Trinity Paper style graphs

library(data.table)
library(readr)

macauResults <- read_delim("~/Documents/RobertsLab/TrinityGraph/10x-result.assoc.txt", 
    "\t", escape_double = FALSE, trim_ws = TRUE)

samples <- c("EPI-103", "EPI-104", "EPI-127", "EPI-128", "EPI-119", "EPI-120", "EPI-135", "EPI-136", "EPI-111", "EPI-113", "EPI-143", "EPI-145")

ambSamps <- c("EPI-119", "EPI-120", "EPI-135", "EPI-136")
  
lowSamps <- c("EPI-103", "EPI-104", "EPI-127", "EPI-128")

slowSamps <- c("EPI-111", "EPI-113", "EPI-143", "EPI-145")

sig.results <- macauResults[macauResults$pvalue < 0.05,]

for(i in 1:length(samples))   {
  
  name <- samples[i]
  temp <- read_delim(paste0("~/Documents/RobertsLab/TrinityGraph/", samples[i], "-percmeth.bedgraph.txt"), 
    "\t", escape_double = FALSE, col_names = FALSE, 
    trim_ws = TRUE)
  assign(name, as.data.table(temp))
}

amb.sig.results <- as.data.table(sig.results$id[sig.results$pvalue < 0.05])
colnames(amb.sig.results) <- "Loc"

for(i in 1:length(ambSamps))   {
  
  temp <- read_delim(paste0("~/Documents/RobertsLab/TrinityGraph/", ambSamps[i], "-percmeth.bedgraph.txt"), 
    "\t", escape_double = FALSE, col_names = FALSE, 
    trim_ws = TRUE) 
  
  temp$Loc <- paste0(temp$X1,"-",temp$X2)
  temp_DMRs <- temp[temp$Loc %in% amb.sig.results$Loc,]
  
  amb.sig.results$temp <- temp_DMRs$X4
  colnames(amb.sig.results)[i + 1] <- ambSamps[i]  
}

low.sig.results <- as.data.table(sig.results$id[sig.results$pvalue < 0.05])
colnames(low.sig.results) <- "Loc"

for(i in 1:length(lowSamps))   {

  temp <- read_delim(paste0("~/Documents/RobertsLab/TrinityGraph/", lowSamps[i], "-percmeth.bedgraph.txt"), 
    "\t", escape_double = FALSE, col_names = FALSE, col_types = cols(X4 = col_double()),
    trim_ws = TRUE) 
  
  temp$Loc <- paste0(temp$X1,"-",temp$X2)
  temp_DMRs <- temp[temp$Loc %in% low.sig.results$Loc,]
  
  low.sig.results$temp <- temp_DMRs$X4
  colnames(low.sig.results)[i + 1] <- lowSamps[i]  
}

slow.sig.results <- as.data.table(sig.results$id[sig.results$pvalue < 0.05])
colnames(slow.sig.results) <- "Loc"

for(i in 1:length(slowSamps))   {
  

  temp <- read_delim(paste0("~/Documents/RobertsLab/TrinityGraph/", slowSamps[i], "-percmeth.bedgraph.txt"), 
    "\t", escape_double = FALSE, col_names = FALSE, col_types = cols(X4 = col_double()),
    trim_ws = TRUE) 
  
  temp$Loc <- paste0(temp$X1,"-",temp$X2)
  temp_DMRs <- temp[temp$Loc %in% slow.sig.results$Loc,]
  
  slow.sig.results$temp <- temp_DMRs$X4
  colnames(slow.sig.results)[i + 1] <- slowSamps[i]  
}

head(amb.sig.results)

head(low.sig.results)

head(slow.sig.results)

for(i in 1:nrow(amb.sig.results))   {
  
  amb.sig.results$Mean[i] <- mean(amb.sig.results$`EPI-119`[i], amb.sig.results$`EPI-120`[i], amb.sig.results$`EPI-135`[i], amb.sig.results$`EPI-136`[i])
  
}
for(i in 1:nrow(slow.sig.results))   {
  
  low.sig.results$Mean[i] <- mean(low.sig.results$`EPI-103`[i], low.sig.results$`EPI-104`[i], low.sig.results$`EPI-127`[i], low.sig.results$`EPI-128`[i])
  
}
for(i in 1:nrow(slow.sig.results))   {
  
  slow.sig.results$Mean[i] <- mean(slow.sig.results$`EPI-111`[i], slow.sig.results$`EPI-113`[i], slow.sig.results$`EPI-143`[i], slow.sig.results$`EPI-145`[i])
  
