Last updated: 2020-03-17
Checks: 7 0
Knit directory: m6AQTL_reproducibleDocument/
This reproducible R Markdown analysis was created with workflowr (version 1.6.0). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(20200317) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility. The version displayed above was the version of the Git repository at the time these results were generated.
Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:
Ignored files:
Ignored: .Rproj.user/
Unstaged changes:
Modified: analysis/index.Rmd
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the R Markdown and HTML files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view them.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| Rmd | 5c01c10 | kevinlkx | 2020-03-17 | wflow_publish(“analysis/cor_effectsize_byRBPs_jointLCLs.Rmd”) |
m6AQTL_workflowr/analysis/cor_effectsize_byRBPs_scatterplots_logOR.adjusted_YangVCF_noPCs_SNPlevelQvalue0.2_APAdist_noTE.Rmd# options(scipen = 999)
suppressPackageStartupMessages(library(GenomicRanges))
library(ggplot2)
library(gplots)
library(RColorBrewer)
library(reshape2)
library(foreach)
library(doParallel)
library(qvalue)
library(BH)
effect_cor <- function(effectsize_joint.df, phenotype_x, phenotype_y, filter_beta = Inf){
x <- effectsize_joint.df[, phenotype_x]
y <- effectsize_joint.df[, phenotype_y]
idx_included <- which(abs(x) <= filter_beta & !is.na(x) & !is.na(y))
x <- x[idx_included]
y <- y[idx_included]
if(length(idx_included) < 2){
cor_Pearson <- NA
slope <- NA
pvalue <- NA
}else{
cor_Pearson <- cor(x, y)
lm.model <- lm(y ~ x)
slope <- summary(lm.model)$coefficients[2, "Estimate"]
pvalue <- summary(lm.model)$coefficients[2, "Pr(>|t|)"]
r.squared = summary(lm.model)$r.squared
}
cor_summary <- c(cor_Pearson = cor_Pearson, r.squared = r.squared, slope = slope, pvalue = pvalue, n = length(x))
return(cor_summary)
}m6A_version <- "jointPeak_threshold5_MeRIP_HISAT2Map"
m6A_phenotype_name <- "m6APeak_logOR_GC.IP.adjusted_qqnorm"
thresh_qvalue <- 0.2
num_PCs_m6AQTL <- 0
num_PCs_joint <- 0
thresh_FDR <- 0.1
type_apaQTL <- "dist"
registerDoParallel(cores = 6)
cat("m6A version: ", m6A_version, "\n")
cat("m6A phenotype: ", m6A_phenotype_name, "\n")
cat("Mapping m6AQTLs: qvalue: ", thresh_qvalue, ", with", num_PCs_m6AQTL, "PCs. \n")
cat("Effect size comparison: ", num_PCs_joint,"PCs, choose SNP-gene pairs for APA-QTLs by: ", type_apaQTL, "\n" )
cat("FDR threshold for testing multiple correlations: ", thresh_FDR*100, "%\n")dir_m6AQTL <- "/project2/xinhe/m6A/m6A_seq/m6A_QTL"
RBPs.gr <- readRDS(paste0(dir_m6AQTL, "/RBP_data/hg19/all.RBP.intersect.hg19.bed.gr.rds"))
RBP_list <- unique(RBPs.gr$name)
cat("list of RBPs with eCLIP data: ", RBP_list)
cat(length(RBP_list), "RBPs \n")dir_gene_info <- "/project2/xinhe/m6A/m6A_seq/m6A_QTL/gene_info/"
gene_list_unique <- read.table(paste0(dir_gene_info, "/geneSymbol_ensembl_unique.txt"),
header = T, sep = "\t", stringsAsFactors = F)
rownames(gene_list_unique) <- gene_list_unique$symboldir_m6AQTL_results <- paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/results/hg19/m6A_QTLs/", m6A_version)
if(thresh_qvalue == 0.2){
