load necessary R packages
#library(vcfR)
#library(vegan)
#library(ggplot2)
#library(ggpubr)Confirm your working directory and location of files
getwd()## [1] "/Users/victorsyi/Desktop/Project"list.files(pattern = "vcf")## [1] "3.29164237-29404237.ALL.chr3_GRCh38.genotypes.20170504.vcf.gz"
## [2] "vcf_num_df.csv"
## [3] "vcf_num_df2.csv"
## [4] "vcf_num.csv"
## [5] "vcf_scaled.csv"Load the vcf file
my_vcf <- "3.29164237-29404237.ALL.chr3_GRCh38.genotypes.20170504.vcf.gz"
vcf <- vcfR::read.vcfR(my_vcf,
convertNA = T)## Scanning file to determine attributes.
## File attributes:
## meta lines: 130
## header_line: 131
## variant count: 7509
## column count: 2513
##
Meta line 130 read in.
## All meta lines processed.
## gt matrix initialized.
## Character matrix gt created.
## Character matrix gt rows: 7509
## Character matrix gt cols: 2513
## skip: 0
## nrows: 7509
## row_num: 0
##
Processed variant 1000
Processed variant 2000
Processed variant 3000
Processed variant 4000
Processed variant 5000
Processed variant 6000
Processed variant 7000
Processed variant: 7509
## All variants processedConvert raw VCF file to genotype scores
vcf_num <- vcfR::extract.gt(vcf,
element = "GT",
IDtoRowNames = F,
as.numeric = T,
convertNA = T)
write.csv(vcf_num, file = "vcf_num.csv", row.names = F)
list.files()## [1] "1000genomes_people_info2-1.csv"
## [2] "3.29164237-29404237.ALL.chr3_GRCh38.genotypes.20170504.vcf.gz"
## [3] "begin.html"
## [4] "begin.Rmd"
## [5] "curious.Rmd"
## [6] "final_report_template.docx"
## [7] "final_report_template.html"
## [8] "final_report_template.Rmd"
## [9] "gwas_pheno_env.csv"
## [10] "pheno.csv"
## [11] "Project.Rproj"
## [12] "rsconnect"
## [13] "vcf_num_df.csv"
## [14] "vcf_num_df2.csv"
## [15] "vcf_num.csv"
## [16] "vcf_scaled.csv"Transpose original VCF orientation to R dataframe orientation
vcf_num_t <- t(vcf_num)Make into a dataframe
vcf_num_df <- data.frame(vcf_num_t)Get person (sample) names
sample <- row.names(vcf_num_df)Add sample info to dataframe
vcf_num_df <- data.frame(sample,
vcf_num_df)Check working directory
getwd()## [1] "/Users/victorsyi/Desktop/Project"Save the csv
write.csv(vcf_num_df,
file = "vcf_num_df.csv",
row.names = F)Confirm presence of file
list.files()## [1] "1000genomes_people_info2-1.csv"
## [2] "3.29164237-29404237.ALL.chr3_GRCh38.genotypes.20170504.vcf.gz"
## [3] "begin.html"
## [4] "begin.Rmd"
## [5] "curious.Rmd"
## [6] "final_report_template.docx"
## [7] "final_report_template.html"
## [8] "final_report_template.Rmd"
## [9] "gwas_pheno_env.csv"
## [10] "pheno.csv"
## [11] "Project.Rproj"
## [12] "rsconnect"
## [13] "vcf_num_df.csv"
## [14] "vcf_num_df2.csv"
## [15] "vcf_num.csv"
## [16] "vcf_scaled.csv"Clean data Load population meta data
pop_meta <- read.csv(file = "1000genomes_people_info2-1.csv")Merge meta data with SNP data
names(pop_meta)## [1] "pop" "super_pop" "sample" "sex" "lat" "lng"names(vcf_num_df)[1:10]## [1] "sample" "X1" "X2" "X3" "X4" "X5" "X6" "X7"
## [9] "X8" "X9"Merge the two sets of data
vcf_num_df2 <- merge(pop_meta,
vcf_num_df,
by = "sample")Check the dimensions before and after merge
nrow(vcf_num_df) == nrow(vcf_num_df2)## [1] TRUECheck the names of the new dataframe
names(vcf_num_df2)[1:15]## [1] "sample" "pop" "super_pop" "sex" "lat" "lng"
## [7] "X1" "X2" "X3" "X4" "X5" "X6"
## [13] "X7" "X8" "X9"Check working directory
getwd()## [1] "/Users/victorsyi/Desktop/Project"save the csv
write.csv(vcf_num_df2, file = "vcf_num_df2.csv", row.names = F)Confirm the presense of the files
list.files()## [1] "1000genomes_people_info2-1.csv"
## [2] "3.29164237-29404237.ALL.chr3_GRCh38.genotypes.20170504.vcf.gz"
## [3] "begin.html"
## [4] "begin.Rmd"
## [5] "curious.Rmd"
## [6] "final_report_template.docx"
## [7] "final_report_template.html"
## [8] "final_report_template.Rmd"
## [9] "gwas_pheno_env.csv"
## [10] "pheno.csv"
## [11] "Project.Rproj"
## [12] "rsconnect"
## [13] "vcf_num_df.csv"
## [14] "vcf_num_df2.csv"
## [15] "vcf_num.csv"
## [16] "vcf_scaled.csv"Omit invariant features
invar_omit <- function(x){
cat("Dataframe of dim", dim(x),"processed...\n")
sds <- apply(x, 2, sd, na.rm = TRUE)
i_var0 <- which(sds == 0)
cat(length(i_var0), "columns removed\n")
if(length(i_var0) > 0){
x <- x[, -i_var0]
}
return(x)
}Omit invariants Check which colums have character data
names(vcf_num_df2)[1:10]## [1] "sample" "pop" "super_pop" "sex" "lat" "lng"
## [7] "X1" "X2" "X3" "X4"New dataframe to store output
vcf_noinvar <- vcf_num_df2vcf_noinvar[, -c(1:6)] <- invar_omit(vcf_noinvar[, -c(1:6)])## Dataframe of dim 2504 7509 processed...
