def clean(self):
cleaned_data = super(MendelianVariantSearchForm, self).clean()
if cleaned_data['search_mode'] not in ['standard_inheritance', 'custom_inheritance', 'gene_burden', 'allele_count', 'all_variants']:
raise forms.ValidationError("Invalid search mode: {}".format(cleaned_data['search_mode']))
if cleaned_data['search_mode'] == 'standard_inheritance' and not cleaned_data.get('inheritance_mode'):
raise forms.ValidationError("Inheritance mode is required for standard search. ")
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
parse_genotype_filter(cleaned_data)
parse_burden_filter(cleaned_data)
parse_allele_count_filter(cleaned_data)
search_spec = MendelianVariantSearchSpec()
search_spec.search_mode = cleaned_data['search_mode']
search_spec.inheritance_mode = cleaned_data.get('inheritance_mode')
search_spec.genotype_inheritance_filter = cleaned_data.get('genotype_filter')
search_spec.gene_burden_filter = cleaned_data.get('burden_filter')
search_spec.allele_count_filter = cleaned_data.get('allele_count_filter')
search_spec.variant_filter = cleaned_data.get('variant_filter')
search_spec.quality_filter = cleaned_data.get('quality_filter')
cleaned_data['search_spec'] = search_spec
return cleaned_data
def clean(self):
cleaned_data = super(CohortVariantSearchForm, self).clean()
if cleaned_data['search_mode'] not in [
'custom_inheritance', 'gene_burden'
]:
raise forms.ValidationError("Invalid search mode: {}".format(
cleaned_data['search_mode']))
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
parse_genotype_filter(cleaned_data)
parse_burden_filter(cleaned_data)
search_spec = MendelianVariantSearchSpec()
search_spec.search_mode = cleaned_data['search_mode']
search_spec.inheritance_mode = cleaned_data.get('inheritance_mode')
search_spec.genotype_inheritance_filter = cleaned_data.get(
'genotype_filter')
search_spec.gene_burden_filter = cleaned_data.get('gene_burden_filter')
search_spec.allele_count_filter = cleaned_data.get(
'allele_count_filter')
search_spec.variant_filter = cleaned_data.get('variant_filter')
search_spec.genotype_quality_filter = cleaned_data.get(
'quality_filter')
cleaned_data['search_spec'] = search_spec
return cleaned_data
def clean(self):
cleaned_data = super(CohortVariantSearchForm, self).clean()
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
search_spec = MendelianVariantSearchSpec()
search_spec.search_mode = cleaned_data['search_mode']
search_spec.inheritance_mode = cleaned_data.get('inheritance_mode')
search_spec.variant_filter = cleaned_data.get('variant_filter')
search_spec.quality_filter = cleaned_data.get('quality_filter')
cleaned_data['search_spec'] = search_spec
return cleaned_data
def clean(self):
cleaned_data = super(MendelianVariantSearchForm, self).clean()
if cleaned_data['search_mode'] not in ['standard_inheritance', 'custom_inheritance', 'gene_burden', 'allele_count', 'all_variants']:
raise forms.ValidationError("Invalid search mode: {}".format(cleaned_data['search_mode']))
if cleaned_data['search_mode'] == 'standard_inheritance' and not cleaned_data.get('inheritance_mode'):
raise forms.ValidationError("Inheritance mode is required for standard search. ")
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
parse_genotype_filter(cleaned_data)
parse_burden_filter(cleaned_data)
parse_allele_count_filter(cleaned_data)
search_spec = MendelianVariantSearchSpec()
search_spec.search_mode = cleaned_data['search_mode']
search_spec.inheritance_mode = cleaned_data.get('inheritance_mode')
search_spec.genotype_inheritance_filter = cleaned_data.get('genotype_filter')
search_spec.gene_burden_filter = cleaned_data.get('gene_burden_filter')
search_spec.allele_count_filter = cleaned_data.get('allele_count_filter')
search_spec.variant_filter = cleaned_data.get('variant_filter')
search_spec.genotype_quality_filter = cleaned_data.get('quality_filter')
