def read_transcripts(genes_fn, seqnames):
''' Parse all the transcripts from a GTF file. '''
stderr.write('parsing gtf ... ')
genes = defaultdict(transcript)
for row in gtf.gtf_file(genes_fn):
if row.seqname not in seqnames: continue
genes[row.attributes['transcript_id']].add_exon(row)
stderr.write('done. ({0} transcripts)\n'.format(len(genes)))
return genes.values()
def read_genes(genes_fn):
'''
For each gene in the GTF file, compute the union of all exons, in sorted
order.
'''
stderr.write('parsing GTF file ... ');
genes = defaultdict(gene)
for row in gtf.gtf_file(genes_fn):
if row.feature != 'exon': continue
if 'gene_id' not in row.attributes: continue
genes[row.attributes['gene_id']].add_row(row)
stderr.write('done. ({0} genes)\n'.format(len(genes)))
for g in genes.itervalues():
g.flatten()
return genes
#!/usr/bin/env python
from gtf import gtf_file
from sys import argv, stdout, stderr, stdin
if len(argv) < 2:
stderr.write('useage: gtf_exons.py genes.gtf')
for row in gtf_file(argv[1]):
if row.feature != 'exon': continue
stdout.write('{seqname}\t{start}\t{end}\t{name}\t{score}\t{strand}\n'.format(
seqname = row.seqname,
start = int(row.start) - 1,
end = row.end,
name = row.attributes['gene_id'],
score = 0,
strand = row.strand))