def run_diamond_on_all_regions(regions: Sequence[secmet.Region],
database: str) -> str:
""" Runs diamond, comparing all features in the given regions to the given database
Arguments:
regions: the regions to use features from
database: the path of the database to compare to
Returns:
diamond's output from stdout
"""
logging.info("Comparing regions to reference database")
extra_args = [
"--compress",
"0",
"--max-target-seqs",
"10000",
"--evalue",
"1e-05",
"--outfmt",
"6", # 6 is blast tabular format, just as in blastp
]
with NamedTemporaryFile() as temp_file:
write_fastas_with_all_genes(regions, temp_file.name)
stdout = subprocessing.run_diamond_search(temp_file.name,
database,
mode="blastp",
opts=extra_args)
return stdout
def find_diamond_matches(
record: Record,
database: str) -> Tuple[HitsByCDS, HitsByReferenceName]:
""" Runs diamond, comparing all features in the record to the given database
Arguments:
record: the record to use as a query
database: the path of the database to compare to
Returns:
a tuple of
a dictionary mapping CDSFeature to
a dictionary mapping reference CDS numeric ID to
a list of Hits for that reference
a dictionary mapping reference region name to
a dictionary mapping reference CDS numeric ID to
a list of Hits for that reference
"""
logging.info("Comparing regions to reference database")
extra_args = [
"--compress",
"0",
"--max-target-seqs",
"10000",
"--evalue",
"1e-05",
"--outfmt",
"6", # 6 is blast tabular format, just as in blastp
]
features = record.get_cds_features_within_regions()
with NamedTemporaryFile() as temp_file:
temp_file.write(
fasta.get_fasta_from_features(features,
numeric_names=True).encode())
temp_file.flush()
raw = subprocessing.run_diamond_search(temp_file.name,
database,
mode="blastp",
opts=extra_args)
return blast_parse(raw, dict(enumerate(features)))