def setup_class(self):
BED4_file = TEST_PATH + "/data/hxb2_pol.bed"
mutation_db_file = TEST_PATH + "/data/mutation_db.tsv"
genes = parse_BED4_file(BED4_file, "hxb2_pol")
self.mutation_db = MutationDB(mutation_db_file, genes)
def cli(ctx, bam, reference, bed4_file, output):
"""This script builds an amino acid census and returns its coverage.
The BAM alignment file corresponds to a pileup of sequences aligned to
the REFERENCE. A BAM index file (.bai) must also be present and, except
for the extension, have the same name as the BAM file. The REFERENCE must
be in FASTA format. The BED4_FILE must be a BED file with at least 4
columns and specify the gene locations within the REFERENCE.
The output is in CSV format."""
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
for r in rs:
# Create a MappedReadCollection object
mapped_read_collection_arr.append(parse_mapped_reads_from_bam(r, bam))
# Parse the genes from the gene file
genes = parse_BED4_file(bed4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]["frame"])
# Create an AACensus object
aa_census = AACensus(reference, mapped_read_collection_arr, genes, frames)
if output:
output.write(aa_census.coverage(frames))
output.close()
else:
click.echo(aa_census.coverage(frames))
def test_invalid_BED4_file(self):
"""Tests to make sure that an exception is raised when attempting to
parse an invalid BED4 file.
"""
# Create an invalid genes file
invalid_BED4_file = os.path.join(
os.path.dirname(os.path.abspath(__file__)), "data",
"invalid_BED4_file.bed")
ref_name = "ref1"
with open(invalid_BED4_file, "w+") as f:
f.write("%s\t0\t100\t0\n" % ref_name)
# Add a genes reference name that doesn't match
# This should raise a ValueError
f.write("different_reference\t101\t200\t2")
with pytest.raises(ValueError):
parse_BED4_file(invalid_BED4_file, ref_name)
os.remove(invalid_BED4_file)
def test_from_aacensus(self):
bam = TEST_PATH + "/data/align.bam"
BED4_file = TEST_PATH + "/data/hxb2_pol.bed"
mapped_read_collection_arr = []
error_rate = 0.0038
# Create a MappedReadCollection object
for r in self.references:
mapped_read_collection_arr.append(
parse_mapped_reads_from_bam(r, bam))
variants = NTVariantCollection.from_mapped_read_collections(
error_rate, self.references, *mapped_read_collection_arr)
variants.filter('q30', 'QUAL<30', True)
variants.filter('ac5', 'AC<5', True)
variants.filter('dp100', 'DP<100', True)
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants)
# Parse the genes from the gene file
genes = parse_BED4_file(BED4_file, self.references[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
aa_census = AACensus(self.reference, mapped_read_collection_arr, genes,
frames)
test_variants = CodonVariantCollection.from_aacensus(aa_census)
ref_seq = self.references[0].seq
for gene in test_variants.variants:
assert gene in genes
for pos in test_variants.variants[gene]:
for frame in frames:
nt_pos = pos / 3 - frame
assert nt_pos >= genes[gene]['start'] or nt_pos <= genes[
gene]['end']
for codon in test_variants.variants[gene][pos]:
ref_codon = ref_seq[(pos):(pos) + 3].lower()
assert codon != ref_codon
def cli(ctx, bam, reference, variants, bed4_file, min_freq, mutation_db,
reporting_threshold, output):
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
for r in rs:
# Create a MappedReadCollection object
mapped_read_collection_arr.append(parse_mapped_reads_from_bam(r, bam))
variants_obj = parse_nt_variants_from_vcf(variants, rs)
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants_obj)
# Parse the genes from the gene file
genes = parse_BED4_file(bed4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
# Create an AACensus object
aa_census = AACensus(reference, mapped_read_collection_arr, genes, frames)
# Create AAVar collection and print the aavf file
aa_vars = AAVariantCollection.from_aacensus(aa_census)
# Filter for mutant frequency
aa_vars.filter('mf' + str(min_freq), 'freq<' + str(min_freq), True)
# Build the mutation database
mutation_db = MutationDB(mutation_db, genes)
# Generate the mutation report
if output:
output.write(
aa_vars.report_dr_mutations(mutation_db, reporting_threshold))
output.close()
else:
click.echo(
aa_vars.report_dr_mutations(mutation_db, reporting_threshold))
def test_valid_BED4_file(self):
"""Tests to make sure that valid BED4+ files (bed files with at least 4
columns) are parsed properly.
