def update_read_dict(read2contig, blast_top_m8, read_dict, accession_dict):
consolidated_dict = read_dict
read2blastm8 = {}
contig2accession = {}
contig2lineage = {}
added_reads = {}
for contig_id, accession_id, _percent_id, _alignment_length, e_value, _bitscore, line in m8.iterate_m8(
blast_top_m8):
contig2accession[contig_id] = (accession_id, line)
contig2lineage[contig_id] = accession_dict[accession_id]
for read_id, contig_id in read2contig.items():
(accession,
m8_line) = contig2accession.get(contig_id, (None, None))
if accession:
(species_taxid, genus_taxid,
family_taxid) = accession_dict[accession]
if consolidated_dict.get(read_id):
consolidated_dict[read_id] += [
contig_id, accession, species_taxid, genus_taxid,
family_taxid
]
consolidated_dict[read_id][2] = species_taxid
else:
added_reads[read_id] = [
read_id, "1", species_taxid, accession, species_taxid,
genus_taxid, family_taxid, contig_id, accession,
species_taxid, genus_taxid, family_taxid,
'from_assembly'
]
if m8_line:
read2blastm8[read_id] = m8_line
return (consolidated_dict, read2blastm8, contig2lineage, added_reads)
def get_top_m8(orig_m8, top_entry_m8):
''' Get top m8 file entry for each read from orig_m8 and output to top_entry_m8 '''
with open(top_entry_m8, 'w') as top_m8f:
top_line = None
top_bitscore = 0
current_read_id = None
for read_id, _accession_id, _percent_id, _alignment_length, e_value, bitscore, line in m8.iterate_m8(
orig_m8):
# Get the top entry of each read_id based on the bitscore
if read_id != current_read_id:
# Different batch start
if current_read_id: # Not the first line
top_m8f.write(top_line)
current_read_id = read_id
top_line = line
top_bitscore = bitscore
elif bitscore > top_bitscore:
top_bitscore = bitscore
top_line = line
top_m8f.write(top_line)
def get_map(m8_file):
return dict((read_id, accession_id)
for read_id, accession_id, _percent_id,
_alignment_length, _e_value, _line in m8.iterate_m8(
m8_file, "annotate_fasta_with_accessions"))