}

head(amb.sig.results)

head(low.sig.results)

head(slow.sig.results)

amb.sig.results$Loc <- as.factor(amb.sig.results$Loc)
low.sig.results$Loc <- as.factor(low.sig.results$Loc)
slow.sig.results$Loc <- as.factor(slow.sig.results$Loc)
colnames(amb.sig.results) <- c("Loc", "EPI_119", "EPI_120", "EPI_135", "EPI_136", "Mean")
colnames(low.sig.results) <- c("Loc", "EPI_103", "EPI_104", "EPI_127", "EPI_128", "Mean")
colnames(slow.sig.results) <- c("Loc", "EPI_111", "EPI_113", "EPI_143", "EPI_145", "Mean")

The below plot shows just the ambient samoples, with a bold line representing the mean and the grey lines showing individual samples.

plot(amb.sig.results$Loc, amb.sig.results$Mean, type = "b")

plot1 <- ggplot(data = amb.sig.results, aes(x = Loc, y = Mean, group = 1))
plot1 + geom_line(aes(size = 2), color = "dark blue") + geom_line(data = amb.sig.results, aes(x = Loc, y = EPI_119, group = 2), position = "jitter", color = "gray") +
  geom_line(data = amb.sig.results, aes(x = Loc, y = EPI_120, group = 2), position = "jitter", color = "gray")+ geom_line(data = amb.sig.results, aes(x = Loc, y = EPI_135, group = 2), position = "jitter", color = "gray") + geom_line(data = amb.sig.results, aes(x = Loc, y = EPI_136, group = 2), position = "jitter", color = "gray") + theme(axis.title.x=element_blank(),     axis.text.x=element_blank(), axis.ticks.x=element_blank()) + ggtitle("Ambient group means by significant location") + scale_y_continuous(limits = c(0.15, 1.01))

Bold line represents mean, grey lines represent individual samples

plot1 <- ggplot(data = low.sig.results, aes(x = Loc, y = Mean, group = 1))
plot1 + geom_line(aes(size = 2), color = "dark green") + geom_line(data = low.sig.results, aes(x = Loc, y = EPI_103, group = 2), position = "jitter", color = "gray") +
  geom_line(data = low.sig.results, aes(x = Loc, y = EPI_104, group = 2), position = "jitter", color = "gray")+ geom_line(data = low.sig.results, aes(x = Loc, y = EPI_127, group = 2), position = "jitter", color = "gray") + geom_line(data = low.sig.results, aes(x = Loc, y = EPI_128, group = 2), position = "jitter", color = "gray") + theme(axis.title.x=element_blank(),     axis.text.x=element_blank(), axis.ticks.x=element_blank()) + ggtitle("Low group means by significant location") + scale_y_continuous(limits = c(0.15, 1.01))

Bold line represents mean, grey lines represent individual samples

plot1 <- ggplot(data = slow.sig.results, aes(x = Loc, y = Mean, group = 1))
plot1 + geom_line(aes(size = 2), color = "maroon") + geom_line(data = slow.sig.results, aes(x = Loc, y = EPI_111), position = "jitter", color = "gray") + geom_line(data = slow.sig.results, aes(x = Loc, y = EPI_113), position = "jitter", color = "gray")+ geom_line(data = slow.sig.results, aes(x = Loc, y = EPI_143), position = "jitter", color = "gray") + geom_line(data = slow.sig.results, aes(x = Loc, y = EPI_145), position = "jitter", color = "gray") + theme(axis.title.x=element_blank(), axis.text.x=element_blank(), axis.ticks.x=element_blank()) + ggtitle("Super Low group means by significant location") + scale_y_continuous(limits = c(0.15, 1.01))

The plot below shows means by treatment group blue is ambient, green is low, and red is super low.

plot1 <- ggplot(data = amb.sig.results, aes(x = Loc, y = Mean, group = 1))
plot1 + geom_line(color = "dark blue") + geom_line(data = low.sig.results, aes(x = Loc, y = Mean, group = 2), position = "jitter", color = "dark green") +
  geom_line(data = slow.sig.results, aes(x = Loc, y = Mean, group = 2), position = "jitter", color = "maroon") + theme(axis.title.x=element_blank(),     axis.text.x=element_blank(), axis.ticks.x=element_blank()) + ggtitle("Means by Treatment Group") + scale_y_continuous(limits = c(0.15, 1.01))

So, all of that was wrong! Nevermind.

