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 get_cDNA_reference():
if valid_file( CDNA_REF ):
logging.info('Using existing cDNA Reference fasta')
return CDNA_REF
else:
logging.info('Creating new cDNA Reference fasta')
create_cDNA_reference()
return get_cDNA_reference()
def get_genomic_reference():
if valid_file( GEN_REF ):
logging.info('Using existing Genomic Reference fasta')
return GEN_REF
else:
logging.info('Creating new Genomic Reference fasta')
create_genomic_reference()
return get_genomic_reference()
def get_exon_reference():
if valid_file( EXON_REF ):
logging.info('Using existing Exon Reference fofn')
return EXON_REF
else:
logging.info('Creating new Exon Reference fofn')
create_exon_fofns()
create_exon_reference()
return get_exon_reference()
def sort_subreads( subread_file, reference_file ):
"""
Aligning
"""
log.info("Aligning subreads to the two best references")
temp = 'temp2.m1'
if valid_file( temp ):
return {hit.qname: hit.tname for hit in BlasrReader(temp)}
align_best_reference( subread_file, reference_file, temp )
return {hit.qname: hit.tname for hit in BlasrReader(temp)}
def order_references( subread_file, reference_file ):
"""
Select the two best reference sequences from a list
"""
log.info("Selecting the best references sequences to use")
temp = 'temp.m1'
if not valid_file( temp ):
align_best_reference( subread_file, reference_file, temp )
c = Counter([hit.tname for hit in BlasrReader(temp)])
return [k for k, v in c.most_common()]
def sort_subreads(subread_file, reference_file):
"""
Aligning
"""
log.info("Aligning subreads to the two best references")
temp = 'temp2.m1'
if valid_file(temp):
return {hit.qname: hit.tname for hit in BlasrReader(temp)}
align_best_reference(subread_file, reference_file, temp)
return {hit.qname: hit.tname for hit in BlasrReader(temp)}
def order_references(subread_file, reference_file):
"""
Select the two best reference sequences from a list
"""
log.info("Selecting the best references sequences to use")
temp = 'temp.m1'
if not valid_file(temp):
align_best_reference(subread_file, reference_file, temp)
c = Counter([hit.tname for hit in BlasrReader(temp)])
return [k for k, v in c.most_common()]
def _align_sequences(query, reference):
"""
Align one fasta file of sequences to another
"""
temp = NamedTemporaryFile(suffix='.m1', delete=False)
align_best_reference(query, reference, output=temp.name)
if valid_file(temp.name):
hits = list(BlasrReader(temp.name))
os.unlink(temp.name)
return hits
os.unlink(temp.name)
return None
def get_input_file( input ):
if os.path.isdir( input ):
log.info("Input appears to be a directory, looking for sequence files")
return get_amplicon_analysis_output( input )
elif valid_file( input ):
if is_fasta( input ):
log.info("Input appears to be a valid Fasta file")
return input
elif is_fastq( input ):
log.info("Input appears to be a valid Fastq file")
return input
else:
msg = "Input is not a valid Fasta or Fastq file!"
log.error( msg )
raise IOError( msg )
else:
msg = "Supplied input does not appear to be a valid AmpliconAnalysis file or directory"
log.error( msg )
raise IOError( msg )
def get_input_file(input):
if os.path.isdir(input):
log.info("Input appears to be a directory, looking for sequence files")
return get_amplicon_analysis_output(input)
elif valid_file(input):
if is_fasta(input):
log.info("Input appears to be a valid Fasta file")
return input
elif is_fastq(input):
log.info("Input appears to be a valid Fastq file")
return input
else:
msg = "Input is not a valid Fasta or Fastq file!"
log.error(msg)
raise IOError(msg)
else:
msg = "Supplied input does not appear to be a valid AmpliconAnalysis file or directory"
log.error(msg)
raise IOError(msg)
def align_best_reference(query, reference, output=None):
"""
Align the output of AA to the references and return
"""
output = _get_output_file(query, output, 'm1')
# Run Blasr
ref_count = fasta_size(reference)
log.info("Aligning %s sequences to %s references" % (query, ref_count))
blasr_args = {'nproc': nproc,
'out': output,
'bestn': 1,
'nCandidates': ref_count,
'noSplitSubreads': True}
if reference_has_index( reference ):
blasr_args['sa'] = reference + '.sa'
run_blasr(query, reference, blasr_args)
# Check the output file
if valid_file( output ):
return output
return None
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 reference_has_index( reference ):
index_file = reference + '.sa'
if valid_file( index_file ):
return index_file
return False