if(num_PCs_m6AQTL == 15){
m6AQTLs_lead.df <- readRDS(paste0(dir_m6AQTL_results, "/fastQTL_YangGeno/", m6A_phenotype_name, "/lead.m6AQTL.", m6A_phenotype_name, ".15PCs.fastQTL.nominals.qvalue_0.2.rds"))
}else{
m6AQTLs_lead.df <- readRDS(paste0(dir_m6AQTL_results, "/fastQTL_YangGeno/", m6A_phenotype_name, "/lead.m6AQTL.", m6A_phenotype_name, ".0PCs.fastQTL.nominals.qvalue_0.2.rds"))
}
}else{
if(num_PCs_m6AQTL == 15){
m6AQTLs_lead.df <- readRDS(paste0(dir_m6AQTL_results, "/fastQTL_YangGeno/", m6A_phenotype_name, "/lead.m6AQTL.", m6A_phenotype_name, ".15PCs.fastQTL.nominals.rds"))
}else{
m6AQTLs_lead.df <- readRDS(paste0(dir_m6AQTL_results, "/fastQTL_YangGeno/", m6A_phenotype_name, "/lead.m6AQTL.", m6A_phenotype_name, ".0PCs.fastQTL.nominals.rds"))
}
}
m6AQTLs_lead.df$peak_snp_pair <- paste(m6AQTLs_lead.df$PEAK, m6AQTLs_lead.df$SNP, sep = "|")
cat(nrow(m6AQTLs_lead.df), "lead m6A-QTLs (", num_PCs_m6AQTL, "PCs in m6A-QTL mapping, qvalue <", thresh_qvalue, ")\n")
cat("Load combined QTL summary stats (Yang's genotype data,", num_PCs_joint, "PCs in joint analysis) ... \n")
if(num_PCs_joint == 0){
dir_combined_data <- paste0(dir_m6AQTL_results, "/jointLCLs_analysis/", m6A_phenotype_name, "/m6AQTLs_full_noPCs_APA", type_apaQTL)
}else{
dir_combined_data <- paste0(dir_m6AQTL_results, "/jointLCLs_analysis/", m6A_phenotype_name, "/m6AQTLs_full_PCs_APA", type_apaQTL)
}
m6AQTLs_info_full.df <- readRDS(paste0(dir_combined_data, "/m6AQTLs_info_full.rds"))
m6AQTLs_info_full.df$peak_snp_pair <- paste(m6AQTLs_info_full.df$PEAK, m6AQTLs_info_full.df$SNP, sep = "|")
idx_sig_peak_snp_pair <- na.omit(match(m6AQTLs_lead.df$peak_snp_pair, m6AQTLs_info_full.df$peak_snp_pair))
m6AQTLs_sig.df <- m6AQTLs_info_full.df[idx_sig_peak_snp_pair, ]
cat(nrow(m6AQTLs_sig.df), "lead m6A-QTLs matched in Yang's genotype data \n")
beta_joint_m6AQTLs_full.df <- readRDS(paste0(dir_combined_data, "/beta_jointLCLs_m6AQTLs_full.rds"))
beta_joint_m6AQTLs_sig.df <- beta_joint_m6AQTLs_full.df[idx_sig_peak_snp_pair, ]
rownames(beta_joint_m6AQTLs_sig.df) <- m6AQTLs_sig.df$peak_snp_pair
pvalue_joint_m6AQTLs_full.df <- readRDS(paste0(dir_combined_data, "/pvalue_jointLCLs_m6AQTLs_full.rds"))
pvalue_joint_m6AQTLs_sig.df <- pvalue_joint_m6AQTLs_full.df[idx_sig_peak_snp_pair, ]
rownames(pvalue_joint_m6AQTLs_sig.df) <- m6AQTLs_sig.df$peak_snp_pair
if(anyDuplicated(m6AQTLs_sig.df$peak_snp_pair)){stop("Duplicated peak-SNP pairs!")}
rm(m6AQTLs_info_full.df)
rm(beta_joint_m6AQTLs_full.df)
rm(pvalue_joint_m6AQTLs_full.df)if(!file.exists(paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.jointPeaks.bed6"))){
cat("load m6A peaks, save in BED12 format and convert to BED6 exons \n")
library(MeRIPtools)
peak_logOR_MeRIPdata <- readRDS(paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.rds"))
peaks_bed12 <- jointPeak(peak_logOR_MeRIPdata$MeRIPdata)
peaks_bed12$name <- paste0(peaks_bed12$chr, ":", peaks_bed12$start, "-",peaks_bed12$end, "_", peaks_bed12$name, "_", peaks_bed12$strand)
peaks_bed12 <- peaks_bed12[peaks_bed12$chr %in% paste0("chr", 1:22),]
if(length(setdiff(m6AQTLs_sig.df$PEAK, peaks_bed12$name)) > 0){
stop("Not all m6A-QTL peaks exist in bed12 peaks!")