"""
# Create a valid BED4 file
valid_BED4_file = os.path.join(
os.path.dirname(os.path.abspath(__file__)), "data",
"valid_BED4_file.bed")
ref_name = "ref1"
genes = {
"gene1": {
"start": 0,
"end": 100
},
"gene 2": {
"start": 101,
"end": 200
}, # Spaces are allowed in the gene name
"gene3": {
"start": 201,
"end": 300
}
}
with open(valid_BED4_file, "w+") as f:
for gene in genes:
f.write(
"%s\t%s\t%s\t%s\n" %
(ref_name, genes[gene]["start"], genes[gene]["end"], gene))
parsed_genes = parse_BED4_file(valid_BED4_file, ref_name)
for gene in parsed_genes:
assert gene in genes
assert parsed_genes[gene]["start"] == genes[gene]["start"]
assert parsed_genes[gene]["end"] == genes[gene]["end"]
assert parsed_genes[gene]["frame"] == genes[gene]["start"] % 3
os.remove(valid_BED4_file)
def codonvar(bam, reference, offset, bed4_file, variants, error_rate, output):
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
# Create a MappedReadCollection object
for r in rs:
mapped_read_collection_arr.append(parse_mapped_reads_from_bam(r, bam))
if variants:
variants_obj = parse_nt_variants_from_vcf(variants, rs)
else:
variants = NTVariantCollection.from_mapped_read_collections(
error_rate, rs, *mapped_read_collection_arr)
variants.filter('q30', 'QUAL<30', True)
variants.filter('ac5', 'AC<5', True)
variants.filter('dp100', 'DP<100', True)
variants_obj = variants
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants_obj)
# Parse the genes from the gene file
genes = parse_BED4_file(bed4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
aa_census = AACensus(reference, mapped_read_collection_arr, genes, frames)
codon_variants = CodonVariantCollection.from_aacensus(aa_census)
if output:
output.write(codon_variants.to_csv_file(offset))
output.close()
else:
click.echo(codon_variants.to_csv_file(offset))
def setup(self):
self.reference = TEST_PATH + "/data/hxb2_pol.fas"
bam = TEST_PATH + "/data/align.bam"
BED4_file = TEST_PATH + "/data/hxb2_pol.bed"
mutation_db = TEST_PATH + "/data/mutation_db.tsv"
min_freq = 0.01
rs = parse_references_from_fasta(self.reference)
mapped_read_collection_arr = []
for r in rs:
# Create a MappedReadCollection object
mapped_read_collection_arr.append(
parse_mapped_reads_from_bam(r, bam))
variants_obj = parse_nt_variants_from_vcf(VARIANTS_FILE, rs)
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants_obj)
# Parse the genes from the gene file
genes = parse_BED4_file(BED4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
# Create an AACensus object
aa_census = AACensus(self.reference, mapped_read_collection_arr, genes,
frames)
# Find the AA mutations
self.aa_collection = AAVariantCollection.from_aacensus(aa_census)
# Build the mutation database
self.mutation_db = MutationDB(mutation_db, genes)
def setup_class(self):
reference = TEST_PATH + "/data/hxb2_pol.fas"
bam = TEST_PATH + "/data/align.bam"
BED4_file = TEST_PATH + "/data/hxb2_pol.bed"
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
for r in rs:
# create MappedReadCollection object
mapped_read_collection_arr.append(
parse_mapped_reads_from_bam(r, bam))
genes = parse_BED4_file(BED4_file, rs[0].name)
# Determine which frames our genes are in
self.frames = set()
for gene in genes:
self.frames.add(genes[gene]["frame"])
