def phylogeny(self) :
num_sequences = self.__count(self.options['cluster-fasta'])
p = Pagan()
if self.options['subset'] :
self.seqdb = self.__read_fasta(self.options['cluster-fasta'], include=self.options['subset'])
self.options['cluster-fasta'] = self.__fasta(self.options['cluster-fasta'] + '.subset', self.seqdb.keys())
self.log.info("read %d cluster centroids using subset(%s)" % (len(self.seqdb), self.options['subset']))
if not self.options['silva-fasta'] :
self.log.info("aligning %s sequences with PAGAN ..." % (num_sequences))
alignment,tree,xmlfile = p.phylogenetic_alignment(self.options['cluster-fasta'])
else :
self.log.info("aligning %s sequences with PAGAN against SILVA ..." % (num_sequences))
alignment,tree,xmlfile = p.silva_phylogenetic_alignment(self.options['silva-fasta'],
self.options['silva-tree'],
self.options['cluster-fasta'])
os.rename(alignment, self.options['phylogeny-fasta'])
os.rename(tree, self.options['phylogeny-tree'])
os.rename(xmlfile, self.options['phylogeny-xml'])
self.log.info("created %s" % self.options['phylogeny-fasta'])
self.log.info("created %s" % self.options['phylogeny-tree'])
if xmlfile :
self.log.info("created %s" % self.options['phylogeny-xml'])
return 0
def alignment_similarity(self, seq1, seq2, homopolymer_correction) :
# write out
f = open(System.tempfilename(ext='cluster'), 'w')
print >> f, seq1.fasta()
print >> f, seq2.fasta()
f.close()
# align
aligned = []
if homopolymer_correction :
fq = Pagan().get_454_alignment(f.name)
else :
fq = Pagan().get_alignment(f.name)
fq.open()
for seq in fq :
if seq.id == ">consensus" :
continue
aligned.append(seq.sequence)
fq.close()
# delete tmp files
os.remove(f.name)
os.remove(fq.get_filename())
# if things are really dissimilar they do not align
# so just give up here for this cluster
if len(aligned) != 2 :
return 0.0
return self.distance2(aligned, homopolymer_correction)