def _run(self) :
global DEBUG
self.query_fname = tmpfasta_orfs(self._queries, strand=True)
#self.query_fname = tmpfasta(self._queries)
self.out_fname = tmpfile()
self.alignment_fname = tmpfasta(self._alignment) # tmpfasta_kill_n(self._alignment)
self.tree_fname = tmpfile(self._tree) if self._tree else None
start_time = time.time()
result = self.pagan.run(self.query_fname,
self.out_fname,
self.alignment_fname,
self.tree_fname,
self.identity,
self.overlap)
elapsed_time = time.time() - start_time
q_count, q_sum, q_min, q_max, q_mean, q_sd = fasta_stats(self.query_fname)
a_count, a_sum, a_min, a_max, a_mean, a_sd = fasta_stats(self.alignment_fname)
if DEBUG :
threadsafe_io('pagan_stats.txt', "%s %d %d %d %d %d %.3f %.3f %d %d %d %d %.3f %.3f %d" % \
(self._genefamily, result, \
q_count, q_sum, q_min, q_max, q_mean, q_sd, \
a_count, a_sum, a_min, a_max, a_mean, a_sd, \
elapsed_time))
return result
def _run(self) :
global DEBUG
self.query_fname = tmpfasta(self.queries)
self.out_fname = tmpfile()
result = self.blastx.run(self.query_fname, self.database, self.out_fname)
q = dict([ (q.id, len(q)) for q in self.input ])
if DEBUG :
for br in self.results :
threadsafe_io('blastx_stats.txt', "%s %s %.3f %d %d %d %d %d %.3e %d %.3f" % \
(br.qseqid,
br.sseqid,
br.pident,
br.length,
br.qstart,
br.qend,
br.sstart,
br.send,
br.evalue,
q[br.qseqid],
((br.pident / 100.0) * (max(br.qstart, br.qend) - min(br.qstart, br.qend))) / float(q[br.qseqid])))
return result
def job_callback(self, job) :
self.log.debug("callback from %s: %s + %s" % (str(job), job.genefamily, ','.join([ i.id for i in job.input ])))
self.log.debug("protein alignment file = %s" % (job.protein_alignment))
if self.db.nucleotide :
self.log.debug("nucleotide alignment file = %s" % (job.nucleotide_alignment))
self._progress()
if job.success() :
dst = tmpfile(directory=self.directory, suffix='.protein')
dst_base = basename(dst)[:-8]
self.log.debug("cp %s %s" % (job.protein_alignment, dst))
shutil.copyfile(job.protein_alignment, dst)
shutil.copyfile(job.query_fname, dst[:-8] + '.queries')
shutil.copyfile(job.alignment_fname, dst[:-8] + '.reference')
if job.tree_fname :
shutil.copyfile(job.tree_fname, dst[:-8] + '.tree')
if self.db.nucleotide :
self.log.debug("cp %s %s" % (job.nucleotide_alignment, dst[:-8] + '.nucleotide'))
shutil.copyfile(job.nucleotide_alignment, dst[:-8] + '.nucleotide')
self.info.put_genefamily2filename(job.genefamily, dst_base)
else :
self.info.put_genefamily2filename(job.genefamily)
def extract_all(self) :
fname = tmpfile()
with open(fname, 'w') as f :
for gf in self.data :
for g in self.data[gf] :
print >> f, g.format('protein' if self.nucleotide else 'fasta').rstrip()
return fname
def _write_to_archive(self, data, zfile, zname) :
fname = tmpfile()
f = open(fname, 'w')
f.write(json.dumps(data))
f.close()
zfile.write(fname, arcname=zname)
os.remove(fname)
