def align_reference_to_contigs( locus, reference, contig_file ):
output_file = 'HLA-%s.m1' % locus
blasr_args = {'nproc': 8,
'out': output_file,
'bestn': 1,
'noSplitSubreads': True}
run_blasr( reference, contig_file, blasr_args )
check_output_file( output_file )
return output_file
def align_amplicons( filetype, sequence_5p, sequence_3p ):
blasr_args = {'bestn': 1,
'out': 'test.m5',
'm': 5,
'noSplitSubreads': True}
if filetype == 'fastq':
temp_5p = write_temp_fasta( sequence_5p )
temp_3p = write_temp_fasta( sequence_3p )
align_left = run_blasr( temp_5p.name, temp_3p.name, blasr_args, verbose=True )
elif filetype == 'fasta':
assert fasta_size( sequence_5p ) == 2
assert fasta_size( sequence_3p ) == 2
align_left = run_blasr( sequence_5p, sequence_3p, blasr_args )
else:
raise ValueError
return align_left
def align_subreads( self, white_list, reference_file ):
"""
Align the subreads in a Whitelist to the created reference
"""
basename = '.'.join( reference_file.split('.')[:-1] )
alignment_file = '%s.m1' % basename
reference_count = fasta_size( reference_file )
blasr_args = { 'nproc': self._nproc,
'out': alignment_file,
'bestn': 1,
'nCandidates': reference_count,
'noSplitSubreads': True }
run_blasr( white_list,
reference_file,
blasr_args )
check_output_file( alignment_file )
return alignment_file
def _align_fasta( query, reference, format ):
"""
Align a single query sequence to all valid references
"""
suffix = '.m%s' % format
temp_align = tempfile.NamedTemporaryFile( suffix=suffix, delete=False )
reference_count = fasta_size( reference )
blasr_args = {'nproc': NPROC,
'out': temp_align.name,
'bestn': reference_count,
'nCandidates': reference_count,
'm': format,
'noSplitSubreads': True}
run_blasr( query, reference, blasr_args )
# Parse the output for return and delete the file
alignments = list( BlasrReader( temp_align.name ))
os.unlink( temp_align.name )
return alignments
def align_amplicons(sequence_file, reference):
sequence_root = '.'.join(sequence_file.split('.')[:-1])
output_file = sequence_root + '.m5'
blasr_args = {
'bestn': 1,
'out': output_file,
'm': 5,
'noSplitSubreads': True
}
return run_blasr(sequence_file, reference, blasr_args, verbose=True)
def _align_subreads( subread_fasta, reference_fasta, locus ):
"""
Align all locus-specific subreads against the appropriate references
"""
location = os.path.dirname( subread_fasta )
alignment_file = os.path.join(location, 'temp.m1')
subread_count = fasta_size( subread_fasta )
reference_count = fasta_size( reference_fasta )
blasr_args = {'nproc': 8,
'out': alignment_file,
'bestn': 1,
'nCandidates': reference_count,
'noSplitSubreads': True}
log.info("Aligning %s reads against %s references for %s" % (subread_count,
reference_count,
locus))
run_blasr( subread_fasta, reference_fasta, blasr_args )
check_output_file( alignment_file )
return alignment_file
def _align_fasta(query, reference, format):
"""
Align a single query sequence to all valid references
"""
suffix = '.m%s' % format
temp_align = tempfile.NamedTemporaryFile(suffix=suffix, delete=False)
reference_count = fasta_size(reference)
blasr_args = {
'nproc': NPROC,
'out': temp_align.name,
'bestn': reference_count,
'nCandidates': reference_count,
'm': format,
'noSplitSubreads': True
}
run_blasr(query, reference, blasr_args)
# Parse the output for return and delete the file
alignments = list(BlasrReader(temp_align.name))
os.unlink(temp_align.name)
return alignments
def _align_subreads(subread_fasta, reference_fasta, locus):
"""
Align all locus-specific subreads against the appropriate references
"""
location = os.path.dirname(subread_fasta)
alignment_file = os.path.join(location, 'temp.m1')
subread_count = fasta_size(subread_fasta)
reference_count = fasta_size(reference_fasta)
blasr_args = {
'nproc': 8,
'out': alignment_file,
'bestn': 1,
'nCandidates': reference_count,
'noSplitSubreads': True
}
log.info("Aligning %s reads against %s references for %s" %
(subread_count, reference_count, locus))
run_blasr(subread_fasta, reference_fasta, blasr_args)
check_output_file(alignment_file)
return alignment_file
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 align_amplicons(filetype, sequence_5p, sequence_3p):
blasr_args = {
'bestn': 1,
'out': 'test.m5',
'm': 5,
'noSplitSubreads': True
}
if filetype == 'fastq':
temp_5p = write_temp_fasta(sequence_5p)
temp_3p = write_temp_fasta(sequence_3p)
align_left = run_blasr(temp_5p.name,
temp_3p.name,
blasr_args,
verbose=True)
elif filetype == 'fasta':
assert fasta_size(sequence_5p) == 2
assert fasta_size(sequence_3p) == 2
align_left = run_blasr(sequence_5p, sequence_3p, blasr_args)
else:
raise ValueError
return align_left
def full_align_best_reference(query, reference, output=None):
"""
Align the output of AA to the references and return
"""
# Figure out the output and remove it if it exists
output = _get_output_file(query, output, 'm5')
# Run Blasr
ref_count = fasta_size(reference)
log.info("Aligning %s sequences to %s references" % (query, ref_count))
blasr_args = {'nproc': nproc,
'out': output,
'm': 5,
'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
check_output_file(output)
return output
def _align_exons( query, exon_fasta, directory ):
"""
Align all supplied exon sequences to a "Query" Fasta
"""
exon_num = exon_fasta[-7]
alignment_file = 'exon%s.m1' % exon_num
alignment_path = os.path.join( directory, alignment_file )
blasr_args = {'nproc': NPROC,
'out': alignment_path,
'minSubreadLength': 20,
'minReadLength': 20,
'maxScore': 0,
'bestn': 1,
'noSplitSubreads': True}
run_blasr( exon_fasta,
query,
blasr_args )
count = count_hits( alignment_path )
if count > 0:
log.info('Found %s hits for Exon #%s' % (count, exon_num))
return alignment_path
log.info('No hits found for Exon #%s' % exon_num)
return None
def _align_exons(query, exon_fasta, directory):
"""
Align all supplied exon sequences to a "Query" Fasta
"""
exon_num = exon_fasta[-7]
alignment_file = 'exon%s.m1' % exon_num
alignment_path = os.path.join(directory, alignment_file)
blasr_args = {
'nproc': NPROC,
'out': alignment_path,
'minSubreadLength': 20,
'minReadLength': 20,
'maxScore': 0,
'bestn': 1,
'noSplitSubreads': True
}
run_blasr(exon_fasta, query, blasr_args)
count = count_hits(alignment_path)
if count > 0:
log.info('Found %s hits for Exon #%s' % (count, exon_num))
return alignment_path
log.info('No hits found for Exon #%s' % exon_num)
return None
def align_amplicons(sequence_file, reference):
sequence_root = ".".join(sequence_file.split(".")[:-1])
output_file = sequence_root + ".m5"
blasr_args = {"bestn": 1, "out": output_file, "m": 5, "noSplitSubreads": True}
return run_blasr(sequence_file, reference, blasr_args, verbose=True)