def output_exons_as_gff(self,
base_diff=6,
const_only=False,
cds_only=False):
"""
Output constitutive exons for all genes as GFF.
- const_only: if True, output only constitutive exons
- cds_only: if True, output CDS only exons
"""
exons_type = "exons"
exons_outdir = self.exons_dir
if const_only:
exons_type = "const_exons"
exons_outdir = self.const_exons_dir
if cds_only:
exons_basename = "%s.cds_only.%s.gff" %(self.source, exons_type)
else:
exons_basename = "%s.%s.gff" %(self.source, exons_type)
gff_output_filename = os.path.join(exons_outdir, exons_basename)
print "Outputting exons..."
print " - Exons type: %s" %(exons_type)
print " - Output file: %s" %(gff_output_filename)
print " - CDS only: %s" %(cds_only)
if os.path.isfile(gff_output_filename):
print "%s exists. Skipping.." %(gff_output_filename)
return
# Output a map from genes to constitutive exons
# for convenience
genes_to_exons_fname = os.path.join(exons_outdir,
exons_basename.replace(".gff",
".to_genes.txt"))
gff_out = gff_utils.Writer(open(gff_output_filename, "w"))
rec_type = "exon"
genes_to_exons = []
genes_to_exons_header = ["gene_id", "exons"]
for gene_id, gene in self.genes.iteritems():
if const_only:
# Get only constitutive exons
exons = gene.compute_const_exons(base_diff=base_diff,
cds_only=cds_only)
elif cds_only:
# Get all CDS exons
exons = gene.cds_parts
else:
# Get all exons
exons = gene.parts
exon_labels = [e.label for e in exons]
if len(exon_labels) == 0:
exon_labels = self.na_val
else:
exon_labels = ",".join(exon_labels)
entry = {"gene_id": gene_id,
"exons": exon_labels}
genes_to_exons.append(entry)
# Output constitutive exons to GFF file
GeneModel.output_parts_as_gff(gff_out,
exons,
gene.chrom,
gene.strand,
source=self.source,
rec_type=rec_type,
gene_id=gene_id)
genes_to_exons = pandas.DataFrame(genes_to_exons)
genes_to_exons.to_csv(genes_to_exons_fname,
cols=genes_to_exons_header,
index=False,
sep="\t")