sig.results2 <- as.data.frame(cbind(amb.sig.results[,1:5], low.sig.results[,2:5], slow.sig.results[,2:5]))
head(sig.results2)

sig.results.t <- as.data.frame(t(sig.results2))
colnames(sig.results.t) <- sig.results2$Loc
sig.results.t <- sig.results.t[2:13,]
sig.results3 <- as.data.frame(colnames(sig.results2)[2:13])
sig.results3 <- as.data.frame(cbind(sig.results3, sig.results.t))
colnames(sig.results3)[1] <- "Sample"

The grah below is… difficult to read. Each line is a DMR and each group of four samples is a Treatment group, first Ambient, second low and third Super Low

sig.results3$Sample <- reorder(sig.results3$Sample, c(1,2,3,4,5,6,7,8,9,10,11,12))
ggplot(data = sig.results3, aes(x = Sample, y = sig.results3[,2], group = 1)) + geom_line() + geom_line(aes(x = Sample, y = sig.results3[,3])) + geom_line(aes(x = Sample, y = sig.results3[,4])) + geom_line(aes(x = Sample, y = sig.results3[,5])) + geom_line(aes(x = Sample, y = sig.results3[,6])) + geom_line(aes(x = Sample, y = sig.results3[,7])) + geom_line(aes(x = Sample, y = sig.results3[,8])) + geom_line(aes(x = Sample, y = sig.results3[,9])) + geom_line(aes(x = Sample, y = sig.results3[,10])) + geom_line(aes(x = Sample, y = sig.results3[,11])) + geom_line(aes(x = Sample, y = sig.results3[,12])) + geom_line(aes(x = Sample, y = sig.results3[,13])) + geom_line(aes(x = Sample, y = sig.results3[,14])) + geom_line(aes(x = Sample, y = sig.results3[,15])) + geom_line(aes(x = Sample, y = sig.results3[,16])) + geom_line(aes(x = Sample, y = sig.results3[,17])) + geom_line(aes(x = Sample, y = sig.results3[,18])) + geom_line(aes(x = Sample, y = sig.results3[,19])) + geom_line(aes(x = Sample, y = sig.results3[,20])) + geom_line(aes(x = Sample, y = sig.results3[,21])) + geom_line(aes(x = Sample, y = sig.results3[,22])) + geom_line(aes(x = Sample, y = sig.results3[,23])) + geom_line(aes(x = Sample, y = sig.results3[,24])) + geom_line(aes(x = Sample, y = sig.results3[,25])) + geom_line(aes(x = Sample, y = sig.results3[,26])) + geom_line(aes(x = Sample, y = sig.results3[,27])) + geom_line(aes(x = Sample, y = sig.results3[,28])) + geom_line(aes(x = Sample, y = sig.results3[,29])) + geom_line(aes(x = Sample, y = sig.results3[,30])) + geom_line(aes(x = Sample, y = sig.results3[,31])) + geom_line(aes(x = Sample, y = sig.results3[,32])) + geom_line(aes(x = Sample, y = sig.results3[,33])) + geom_line(aes(x = Sample, y = sig.results3[,34])) + geom_line(aes(x = Sample, y = sig.results3[,35])) + geom_line(aes(x = Sample, y = sig.results3[,36])) + geom_line(aes(x = Sample, y = sig.results3[,37])) + geom_line(aes(x = Sample, y = sig.results3[,38])) + geom_line(aes(x = Sample, y = sig.results3[,39])) + geom_line(aes(x = Sample, y = sig.results3[,40])) + geom_line(aes(x = Sample, y = sig.results3[,41])) + geom_line(aes(x = Sample, y = sig.results3[,42])) + theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank()) +
  ggtitle("Proportion of Methylation by sample. Each line represents a DMR") + geom_vline(xintercept = 8.1, linetype = "longdash", color = "blue") + ggtitle(colnames(sig.results3[i])) + geom_vline(xintercept = 4.1, linetype = "longdash", color = "blue")

The graphs below look at individual DMRs by samples, the blue lines are there to just serve to mark between Ambient, Low, and Super Low groups.

for(i in 2:41)   {
  
  graph <- ggplot(data = sig.results3, aes(x = Sample, y = sig.results3[,i], group = 1)) + geom_line() + geom_vline(xintercept = 4.1, linetype = "longdash", color = "blue") + geom_vline(xintercept = 8.1, linetype = "longdash", color = "blue") + ggtitle(colnames(sig.results3[i]))
  
  print(graph)
  
}

Add a new chunk by clicking the Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).