def test_cnv_simple(genomic_sequence_2013, gene_models_2013, variant_type,
location, effect_type, effect_genes):
effects = VariantAnnotator.annotate_variant(gene_models_2013,
genomic_sequence_2013,
loc=location,
variant_type=variant_type)
assert effects
et, eg, _ = \
VariantAnnotator.effect_simplify(effects)
assert et == effect_type
assert set(eg) == set(effect_genes)
def test_chr2_32853362_ins_var(genomic_sequence_2013, gene_models_2013):
effects = VariantAnnotator.annotate_variant(
gene_models_2013,
genomic_sequence_2013,
loc="6:157527729",
var="complex(CTGG->ATAG)",
)
assert len(effects) == 2
effects_sorted = sorted(effects, key=lambda k: k.transcript_id)
assert effects_sorted[0].gene == "ARID1B"
assert effects_sorted[0].transcript_id == "NM_017519_1"
assert effects_sorted[0].strand == "+"
assert effects_sorted[0].effect == "nonsense"
# assert effects_sorted[0].prot_pos is None
# assert effects_sorted[0].prot_length is None
assert effects_sorted[0].aa_change == "His,Trp->Gln,End"
assert effects_sorted[1].gene == "ARID1B"
assert effects_sorted[1].transcript_id == "NM_020732_1"
assert effects_sorted[1].strand == "+"
assert effects_sorted[1].effect == "nonsense"
assert effects_sorted[1].prot_pos, 1
assert effects_sorted[1].prot_length, 843
assert effects_sorted[1].aa_change == "His,Trp->Gln,End"
def test_chr1_802610_867930_CNV_var(genome_2013, gene_models_2013):
effects = VariantAnnotation.annotate_variant(gene_models_2013,
genome_2013,
loc="1:802610-867930",
var="CNV+")
assert len(effects) == 3
effects_sorted = sorted(effects, key=lambda k: k.transcript_id)
assert effects_sorted[0].gene == "SAMD11"
assert effects_sorted[0].transcript_id == "NM_152486_1"
assert effects_sorted[0].strand == "+"
assert effects_sorted[0].effect == "unknown"
# assert effects_sorted[0].prot_pos is None
# assert effects_sorted[0].prot_length is None
assert effects_sorted[0].aa_change is None
assert effects_sorted[1].gene == "LOC100130417"
assert effects_sorted[1].transcript_id == "NR_026874_1"
assert effects_sorted[1].strand == "-"
assert effects_sorted[1].effect == "unknown"
# assert effects_sorted[1].prot_pos is None
# assert effects_sorted[1].prot_length is None
assert effects_sorted[1].aa_change is None
assert effects_sorted[2].gene == "FAM41C"
assert effects_sorted[2].transcript_id == "NR_027055_1"
assert effects_sorted[2].strand == "-"
assert effects_sorted[2].effect == "unknown"
# assert effects_sorted[2].prot_pos is None
# assert effects_sorted[2].prot_length is None
assert effects_sorted[2].aa_change is None
def __init__(self, config, genomes_db, **kwargs):
super(EffectAnnotatorBase, self).__init__(config, genomes_db)
self.effect_annotator = VariantAnnotator(
self.genomic_sequence,
self.gene_models,
promoter_len=self.config.options.prom_len,
)
self.columns = OrderedDict()
for col_name, _col_type in self.COLUMNS_SCHEMA:
self.columns[col_name] = getattr(self.config.columns, col_name)
def test_chr1_120387132_del_var(genome_2013, gene_models_2013):
[effect] = VariantAnnotation.annotate_variant(gene_models_2013,
genome_2013,
loc="1:120387132",
var="del(71)")
assert effect.gene == "NBPF7"
assert effect.transcript_id == "NM_001047980_1"
assert effect.strand == "-"
assert effect.effect == "noStart"
assert effect.prot_pos == 1
assert effect.prot_length == 421
assert effect.aa_change is None
def test_chr2_237172988_ins_var(genome_2013, gene_models_2013):
[effect] = VariantAnnotation.annotate_variant(gene_models_2013,
genome_2013,
loc="2:237172988",
var="ins(TTGTTACG)")
assert effect.gene == "ASB18"
assert effect.transcript_id == "NM_212556_1"
assert effect.strand == "-"
assert effect.effect == "noStart"
assert effect.prot_pos == 1
assert effect.prot_length == 466
assert effect.aa_change is None
def test_chr5_75902128_sub_var(genomic_sequence_2013, gene_models_2013):
[effect] = VariantAnnotator.annotate_variant(
gene_models_2013,
genomic_sequence_2013,
loc="5:75902128",
var="sub(C->T)",
)
assert effect.gene == "IQGAP2"
assert effect.transcript_id == "NM_006633_1"
assert effect.strand == "+"
assert effect.effect == "nonsense"
# assert effect.prot_pos is None
# assert effect.prot_length is None
assert effect.aa_change == "Arg->End"
def test_synonymous_complex_var(genomic_sequence_2013, gene_models_2013):
[effect] = VariantAnnotator.annotate_variant(
gene_models_2013,
genomic_sequence_2013,
loc="1:897349",
var="complex(GG->AA)",
)
assert effect.gene == "KLHL17"
assert effect.transcript_id == "NM_198317_1"
assert effect.strand == "+"
assert effect.effect == "missense"
assert effect.prot_pos == 211
assert effect.prot_length == 642
assert effect.aa_change == "Lys,Ala->Lys,Thr"
def test_just_next_to_splice_site_var(genomic_sequence_2013, gene_models_2013):
effects = VariantAnnotator.annotate_variant(
gene_models_2013, genomic_sequence_2013, loc="5:86705101", var="del(4)"
)
assert len(effects) == 2
effects_sorted = sorted(effects, key=lambda k: k.transcript_id)
assert effects_sorted[0].gene == "CCNH"
assert effects_sorted[0].transcript_id == "NM_001199189_1"
