Parses output from Cufflinks' cuffdiff averaging two samples. Outputs tab-delimited csv files (log2 transformed and not).

cutdiff_output_parser.py

#!/usr/bin/python

from sys import argv
import csv,StringIO
import math

expr = open(argv[1], 'r')

output = open(argv[1] + '.bed','w')
output_log = open(argv[1] + '_log2.bed','w')

for i, line in enumerate(expr):
    if i == 0:
		print "Doing...."
	else:
		row = line.split('\t')
		gene_id = row[1]
		locus = row[3]
		value_1 = row[7]
		value_2 = row[8]
		ave = ((float(value_1) + float(value_2)) / 2)
		ave_log = (math.log(1+(float(value_1) + float(value_2)) / 2,2))
		
		chrom, coord = locus.split(':')
		start, end = coord.split('-')
		
		out = [chrom, start, end, gene_id, ave]
		out_log = [chrom, start, end, gene_id, ave_log]
		
	wr = csv.writer(output, delimiter='\t')
	wr.writerow(out)

output.close()

print """Done.

Created .bed files in current directory.
"""