def assign(self):
# Prepare #
self.tmp_dir = tempfile.mkdtemp() + '/'
os.symlink(self.fasta.path, self.tmp_dir + 'reads.fasta')
# Run #
start_time = time.asctime()
sh.mothur("#classify.seqs(fasta=%s, reference=%s, taxonomy=%s, cutoff=%i, processors=8, probs=F)" %
(self.tmp_dir + 'reads.fasta', train_db_path, train_tax_path, 80))
process_log_file('mothur.classify.seqs.logfile', self.base_dir, start_time)
def assign(self):
# Prepare #
self.tmp_dir = tempfile.mkdtemp() + '/'
os.symlink(self.fasta.path, self.tmp_dir + 'reads.fasta')
# Run #
start_time = time.asctime()
sh.mothur("#classify.seqs(fasta=%s, reference=%s, taxonomy=%s, cutoff=%i, processors=8, probs=F)" %
(self.tmp_dir + 'reads.fasta', train_db_path, train_tax_path, 80))
process_log_file('mothur.classify.seqs.logfile', self.base_dir, start_time)
def align(self, ref_path):
# Run it #
sh.mothur("#align.seqs(candidate=%s, template=%s, search=blast, flip=false, processors=8);" % (self.path, ref_path))
# Move things #
shutil.move(self.path[:-6] + '.align', self.aligned_path)
shutil.move(self.path[:-6] + '.align.report', self.report_path)
shutil.move(self.path[:-6] + '.flip.accnos', self.accnos_path)
# Clean up #
if os.path.exists('formatdb.log'): os.remove('formatdb.log')
if os.path.exists('error.log') and os.path.getsize('error.log') == 0: os.remove('error.log')
for p in sh.glob('mothur.*.logfile'): os.remove(p)
def template_align(self, ref_path):
# Run it #
sh.mothur("#align.seqs(candidate=%s, template=%s, search=blast, flip=false, processors=8);" % (self.path, ref_path))
# Move things #
shutil.move(self.path[:-6] + '.align', self.aligned_path)
shutil.move(self.path[:-6] + '.align.report', self.report_path)
shutil.move(self.path[:-6] + '.flip.accnos', self.accnos_path)
# Clean up #
if os.path.exists('formatdb.log'): os.remove('formatdb.log')
if os.path.exists('error.log') and os.path.getsize('error.log') == 0: os.remove('error.log')
for p in sh.glob('mothur.*.logfile'): os.remove(p)
def align_mothur(self):
"""Step 1 with mothur"""
# Run it #
sh.mothur("#align.seqs(candidate=%s, template=%s, search=kmer, flip=false, processors=%s);" \
% (self.tax.centers, reference, 16))
# Move things #
shutil.move(self.tax.centers.prefix_path + '.align', self.mothur_aligned)
shutil.move(self.tax.centers.prefix_path + '.align.report', self.p.mothur_report)
path = self.tax.centers.prefix_path + '.flip.accnos'
if os.path.exists(path): shutil.move(path, self.p.mothur_accnos)
# Clean up #
for p in glob.glob('mothur.*.logfile'): os.remove(p)
def align_mothur(self):
"""Step 1 with mothur"""
# Run it #
sh.mothur("#align.seqs(candidate=%s, template=%s, search=kmer, flip=false, processors=%s);" \
% (self.tax.centers, reference, 16))
# Move things #
shutil.move(self.tax.centers.prefix_path + '.align',
self.mothur_aligned)
shutil.move(self.tax.centers.prefix_path + '.align.report',
self.p.mothur_report)
path = self.tax.centers.prefix_path + '.flip.accnos'
if os.path.exists(path): shutil.move(path, self.p.mothur_accnos)
# Clean up #
for p in glob.glob('mothur.*.logfile'):
os.remove(p)
def template_align(self, ref_path):
"""We align the sequences in the fasta file with mothur and a template."""
# Run it #
msg = "#align.seqs(candidate=%s, template=%s, search=blast," \
"flip=false, processors=8);"
sh.mothur(msg % (self.path, ref_path))
# Move things #
shutil.move(self.path[:-6] + '.align', self.p.aligned)
shutil.move(self.path[:-6] + '.align.report', self.p.report)
shutil.move(self.path[:-6] + '.flip.accnos', self.p.accnos)
# Clean up #
if os.path.exists('formatdb.log'):
os.remove('formatdb.log')
if os.path.exists('error.log') and os.path.getsize('error.log') == 0:
os.remove('error.log')
for path in sh.glob('mothur.*.logfile'):
os.remove(path)
# Return #
return self.p.aligned
def join(self):
"""Uses pandaseq 2.7 to join the foward and reverse reads together.
See https://github.com/neufeld/pandaseq"""
# Special case for new primers that don't join #
rev_primer_name = self.info['primers']['reverse']['name']
not_joining_primers = ("1132R", "1000R")
if rev_primer_name in not_joining_primers:
print "No overlap special case"
self.trim_and_concat.run()
return
# Special case for primers that highly overlap #
high_overlap_primers = ("806R",)
if rev_primer_name in high_overlap_primers:
print "High overlap special case, using mothur"
result = sh.mothur("#make.contigs(ffastq=%s, rfastq=%s);" % (self.uncomrpessed_pair.fwd, self.uncomrpessed_pair.rev))
if "ERROR" in result.stdout: raise Exception("Mothur didn't run correctly")
# Move things #
#shutil.move(self.tax.centers.prefix_path + '.align', self.mothur_aligned)
#shutil.move(self.tax.centers.prefix_path + '.align.report', self.p.mothur_report)
return
# Default case #
command = 'pandaseq27 -T 1 -f %s -r %s -u %s -F 1> %s 2> %s'
command = command % (self.fwd, self.rev, self.unassembled.path, self.assembled.path, self.assembled.p.out)
shell_call(command) # Because it exits with status 1 https://github.com/neufeld/pandaseq/issues/40
def check(self):
start_time = time.asctime()
sh.mothur("#chimera.uchime(fasta=%s, dereplicate=t, processors=8)" %
(self.p.len_filtered_fasta))
process_log_file('mothur.chimera.uchime.logfile', self.p.orig_dir, start_time)
def check(self):
start_time = time.asctime()
sh.mothur("#chimera.uchime(fasta=%s, dereplicate=t, processors=8)" %
(self.p.len_filtered_fasta))
process_log_file('mothur.chimera.uchime.logfile', self.p.orig_dir, start_time)
