def orient_sequences(input_file,
reference_file=None,
alignment_file=None,
output_file=None):
"""
Reorient a fasta file so all sequences are in the same direction as their reference
"""
log.info(
"Reorienting all sequences in %s to the direction of their reference" %
input_file)
# Set the output file and type
output_file = output_file or _get_output_file(input_file)
output_type = _get_output_type(output_file)
if valid_file(output_file):
log.info("Found existing output file %s, skipping orientation step" %
output_file)
return output_file
# Check the input files, and align the input file if needed
alignment_file = get_alignment_file(input_file, reference_file,
alignment_file)
reversed_seqs = _identify_reversed_sequences(alignment_file)
log.info("Identified %s sequences needing Reverse Complementation" %
len(reversed_seqs))
input_records = _parse_input_records(input_file)
reversed_records = _reverse_records(input_records, reversed_seqs)
log.info("Writing out sequences to %s" % output_file)
_write_output(reversed_records, output_file, output_type)
return output_file
def extract_alleles(input_file,
output_file=None,
reference_file=None,
alignment_file=None,
method=METHOD,
sort=SORT,
loci=LOCI):
"""Pick the top 2 Amplicon Analysis consensus seqs per group from a Fasta"""
method = method or METHOD
loci = loci or LOCI
# Set the output file if not specified
output_file = output_file or _get_output_file(input_file)
output_type = get_file_type(output_file)
# If align to reference for breaking ties
alignment_file = get_alignment_file(input_file, reference_file,
alignment_file)
alignments = list(BlasrReader(alignment_file))
# Run the appropriate grouping
if method == 'locus':
groups = _group_by_locus(alignments, loci)
elif method == 'barcode':
groups = _group_by_barcode(alignments)
elif method == 'both':
groups = _group_by_both(alignments, loci)
elif method == 'all':
groups = {a.qname: [a] for a in alignments}
else:
msg = "Invalid Selection Metric: %s" % method
log.error(msg)
raise ValueError(msg)
# Read the input sequences and use them to generate our sorting data
sequences = read_sequences(input_file)
if sort == 'num_reads':
sorting_data = {s.name: consensus_size(s) for s in sequences}
elif sort == 'accuracy':
assert get_file_type(input_file) == 'fastq'
sorting_data = {s.name: record_accuracy(s) for s in sequences}
else:
msg = "Invalid Sorting Metric: %s" % sort
log.error(msg)
raise ValueError(msg)
log.info('Sorting sequences for selection according to "%s"' % sort)
ordered = _sort_groups(groups, sorting_data)
log.info('Selecting top sequences from %s according to the "%s" policy' %
(input_file, method))
selected = list(_select_sequences(ordered))
log.info('Selected %s sequences from %s total for further analysis' %
(len(selected), len(sequences)))
log.info('Writing the selected sequences out to %s' % output_file)
subset = list(_subset_sequences(sequences, selected))
_write_output(subset, output_file, output_type)
return output_file
def extract_alleles( input_file, output_file=None, reference_file=None,
alignment_file=None,
method=METHOD,
sort=SORT,
loci=LOCI ):
"""Pick the top 2 Amplicon Analysis consensus seqs per group from a Fasta"""
method = method or METHOD
loci = loci or LOCI
# Set the output file if not specified
output_file = output_file or _get_output_file( input_file )
output_type = get_file_type( output_file )
# If align to reference for breaking ties
alignment_file = get_alignment_file( input_file, reference_file, alignment_file )
alignments = list( BlasrReader( alignment_file ))
# Run the appropriate grouping
if method == 'locus':
groups = _group_by_locus( alignments, loci )
elif method == 'barcode':
groups = _group_by_barcode( alignments )
elif method == 'both':
groups = _group_by_both( alignments, loci )
elif method == 'all':
groups = {a.qname: [a] for a in alignments}
else:
msg = "Invalid Selection Metric: %s" % method
log.error( msg )
raise ValueError( msg )
# Read the input sequences and use them to generate our sorting data
sequences = read_sequences( input_file )
if sort == 'num_reads':
sorting_data = {s.name: consensus_size(s) for s in sequences}
elif sort == 'accuracy':
assert get_file_type(input_file) == 'fastq'
sorting_data = {s.name: record_accuracy(s) for s in sequences}
else:
msg = "Invalid Sorting Metric: %s" % sort
log.error( msg )
raise ValueError( msg )
log.info('Sorting sequences for selection according to "%s"' % sort)
ordered = _sort_groups( groups, sorting_data )
log.info('Selecting top sequences from %s according to the "%s" policy' % (input_file, method))
selected = list( _select_sequences( ordered ))
log.info('Selected %s sequences from %s total for further analysis' % (len(selected), len(sequences)))
log.info('Writing the selected sequences out to %s' % output_file)
subset = list( _subset_sequences( sequences, selected ))
_write_output( subset, output_file, output_type )
return output_file
def cdna_from_file( input_file, hmm_fofn, output=None,
reference=None,
alignment=None ):
"""
Extract the cDNA sequences from a mixed Fasta or Fastq
"""
# Check the input files, and align the input file if needed
alignment_file = get_alignment_file( input_file, reference, alignment )
output_file = output or get_output_file( input_file, 'cDNA' )
# Prepare the Fasta by orienting and subsetting it
records = _parse_input_records( input_file )
hmms = parse_locus_dict( hmm_fofn )
loci = _parse_loci( alignment_file )
# Compose and output the records
cdna_records = list( cdna_from_records( records, loci, hmms ))
write_records( cdna_records, output_file )
return output_file
def orient_sequences( input_file, reference_file=None, alignment_file=None, output_file=None ):
"""
Reorient a fasta file so all sequences are in the same direction as their reference
"""
log.info("Reorienting all sequences in %s to the direction of their reference" % input_file)
# Set the output file and type
output_file = output_file or _get_output_file( input_file )
output_type = _get_output_type( output_file )
if valid_file( output_file ):
log.info("Found existing output file %s, skipping orientation step" % output_file)
return output_file
# Check the input files, and align the input file if needed
alignment_file = get_alignment_file( input_file, reference_file, alignment_file )
reversed_seqs = _identify_reversed_sequences( alignment_file )
log.info("Identified %s sequences needing Reverse Complementation" % len(reversed_seqs))
input_records = _parse_input_records( input_file )
reversed_records = _reverse_records( input_records, reversed_seqs )
log.info("Writing out sequences to %s" % output_file)
_write_output( reversed_records, output_file, output_type )
return output_file
def cdna_from_file(input_file,
hmm_fofn,
output=None,
reference=None,
alignment=None):
"""
Extract the cDNA sequences from a mixed Fasta or Fastq
"""
# Check the input files, and align the input file if needed
alignment_file = get_alignment_file(input_file, reference, alignment)
output_file = output or get_output_file(input_file, 'cDNA')
# Prepare the Fasta by orienting and subsetting it
records = _parse_input_records(input_file)
hmms = parse_locus_dict(hmm_fofn)
loci = _parse_loci(alignment_file)
# Compose and output the records
cdna_records = list(cdna_from_records(records, loci, hmms))
write_records(cdna_records, output_file)
return output_file
