def main():
# configure the logger here
# default to error, --verbose flag will set to INFO
log = logging.getLogger(__name__)
log.setLevel(logging.ERROR)
log_formatter = logging.Formatter(
'CRISPR_SCORE|%(levelname)s|: %(message)s')
stream_handler = logging.StreamHandler()
stream_handler.setLevel(logging.ERROR)
stream_handler.setFormatter(log_formatter)
log.addHandler(stream_handler)
# import the custom parser
parser = CRISPRArgumentParser(
description=("Find & profile CRISPR targets in a fasta file"), )
args = parser.parse_args()
if args.verbose:
log.setLevel(logging.INFO)
stream_handler.setLevel(logging.INFO)
# validate files
# fasta must be given
if not file_exists(args.fasta):
log.error("fasta file %s does not exist -- Exiting", args.fasta)
raise SystemExit
# GFF is optional
if args.gff is not None and not file_exists(args.gff):
log.error("gff file %s does not exist -- Exiting", args.gff)
raise SystemExit
# set up data
fasta = read_fasta_file(args.fasta)
# the kmer_spectra is a dictionary mapping kmers to a count score
# the score = (1 / count) * multiplyer, so a kmer which appears once
# has a base score of one, whereas a more common one (5) would have .2
kmer_size = args.seed_end - args.seed_start + 1
kmer_specta = build_kmer_count(fasta, k=kmer_size)
targets = []
# yields a Crispir target seq instance
for target in generate_targets(fasta, target_length=args.length):
log.info("Target found: %s", target)
# filter and score
if filter_target(target,
gc_low=args.gc_low,
gc_high=args.gc_high,
homopolymer_length=args.homopolymer):
# pass the scoring parameters from args Namespace as a dict
target.score = score_target(target, kmer_specta, **args.__dict__)
targets.append(target)
log.info("Filtering/Scoring complete, %i targets passed", len(targets))
if len(targets) == 0:
log.warn("No targets passed filtering, exiting!")
return
# optionally check the gff file
# overlaps are appened to the CrisprTarget genes[] data member
if args.gff:
log.info("Reading in GFF annotations")
gff = read_gff_file(args.gff)
log.info("Found %i genes from %s", len(gff), args.gff)
if len(gff) > 0:
targets = annotate_gene_overlaps(targets, gff)
# write the output as a bed file
write_bed_file(targets, fasta, args.output)
def test_reading_invalid_gff(self):
self.assertTrue(file_exists(self.invalid_gff_file))
gff = read_gff_file(self.invalid_gff_file)
# the gene line is invalid so its skipped
self.assertEqual(len(gff), 0)
def test_reading_valid_gff(self):
self.assertTrue(file_exists(self.valid_gff_file))
gff = read_gff_file(self.valid_gff_file)
self.assertEqual(len(gff), 1)
self.assertEqual(gff["ID=gene1;Name=GENE1"], (1000, 7000))
def test_reading_invalid_fasta(self):
self.assertTrue(file_exists(self.invalid_fasta_file))
self.assertRaises(ValueError, read_fasta_file, self.invalid_fasta_file)
def test_reading_valid_fasta(self):
self.assertTrue(file_exists(self.valid_fasta_file))
rec = read_fasta_file(self.valid_fasta_file)
self.assertEqual(rec.id, "valid_fasta")
def test_file_doesnt_exist(self):
self.assertFalse(file_exists(self.bad_file))