def write_sequences(filetype, merged, output):
if filetype == 'fasta':
write_fasta(merged, output)
elif filetype == 'fastq':
write_fastq(merged, output)
else:
raise ValueError
def create_chimeras(input_file,
output=None,
reference_file=None,
alignment_file=None):
"""
Pick the top N Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
# Check the input files, and align the input file if needed
if reference_file and alignment_file is None:
alignment_file = align_best_reference(input_file, reference_file)
elif reference_file is None and alignment_file is None:
msg = "extract_alleles requires either an Alignment or a Reference!"
log.error(msg)
raise IOError(msg)
# Set the output file if not specified
if output is None:
basename = '.'.join(input_file.split('.')[:-1])
output = '%s.chimeras.fasta' % basename
# Parse the alignment data and extract the target sequences
alignments = list(BlasrReader(alignment_file))
groups = _group_by_locus(alignments)
groups = _filter_groups(groups)
sequences = list(FastaReader(input_file))
chimeras = list(_create_chimeras(groups, sequences))
write_fasta(chimeras, output)
return output
def write_sequences( filetype, merged, output ):
if filetype == 'fasta':
write_fasta( merged, output )
elif filetype == 'fastq':
write_fastq( merged, output )
else:
raise ValueError
def _write_output( records, output_file, output_type ):
"""Write the records out to file"""
if output_type == 'fasta':
write_fasta( records, output_file )
else:
with FastqWriter( output_file ) as writer:
for record in records:
writer.writeRecord( record )
check_output_file( output_file )
def _write_output(records, output_file, output_type):
"""Write the records out to file"""
if output_type == 'fasta':
write_fasta(records, output_file)
else:
with FastqWriter(output_file) as writer:
for record in records:
writer.writeRecord(record)
check_output_file(output_file)
def pair_exon_files( fofn_file, overlap_exon ):
"""
Pair the 5' and 3' amplicons of a gene base on 1 overlapping exon
"""
exon_files = list( _parse_fofn( fofn_file ))
output_file = _get_output_file( exon_fasta )
fasta_records = list( FastaReader( exon_fasta ))
sorted_records = _sort_fasta_records( fasta_records )
cDNA_record = _combine_records( sorted_records )
write_fasta( [cDNA_record], output_file )
def from_assembly( contig_file, reference_fofn ):
contigs = read_fasta_dict( contig_file )
references = read_reference_fofn( reference_fofn )
all_genes = []
for locus, reference in references.iteritems():
alignment = align_reference_to_contigs( locus, reference, contig_file )
hits = read_blasr_hits( alignment )
hits = pick_blasr_hits( hits )
genes = extract_genes( contigs, hits )
all_genes += genes
write_fasta( all_genes, 'output.fasta' )
def parse_reference(self):
"""
Parse HLA data from the configured reference FOFN
"""
log.info("Parsing the supplied FOFN of HLA reference data")
hla_reference_seqs = self.get_filepath( "references", "HLA_references.fasta" )
sequences, metadata, loci = parse_reference_fofn( self.hla_reference )
log.info("Writing collected HLA reference sequences to file")
write_fasta( sequences, hla_reference_seqs )
check_output_file( hla_reference_seqs )
log.info("Finished parsing the HLA reference data\n")
return hla_reference_seqs, metadata, loci
def trim_fasta( fasta_file, blasr_file, output_file, locus_dict, window=WINDOW, loci=LOCI ):
log.info('Trimming sequences in "%s"' % fasta_file)
log.debug("\tWindow Size:\t%s" % window)
records = list( FastaReader( fasta_file ) )
trims = parse_trims( blasr_file, window )
trims = filter_trims_on_loci( trims, locus_dict, loci )
trimmed_records = apply_trims( records, trims )
write_fasta( trimmed_records, output_file )
log.info('Finished trimming the supplied sequencs\n')
return
def parse_reference(self):
"""
Parse HLA data from the configured reference FOFN
"""
log.info("Parsing the supplied FOFN of HLA reference data")
hla_reference_seqs = self.get_filepath("references",
"HLA_references.fasta")
sequences, metadata, loci = parse_reference_fofn(self.hla_reference)
log.info("Writing collected HLA reference sequences to file")
write_fasta(sequences, hla_reference_seqs)
check_output_file(hla_reference_seqs)
log.info("Finished parsing the HLA reference data\n")
return hla_reference_seqs, metadata, loci
def _write_temp_fasta( record ):
"""
Write a sequence record out to a temporary Fasta file
"""
temp = tempfile.NamedTemporaryFile( suffix='.fasta', delete=False )
if isinstance( record, FastaRecord ):
write_fasta( [record], temp.name )
elif isinstance( record, FastqRecord ):
temp_record = FastaRecord(record.name, record.sequence)
write_fasta( [temp_record], temp.name )
else:
msg = 'Record must be either FastaRecord or FastqRecord'
log.error( msg )
raise TypeError( msg )
return temp.name
def _write_temp_fasta(record):
"""
Write a sequence record out to a temporary Fasta file
"""
temp = tempfile.NamedTemporaryFile(suffix='.fasta', delete=False)
if isinstance(record, FastaRecord):
write_fasta([record], temp.name)
elif isinstance(record, FastqRecord):
temp_record = FastaRecord(record.name, record.sequence)
write_fasta([temp_record], temp.name)
else:
msg = 'Record must be either FastaRecord or FastqRecord'
log.error(msg)
raise TypeError(msg)
return temp.name
def create_chimeras( input_file, output=None, reference_file=None, alignment_file=None ):
"""
Pick the top N Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
# Check the input files, and align the input file if needed
if reference_file and alignment_file is None:
alignment_file = align_best_reference( input_file, reference_file )
elif reference_file is None and alignment_file is None:
msg = "extract_alleles requires either an Alignment or a Reference!"
log.error( msg )
raise IOError( msg )
# Set the output file if not specified
if output is None:
basename = '.'.join( input_file.split('.')[:-1] )
output = '%s.chimeras.fasta' % basename
# Parse the alignment data and extract the target sequences
alignments = list( BlasrReader( alignment_file ))
groups = _group_by_locus( alignments )
groups = _filter_groups( groups )
sequences = list( FastaReader( input_file ))
chimeras = list( _create_chimeras( groups, sequences ))
write_fasta( chimeras, output )
return output
def extract_best_reads(input_file,
output_file=None,
min_length=MIN_LENGTH,
min_score=MIN_SCORE):
"""
Extract, filter and subset subreads from Bas/Bax/Fofn Files
"""
if output_file is None:
basename = '.'.join( input_file.split('.')[:-1] )
output_file = '%s.best.fasta' % basename
log.info('Extracting subreads from %s' % os.path.basename(input_file))
log.debug('\tMinimum Length:\t%s' % min_length)
log.debug('\tMinimum Score:\t%s' % min_score)
reads = []
for i, filename in enumerate(_iterate_input_files( input_file )):
reads += list( _extract_from_bash5( filename, min_length, min_score ))
log.info("Extracted %s subreads from %s files" % (len(reads), i+1))
write_fasta( reads, output_file )
check_output_file( output_file )
log.info("Finished extracting subreads")
return output_file
def subset_sequences( fasta_file, summary_file, output_file ):
seq_ids = identify_sequences( summary_file )
sequences = subset_sequence_records( fasta_file, seq_ids )
write_fasta( sequences, output_file )
