def run_nrpspks_specific_hmmer(seq_record, withinclustergenes, pksnrpsvars):
nrpspksfasta = utils.get_specific_multifasta(withinclustergenes)
#Analyse for abMotifs
abmotif_opts = ["-E", "0.25"]
abmotif_results = utils.run_hmmscan(
utils.get_full_path(__file__, "abmotifs.hmm"), nrpspksfasta,
abmotif_opts)
mhmmlengthsdict = utils.hmmlengths(
utils.get_full_path(__file__, "abmotifs.hmm"))
pksnrpsvars.motifdict = parse_hmmscan_results(abmotif_results,
mhmmlengthsdict)
#Analyse for C/A/PCP/E/KS/AT/ATd/DH/KR/ER/ACP/TE/TD/COM/Docking/MT/CAL domains
nrpspksdomain_opts = ["--cut_tc"]
nrpspksdomain_results = utils.run_hmmscan(
utils.get_full_path(__file__, "nrpspksdomains.hmm"), nrpspksfasta,
nrpspksdomain_opts)
hmmlengthsdict = utils.hmmlengths(
utils.get_full_path(__file__, "nrpspksdomains.hmm"))
pksnrpsvars.domaindict = parse_hmmscan_results(nrpspksdomain_results,
hmmlengthsdict)
filter_nonterminal_docking_domains(seq_record, pksnrpsvars)
#Analyse KS domains & PKS/NRPS protein domain composition to detect NRPS/PKS types
kshmmlengthsdict = utils.hmmlengths(
utils.get_full_path(__file__, "ksdomains.hmm"))
ksdomain_results = utils.run_hmmscan(
utils.get_full_path(__file__, "ksdomains.hmm"), nrpspksfasta,
nrpspksdomain_opts)
pksnrpsvars.ksdomaindict = parse_hmmscan_results(ksdomain_results,
kshmmlengthsdict)
def run(seq_record, options):
"run hmmsearch against PFAM for all CDS features"
if 'pfamdir' not in options:
options.pfamdir = utils.get_full_path(__file__, '')
query_sequence = utils.get_multifasta(seq_record)
target_hmmfile = path.join(options.pfamdir, 'Pfam-A.hmm')
logging.info('Running whole-genome pfam search')
if options.skip_cleanup:
results_file = path.join(options.full_outputfolder_path, 'fullhmmer.txt')
if path.exists(results_file):
results = list(SearchIO.parse(results_file, 'hmmer3-text'))
else:
results = utils.run_hmmscan(target_hmmfile, query_sequence, results_file=results_file)
else:
results = utils.run_hmmscan(target_hmmfile, query_sequence)
_annotate(seq_record, options, results)
def run(seq_record, options):
"run hmmsearch against PFAM for all CDS features"
if 'pfamdir' not in options:
options.pfamdir = utils.get_full_path(__file__, '')
query_sequence = utils.get_multifasta(seq_record)
target_hmmfile = path.join(options.pfamdir, 'Pfam-A.hmm')
logging.info('Running whole-genome pfam search')
results = utils.run_hmmscan(target_hmmfile, query_sequence)
_annotate(seq_record, options, results)
def test_run_hmmscan(self):
"Test utils.run_hmmscan()"
mock('Bio.SearchIO.parse', tracker=self.tt, returns=['mock result'])
mock('utils.execute', tracker=self.tt, returns=('output', 'error', 0))
expected = r""" Called utils.execute(
['hmmscan', '--cpu', '2', '--nobias', 'fake.hmm', '-'],
input='>testinput\nMADEUP')
Called Bio.SearchIO.parse(
<cStringIO.StringI object at ...>,
'hmmer3-text')"""
hits = utils.run_hmmscan('fake.hmm', ">testinput\nMADEUP")
self.assertEqual(len(hits), 1)
hit = hits.pop()
self.assertEqual('mock result', hit)
assert_same_trace(self.tt, expected)
def test_run_hmmscan_write_resultfile(self):
"""Test utils.run_hmmscan() writing a results file"""
mock('Bio.SearchIO.parse', tracker=self.tt, returns=['mock result'])
mock('utils.execute', tracker=self.tt, returns=('output', 'error', 0))
expected = r""" Called utils.execute(
['hmmscan', '--cpu', '2', '--nobias', 'fake.hmm', '-'],
