setwd("~/Desktop/ENCODE")
#read metadata
meta<-read.delim("metadata.tsv")
head(meta)

#remove mock data
mock<-grep("mock",meta$Experiment.target)
meta<-meta[-mock,]
head(meta)

#select required column only
meta1<-meta[,c(2,3,7,17,30,42,43)]
head(meta1)

#change names
meta1$File.format<-sub("bed narrowPeak","bed",meta1$File.format)
meta1$Output.type<-sub("minus strand signal of unique reads","minus",meta1$Output.type)
meta1$Output.type<-sub("plus strand signal of unique reads","plus",meta1$Output.type)
meta1$Experiment.target<-sub("-human","",meta1$Experiment.target)
head(meta1)

#select bigWig
bigWig<-subset(meta1,meta1$File.format=="bigWig")
head(bigWig)

#select bed
bed<-subset(meta1,meta1$File.format=="bed")
head(bed)

#create bed download command
bed$download<-paste("wget ",bed$File.download.URL," -O ",bed$Biosample.term.name,"_",bed$Experiment.target,"_",bed$Biological.replicate.s.,".bed.gz",sep="")
head(bed)

#create bigWig download command
bigWig$download<-paste("wget ",bed$File.download.URL," -O ",bed$Biosample.term.name,"_",bigWig$Experiment.target,"_",bigWig$Output.type,"_",bigWig$Biological.replicate.s.,".bigWig",sep="")
head(bigWig)

#select genome assembly
bigWig_hg19<-subset(bigWig,bigWig$Assembly=="hg19")
bigWig_GRCh38<-subset(bigWig,bigWig$Assembly=="GRCh38")
bed_hg19<-subset(bed,bed$Assembly=="hg19")
bed_GRCh38<-subset(bed,bed$Assembly=="GRCh38")
head(bigWig_hg19)
head(bigWig_GRCh38)
head(bed_hg19)
head(bed_GRCh38)

#create download command file
write(bigWig_hg19$download,"download_bigWig_hg19.txt")
write(bigWig_GRCh38$download,"download_bigWig_GRCh38.txt")
write(bed_hg19$download,"download_bed_hg19.txt")
write(bed_GRCh38$download,"download_bed_GRCh38.txt")