}
colnames(peaks_bed12)[1] <- "#chr"
write.table(peaks_bed12, paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.jointPeaks.bed12"), col.names = T, row.names = F, quote = F, sep = "\t")
system(paste("bed12tobed6 -i", paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.jointPeaks.bed12"), ">", paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.jointPeaks.bed6")))
}
m6A_peaks_bed6 <- read.table(paste0("/project2/xinhe/m6A/m6A_seq/m6A_QTL/peakcalling/", m6A_version, "/peak_logOR_MeRIPdata.jointPeaks.bed6"), header = F, sep = "\t", comment.char = "", stringsAsFactors = F)
colnames(m6A_peaks_bed6) <- c("chr", "start", "end", "PEAK", "score", "strand")
## keep sig peaks
m6A_sig_peaks_bed6 <- m6A_peaks_bed6[m6A_peaks_bed6$PEAK %in% m6AQTLs_sig.df$PEAK, ]
m6A_sig_peaks_bed6.gr <- makeGRangesFromDataFrame(m6A_sig_peaks_bed6, keep.extra.columns = T)
cat(length(unique(m6A_sig_peaks_bed6.gr$PEAK)), "(", length(unique(m6A_sig_peaks_bed6.gr$PEAK))/nrow(m6AQTLs_sig.df)*100, "% )", "sig.m6A peaks in BED6 list \n")## overlap m6A peaks with RBP (RBP within m6A peaks)
RBP_peak_overlaps_within <- as.data.frame(findOverlaps(query = RBPs.gr, subject = m6A_sig_peaks_bed6.gr, ignore.strand = F, type = "within"))
colnames(RBP_peak_overlaps_within) <- c("idx_RBP", "idx_peak")
RBP_peak_overlaps_within$RBP <- RBPs.gr[RBP_peak_overlaps_within$idx_RBP]$name
RBP_peak_overlaps_within$PEAK <- m6A_sig_peaks_bed6.gr[RBP_peak_overlaps_within$idx_peak]$PEAK
if(any(setdiff(m6AQTLs_sig.df$PEAK, m6A_sig_peaks_bed6$PEAK))){
cat("Peaks not in bed6 peak list:", setdiff(m6AQTLs_sig.df$PEAK, m6A_sig_peaks_bed6$PEAK), "\n")
}
m6AQTLs_sig.df$RBPs_overlapped_within <- foreach(i=1:nrow(m6AQTLs_sig.df), .combine=c) %dopar% {
if(m6AQTLs_sig.df$PEAK[i] %in% RBP_peak_overlaps_within$PEAK){
RBPs_overlapped <- unique(RBP_peak_overlaps_within[RBP_peak_overlaps_within$PEAK == m6AQTLs_sig.df$PEAK[i], "RBP"])
return(paste(c(RBPs_overlapped,""), collapse = ","))
}else{
return("")
}
}cat(length(which(m6AQTLs_sig.df$RBPs_overlapped_within != "")), "out of", length(m6AQTLs_sig.df$PEAK), "(",
length(which(m6AQTLs_sig.df$RBPs_overlapped_within != ""))/length(m6AQTLs_sig.df$PEAK)*100,"% ) sig peaks have RBP binding sites within peaks. \n\n")
## count the number of peaks and genes that each RBP bind to
m6AsigPeaks_RBP_overlap_counts <- sapply(RBP_list, function(x){
idx_RBP_overlapped_within <- grep(paste0(x, ","), m6AQTLs_sig.df$RBPs_overlapped_within)
num_peaks <- length(unique(m6AQTLs_sig.df[idx_RBP_overlapped_within, "PEAK"]))
num_genes <- length(unique(m6AQTLs_sig.df[idx_RBP_overlapped_within, "gene_name"]))
return(c(num_peaks, num_genes))}
)
