def create_m5_reference(m5_file):
log.info('Parsing Blasr M5 results from "{0}"'.format(m5_file))
results = {}
diffs = {}
for record in BlasrReader(m5_file):
qname = get_base_sequence_name(record.qname)
tname = get_base_sequence_name(record.tname)
diff_count = int(record.nmis) + int(record.nins) + int(record.ndel)
if qname not in diffs:
results[qname] = tname
diffs[qname] = diff_count
elif diffs[qname] > diff_count:
results[qname] = tname
diffs[qname] = diff_count
log.info('Finished reading Blasr results')
return results
def create_m1_reference( m1_file, reference=None ):
log.info('Parsing Blasr M1 results from "{0}"'.format( m1_file ))
results = {}
for record in BlasrReader( m1_file ):
qname = get_base_sequence_name( record.qname )
tname = get_base_sequence_name( record.tname )
if qname in results:
msg = 'Duplicate sequence ids found! "{0}"'.format( qname )
log.info( msg )
raise KeyError( msg )
if reference:
results[qname] = reference[tname]
else:
results[qname] = tname
log.info('Finished reading Blasr results')
return results
def create_m5_reference( m5_file ):
log.info('Parsing Blasr M5 results from "{0}"'.format( m5_file ))
results = {}
diffs = {}
for record in BlasrReader( m5_file ):
qname = get_base_sequence_name( record.qname )
tname = get_base_sequence_name( record.tname )
diff_count = int(record.nmis) + int(record.nins) + int(record.ndel)
if qname not in diffs:
results[qname] = tname
diffs[qname] = diff_count
elif diffs[qname] > diff_count:
results[qname] = tname
diffs[qname] = diff_count
log.info('Finished reading Blasr results')
return results
def create_m1_reference(m1_file, reference=None):
log.info('Parsing Blasr M1 results from "{0}"'.format(m1_file))
results = {}
for record in BlasrReader(m1_file):
qname = get_base_sequence_name(record.qname)
tname = get_base_sequence_name(record.tname)
if qname in results:
msg = 'Duplicate sequence ids found! "{0}"'.format(qname)
log.info(msg)
raise KeyError(msg)
if reference:
results[qname] = reference[tname]
else:
results[qname] = tname
log.info('Finished reading Blasr results')
return results
def append_typing_results(summary_file, combined_typings, output_file):
typing_header = '\tGenType\tGenPctId\tExonType\tExonPctId\tType\t'
with open(output_file, 'w') as output:
with open(summary_file, 'r') as handle:
header = handle.next().strip()
output.write(header + typing_header)
for line in handle:
parts = line.strip().split()
name = get_base_sequence_name(parts[1])
parts += combined_typings[name]
output.write('\t'.join(parts) + '\n')
def _parse_blasr_alignment( blasr_file ):
results = {}
for entry in BlasrReader( blasr_file ):
name = get_base_sequence_name( entry.qname )
if isinstance(entry, BlasrM1):
results[name] = [entry.tname, entry.pctsimilarity]
elif isinstance(entry, BlasrM5):
diffs = int(entry.nmis) + int(entry.nins) + int(entry.ndel)
pctid = 100 * int(entry.nmat) / float(int(entry.nmat) + diffs)
results[name] = [entry.tname, pctid]
return results
def _parse_blasr_alignment(blasr_file):
results = {}
for entry in BlasrReader(blasr_file):
name = get_base_sequence_name(entry.qname)
if isinstance(entry, BlasrM1):
results[name] = [entry.tname, entry.pctsimilarity]
elif isinstance(entry, BlasrM5):
diffs = int(entry.nmis) + int(entry.nins) + int(entry.ndel)
pctid = 100 * int(entry.nmat) / float(int(entry.nmat) + diffs)
results[name] = [entry.tname, pctid]
return results
def append_typing_results( summary_file, combined_typings, output_file):
typing_header = '\tGenType\tGenPctId\tExonType\tExonPctId\tType\t'
with open(output_file, 'w') as output:
with open(summary_file, 'r') as handle:
header = handle.next().strip()
output.write(header + typing_header)
for line in handle:
parts = line.strip().split()
name = get_base_sequence_name( parts[1] )
parts += combined_typings[name]
output.write('\t'.join(parts) + '\n')
def parse_typing( typing_file ):
results = {}
with open( typing_file) as handle:
for line in handle:
if line.startswith('Locus'):
continue
parts = line.strip().split()
name = get_base_sequence_name( parts[1] )
typing = parts[2]
pctid = parts[5]
results[name] = [typing, pctid]
return results
