def test_variant_case_no_genes(adapter, case_obj, variant_obj):
"""Test to preprocess a variant"""
# GIVEN a variant wihtout gene info
assert variant_obj.get("genes") is None
# GIVEN that no region vcf exists
assert "region_vcf_file" not in case_obj
# WHEN adding info
variant_case(adapter, case_obj, variant_obj)
# THEN assert no region vcf was added since there where no gene info
assert "region_vcf_file" not in case_obj
def test_variant_case(adapter, case_obj, variant_obj):
"""Test to preprocess a variant"""
# GIVEN a variant WITH gene info
variant_obj["genes"] = [
{
"hgnc_id": 1
},
{
"hgnc_id": 2,
"common": {
"chromosome": "1",
"start": "10",
"end": "100"
}
},
]
# GIVEN a variant without gene info
assert case_obj.get("region_vcf_file") is None
variant_case(adapter, case_obj, variant_obj)
# THEN assert that the region VCF was created
assert case_obj.get("region_vcf_file") is not None
def variant(
store,
institute_id,
case_name,
variant_id=None,
variant_obj=None,
add_case=True,
add_other=True,
get_overlapping=True,
add_compounds=True,
variant_category=None,
variant_type=None,
case_obj=None,
institute_obj=None,
):
"""Pre-process a single variant for the detailed variant view.
Adds information from case and institute that is not present on the variant
object
Args:
store(scout.adapter.MongoAdapter)
institute_id(str)
case_name(str)
variant_id(str)
variant_obj(dict)
add_case(bool): If info about files case should be added
add_other(bool): If information about other causatives should be added
get_overlapping(bool): If overlapping variants should be collected
variant_type(str): in ["clinical", "research"]
variant_category(str): ["snv", "str", "sv", "cancer", "cancer_sv"]
institute_obj(scout.models.Institute)
case_obj(scout.models.Case)
Returns:
variant_info(dict): {
'variant': <variant_obj>,
'causatives': <list(other_causatives)>,
'events': <list(events)>,
'overlapping_svs': <list(overlapping svs)>,
'manual_rank_options': MANUAL_RANK_OPTIONS,
'cancer_tier_options': CANCER_TIER_OPTIONS,
'dismiss_variant_options': DISMISS_VARIANT_OPTIONS,
'ACMG_OPTIONS': ACMG_OPTIONS,
'igv_tracks': IGV_TRACKS,
'evaluations': <list(evaluations)>,
}
"""
if not (institute_obj and case_obj):
institute_obj, case_obj = institute_and_case(store, institute_id,
case_name)
# If the variant is already collected we skip this part
if not variant_obj:
# NOTE this will query with variant_id == document_id, not the variant_id.
variant_obj = store.variant(variant_id)
if variant_obj is None:
return None
variant_type = variant_type or variant_obj.get("variant_type", "clinical")
# request category specific variant display
variant_category = variant_obj.get("category", "snv")
LOG.debug("Variant category {}".format(variant_category))
variant_id = variant_obj["variant_id"]
genome_build = str(case_obj.get("genome_build", "37"))
if genome_build not in ["37", "38"]:
genome_build = "37"
panels = default_panels(store, case_obj)
variant_obj = add_gene_info(store,
variant_obj,
gene_panels=panels,
genome_build=genome_build)
# Add information about bam files and create a region vcf
if add_case:
variant_case(store, case_obj, variant_obj)
# Collect all the events for the variant
events = list(
store.events(institute_obj, case=case_obj, variant_id=variant_id))
def variant(
store,
institute_id,
case_name,
variant_id=None,
variant_obj=None,
add_case=True,
add_other=True,
get_overlapping=True,
add_compounds=True,
variant_category=None,
variant_type=None,
case_obj=None,
institute_obj=None,
):
"""Pre-process a single variant for the detailed variant view.
Adds information from case and institute that is not present on the variant
object
Args:
store(scout.adapter.MongoAdapter)
institute_id(str)
case_name(str)
variant_id(str)
variant_obj(dict)
add_case(bool): If info about files case should be added
add_other(bool): If information about other causatives should be added
get_overlapping(bool): If overlapping variants should be collected
variant_type(str): in ["clinical", "research"]
variant_category(str): ["snv", "str", "sv", "cancer", "cancer_sv"]
institute_obj(scout.models.Institute)
case_obj(scout.models.Case)
Returns:
variant_info(dict): {
'variant': <variant_obj>,
'causatives': <list(other_causatives)>,
'events': <list(events)>,
'overlapping_svs': <list(overlapping svs)>,
'manual_rank_options': MANUAL_RANK_OPTIONS,
'cancer_tier_options': CANCER_TIER_OPTIONS,
'dismiss_variant_options': DISMISS_VARIANT_OPTIONS,
'ACMG_OPTIONS': ACMG_OPTIONS,
'evaluations': <list(evaluations)>,
}
"""
if not (institute_obj and case_obj):
institute_obj, case_obj = institute_and_case(store, institute_id,
case_name)
# If the variant is already collected we skip this part
if not variant_obj:
