def test_add_second_vcf_variant_set(self):
# This VCF only has one Variant which is not in the first VCF
vcf = VCF(f="tests/vcf_tests/test2.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantSet.objects().count() == 3
assert VariantCallSet.objects().count() == 2
assert VariantCall.objects().count() == 42
assert Variant.objects().count() == 22
assert len(Variant.objects()[0].variant_sets) == 3
assert len(
Variant.objects.get(
names="UNION_BC_k31_var_147").variant_sets) == 3
def _remove_variant_set(self, variant_set_name):
vs = VariantSet.objects.get(name=variant_set_name,
reference_set=self.reference_set)
for call_set in VariantCallSet.objects(variant_sets=vs):
call_set.variant_sets.remove(vs)
call_set.save()
# Remove calls from callsets that only have this variantset
if len(call_set.variant_sets) < 2:
VariantCall.objects(call_set=call_set).delete()
call_set.delete()
# Remove variants that are ONLY from this variant set
Variant.objects(variant_sets=vs, variant_sets__size=2).delete()
VariantSetMetadata.objects(variant_set=vs).delete()
vs.delete()
def test_add_add_variants_and_calls(self):
vcf = VCF(f="tests/vcf_tests/test.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantCall.objects().count() == 21
assert Variant.objects().count() == 21
def get_context(pos, kmer):
context = []
for variant in Variant.objects(start__ne=pos,
start__gt=pos - kmer,
start__lt=pos + kmer):
for split_variant in variant.split():
context.append(split_variant)
return context
def test_add_new_vcf_variant_set(self):
vcf = VCF(f="tests/vcf_tests/test.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
# We create a global variant set as well as one for the individual VCF
assert VariantSet.objects().count() == 2
vs = VariantSet.objects()[0]
assert len(Variant.objects()[0].variant_sets) == 2
assert vs.name == "test.vcf"
def test_add_second_vcf_variant_set(self):
# This VCF only has one Variant which is not in the first VCF
vcf = VCF(f="tests/vcf_tests/test3.vcf",
reference_set_id=self.reference_set.id,
method="CORTEX")
vcf.add_to_database()
assert VariantSet.objects().count() == 2
assert VariantCallSet.objects().count() == 1
assert VariantCall.objects().count() == 106
assert Variant.objects().count() == 106
assert Variant.snps().count() == 89
assert Variant.indels().count() == 17
assert Variant.insertions().count() == 8
assert Variant.deletions().count() == 8
assert Variant.ph_snps.count() == 1
def run(parser, args):
DB = connect('mykrobe-%s' % (args.db_name))
if DB is not None:
try:
Variant.objects()
logging.info(
"Connected to mykrobe-%s" % (args.db_name))
except (ServerSelectionTimeoutError):
DB = None
logging.warning(
"Could not connect to database. Continuing without using genetic backgrounds")
mutations = []
reference = os.path.basename(args.reference_filepath).split('.fa')[0]
if args.vcf:
run_make_probes_from_vcf_file(args)
elif args.genbank:
aa2dna = GeneAminoAcidChangeToDNAVariants(
args.reference_filepath,
args.genbank)
if args.text_file:
with open(args.text_file, 'r') as infile:
reader = csv.reader(infile, delimiter="\t")
for row in reader:
gene, mutation_string, alphabet = row
if alphabet == "DNA":
protein_coding_var = False
else:
protein_coding_var = True
for var_name in aa2dna.get_variant_names(
gene, mutation_string, protein_coding_var):
mutation = Mutation(reference=reference,
var_name=var_name,
gene=aa2dna.get_gene(gene),
mut=mutation_string)
mutations.append(mutation)
else:
for variant in args.variants:
gene, mutation = variant.split("_")
for var_name in aa2dna.get_variant_names(gene, mutation):
mutations.append(
Mutation(reference=reference,
var_name=var_name,
gene=gene,
mut=mutation))
else:
if args.text_file:
with open(args.text_file, 'r') as infile:
reader = csv.reader(infile, delimiter="\t")
for row in reader:
gene_name, pos, ref, alt, alphabet = row
if gene_name == "ref":
mutations.append(
Mutation(
reference=reference,
var_name="".join([ref, pos, alt])))
else:
mutations.append(
Mutation(
reference=reference,