assert effects_sorted[0].strand == "-"
assert effects_sorted[0].effect == "intron"
# assert effects_sorted[0].prot_pos is None
# assert effects_sorted[0].prot_length is None
assert effects_sorted[0].aa_change is None
assert effects_sorted[1].gene == "CCNH"
assert effects_sorted[1].transcript_id == "NM_001239_1"
assert effects_sorted[1].strand == "-"
assert effects_sorted[1].effect == "intron"
# assert effects_sorted[1].prot_pos is None
# assert effects_sorted[1].prot_length is None
assert effects_sorted[1].aa_change is None
def cli_vcf(argv=sys.argv[1:]):
parser = argparse.ArgumentParser(
description="VCF variants effect annotator",
conflict_handler="resolve",
formatter_class=argparse.RawDescriptionHelpFormatter,
)
cli_genome_options(parser)
parser.add_argument("input_filename", help="input VCF variants file name")
parser.add_argument("output_filename",
nargs="?",
help="output file name (default: stdout)")
args = parser.parse_args(argv)
genomic_sequence, gene_models = parse_cli_genome_options(args)
assert genomic_sequence is not None
assert gene_models is not None
annotator = VariantAnnotation(genomic_sequence,
gene_models,
promoter_len=args.promoter_len)
assert os.path.exists(args.input_filename), args.input_filename
infile = pysam.VariantFile(args.input_filename)
if args.output_filename is None:
outfile = sys.stdout
else:
outfile = open(args.output_filename, "w")
start = time.time()
# Transfer VCF header
header = infile.header
header.add_meta("variant_effect_annotation",
"GPF variant effects annotation")
header.add_meta("variant_effect_annotation_command",
'"{}"'.format(" ".join(sys.argv)))
header.info.add("ET", ".", "String", "effected type")
header.info.add("EG", ".", "String", "effected gene")
header.info.add("ED", ".", "String", "effect details")
print(str(header), file=outfile, end="")
counter = 0
for counter, variant in enumerate(infile):
effect_types = []
effect_genes = []
effect_details = []
eg = ""
ed = ""
for alt in variant.alts:
effects = annotator.do_annotate_variant(
chrom=variant.chrom,
position=variant.pos,
ref=variant.ref,
alt=alt,
)
et, eg, ed = annotator.effect_description(effects)
ed = ed.replace(";", "|")
effect_types.append(et)
effect_genes.append(eg)
effect_details.append(ed)
effect_types = ",".join(effect_types)
effect_genes = ",".join(effect_genes)
effect_details = ",".join(effect_details)
variant.info["ET"] = effect_types
variant.info["EG"] = eg
variant.info["ED"] = ed
print(str(variant), file=outfile, end="")
if (counter + 1) % 1000 == 0:
elapsed = time.time() - start
print(
f"processed {counter + 1} variants in {elapsed:0.2f} sec",
file=sys.stderr,
)
infile.close()
if args.output_filename:
outfile.close()
elapsed = time.time() - start
print(80 * "=", file=sys.stderr)
print(
f"DONE: {counter + 1} variants in {elapsed:0.2f} sec",
file=sys.stderr,
)
print(80 * "=", file=sys.stderr)
def cli(argv=sys.argv[1:]):
parser = argparse.ArgumentParser(
description="variants effect annotator",
conflict_handler="resolve",
formatter_class=argparse.RawDescriptionHelpFormatter,
)
cli_genome_options(parser)
cli_variants_options(parser)
parser.add_argument("input_filename",
nargs="?",
help="input variants file name")
parser.add_argument("output_filename",
nargs="?",
help="output file name (default: stdout)")
args = parser.parse_args(argv)
genomic_sequence, gene_models = parse_cli_genome_options(args)
assert genomic_sequence is not None
assert gene_models is not None
annotator = VariantAnnotation(genomic_sequence,
gene_models,
promoter_len=args.promoter_len)
variant_columns = parse_cli_variants_options(args)
if args.input_filename == "-" or args.input_filename is None:
infile = sys.stdin
else:
assert os.path.exists(args.input_filename), args.input_filename
infile = open(args.input_filename, "r")
if args.output_filename is None:
outfile = sys.stdout
else:
outfile = open(args.output_filename, "w")
start = time.time()
header = None
if args.no_header:
for key, value in variant_columns.items():
variant_columns[key] = int(value)
else:
line = infile.readline().strip()
header = [c.strip() for c in line.split("\t")]
for key, value in variant_columns.items():
assert value in header
variant_columns[key] = header.index(value)
header.extend(["effectType", "effectGene", "effectDetails"])
print("\t".join(header), file=outfile)
counter = 0
for counter, line in enumerate(infile):
if line[0] == "#":
continue
columns = [c.strip() for c in line.split("\t")]
variant = {
key: columns[value]
for key, value in variant_columns.items()
}
effects = annotator.do_annotate_variant(**variant)
desc = annotator.effect_description(effects)
columns.extend(desc)
print("\t".join(columns), file=outfile)
if (counter + 1) % 1000 == 0:
elapsed = time.time() - start
print(
f"processed {counter + 1} lines in {elapsed:0.2f} sec",
file=sys.stderr,
)
infile.close()
if args.output_filename:
outfile.close()
elapsed = time.time() - start
print(80 * "=", file=sys.stderr)
print(
f"DONE: {counter + 1} variants in {elapsed:0.2f} sec",
file=sys.stderr,
)
print(80 * "=", file=sys.stderr)