def separate_listed_sequences( fasta_file, good_values, good_output, bad_output ):
"""
Separate a fasta file into two based on a supplied value list
"""
with FastaWriter( good_output ) as good_handle:
with FastaWriter( bad_output ) as bad_handle:
for record in FastaReader( fasta_file ):
name = get_base_sequence_name( record.name )
if name in good_values:
good_handle.writeRecord( record )
else:
bad_handle.writeRecord( record )
def parse_typing(typing_file):
results = {}
with open(typing_file) as handle:
for line in handle:
if line.startswith('Locus'):
continue
parts = line.strip().split()
name = get_base_sequence_name(parts[1])
typing = parts[2]
pctid = parts[5]
results[name] = [typing, pctid]
return results
def separate_aligned_sequences( fasta_file, dictionary, good_values, good_output, bad_output ):
"""
Separate a fasta file into two based on a supplied dictionary and value list
"""
with FastaWriter( good_output ) as good_handle:
with FastaWriter( bad_output ) as bad_handle:
for record in FastaReader( fasta_file ):
name = get_base_sequence_name( record.name )
value = dictionary.get(name, "Unmapped")
if value in good_values:
good_handle.writeRecord( record )
else:
bad_handle.writeRecord( record )
def separate_listed_sequences(fasta_file, good_values, good_output,
bad_output):
"""
Separate a fasta file into two based on a supplied value list
"""
with FastaWriter(good_output) as good_handle:
with FastaWriter(bad_output) as bad_handle:
for record in FastaReader(fasta_file):
name = get_base_sequence_name(record.name)
if name in good_values:
good_handle.writeRecord(record)
else:
bad_handle.writeRecord(record)
def separate_aligned_sequences(fasta_file, dictionary, good_values,
good_output, bad_output):
"""
Separate a fasta file into two based on a supplied dictionary and value list
"""
with FastaWriter(good_output) as good_handle:
with FastaWriter(bad_output) as bad_handle:
for record in FastaReader(fasta_file):
name = get_base_sequence_name(record.name)
value = dictionary.get(name, "Unmapped")
if value in good_values:
good_handle.writeRecord(record)
else:
bad_handle.writeRecord(record)
def create_sam_reference( sam_file, reference=None ):
log.info('Parsing SAM alignments from "{0}"'.format(sam_file))
results = {}
for record in SamReader(sam_file):
name = get_base_sequence_name( record.rname )
if record.qname in results:
msg = 'Duplicate sequence ids found! "{0}"'.format( record.qname )
log.info( msg )
raise KeyError( msg )
if reference:
results[record.qname] = reference[name]
else:
results[record.qname] = name
log.info('Finished reading SAM file results')
return results
def create_sam_reference(sam_file, reference=None):
log.info('Parsing SAM alignments from "{0}"'.format(sam_file))
results = {}
for record in SamReader(sam_file):
name = get_base_sequence_name(record.rname)
if record.qname in results:
msg = 'Duplicate sequence ids found! "{0}"'.format(record.qname)
log.info(msg)
raise KeyError(msg)
if reference:
results[record.qname] = reference[name]
else:
results[record.qname] = name
log.info('Finished reading SAM file results')
return results
def separate_sequences( fasta_file, dictionary, prefix='' ):
"""
Separate a fasta file into multiple groups based on some dict
"""
file_handles = {}
for record in FastaReader( fasta_file ):
name = get_base_sequence_name( record.name )
group = dictionary.get( name, "Unmapped" )
group_file = prefix + '_' + group + '.fasta'
try:
file_handles[group_file].writeRecord( record )
except KeyError:
file_handles[group_file] = FastaWriter( group_file )
file_handles[group_file].writeRecord( record )
return closed_file_handles( file_handles )
def separate_sequences(fasta_file, dictionary, prefix=''):
"""
Separate a fasta file into multiple groups based on some dict
"""
file_handles = {}
for record in FastaReader(fasta_file):
name = get_base_sequence_name(record.name)
group = dictionary.get(name, "Unmapped")
group_file = prefix + '_' + group + '.fasta'
try:
file_handles[group_file].writeRecord(record)
except KeyError:
file_handles[group_file] = FastaWriter(group_file)
file_handles[group_file].writeRecord(record)
return closed_file_handles(file_handles)