var_name=row[0]))
else:
mutations.extend(Mutation(reference=reference, var_name=v)
for v in args.variants)
al = AlleleGenerator(
reference_filepath=args.reference_filepath,
kmer=args.kmer)
for enum, mut in enumerate(mutations):
if enum % 100 == 0:
logger.info(
"%i of %i - %f%%" % (enum, len(mutations), round(100*enum/len(mutations), 2)))
variant_panel = make_variant_probe(
al, mut.variant, args.kmer, DB=DB, no_backgrounds=args.no_backgrounds)
if variant_panel is not None:
for i, ref in enumerate(variant_panel.refs):
try:
gene_name = mut.gene.name
except AttributeError:
gene_name = "NA"
sys.stdout.write(
">ref-%s?var_name=%s&num_alts=%i&ref=%s&enum=%i&gene=%s&mut=%s\n" %
(mut.mut, mut.variant.var_name, len(
variant_panel.alts), mut.reference, i, gene_name, mut.mut))
sys.stdout.write("%s\n" % ref)
for i, a in enumerate(variant_panel.alts):
sys.stdout.write(">alt-%s?var_name=%s&enum=%i&gene=%s&mut=%s\n" %
(mut.mut, mut.variant.var_name, i, gene_name, mut.mut))
sys.stdout.write("%s\n" % a)
else:
logging.warning(
"All variants failed for %s_%s - %s" %
(mut.gene, mut.mut, mut.variant))
def run(parser, args):
# There's no need to try to connect to database if we're not doing backgrounds
if args.no_backgrounds:
logger.info(
"Not connecting to database, because --no-backgrounds option used")
DB = None
else:
DB = connect("%s-%s" % (DB_PREFIX, args.db_name))
if DB is not None:
try:
Variant.objects()
logger.info("Connected to %s-%s" % (DB_PREFIX, args.db_name))
except (ServerSelectionTimeoutError):
DB = None
logger.warning(
"Could not connect to database. Continuing without using genetic backgrounds"
)
mutations = []
lineages = set()
reference = os.path.basename(args.reference_filepath).split(".fa")[0]
if args.vcf:
run_make_probes_from_vcf_file(args)
elif args.genbank:
aa2dna = GeneAminoAcidChangeToDNAVariants(args.reference_filepath,
args.genbank)
if args.text_file:
with open(args.text_file, "r") as infile:
reader = csv.reader(infile, delimiter="\t")
for row in reader:
gene, mutation_string, alphabet = row
if alphabet == "DNA":
protein_coding_var = False
else:
protein_coding_var = True
for var_name in aa2dna.get_variant_names(
gene, mutation_string, protein_coding_var):
mutation = Mutation(
reference=reference,
var_name=var_name,
gene=aa2dna.get_gene(gene),
mut=mutation_string,
protein_coding_var=protein_coding_var,
)
mutations.append(mutation)
else:
for variant in args.variants:
gene, mutation = variant.split("_")
for var_name in aa2dna.get_variant_names(gene, mutation):
mutations.append(
Mutation(
reference=reference,
var_name=var_name,
gene=gene,
mut=mutation,
))
else:
if args.text_file:
mutations, lineages = load_dna_vars_txt_file(
args.text_file, reference)
if args.lineage:
with open(args.lineage, "w") as f:
json.dump(lineages, f, sort_keys=True, indent=2)
else:
mutations.extend(
Mutation(reference=reference, var_name=v)
for v in args.variants)
al = AlleleGenerator(reference_filepath=args.reference_filepath,
kmer=args.kmer)
for enum, mut in enumerate(mutations):
if enum % 100 == 0:
logger.info(
"%i of %i - %f%%" %
(enum, len(mutations), round(100 * enum / len(mutations), 2)))
variant_panel = make_variant_probe(al,
mut.variant,
args.kmer,
DB=DB,
no_backgrounds=args.no_backgrounds)
if variant_panel is not None:
for i, ref in enumerate(variant_panel.refs):
try:
gene_name = mut.gene.name
except AttributeError:
gene_name = "NA"
sys.stdout.write(
">ref-%s?var_name=%s&num_alts=%i&ref=%s&enum=%i&gene=%s&mut=%s\n"
% (
mut.mutation_output_name,
mut.variant.var_name,
len(variant_panel.alts),
mut.reference,
i,
gene_name,
mut.mutation_output_name,
))
sys.stdout.write("%s\n" % ref)
for i, a in enumerate(variant_panel.alts):
sys.stdout.write(
">alt-%s?var_name=%s&enum=%i&gene=%s&mut=%s\n" % (
mut.mutation_output_name,
mut.variant.var_name,
i,
gene_name,
mut.mutation_output_name,
))
sys.stdout.write("%s\n" % a)
else:
logger.warning("All variants failed for %s_%s - %s" %
(mut.gene, mut.mutation_output_name, mut.variant))