subset(pc3_217_LOSS, chr == “chr17”) pc3_220_chr17_LOSS <- subset(pc3_220_LOSS, chr == “chr17”)

common_genes_217vs220_chr17 <- intersect(pc3_217_chr17_LOSS$gene, pc3_220_chr17_LOSS$gene)

pc3_217_chr17_LOSS <- pc3_217_chr17_LOSS[pc3_217_chr17_LOSS$gene %in% common_genes_217vs220_chr17, ] pc3_220_chr17_LOSS <- pc3_220_chr17_LOSS[pc3_220_chr17_LOSS$gene %in% common_genes_217vs220_chr17, ]

pc3_217vs220_chr17 <- merge(pc3_217_chr17_LOSS, pc3_220_chr17_LOSS, by = “gene”)

pc3_217vs220_chr17$fold_change <- pc3_217vs220_chr17$TPM.x / pc3_217vs220_chr17$TPM.y

pc3_217vs220_chr17_genes_06 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change < 0.6] pc3_217vs220_chr17_genes_15 <- pc3_217vs220_chr17$gene[pc3_217vs220_chr17$fold_change > 1.5]

ggplot(pc3_217vs220_chr17, aes(x = TPM.x, y = TPM.y)) + geom_point(aes(color = factor( case_when( fold_change < 0.6 ~ “< 0.6”, fold_change > 1.5 ~ “> 1.5”, TRUE ~ “Other” ) ))) + geom_text_repel(data = subset(pc3_217vs220_chr17, fold_change < 0.6 | fold_change > 1.5), aes(label = gene), size = 3, nudge_x = 0.5, nudge_y = 0.5) + scale_color_manual(values = c(“< 0.6” = “#8c5c47”, “> 1.5” = “#4a8cb0”, “Other” = “black”)) + labs(x = “pc3_217_chr17_LOSS (TPM)”, y = “pc3_220_chr17_LOSS (TPM)”, title = “pc3_chr17: 217 (LOSS) vs 220 (LOSS)”, color = “Fold Change”) + theme(legend.position = “right”) + theme_bw()

pc3_217vs220_chr17_genes_06 <- data.frame(gene = na.omit(pc3_217vs220_chr17_genes_06))

pc3_217vs220_chr17_genes_15 <- data.frame(gene = na.omit(pc3_217vs220_chr17_genes_15))

CNS

Monica Rodriguez

2024-06-07

Cut offs:

File_neutral: 0.8 – 1.2

File_gain: > 1.2

File_loss: < 0.8

Combine rows w/ (1) same seg/means & (2) sequential end/start

Load the dplyr package

Merge the data frames based on the ‘gene’ column

Select the desired columns

Save the final merged data frame to a new file

Merge the data frames based on the ‘gene’ column

Import TPM files

Merge files w/ TPM to add TPM and length columns

Select the UPDATED desired columns

Save as a csv

Merge the data frames based on the ‘gene’ column

Import TPM files

Merge files w/ TPM to add TPM and length columns

Select the UPDATED desired columns

Load the necessary packages

Create separate layers for each data frame

Create a combined volcano plot by adding the layers together

Save the volcano plot as a PNG file

Print the combined volcano plot

Create separate layers for each data frame

Create a combined volcano plot by adding the layers together

Print the combined volcano plot

Create the violin plots

Save the violin plot as a PNG file

pc3_217_violin_plot.png

Save the violin plot as a PNG file

pc3_2170_violin_plot.png

Should look like CNV Plot

Y Axis: TPM

X Axis: Abspos

color code

2 different plots to make

- all gains, neutrals, and losses (use data WITH 0’s)

- individual plots for each gain, neutral, and loss (use data WITH 0’s)

Self Note:

Either reupload OG csv and assign to variable

OR

Make sure to use variable that WASN’T subsetted w/o 0’s

ex: pc3_224_GAIN was subsetted w/o 0’s and pc3_2240_GAIN contains all data even / 0’s

TPM - Y Axis

Seg.mean - X Axis

2 plots

all data (gain, neutral, loss)

individuals (gain, neutral, loss)

w/ or w/o 0’s??

count bins first

greater than or equal to 1000 (1000 ≥)

mildly: between 10 and 1000 (10 ≤ x < 1000)

lowly: between 0.5 and 10 (0.5 ≤ x < 10)

not expressed: lower than 0.5 (x < 0.5)

greater than or equal to 1000 (1000 ≥)

mildly: between 10 and 1000 (10 ≤ x < 1000)

lowly: between 0.5 and 10 (0.5 ≤ x < 10)

not expressed: lower than 0.5 (x < 0.5)

adjust TPM: everything over 1000 is set to 1000

make a copy of data frame

grouped bar plot

stacked bar plot

^ done with samples 217-224

to count occurences

Step 2: Use the table() function to count the occurrences of each gene

Step 3: Create a new dataframe from the result

compare TPMs in chromosomes

Chromosome 1 (beginning)

pc3_ctrl_217 = baseline

^neutral

download the neutral/gain/loss file for the sample

subset for genes on the specific chromosome

once you have your comparisons (from excel sheet), only keep genes in BOTH data sets

make list of common names

subset data using list of common names

plot TPM_neutral (217_neutral on y axis vs 219_gain on x axis)

create new column for fold change