rownames(m6AsigPeaks_RBP_overlap_counts) <- c("num_peaks", "num_genes")
RBP_m6AsigPeaks_within <- sort(m6AsigPeaks_RBP_overlap_counts["num_peaks",], decreasing = T)
print(RBP_m6AsigPeaks_within)thresh_RBP <- 50
cat("Show RBPs with at least", thresh_RBP, "gene(peak)-SNP pairs (overlap == within). \n\n")
RBP_m6AsigPeaks_within[RBP_m6AsigPeaks_within >= thresh_RBP]
RBPnames_filtered_m6AsigPeaks_within <- names(RBP_m6AsigPeaks_within[RBP_m6AsigPeaks_within >= thresh_RBP])dir_combined_data_RBPs <- paste0(dir_combined_data, "/effectsize_jointLCLs_byRBPs/")
dir.create(dir_combined_data_RBPs, showWarnings = F, recursive = T)
RBP_list_included <- sort(RBPnames_filtered_m6AsigPeaks_within)
cat(length(RBP_list_included), "RBPs included. \n ")
phenotype_list <- c("Expression", "Ribosome", "Protein", "Decay", "APA")
cor_m6APeakAnno.m <- matrix(NA, nrow = length(RBP_list_included), ncol = length(phenotype_list))
colnames(cor_m6APeakAnno.m) <- phenotype_list
rownames(cor_m6APeakAnno.m) <- RBP_list_included
Cor_ByRBP <- melt(cor_m6APeakAnno.m)
Cor_ByRBP <- data.frame(anno = Cor_ByRBP[,1], phenotype = Cor_ByRBP[,2],
cor = NA, r.squared = NA, slope = NA, pvalue = NA, n = NA)
for(i in 1:length(RBP_list_included)){
RBP_name <- RBP_list_included[i]
idx_matched <- grep(paste0(RBP_name, ","), m6AQTLs_sig.df$RBPs_overlapped_within)
# idx_matched <- sort(which(m6AQTLs_sig.df$PEAK %in% RBP_peak_overlaps_within[RBP_peak_overlaps_within$RBP == RBP_name, "PEAK"]))
m6AQTLs_sig_RBP.df <- m6AQTLs_sig.df[idx_matched, ]
m6AQTLs_sig_RBP.df <- m6AQTLs_sig_RBP.df[order(m6AQTLs_sig_RBP.df$pvalue), ]
## Select lead SNPs
m6AQTLs_sig_RBP.df <- m6AQTLs_sig_RBP.df[!duplicated(m6AQTLs_sig_RBP.df$PEAK), ]
## Select the strongest associated peak-snp pair for each gene-snp pair
m6AQTLs_sig_RBP.df <- m6AQTLs_sig_RBP.df[!duplicated(m6AQTLs_sig_RBP.df$gene_snp_pair), ]
if( nrow(m6AQTLs_sig.df[idx_matched, ]) != nrow(m6AQTLs_sig_RBP.df) ){
cat(RBP_name, ": ", length(idx_matched), "peak-SNP pairs --> ", nrow(m6AQTLs_sig_RBP.df), "gene-SNP pairs \n")
}else{
cat(RBP_name, ": ", nrow(m6AQTLs_sig_RBP.df), "gene-snp pairs \n")
}
effectsize_joint.df <- beta_joint_m6AQTLs_sig.df[m6AQTLs_sig_RBP.df$peak_snp_pair,]
pvalue_joint.df <- pvalue_joint_m6AQTLs_sig.df[m6AQTLs_sig_RBP.df$peak_snp_pair,]
saveRDS(effectsize_joint.df, paste0(dir_combined_data_RBPs, "/effectjoint_", RBP_name, "_overlapWithin.rds"))
saveRDS(pvalue_joint.df, paste0(dir_combined_data_RBPs, "/pvaluejoint_", RBP_name, "_overlapWithin.rds"))
for(phenotype in phenotype_list){
cor_summary <- effect_cor(effectsize_joint.df, "m6A", phenotype)
Cor_ByRBP[Cor_ByRBP$anno == RBP_name & Cor_ByRBP$phenotype == phenotype, 3:ncol(Cor_ByRBP)] <- cor_summary