## 1920 columns removedcreate an object to store the number of invariant columns removed
my_meta_N_invar_cols <- 1920Remove low quality data Load find_NAs()
find_NAs <- function(x){
NAs_TF <- is.na(x)
i_NA <- which(NAs_TF == TRUE)
N_NA <- length(i_NA)
return(i_NA)
}N_rows <- nrow(vcf_noinvar)
N_NA <- rep(x = 0, times = N_rows)
N_SNPs <- ncol(vcf_noinvar)
cat("This may take a minute...")## This may take a minute...for(i in 1:N_rows){
i_NA <- find_NAs(vcf_noinvar[i,])
N_NA_i <- length(i_NA)
N_NA[i] <- N_NA_i
}Check if any row has >50% NAs
cutoff50 <- N_SNPs*0.5
percent_NA <- N_NA/N_SNPs*100
any(percent_NA > 50)## [1] FALSEWhat is the average number of NAs per row?
mean(percent_NA)## [1] 0Save the mean percent of NAs per row
my_meta_N_meanNA_rows <- mean(percent_NA)Imputation of NAs Mean imputation
mean_imputation <- function(df){
cat("This make take some time...")
n_cols <- ncol(df)
for(i in 1:n_cols){
column_i <- df[,i]
mean_i <- mean(column_i, na.rm = TRUE)
NAs_i <- which(is.na(column_i))
N_NAs <- length(NAs_i)
column_i[NAs_i] <- mean_i
df[, i] <- column_i
}
return(df)
}names(vcf_noinvar)[1:10]## [1] "sample" "pop" "super_pop" "sex" "lat" "lng"
## [7] "X1" "X2" "X3" "X4"vcf_noNA <- vcf_noinvar
vcf_noNA[,-c(1:6)] <- mean_imputation(vcf_noinvar[,-c(1:6)])## This make take some time...Prepare for PCA Scale my data
#new copy of data
vcf_scaled <- vcf_noNA
#scale
vcf_scaled[,-c(1:6)] <- scale(vcf_noNA[,-c(1:6)])Run the PCA
vcf_pca <- prcomp(vcf_scaled[,-c(1:6)])write.csv(vcf_scaled, file = "vcf_scaled.csv")PCA diagnostics Examine the default screeplot
screeplot(vcf_pca)
Calculate explained variation
#Load PCA variation function
PCA_variation <- function(pca_summary, PCs = 2){
var_explained <- pca_summary$importance[2,1:PCs]*100
var_explained <- round(var_explained,3)
return(var_explained)
}Extract PCA variation data and calculate percentage variation
#Get summary information
vcf_pca_summary <- summary(vcf_pca)
#Extract raw variation data
var_out <- PCA_variation(vcf_pca_summary,
PCs = 1500)Calculate the cut off for the rule of thumb
#number of dimensions in the data
N_columns <- ncol(vcf_scaled)
#The value of the cutoff
cut_off <- 1/N_columns*100Calculate the number PCs which exceed the cut off
#which values below the cutoff
i_cut_off <- which(var_out < cut_off)
#what is first value below cutoff
i_cut_off <- min(i_cut_off)Save the first value below the cutoff
my_meta_N_meanNA_rowsPCs <- i_cut_offExtract the amount of variation explained by the first 3 PCs
my_meta_var_PC123 <- var_out[c(1,2,3)]Plot percent variation
#make barplot
barplot(var_out,
main = "Percent variation (%) Scree plot",
ylab = "Percent variation (%) explained",
names.arg = 1:length(var_out))
abline(h = cut_off, col = 2, lwd = 2)
abline(v = i_cut_off)
legend("topright",
col = c(2,1),
lty = c(1,1),
legend = c("Vertical line: cutoff",
"Horizontal line: 1st value below cur off"))
Plot cumulative percentage variation
cumulative_variation <- cumsum(var_out)
plot(cumulative_variation, type = "l")
Plot PCA results calculate scores
#Get the scores
#call veggan::scores()
vcf_pca_scores <- vegan::scores(vcf_pca)Combine the scores with the species information into the dataframe
#call data.frames()
vcf_pca_scores2 <- data.frame(super_pop = vcf_noNA$super_pop,
vcf_pca_scores)my_meta_var_PC123## PC1 PC2 PC3
## 1.861 1.442 1.251Plot the results
ggpubr::ggscatter(data = vcf_pca_scores2,
y = "PC2",
x = "PC1",
color = "super_pop",
shape = "super_pop",
main = "PCA Scatterplot",
xlab = "PC1 (1.861% of variation)",
ylab = "PC2 (1.442% of variation)")
ggpubr::ggscatter(data = vcf_pca_scores2,
y = "PC3",
x = "PC2",
color = "super_pop",
shape = "super_pop",
main = "PCA Scatterplot",
xlab = "PC2 (1.442% of variation)",
ylab = "PC3 (1.251% of variation)")
ggpubr::ggscatter(data = vcf_pca_scores2,
y = "PC3",
x = "PC1",
color = "super_pop",
shape = "super_pop",
main = "PCA Scatterplot",
xlab = "PC1 (1.861% of variation)",
ylab = "PC3 (1.251% of variation)")