如何获得FPKM/RPKM计算需要的基因长度（考虑exon之间的overlap）

install.packages("data.table")
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("rtracklayer")
BiocManager::install("IRanges")

library(data.table)
library("IRanges")
require("rtracklayer")

hg19 <- readGFF("hg19.gencodev27.gtf")
anno <- setDT(hg19)
anno <- anno[type=="exon",]
setnames(anno,c("seqid","start","end","gene_name","exon_number"),c("Chr","ExonStart","ExonEnd","Gene","Exon_number"))
#mkdir bin and mean by bin
Exon_region <- unique(anno[,.(Chr,ExonStart,ExonEnd,Exon_number,Gene)])
Exon_region <- Exon_region[,{x <- IRanges(ExonStart,ExonEnd);y <- reduce(x); list(ExonStart=y@start,ExonEnd=y@start+y@width-1)},by=.(Gene,Chr)]
Exon_region[,Exon_num:=1:.N,by=Gene]
Exon_region <- Exon_region[,.(Chr,ExonStart,ExonEnd,Exon_num,Gene)]
Exon_len <- Exon_region[,.(ExonLen = ExonEnd - ExonStart + 1),by=.(Exon_num,Gene)]
gene_len <- Exon_len[,.(Length = sum(ExonLen)),by=Gene]


# write out
fwrite(Exon_region,file="All_hg19gene_exon.bed", sep = "\t", col.names = T)
fwrite(gene_len, file = "All_hg19gene_len.txt", sep = "\t", col.names = T)
~