BedInverseExons

When you export exons as bed with the table browser, they are numbered 0,1,2,3... by default, but always 5' to 3'. This is uncommon for biologists, they usually call the most upstream one exon 1. This scripts inverses the numbers.

#!/usr/bin/python 

# input: bed file with exons as exported from ucsc
# output: bed file with exons numbers reversed if strand="-"

from sys import *
    
f = open(argv[1], "r")
maxExon = {}
for l in f:
    fs = l.split()
    (gene,dummy,exon) = fs[3].split("_")
    maxExon[gene] = int(exon)
    
f = open(argv[1], "r")
    
for l in f:
    fs = l.split()
    if fs[5]=="-":
        (gene,dummy,exon) = fs[3].split("_")
        name = gene + "_exon_" + str(maxExon[gene] - int(exon))
        print "\t".join([fs[0], fs[1], fs[2],name,fs[4],fs[5]])
    else:
        print l,