def test_sort_ids(self):
"""sort_ids sorts by abundance"""
mapping = {"1":["0","2","5","6"],
"3":[],
"4":[],
"11":[1,2,3,4,5,6,7,8,9],
"8":["7"]}
self.assertEqual(sort_ids(["1","3","4","8","11"], mapping),\
["11","1","8","4","3"])
def test_sort_ids(self):
"""sort_ids sorts by abundance"""
mapping = {"1":["0","2","5","6"],
"3":[],
"4":[],
"11":[1,2,3,4,5,6,7,8,9],
"8":["7"]}
self.assertEqual(sort_ids(["1","3","4","8","11"], mapping),\
["11","1","8","4","3"])
def combine_mappings(fasta_fh, mapping_fh, denoised_seqs_fh,
otu_picker_otu_map_fh, out_dir):
"""Combine denoiser and OTU picker mapping file, replace flowgram IDs.
fasta_fh: a fasta file with labels as produced by Qiime's split_libraries.py
used to replace flowgram id with the unique se_sample_id
mapping_fh: The cluster mapping from the denoiser.py
denoised_seqs_fh: the Fasta output files from denoiser.py
otu_picker_map_fh: cluster map from otu picker on denoised_seqs_fh
out_dir: output directory
"""
# read in mapping from split_library file
labels = imap(lambda a_b: a_b[0], MinimalFastaParser(fasta_fh))
# mapping from seq_id to sample_id
sample_id_mapping = extract_read_to_sample_mapping(labels)
denoiser_mapping = read_denoiser_mapping(mapping_fh)
# read in cd_hit otu map
# and write out combined otu_picker+denoiser map
otu_fh = open(out_dir + "/denoised_otu_map.txt", "w")
for otu_line in otu_picker_otu_map_fh:
otu_split = otu_line.split()
otu = otu_split[0]
ids = otu_split[1:]
get_sample_id = sample_id_mapping.get
# concat lists
# make sure the biggest one is first for pick_repr
all_ids = sort_ids(ids, denoiser_mapping)
all_ids.extend(sum([denoiser_mapping[id] for id in ids], []))
try:
otu_fh.write("%s\t" % otu +
"\t".join(map(get_sample_id, all_ids)) + "\n")
except TypeError:
# get returns Null if denoiser_mapping id not present in
# sample_id_mapping
print "Found id in denoiser output, which was not found in split_libraries " +\
"output FASTA file. Wrong file?"
exit()
fasta_out_fh = open(out_dir + "/denoised_all.fasta", "w")
for label, seq in MinimalFastaParser(denoised_seqs_fh):
id = label.split()[0]
newlabel = "%s %s" % (sample_id_mapping[id], id)
fasta_out_fh.write(Sequence(name=newlabel, seq=seq).toFasta() + "\n")
def combine_mappings(fasta_fh, mapping_fh, denoised_seqs_fh,
otu_picker_otu_map_fh, out_dir):
"""Combine denoiser and OTU picker mapping file, replace flowgram IDs.
fasta_fh: a fasta file with labels as produced by Qiime's split_libraries.py
used to replace flowgram id with the unique se_sample_id
mapping_fh: The cluster mapping from the denoiser.py
denoised_seqs_fh: the Fasta output files from denoiser.py
otu_picker_map_fh: cluster map from otu picker on denoised_seqs_fh
out_dir: output directory
"""
# read in mapping from split_library file
labels = imap(lambda a_b: a_b[0], parse_fasta(fasta_fh))
# mapping from seq_id to sample_id
sample_id_mapping = extract_read_to_sample_mapping(labels)
denoiser_mapping = read_denoiser_mapping(mapping_fh)
# read in cd_hit otu map
# and write out combined otu_picker+denoiser map
otu_fh = open(out_dir + "/denoised_otu_map.txt", "w")
for otu_line in otu_picker_otu_map_fh:
otu_split = otu_line.split()
otu = otu_split[0]
ids = otu_split[1:]
get_sample_id = sample_id_mapping.get
# concat lists
# make sure the biggest one is first for pick_repr
all_ids = sort_ids(ids, denoiser_mapping)
all_ids.extend(sum([denoiser_mapping[id] for id in ids], []))
try:
otu_fh.write("%s\t" % otu +
"\t".join(map(get_sample_id, all_ids)) + "\n")
except TypeError:
# get returns Null if denoiser_mapping id not present in
# sample_id_mapping
print "Found id in denoiser output, which was not found in split_libraries " +\
"output FASTA file. Wrong file?"
exit()
fasta_out_fh = open(out_dir + "/denoised_all.fasta", "w")
for label, seq in parse_fasta(denoised_seqs_fh):
id = label.split()[0]
newlabel = "%s %s" % (sample_id_mapping[id], id)
fasta_out_fh.write(BiologicalSequence(seq, id=newlabel).to_fasta())