#setup
library(data.table)
library(ggplot2)
setwd("~/Desktop/ENCODE")
data=fread("shRNA_RNAseq_FPKM.txt")
setkey(data,"gene_id")
meta=read.delim("shRNA_RNAseq_meta.txt")

while (TRUE){
  
 #input gene of interest ID
 gi=readline("enter GENCODEv24 ID: ") # set the ENCODEv24_ID for the gene of interest
 
 #create data table for ggplot2
  FPKM_gi<-data[gi] # select FPKM data for the gene of interest
  FPKM_gi<-as.data.table(t(FPKM_gi),keep.rownames=TRUE) #exchange the row and column
  FPKM_gi<-FPKM_gi[-1,] #delete row containing extra information
  FPKM_meta=merge(meta,FPKM_gi,by.x="File.accession",by.y="rn") #assign metadata
  FPKM_meta$V1<-as.numeric(FPKM_meta$V1) #change data type for ggplot2
  
  FPKM_HepG2=FPKM_meta[FPKM_meta$Biosample.term.name=="HepG2",]
  FPKM_K562=FPKM_meta[FPKM_meta$Biosample.term.name=="K562",]
  
  #plot data HepG2
  ggplot(FPKM_HepG2,aes(x=reorder(x=Experiment.target,X=V1),y=V1)) +
    geom_boxplot() +
    theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
    ylab(paste("FPKM",gi)) +
    xlab("KD gene")
  ggsave(filename=paste("plot_HepG2_",gi,".png",sep=""),height=5,width=nrow(FPKM_HepG2)/15,limitsize=FALSE)
  
  #plot data K562
  ggplot(FPKM_K562,aes(x=reorder(x=Experiment.target,X=V1),y=V1)) +
    geom_boxplot() +
    theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
    ylab(paste("FPKM",gi)) +
    xlab("KD gene")
  ggsave(filename=paste("plot_K562_",gi,".png",sep=""),height=5,width=nrow(FPKM_K562)/15,limitsize=FALSE)

  #continue?
  x=readline("continue? (y/n): ")  
  if (x=="n") {
    break
  }
  }