cleaned_data['search_spec'] = search_spec
return cleaned_data
def mendelian_variant_search_spec(request):
project, family = get_project_and_family_for_user(request.user, request.GET)
# TODO: use form
search_hash = request.GET.get('search_hash')
search_spec_dict, variants = cache_utils.get_cached_results(project.project_id, search_hash)
search_spec = MendelianVariantSearchSpec.fromJSON(search_spec_dict)
if variants is None:
variants = api_utils.calculate_mendelian_variant_search(search_spec, family.xfamily())
else:
variants = [Variant.fromJSON(v) for v in variants]
add_extra_info_to_variants_family(get_reference(), family, variants)
return_type = request.GET.get('return_type')
if return_type == 'json' or not return_type:
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_spec': search_spec_dict,
})
elif request.GET.get('return_type') == 'csv':
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
indiv_ids = family.indiv_ids_with_variant_data()
headers = xbrowse_displays.get_variant_display_headers(get_mall(project.project_id), project, indiv_ids)
writer.writerow(headers)
for variant in variants:
fields = xbrowse_displays.get_display_fields_for_variant(get_mall(project.project_id), project, variant, indiv_ids)
writer.writerow(fields)
return response
def mendelian_variant_search_spec(request):
project, family = get_project_and_family_for_user(request.user, request.GET)
search_hash = request.GET.get('search_hash')
search_spec_dict, variants = cache_utils.get_cached_results(project.project_id, search_hash)
search_spec = MendelianVariantSearchSpec.fromJSON(search_spec_dict)
if variants is None:
variants = api_utils.calculate_mendelian_variant_search(search_spec, family.xfamily())
else:
variants = [Variant.fromJSON(v) for v in variants]
add_extra_info_to_variants_family(get_reference(), family, variants)
return_type = request.GET.get('return_type')
if return_type == 'json' or not return_type:
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_spec': search_spec_dict,
})
elif request.GET.get('return_type') == 'csv':
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
indiv_ids = family.indiv_ids_with_variant_data()
headers = xbrowse_displays.get_variant_display_headers(get_mall(project.project_id), project, indiv_ids)
writer.writerow(headers)
for variant in variants:
fields = xbrowse_displays.get_display_fields_for_variant(get_mall(project.project_id), project, variant, indiv_ids)
writer.writerow(fields)
return response
def clean(self):
cleaned_data = super(CohortVariantSearchForm, self).clean()
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
search_spec = MendelianVariantSearchSpec()
search_spec.search_mode = cleaned_data['search_mode']
search_spec.inheritance_mode = cleaned_data.get('inheritance_mode')
search_spec.variant_filter = cleaned_data.get('variant_filter')
search_spec.quality_filter = cleaned_data.get('quality_filter')
cleaned_data['search_spec'] = search_spec
return cleaned_data
def cohort_variant_search_spec(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
# TODO: use form
search_spec_dict, variants = cache_utils.get_cached_results(project.project_id, request.GET.get('search_hash'))
search_spec = MendelianVariantSearchSpec.fromJSON(search_spec_dict)
if variants is None:
variants = api_utils.calculate_mendelian_variant_search(search_spec, cohort.xfamily())
else:
variants = [Variant.fromJSON(v) for v in variants]
api_utils.add_extra_info_to_variants_cohort(get_reference(), cohort, variants)
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_spec': search_spec.toJSON(),
})
def cohort_variant_search_spec(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
# TODO: use form
search_spec_dict, variants = cache_utils.get_cached_results(project.project_id, request.GET.get('search_hash'))
search_spec = MendelianVariantSearchSpec.fromJSON(search_spec_dict)
