def make_DF(self):
df = DF(self.name, self.aln_length)
for id in self.fasta_reader.iterkeys():
r = self.fasta_reader[id]
for nt in DF.nucleotides():
# TODO: make find_all_indices iterative to be mem-efficient
seq = r.seq.tostring().replace("U", "T")
positions = find_all_indices(seq, nt)
df.add_to_vec(nt=nt, positions=positions, counts=[1] * len(positions))
# for gapped_pos,nt in enumerate(r.seq):
# df.add_to_vec(nt=nt, positions=[gapped_pos], counts=[1])
return df
def subsample(self, se):
df = DF(self.pyro.name, self.pyro.aln_length)
keys = self.pyro.keys()
for id in random.sample(keys, min(se, len(keys))):
# to prevent "sample larger than population error" use min()
r = self.pyro[id]
for nt in DF.nucleotides():
seq = r.seq.tostring().replace("U", "T")
positions = find_all_indices(seq, nt)
df.add_to_vec(nt=nt, positions=positions, counts=[1] * len(positions))
# for i,ecoli_pos in enumerate(SILVA.Ecoli1542_SILVA100):
# df.add_to_vec(nt=r.seq[ecoli_pos], positions=[i], counts=[1])
return df