}
}
Cor_ByRBP$anno <- factor(Cor_ByRBP$anno, levels = rev(RBP_list_included))
Cor_ByRBP$phenotype <- factor(Cor_ByRBP$phenotype, levels = phenotype_list)
saveRDS(Cor_ByRBP, paste0(dir_combined_data, "/cor_effectsize_jointLCLs_RBPs_overlapWithin.rds"))Cor_ByRBP <- readRDS(paste0(dir_combined_data, "/cor_effectsize_jointLCLs_RBPs_overlapWithin.rds"))
Cor_ByRBP$sign_nlogP <- sign(Cor_ByRBP$cor) * -log10(Cor_ByRBP$pvalue)
phenotype_selected <- c("APA", "Expression", "Decay", "Ribosome", "Protein")
cat("selected phenotypes: ",phenotype_selected, "\n")
Cor_ByRBP <- Cor_ByRBP[Cor_ByRBP$phenotype %in% phenotype_selected, ]
Cor_ByRBP$phenotype <- factor(Cor_ByRBP$phenotype, levels = phenotype_selected )
cat("FDR threshold for testing multiple correlations: ", thresh_FDR*100, "%\n")
Cor_ByRBP$p_bonferroni <- p.adjust(Cor_ByRBP$pvalue, method = "bonferroni")
Cor_ByRBP$p_BH <- p.adjust(Cor_ByRBP$pvalue, method = "BH")
Cor_ByRBP$qvalue <- qvalue(Cor_ByRBP$pvalue)$qvalues
Cor_ByRBP_sig_bonferroni <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[which(Cor_ByRBP$p_bonferroni < thresh_FDR),"anno"],]
cat(length(unique(Cor_ByRBP_sig_bonferroni$anno)), "RBPs with adjusted pvalue (Bonferroni method) <", thresh_FDR, "\n")
print(unique(as.character(Cor_ByRBP_sig_bonferroni$anno)))
Cor_ByRBP_sig_BH <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[which(Cor_ByRBP$p_BH < thresh_FDR),"anno"],]
cat(length(unique(Cor_ByRBP_sig_BH$anno)), "RBPs with BH FDR <", thresh_FDR, "\n")
print(unique(as.character(Cor_ByRBP_sig_BH$anno)))
Cor_ByRBP_sig_qvalue <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[which(Cor_ByRBP$qvalue < thresh_FDR),"anno"],]
cat(length(unique(Cor_ByRBP_sig_qvalue$anno)), "RBPs with qvalue <", thresh_FDR, "\n")
print(unique(as.character(Cor_ByRBP_sig_qvalue$anno)))Cor_ByRBP_sig <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[ which(Cor_ByRBP$p_BH < thresh_FDR),"anno"] ,]
cat(length(unique(Cor_ByRBP_sig$anno)), "significant RBPs with BH FDR <", thresh_FDR, "\n")
Cor_ByRBP_sig$phenotype <- factor(Cor_ByRBP_sig$phenotype, levels = phenotype_selected)
if(nrow(Cor_ByRBP_sig) > 0){
# print(Cor_ByRBP_sig[which(Cor_ByRBP_sig$p_BH < thresh_FDR), ])
cat("Range of correlations:")
print(round(range(Cor_ByRBP_sig$cor),2))
cor_limit <- max(abs(Cor_ByRBP_sig$cor)) + 0.2
ggplot(Cor_ByRBP_sig)+
geom_point(aes(x = phenotype, y = anno, colour = cor, size = -log10(pvalue)) )+
geom_point(data = Cor_ByRBP_sig[which(Cor_ByRBP_sig$p_BH < thresh_FDR), ], aes(x = phenotype, y = anno), size = 8,shape = 1 )+
scale_color_gradient2(midpoint=0, limit = c(-cor_limit, cor_limit) , low="blue", mid="white",
high="red", space ="Lab" ) +