if variants is None:
variants = api_utils.calculate_mendelian_variant_search(search_spec, cohort.xfamily())
else:
variants = [Variant.fromJSON(v) for v in variants]
api_utils.add_extra_info_to_variants_cohort(get_reference(), cohort, variants)
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_spec': search_spec.toJSON(),
})
def combine_mendelian_families_spec(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
search_hash = request.GET.get('search_hash')
search_spec, genes = cache_utils.get_cached_results(project.project_id, search_hash)
search_spec_obj = MendelianVariantSearchSpec.fromJSON(search_spec)
if request.GET.get('return_type') != 'csv' or not request.GET.get('group_by_variants'):
if genes is None:
genes = api_utils.calculate_combine_mendelian_families(family_group, search_spec)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
if request.GET.get('return_type') != 'csv':
return JSONResponse({
'is_error': False,
'genes': genes,
'search_spec': search_spec,
})
else:
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="family_group_results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
writer.writerow(["gene", "# families", "family list", "chrom", "start", "end"])
for gene in genes:
family_id_list = [family_id for (project_id, family_id) in gene["family_id_list"]]
writer.writerow(map(str, [gene["gene_name"], len(family_id_list), " ".join(family_id_list), gene["chr"], gene["start"], gene["end"], ""]))
return response
else:
# download results grouped by variant
indiv_id_list = []
for family in family_group.get_families():
indiv_id_list.extend(family.indiv_ids_with_variant_data())
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
headers = ['genes','chr','pos','ref','alt','worst_annotation' ]
headers.extend(project.get_reference_population_slugs())
headers.extend([ 'polyphen','sift','muttaster','fathmm'])
for indiv_id in indiv_id_list:
headers.append(indiv_id)
headers.append(indiv_id+'_gq')
headers.append(indiv_id+'_dp')
writer.writerow(headers)
mall = get_mall(project.project_id)
variant_key_to_individual_id_to_variant = defaultdict(dict)
variant_key_to_variant = {}
for family in family_group.get_families():
for variant in get_variants_with_inheritance_mode(
mall,
family.xfamily(),
search_spec_obj.inheritance_mode,
search_spec_obj.variant_filter,
search_spec_obj.quality_filter,
):
if len(variant.coding_gene_ids) == 0:
continue
variant_key = (variant.xpos, variant.ref, variant.alt)
variant_key_to_variant[variant_key] = variant
for indiv_id in family.indiv_ids_with_variant_data():
variant_key_to_individual_id_to_variant[variant_key][indiv_id] = variant
for variant_key in sorted(variant_key_to_individual_id_to_variant.keys()):
variant = variant_key_to_variant[variant_key]
individual_id_to_variant = variant_key_to_individual_id_to_variant[variant_key]
genes = [mall.reference.get_gene_symbol(gene_id) for gene_id in variant.coding_gene_ids]
fields = []
fields.append(','.join(genes))
fields.extend([
variant.chr,
str(variant.pos),
variant.ref,
variant.alt,
variant.annotation.get('vep_group', '.'),
])
for ref_population_slug in project.get_reference_population_slugs():
fields.append(variant.annotation['freqs'][ref_population_slug])
for field_key in ['polyphen', 'sift', 'muttaster', 'fathmm']:
fields.append(variant.annotation[field_key])
for indiv_id in indiv_id_list:
variant = individual_id_to_variant.get(indiv_id)
genotype = None
if variant is not None:
genotype = variant.get_genotype(indiv_id)
if genotype is None:
fields.extend(['.', '.', '.'])
else:
if genotype.num_alt == 0:
fields.append("%s/%s" % (variant.ref, variant.ref))
elif genotype.num_alt == 1:
fields.append("%s/%s" % (variant.ref, variant.alt))
elif genotype.num_alt == 2:
fields.append("%s/%s" % (variant.alt, variant.alt))
else:
fields.append("./.")
fields.append(str(genotype.gq) if genotype.gq is not None else '.')
