def type_sequences( input, grouping=GROUPING,
exon_fofn=None,
genomic_reference=None,
cDNA_reference=None,
loci=None):
"""
Pick the top Amplicon Analysis consensus seqs from a Fasta by Nreads
"""
log_file = get_log_file( input )
initialize_logger( log, log_file=log_file )
# First, get any references not specified by the user
grouping = grouping or GROUPING
exon_fofn = exon_fofn or get_exon_reference()
genomic_reference = genomic_reference or get_genomic_reference()
cDNA_reference = cDNA_reference or get_cDNA_reference()
# Second, get the input file if a directory was specified
sequence_file = get_input_file( input )
# Finally, run the Typing procedure
renamed_file = rename_sequences( sequence_file )
raw_alignment = full_align_best_reference( renamed_file, genomic_reference )
reoriented = orient_sequences( renamed_file, alignment_file=raw_alignment )
selected = extract_alleles( reoriented, alignment_file=raw_alignment,
method=grouping,
loci=loci)
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 )
typing = summarize_typing( gDNA_alignment, cDNA_alignment )
return typing
def fetch_chimera_scores(folder, reference):
# Check and read the reference data
reference_files = find_reference_files(reference)
# Check and read the query data
barcode_files = split_results(folder)
# Check and read the log data
log_data = read_log_data(folder)
good = []
bad = []
for barcode, reference in reference_files.iteritems():
sample_file = barcode_files[barcode]
alignments = full_align_best_reference(sample_file, reference)
by_reference = hits_by_reference(alignments)
best, rest = separate_best_hits(by_reference)
good += best
bad += rest
print "Found %s 'good' and %s 'bad' consensus sequences" % (len(good),
len(bad))
good_scores = sorted({r.qname: log_data[r.qname]
for r in good}.iteritems(),
key=itemgetter(1),
reverse=True)
bad_scores = sorted({r.qname: log_data[r.qname]
for r in bad}.iteritems(),
key=itemgetter(1))
print[k for k in good_scores[:5]]
print[k for k in bad_scores]
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 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 fetch_chimera_scores(folder, reference):
# Check and read the reference data
reference_files = find_reference_files(reference)
# Check and read the query data
barcode_files = split_results(folder)
# Check and read the log data
log_data = read_log_data(folder)
good = []
bad = []
for barcode, reference in reference_files.iteritems():
sample_file = barcode_files[barcode]
alignments = full_align_best_reference(sample_file, reference)
by_reference = hits_by_reference(alignments)
best, rest = separate_best_hits(by_reference)
good += best
bad += rest
print "Found %s 'good' and %s 'bad' consensus sequences" % (len(good), len(bad))
good_scores = sorted({r.qname: log_data[r.qname] for r in good}.iteritems(), key=itemgetter(1), reverse=True)
bad_scores = sorted({r.qname: log_data[r.qname] for r in bad}.iteritems(), key=itemgetter(1))
print [k for k in good_scores[:5]]
