def run_on_record(record: Record, results: Optional[hmmer.HmmerResults],
options: ConfigType) -> hmmer.HmmerResults:
""" Run hmmsearch against PFAM for all CDS features within the record """
if options.fullhmmer_pfamdb_version == "latest":
database_version = pfamdb.find_latest_database_version(
options.database_dir)
else:
database_version = options.fullhmmer_pfamdb_version
if results:
previous_db = pfamdb.get_db_version_from_path(results.database)
# same version requested, so reuse the results
if database_version == previous_db:
return results
else:
logging.debug("Replacing fullhmmer results from %s with %s",
previous_db, database_version)
logging.info('Running whole-genome PFAM search')
database = os.path.join(options.database_dir, 'pfam', database_version,
'Pfam-A.hmm')
return hmmer.run_hmmer(record, record.get_cds_features(), MAX_EVALUE,
MIN_SCORE, database, "fullhmmer")
def run_on_record(record: Record, results: Optional[hmmer.HmmerResults],
options: ConfigType) -> hmmer.HmmerResults:
""" Run hmmsearch against PFAM for all CDS features within the record """
if options.clusterhmmer_pfamdb_version == "latest":
database_version = pfamdb.find_latest_database_version(
options.database_dir)
else:
database_version = options.clusterhmmer_pfamdb_version
if results:
previous_db = pfamdb.get_db_version_from_path(results.database)
# same version requested, so reuse the results
if database_version == previous_db:
return results
logging.debug("Replacing clusterhmmer results from %s with %s",
previous_db, database_version)
logging.info('Running cluster PFAM search')
features = []
for region in record.get_regions():
features.extend(list(region.cds_children))
database = os.path.join(options.database_dir, 'pfam', database_version,
'Pfam-A.hmm')
return hmmer.run_hmmer(record, features, MAX_EVALUE, MIN_SCORE, database,
"clusterhmmer")
def run_on_record(record, results, options) -> hmmer.HmmerResults:
""" Run hmmsearch against PFAM for all CDS features within the record """
if results:
return results
logging.info('Running whole-genome PFAM search')
if options.fullhmmer_pfamdb_version == "latest":
database_version = pfamdb.find_latest_database_version(
options.database_dir)
else:
database_version = options.fullhmmer_pfamdb_version
database = os.path.join(options.database_dir, 'pfam', database_version,
'Pfam-A.hmm')
return hmmer.run_hmmer(record, record.get_cds_features(), MAX_EVALUE,
MIN_SCORE, database, "fullhmmer")
def run_rrefinder(record: Record, bitscore_cutoff: float, min_length: int, database: str) -> RREFinderResults:
"""Run RREFinder on a given record
"""
# Gather all RRE candidates
candidates_by_protocluster, cds_info = gather_rre_candidates(record)
# Run hmmscan per protocluster and gather the hits
if cds_info == {}:
filtered_hits_by_protocluster: Dict[int, List[str]] = {}
filtered_hits_by_cds: Dict[str, List[HmmerHit]] = {}
else:
hmm_results = run_hmmer(record, cds_info.values(), max_evalue=1, min_score=bitscore_cutoff,
database=database, tool='rrefinder', use_cut_tc=False,
filter_overlapping=False)
# Extract the RRE hits
hits_by_cds = extract_rre_hits(hmm_results)
# Filter the hits
filtered_hits_by_cds, filtered_hits_by_protocluster = filter_hits(hits_by_cds, candidates_by_protocluster,
min_length, bitscore_cutoff)
return RREFinderResults(record.id, bitscore_cutoff, min_length,
filtered_hits_by_protocluster, filtered_hits_by_cds)
def run_on_record(record, results, options) -> hmmer.HmmerResults:
""" Run hmmsearch against PFAM for all CDS features within the record """
if results:
return results
logging.info('Running cluster PFAM search')
if options.clusterhmmer_pfamdb_version == "latest":
database_version = pfamdb.find_latest_database_version(
options.database_dir)
else:
database_version = options.clusterhmmer_pfamdb_version
database = os.path.join(options.database_dir, 'pfam', database_version,
'Pfam-A.hmm')
features = []
for cluster in record.get_clusters():
features.extend(list(cluster.cds_children))
return hmmer.run_hmmer(record, features, MAX_EVALUE, MIN_SCORE, database,
"clusterhmmer")
def run_on_record(record: Record, results: Optional[TIGRFamResults],
options: ConfigType) -> TIGRFamResults:
""" Run hmmsearch against TIGRFam for all CDS features within the record """
logging.info('Running TIGRFam search')
if results:
return results
features = []
for region in record.get_regions():
features.extend(list(region.cds_children))
tigr_db = os.path.join(options.database_dir, "tigrfam", "TIGRFam.hmm")
hmmer_results = hmmer.run_hmmer(record,
features,
MAX_EVALUE,
MIN_SCORE,
tigr_db,
"tigrfam",
filter_overlapping=False)
return TIGRFamResults.from_hmmer_results(hmmer_results)
def run_rrefinder(record: Record, bitscore_cutoff: float, min_length: int,
database: str) -> RREFinderResults:
"""Run RREFinder on a given record
"""
# Gather all RRE candidates
candidates_per_protocluster, cds_info = gather_rre_candidates(record)
# Run hmmscan per protocluster and gather the hits
hmm_results = run_hmmer(record,
cds_info.values(),
max_evalue=1,
min_score=bitscore_cutoff,
database=database,
tool='rrefinder',
use_cut_tc=False)
# Extract the RRE hits
hit_info = extract_rre_hits(hmm_results)
# Filter the hits
filtered_hit_info, filtered_hits_per_protocluster = filter_hits(
hit_info, candidates_per_protocluster, min_length, bitscore_cutoff)
# Convert to RREFinderResults object
RRE_results = RREFinderResults(record.id, bitscore_cutoff, min_length,
filtered_hits_per_protocluster,
filtered_hit_info)
return RRE_results