Flattened CCDS

flatten_transcripts.py

import pybedtools as pbt
import sys

def merge_gene(lines):
	tmp = pbt.BedTool(lines, from_string=True).merge(nms=True)
	print tmp

gene_lines = ''
curr_gene = None
prev_gene = None
for line in sys.stdin:
	fields = line.strip().split()
	curr_gene = fields[3]
	if curr_gene != prev_gene and prev_gene is not None:
		merge_gene(gene_lines)
		gene_lines = line
	else:
		gene_lines += line
	prev_gene = curr_gene

merge_gene(gene_lines)

gistfile1.txt

awk '{OFS="\t"; print $1,$2,$3,$5"_"$6,$7,$4}' hgncgenes.exons.bed | awk 'NR>1' > temp
cat temp | python flatten_transcripts.py | awk 'length($0) > 0' | tr ";" "\t" | cut -f 1-4 > flattened_genes.bed