library(reshape2)

library(ggplot2)

setwd("D:/BaiduYunDownload/01_¿ÆÑй¤×÷/03_Machine learning in proteomics/Metabolism")

rm(list=ls())

table=read.csv("786-QE-GLN-pos-final.csv",header=T,sep=";")

compoundlist=colnames(table)

compound=table[,"Compound"]

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

compoundname=as.character(compound[i])

A=strsplit(compoundname, "\\*")

formula=unlist(A)[2]

B=unlist(strsplit(formula,""))[1]

com=NULL

if(B!='m') #It is the first in such chemical

{

com=unlist(A)[1]

j=i

kn=NULL

nc=NULL

labeling=NULL

while(unlist(strsplit(as.character(compound[j]),"\\*"))[1]==com)

{

labeling=append(labeling,unlist(strsplit(as.character(compound[j]),"\\*"))[2],after=length(labeling))

kn=append(kn,table[,"avrage.kn"][j],after=length(kn))

nc=append(nc,table[,"avrage.nc"][j],after=length(nc))

j=j+1

}

labeling[1]='m0'

picdata=data.frame(com=labeling,HSP60.NC=nc,HSP60.KN=kn)

picdata=melt(picdata,id.vars = "com",varible.name="type",value.name = "abundance")

pic=ggplot(picdata,aes(com,abundance,fill=variable))+geom_bar(stat="identity",position="dodge")+labs(y='Intensity',title=com)

savename=gsub("/", "_", com, fixed = TRUE)

file=paste(savename,".jpg",sep='')

ggsave(sprintf(file),pic,dpi=1200)

}

}