self.aa_census = AACensus(reference, mapped_read_collection_arr, genes,
self.frames)
def setup(self):
bam = TEST_PATH + "/data/align.bam"
reference = TEST_PATH + "/data/hxb2_pol.fas"
BED4_file = TEST_PATH + "/data/hxb2_pol.bed"
error_rate = 0.0038
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
# Create a MappedReadCollection object
for r in rs:
mapped_read_collection_arr.append(
parse_mapped_reads_from_bam(r, bam))
variants = NTVariantCollection.from_mapped_read_collections(
error_rate, rs, *mapped_read_collection_arr)
variants.filter('q30', 'QUAL<30', True)
variants.filter('ac5', 'AC<5', True)
variants.filter('dp100', 'DP<100', True)
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants)
# Parse the genes from the gene file
genes = parse_BED4_file(BED4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
aa_census = AACensus(reference, mapped_read_collection_arr, genes,
frames)
self.codon_variants = CodonVariantCollection.from_aacensus(aa_census)
def __init__(self, id, output_dir, reads, reference, BED4_file,
mutation_db, quiet, consensus_pct):
self.id = id
self.output_dir = output_dir
self.reads = reads
self.reference = reference
self.mutation_db = mutation_db
self.BED4_file = BED4_file
self.quiet = quiet
self.consensus_pct = consensus_pct
self.input_size = 0
self.determine_input_size()
self.references = parse_references_from_fasta(self.reference)
self.genes = parse_BED4_file(BED4_file, self.references[0].name)
self.quality = QualityControl()
if not os.path.isdir(output_dir):
os.mkdir(output_dir)
self.filtered_reads_dir = "%s/filtered.fastq" % output_dir
def aavar(bam, reference, bed4_file, variants, mutation_db, min_freq,
error_rate, output):
rs = parse_references_from_fasta(reference)
mapped_read_collection_arr = []
for r in rs:
# Create a MappedReadCollection object
mapped_read_collection_arr.append(parse_mapped_reads_from_bam(r, bam))
if variants:
variants_obj = parse_nt_variants_from_vcf(variants, rs)
else:
variants = NTVariantCollection.from_mapped_read_collections(
error_rate, rs, *mapped_read_collection_arr)
variants.filter('q30', 'QUAL<30', True)
variants.filter('ac5', 'AC<5', True)
variants.filter('dp100', 'DP<100', True)
variants_obj = variants
# Mask the unconfident differences
for mrc in mapped_read_collection_arr:
mrc.mask_unconfident_differences(variants_obj)
# Parse the genes from the gene file
genes = parse_BED4_file(bed4_file, rs[0].name)
# Determine which frames our genes are in
frames = set()
for gene in genes:
frames.add(genes[gene]['frame'])
# Create an AACensus object
aa_census = AACensus(reference, mapped_read_collection_arr, genes, frames)
# Create AAVar collection and print the aavf file
aa_vars = AAVariantCollection.from_aacensus(aa_census)
# Filter for mutant frequency
aa_vars.filter('mf0.01', 'freq<0.01', True)
# Build the mutation database and update collection
if mutation_db is not None:
mutation_db = MutationDB(mutation_db, genes)
aa_vars.apply_mutation_db(mutation_db)
aavf_obj = aa_vars.to_aavf_obj("aavar", os.path.basename(reference),
CONFIDENT)
records = list(aavf_obj)
if output:
writer = parser.Writer(output, aavf_obj)
else:
writer = parser.Writer(sys.stdout, aavf_obj)
for record in records:
writer.write_record(record)
if output:
output.close
writer.close()
