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 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_contigs_to_genome(self, contig_file):
log.info("Looking for Contig-to-Genome alignment data")
contig_genome_align = self.get_filepath( 'alignments', 'contigs_to_genome.m1' )
if valid_file( contig_genome_align ):
log.info("Using existing Contig->Genome alignment file\n")
else:
log.info("No Contig->Genome alignment found, creating...")
align_best_reference( contig_file, self.human_reference, output=contig_genome_align )
check_output_file( contig_genome_align )
log.info("Finished aligning contigs to the genomic reference\n")
return create_m1_reference( contig_genome_align )
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 align_contigs_to_reference(self, contig_file, reference_file):
"""
Align HBAR contigs to an HLA reference Fasta
"""
log.info("Looking for Contig-to-Reference alignment data")
contig_reference_align = self.get_filepath( 'alignments', 'contigs_to_reference.m1' )
if valid_file( contig_reference_align ):
log.info("Using an existing Contig->Reference alignment file\n")
else:
log.info("No Contig->Reference alignment found, creating...")
align_best_reference( contig_file, reference_file, output=contig_reference_align )
check_output_file( contig_reference_align )
log.info("Finished aligning contigs to the HLA reference data\n")
return create_m1_reference( contig_reference_align )
def align_subreads_to_contigs(self, subread_file, contig_file ):
"""
Align the subreads to the contigs assembled by HBAR
"""
log.info("Looking for Subread-to-Contig alignment data")
subread_contig_align = self.get_filepath( 'alignments', 'subreads_to_contigs.m1' )
if valid_file( subread_contig_align ):
log.info("Using existing Subread->Contig alignment file\n")
else:
log.info("No Subread->Contig alignment found, creating...")
align_best_reference( subread_file, contig_file, output=subread_contig_align )
check_output_file( subread_contig_align )
log.info("Finished aligning subreads to the HBAR contigs\n")
return create_m1_reference( subread_contig_align )
def align_contigs_to_genome(self, contig_file):
log.info("Looking for Contig-to-Genome alignment data")
contig_genome_align = self.get_filepath('alignments',
'contigs_to_genome.m1')
if valid_file(contig_genome_align):
log.info("Using existing Contig->Genome alignment file\n")
else:
log.info("No Contig->Genome alignment found, creating...")
align_best_reference(contig_file,
self.human_reference,
output=contig_genome_align)
check_output_file(contig_genome_align)
log.info("Finished aligning contigs to the genomic reference\n")
return create_m1_reference(contig_genome_align)
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 align_contigs_to_reference(self, contig_file, reference_file):
"""
Align HBAR contigs to an HLA reference Fasta
"""
log.info("Looking for Contig-to-Reference alignment data")
contig_reference_align = self.get_filepath('alignments',
'contigs_to_reference.m1')
if valid_file(contig_reference_align):
log.info("Using an existing Contig->Reference alignment file\n")
else:
log.info("No Contig->Reference alignment found, creating...")
align_best_reference(contig_file,
reference_file,
output=contig_reference_align)
check_output_file(contig_reference_align)
log.info("Finished aligning contigs to the HLA reference data\n")
return create_m1_reference(contig_reference_align)
def align_subreads_to_contigs(self, subread_file, contig_file):
"""
Align the subreads to the contigs assembled by HBAR
"""
log.info("Looking for Subread-to-Contig alignment data")
subread_contig_align = self.get_filepath('alignments',
'subreads_to_contigs.m1')
if valid_file(subread_contig_align):
log.info("Using existing Subread->Contig alignment file\n")
else:
log.info("No Subread->Contig alignment found, creating...")
align_best_reference(subread_file,
contig_file,
output=subread_contig_align)
check_output_file(subread_contig_align)
log.info("Finished aligning subreads to the HBAR contigs\n")
return create_m1_reference(subread_contig_align)
def type_fasta( input_fofn, input_fasta, exon_fofn, genomic_reference, cDNA_reference ):
"""
Pick the top N Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
# First we align the sequences to the reference and annotate typing
raw_alignment = align_best_reference( input_fasta, genomic_reference )
reoriented = orient_fasta( input_fasta, alignment_file=raw_alignment )
selected = extract_alleles( reoriented, alignment_file=raw_alignment )
gDNA_alignment = full_align_best_reference( selected, genomic_reference )
cDNA_file = extract_cDNA( selected, exon_fofn, alignment_file=gDNA_alignment )
cDNA_alignment = align_by_identity( cDNA_file, cDNA_reference )
summarize_typing( gDNA_alignment, cDNA_alignment )
# Next we generate some mock chimera sequences
chimera_file = create_chimeras( selected, alignment_file=gDNA_alignment )
basename = '.'.join( chimera_file.split('.')[:-2] )
combined_file = '%s.combined.fasta' % basename
combine_fasta( [input_fasta, chimera_file], combined_file )
# Finally we use a competetive alignment of best-reads to summarize the allelic breakdown
dirname = os.path.dirname( input_fasta )
best_reads = os.path.join( dirname, 'reads_of_insert.fasta' )
extract_best_reads( input_fofn, best_reads )
best_alignment = align_best_reference( best_reads, combined_file )
summarize_alleles( best_alignment, raw_alignment, selected )
def type_sequences( input_folder, exon_fofn, genomic_reference, cDNA_reference ):
"""
Pick the top N Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
sequence_file = os.path.join( input_folder, 'amplicon_analysis.fastq' )
csv_file = os.path.join( input_folder, 'amplicon_analysis.csv' )
# First we align the sequences to the reference and annotate typing
raw_alignment = align_best_reference( sequence_file, genomic_reference )
reoriented = orient_sequences( sequence_file, alignment_file=raw_alignment )
reoriented_csv = orient_amp_analysis( csv_file, raw_alignment )
selected = extract_alleles( reoriented, alignment_file=raw_alignment )
selected_csv = subset_amp_analysis( reoriented_csv, selected )
gDNA_alignment = full_align_best_reference( selected, genomic_reference )
cDNA_file = extract_cDNA( selected, exon_fofn, alignment_file=gDNA_alignment )
cDNA_alignment = align_by_identity( cDNA_file, cDNA_reference )
summarize_typing( gDNA_alignment, cDNA_alignment )
def type_sequences(input_folder, exon_fofn, genomic_reference, cDNA_reference):
"""
Pick the top N Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
sequence_file = os.path.join(input_folder, 'amplicon_analysis.fastq')
csv_file = os.path.join(input_folder, 'amplicon_analysis.csv')
# First we align the sequences to the reference and annotate typing
raw_alignment = align_best_reference(sequence_file, genomic_reference)
reoriented = orient_sequences(sequence_file, alignment_file=raw_alignment)
reoriented_csv = orient_amp_analysis(csv_file, raw_alignment)
selected = extract_alleles(reoriented, alignment_file=raw_alignment)
selected_csv = subset_amp_analysis(reoriented_csv, selected)
gDNA_alignment = full_align_best_reference(selected, genomic_reference)
cDNA_file = extract_cDNA(selected,
exon_fofn,
alignment_file=gDNA_alignment)
cDNA_alignment = align_by_identity(cDNA_file, cDNA_reference)
summarize_typing(gDNA_alignment, cDNA_alignment)
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