theme_minimal() + # minimal theme
theme(text = element_text(face = "bold", size = 14),panel.grid = element_blank(),
axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),
axis.title.x = element_blank(),
axis.title.y = element_blank(),
axis.text = element_text(size = 16, color = "black"))
}Cor_ByRBP_sig <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[ which(Cor_ByRBP$qvalue < thresh_FDR),"anno"] ,]
cat(length(unique(Cor_ByRBP_sig$anno)), "significant RBPs with qvalue <", thresh_FDR, "\n")
Cor_ByRBP_sig$phenotype <- factor(Cor_ByRBP_sig$phenotype, levels = phenotype_selected )
if(nrow(Cor_ByRBP_sig) > 0){
# print(Cor_ByRBP_sig[which(Cor_ByRBP_sig$qvalue < thresh_FDR), ])
cat("Range of correlations:")
print(round(range(Cor_ByRBP_sig$cor),2))
cor_limit <- max(abs(Cor_ByRBP_sig$cor)) + 0.2
ggplot(Cor_ByRBP_sig)+
geom_point(aes(x = phenotype, y = anno, colour = cor , size = -log10(pvalue)) )+
geom_point(data = Cor_ByRBP_sig[which(Cor_ByRBP_sig$qvalue < thresh_FDR), ], aes(x = phenotype, y = anno), size = 8,shape = 1)+
scale_color_gradient2(midpoint=0, limit = c(-cor_limit, cor_limit) , low="blue", mid="white",
high="red", space ="Lab" ) +
theme_minimal() + # minimal theme
theme(text = element_text(face = "bold", size = 14),panel.grid = element_blank(),
axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),
axis.title.x = element_blank(),
axis.title.y = element_blank(),
axis.text = element_text(size = 16, color = "black"))
}Cor_ByRBP_sig <- Cor_ByRBP[Cor_ByRBP$anno %in% Cor_ByRBP[ which(Cor_ByRBP$p_bonferroni < thresh_FDR),"anno"],]
cat(length(unique(Cor_ByRBP_sig$anno)), "significant RBPs with Bonferroni adjusted pvalue <", thresh_FDR, "\n")
Cor_ByRBP_sig$phenotype <- factor(Cor_ByRBP_sig$phenotype, levels = phenotype_selected )
if(nrow(Cor_ByRBP_sig) > 0){
# print(Cor_ByRBP_sig[which(Cor_ByRBP_sig$p_bonferroni < thresh_FDR), ])
cat("Range of correlations:")
print(round(range(Cor_ByRBP_sig$cor),2))
cor_limit <- max(abs(Cor_ByRBP_sig$cor)) + 0.2
ggplot(Cor_ByRBP_sig)+
geom_point(aes(x = phenotype, y = anno, colour = cor, size = -log10(pvalue)) )+
geom_point(data = Cor_ByRBP_sig[which(Cor_ByRBP_sig$p_bonferroni < thresh_FDR), ], aes(x = phenotype, y = anno), size = 8,shape = 1 )+
scale_color_gradient2(midpoint=0, limit = c(-cor_limit, cor_limit) , low="blue", mid="white",
high="red", space ="Lab" ) +
theme_minimal() + # minimal theme
theme(text = element_text(face = "bold", size = 14),panel.grid = element_blank(),
axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),
axis.title.x = element_blank(),
axis.title.y = element_blank(),
axis.text = element_text(size = 16, color = "black"))
}
sessionInfo()