# NOTE this will query with variant_id == document_id, not the variant_id.
variant_obj = store.variant(variant_id)
if variant_obj is None:
return None
variant_type = variant_type or variant_obj.get("variant_type", "clinical")
# request category specific variant display
variant_category = variant_obj.get("category", "snv")
LOG.debug("Variant category {}".format(variant_category))
variant_id = variant_obj["variant_id"]
genome_build = str(case_obj.get("genome_build", "37"))
if genome_build not in ["37", "38"]:
genome_build = "37"
panels = default_panels(store, case_obj)
variant_obj = add_gene_info(store,
variant_obj,
gene_panels=panels,
genome_build=genome_build)
# Add information about bam files and create a region vcf
if add_case:
variant_case(store, case_obj, variant_obj)
# Collect all the events for the variant
events = store.events(institute_obj, case=case_obj, variant_id=variant_id)
for event in events:
event["verb"] = VERBS_MAP[event["verb"]]
# Comments are not on case level so these needs to be fetched on their own
variant_obj["comments"] = store.events(institute_obj,
case=case_obj,
variant_id=variant_id,
comments=True)
# Adds information about other causative variants
other_causatives = []
if add_other:
other_causatives = [
causative
for causative in store.other_causatives(case_obj, variant_obj)
]
# Gather display information for the genes
variant_obj.update(predictions(variant_obj.get("genes", [])))
# Prepare classification information for visualisation
classification = variant_obj.get("acmg_classification")
if isinstance(classification, int):
acmg_code = ACMG_MAP[variant_obj["acmg_classification"]]
variant_obj["acmg_classification"] = ACMG_COMPLETE_MAP[acmg_code]
# sort compounds on combined rank score
compounds = variant_obj.get("compounds", [])
if compounds:
# Gather display information for the compounds
for compound_obj in compounds:
compound_obj.update(predictions(compound_obj.get("genes", [])))
variant_obj["compounds"] = sorted(
variant_obj["compounds"],
key=lambda compound: -compound["combined_score"])
variant_obj["end_position"] = end_position(variant_obj)
# Add general variant links
variant_obj.update(get_variant_links(variant_obj, int(genome_build)))
variant_obj["frequencies"] = frequencies(variant_obj)
if variant_category in ["snv", "cancer"]:
# This is to convert a summary of frequencies to a string
variant_obj["frequency"] = frequency(variant_obj)
# Format clinvar information
variant_obj["clinsig_human"] = (clinsig_human(variant_obj)
if variant_obj.get("clnsig") else None)
# Add display information about callers
variant_obj["callers"] = callers(variant_obj, category=variant_category)
# Convert affection status to strings for the template
is_affected(variant_obj, case_obj)
if variant_obj.get("genetic_models"):
variant_models = set(
model.split("_", 1)[0] for model in variant_obj["genetic_models"])
all_models = variant_obj.get("all_models", set())
variant_obj["is_matching_inheritance"] = set.intersection(
variant_models, all_models)
# Prepare classification information for visualisation
classification = variant_obj.get("acmg_classification")
if isinstance(classification, int):
acmg_code = ACMG_MAP[variant_obj["acmg_classification"]]
variant_obj["acmg_classification"] = ACMG_COMPLETE_MAP[acmg_code]
evaluations = []
for evaluation_obj in store.get_evaluations(variant_obj):
evaluation(store, evaluation_obj)
evaluations.append(evaluation_obj)
case_clinvars = store.case_to_clinVars(case_obj.get("display_name"))
if variant_id in case_clinvars:
variant_obj["clinvar_clinsig"] = case_clinvars.get(
variant_id)["clinsig"]
overlapping_vars = []
if get_overlapping:
for var in store.overlapping(variant_obj):
var.update(predictions(var.get("genes", [])))
overlapping_vars.append(var)
variant_obj["end_chrom"] = variant_obj.get("end_chrom",
variant_obj["chromosome"])
dismiss_options = DISMISS_VARIANT_OPTIONS
if case_obj.get("track") == "cancer":
dismiss_options = {
**DISMISS_VARIANT_OPTIONS,
**CANCER_SPECIFIC_VARIANT_DISMISS_OPTIONS,
}
return {
"institute": institute_obj,
"case": case_obj,
"variant": variant_obj,
variant_category: True,
"causatives": other_causatives,
"events": events,
"overlapping_vars": overlapping_vars,
"manual_rank_options": MANUAL_RANK_OPTIONS,
"cancer_tier_options": CANCER_TIER_OPTIONS,
"dismiss_variant_options": dismiss_options,
"mosaic_variant_options": MOSAICISM_OPTIONS,
"ACMG_OPTIONS": ACMG_OPTIONS,
"evaluations": evaluations,
}