def perform_hmmer(fasta, query_profiles, pfam, session):
"""Main of running a hmmer search
Args:
database_pfam: String, Path to pfam db
database: String, Path to seqeunce db, in fasta or gbk format
query_profiles: List, Pfam profiles needed to be searched
Returns:
hit_res: List of class objects with the hits
"""
LOG.info("Starting hmmer search")
# Make sure we can find hmmfetch and hmmsearch on PATH
helpers.get_program_path(["hmmfetch", "hmmsearch"])
# Find Pfam database (.dat and .hmm)
hmm, dat = check_pfam_db(pfam)
# Find real Pfam accessions
session.queries = get_full_accession_number(dat, query_profiles)
if not session.queries:
LOG.error("No valid profiles could be selected")
return
# Extract HMM profiles from database
query = fetch_profiles(hmm, dat, session.queries)
# Run search
results = run_hmmsearch(hmm, fasta, query)
# Parse results and return
return parse_hmmer_output(results)
def preform_hmmer(
database,
query_profiles,
database_pfam,
):
"""Main of running a hmmer search
Args:
database_pfam: String, Path to pfam db
database: String, Path to seqeunce db, in fasta or gbk format
query_profiles: List, Pfam profiles needed to be searched
Returns:
hit_res: List of class objects with the hits
"""
#1. Check if program exist else give error message and stop program
helpers.get_program_path(["hmmfetch", "hmmsearch"])
LOG.info("Starting hmmer search")
#2. run check_pfam_d
check_pfam_db(database_pfam)
#3. get_full_acc_number and run hmmfetch
ls_keys = fetch_profiles(database_pfam, query_profiles)
#4. run hmmsearch
ls_res = run_hmmsearch(database_pfam, database, ls_keys)
#5. Parse hmm output, needs to be the same as blast output
hit_res = parse_hmmer_output(ls_res)
return hit_res
def test_get_program_path_not_found(monkeypatch):
def return_none(alias):
return
monkeypatch.setattr(shutil, "which", return_none)
with pytest.raises(ValueError):
helpers.get_program_path(["alias"])
def test_get_program_path(monkeypatch):
def return_path(alias):
return "test_path"
monkeypatch.setattr(shutil, "which", return_path)
assert helpers.get_program_path(["alias"]) == "test_path"
def diamond(fasta,
database,
max_evalue=0.01,
min_identity=30,
min_coverage=50,
cpus=1):
"""Launch a local DIAMOND search against a database.
Arguments:
fasta (str): Path to FASTA format query file
database (str): Path to DIAMOND database generated with cblaster makedb
max_evalue (float): Maximum e-value threshold
min_identity (float): Minimum identity (%) cutoff
min_coverage (float): Minimum coverage (%) cutoff
cpus (int): Number of CPU threads for DIAMOND to use
Returns:
list: Rows from DIAMOND search result table (split by newline)
"""
diamond = helpers.get_program_path(["diamond", "diamond-aligner"])
LOG.debug("diamond path: %s", diamond)
parameters = {
"args": [diamond, "blastp"],
"--query":
fasta,
"--db":
database,
"--id":
str(min_identity),
"--evalue":
str(max_evalue),
"--outfmt": [
"6",
"qseqid",
"sseqid",
"pident",
"qcovhsp",
"evalue",
"bitscore",
],
"--threads":
str(cpus),
"--query-cover":
str(min_coverage),
"--max-hsps":
"1",
}
command = helpers.form_command(parameters)
LOG.debug("Parameters: %s", command)
results = subprocess.run(command,
stderr=subprocess.DEVNULL,
stdout=subprocess.PIPE,
check=True)
return results.stdout.decode().split("\n")
def diamond_makedb(fasta, name):
"""Builds a DIAMOND database from JSON.
Args:
fasta (str): Path to FASTA file containing protein sequences.
name (str): Name for DIAMOND database.
"""
diamond = helpers.get_program_path(["diamond", "diamond-aligner"])
subprocess.run(
[diamond, "makedb", "--in", fasta, "--db", name],
stdout=subprocess.DEVNULL,
stderr=subprocess.DEVNULL,
)