fields.append(genotype.extras['dp'] if genotype.extras.get('dp') is not None else '.')
writer.writerow(fields)
return response
def combine_mendelian_families_spec(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
search_hash = request.GET.get('search_hash')
search_spec, genes = cache_utils.get_cached_results(project.project_id, search_hash)
search_spec_obj = MendelianVariantSearchSpec.fromJSON(search_spec)
if request.GET.get('return_type') != 'csv' or not request.GET.get('group_by_variants'):
if genes is None:
genes = api_utils.calculate_combine_mendelian_families(family_group, search_spec)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
if request.GET.get('return_type') != 'csv':
return JSONResponse({
'is_error': False,
'genes': genes,
'search_spec': search_spec,
})
else:
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="family_group_results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
writer.writerow(["gene", "# families", "family list", "chrom", "start", "end"])
for gene in genes:
family_id_list = [family_id for (project_id, family_id) in gene["family_id_list"]]
writer.writerow(map(str, [gene["gene_name"], len(family_id_list), " ".join(family_id_list), gene["chr"], gene["start"], gene["end"], ""]))
return response
else:
# download results grouped by variant
indiv_id_list = []
for family in family_group.get_families():
indiv_id_list.extend(family.indiv_ids_with_variant_data())
response = HttpResponse(content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename="results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
headers = ['genes','chr','pos','ref','alt','worst_annotation' ]
headers.extend(project.get_reference_population_slugs())
headers.extend([ 'polyphen','sift','muttaster','fathmm'])
for indiv_id in indiv_id_list:
headers.append(indiv_id)
headers.append(indiv_id+'_gq')
headers.append(indiv_id+'_dp')
writer.writerow(headers)
mall = get_mall(project.project_id)
variant_key_to_individual_id_to_variant = defaultdict(dict)
variant_key_to_variant = {}
for family in family_group.get_families():
for variant in get_variants_with_inheritance_mode(
mall,
family.xfamily(),
search_spec_obj.inheritance_mode,
search_spec_obj.variant_filter,
search_spec_obj.quality_filter,
):
if len(variant.coding_gene_ids) == 0:
continue
variant_key = (variant.xpos, variant.ref, variant.alt)
variant_key_to_variant[variant_key] = variant
for indiv_id in family.indiv_ids_with_variant_data():
variant_key_to_individual_id_to_variant[variant_key][indiv_id] = variant
for variant_key in sorted(variant_key_to_individual_id_to_variant.keys()):
variant = variant_key_to_variant[variant_key]
individual_id_to_variant = variant_key_to_individual_id_to_variant[variant_key]
genes = [mall.reference.get_gene_symbol(gene_id) for gene_id in variant.coding_gene_ids]
fields = []
fields.append(','.join(genes))
fields.extend([
variant.chr,
str(variant.pos),
variant.ref,
variant.alt,
variant.annotation.get('vep_group', '.'),
])
for ref_population_slug in project.get_reference_population_slugs():
fields.append(variant.annotation['freqs'][ref_population_slug])
for field_key in ['polyphen', 'sift', 'muttaster', 'fathmm']:
fields.append(variant.annotation[field_key])
for indiv_id in indiv_id_list:
variant = individual_id_to_variant.get(indiv_id)
genotype = None
if variant is not None:
genotype = variant.get_genotype(indiv_id)
if genotype is None:
fields.extend(['.', '.', '.'])
else:
if genotype.num_alt == 0:
fields.append("%s/%s" % (variant.ref, variant.ref))
elif genotype.num_alt == 1:
fields.append("%s/%s" % (variant.ref, variant.alt))
elif genotype.num_alt == 2:
fields.append("%s/%s" % (variant.alt, variant.alt))
else:
fields.append("./.")
fields.append(str(genotype.gq) if genotype.gq is not None else '.')
fields.append(genotype.extras['dp'] if genotype.extras.get('dp') is not None else '.')
writer.writerow(fields)
return response