infer GTF genes, for biostars #99287

gffutils-gtf.py

import gffutils

# Import the GTF file into a sqlite3 database.
# This only ever has to be done once.
db = gffutils.create_db("example.gtf", dbfn='example.gtf.db')

# In other scripts, you can connect to the database like this:
db = gffutils.FeatureDB('example.gtf.db')

# Note that gene and transcript have been inferred
print list(db.featuretypes())
# ['CDS', 'exon', 'gene', 'start_codon', 'stop_codon', 'transcript']

# Here's how to write genes out to file
with open('example_genes.gtf', 'w') as fout:
  for gene in db.features_of_type('gene'):
    fout.write(str(gene) + '\n')