input='>testinput\nMADEUP')
Called Bio.SearchIO.parse(
<cStringIO.StringI object at ...>,
'hmmer3-text')"""
results_file = path.join(self.tmpdir, 'fake_hmmscan_output.txt')
hits = utils.run_hmmscan('fake.hmm',
">testinput\nMADEUP",
results_file=results_file)
self.assertEqual(len(hits), 1)
hit = hits.pop()
self.assertEqual('mock result', hit)
assert_same_trace(self.tt, expected)
self.assertTrue(path.exists(results_file))
self.assertEqual(open(results_file).read(), 'output')
def run_smcog_analysis(seq_record, options):
#run_smcog_analysis(opts, globalvars, geneclustervars, pksnrpscoregenes)
logging.info('Running smCOG analysis')
smcogvars = utils.Storage()
smcogvars.smcogtreedict = {}
smcogvars.smcogdict = {}
geneclustergenes = utils.get_withincluster_cds_features(seq_record)
pksnrpscoregenes = utils.get_pksnrps_cds_features(seq_record)
logging.info("Performing smCOG analysis")
smcogs_fasta = utils.get_specific_multifasta(geneclustergenes)
smcogs_opts = ["-E", "1E-6"]
smcogs_results = utils.run_hmmscan(utils.get_full_path(__file__, "smcogs.hmm"), smcogs_fasta, smcogs_opts)
hmmlengthsdict = utils.hmmlengths(utils.get_full_path(__file__, "smcogs.hmm"))
smcogvars.smcogdict = parse_hmmscan_results(smcogs_results, hmmlengthsdict)
#Write output
options.smcogsfolder = path.abspath(path.join(options.outputfoldername, "smcogs"))
if not os.path.exists(options.smcogsfolder):
os.mkdir(options.smcogsfolder)
originaldir = os.getcwd()
os.chdir(options.smcogsfolder)
smcogfile = open("smcogs.txt","w")
pksnrpscoregenenames = [utils.get_gene_id(feature) for feature in pksnrpscoregenes]
for feature in geneclustergenes:
k = utils.get_gene_id(feature)
if k not in pksnrpscoregenenames:
if smcogvars.smcogdict.has_key(k):
l = smcogvars.smcogdict[k]
smcogfile.write(">> " + k + "\n")
smcogfile.write("name\tstart\tend\te-value\tscore\n")
smcogfile.write("** smCOG hits **\n")
for i in l:
smcogfile.write(str(i[0]) + "\t" + str(i[1]) + "\t" + str(i[2]) + "\t" + str(i[3]) + "\t" + str(i[4]) + "\n")
smcogfile.write("\n\n")
smcogfile.close()
#smCOG phylogenetic tree construction
logging.info("Calculating and drawing phylogenetic trees of cluster genes "
"with smCOG members")
with TemporaryDirectory(change=True):
smcoganalysisgenes = []
for feature in geneclustergenes:
k = utils.get_gene_id(feature)
if k not in pksnrpscoregenenames:
smcoganalysisgenes.append(feature)
smcogsets = []
equalpartsizes = int(len(smcoganalysisgenes)/options.cpus)
for i in range(options.cpus):
if i == 0:
geneslist = smcoganalysisgenes[:equalpartsizes]
elif i == (options.cpus - 1):
geneslist = smcoganalysisgenes[(i*equalpartsizes):]
else:
geneslist = smcoganalysisgenes[(i*equalpartsizes):((i+1)*equalpartsizes)]
smcogsets.append(geneslist)
processes = []
z = 0
for k in smcogsets:
processes.append(Process(target=smcog_analysis,
args=[k, z, seq_record,
smcogvars.smcogdict, options.smcogsfolder]))
z += 1
for k in processes:
k.start()
time.sleep(1)
while True:
processrunning = "n"
for k in processes:
if k.is_alive():
processrunning = "y"
if processrunning == "y":
time.sleep(5)
else:
break
for k in processes:
k.join()
os.chdir(options.smcogsfolder)
dircontents = os.listdir(os.getcwd())
for k in dircontents:
if ".png" in k:
tag = k.split(".png")[0]
smcogvars.smcogtreedict[tag] = tag + ".png"
os.chdir(originaldir)
_annotate(geneclustergenes, smcogvars, options)