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="xbrowse_results_{}.csv"'.format(search_hash)
writer = csv.writer(response)
indiv_ids = family.indiv_ids_with_variant_data()
headers = xbrowse_displays.get_variant_display_headers(indiv_ids)
writer.writerow(headers)
for variant in variants:
fields = xbrowse_displays.get_display_fields_for_variant(variant, get_reference(), indiv_ids)
writer.writerow(fields)
return response
def family_group_gene(request, project_id, family_group_slug, gene_id):
project = get_object_or_404(Project, project_id=project_id)
family_group = get_object_or_404(FamilyGroup,
project=project,
slug=family_group_slug)
if not project.can_view(request.user):
return HttpResponse('unauthorized')
gene_id = get_gene_id_from_str(gene_id, get_reference())
gene = get_reference().get_gene(gene_id)
varfilter = get_default_variant_filter(
'all_coding',
mall.get_annotator().reference_population_slugs)
variants_by_family = family_group_analysis.get_variants_in_gene(
family_group, gene_id, variant_filter=varfilter)
return render(
request, 'family_group/family_group_gene.html', {
'project': project,
'family_group': family_group,
'family_group_json': json.dumps(family_group.toJSON()),
'gene_json': json.dumps(gene),
'gene': gene,
'variants_by_family_json': json.dumps(variants_by_family),
})
def calculate_mendelian_variant_search(search_spec, xfamily):
variants = None
if search_spec.search_mode == 'standard_inheritance':
variants = list(get_variants_with_inheritance_mode(
get_mall(),
xfamily,
search_spec.inheritance_mode,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.genotype_quality_filter,
))
elif search_spec.search_mode == 'custom_inheritance':
variants = list(get_variants_family(
get_datastore(),
xfamily,
genotype_filter=search_spec.genotype_inheritance_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.genotype_quality_filter,
))
elif search_spec.search_mode == 'gene_burden':
gene_stream = get_genes_family(
get_datastore(),
get_reference(),
xfamily,
burden_filter=search_spec.gene_burden_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.genotype_quality_filter,
)
variants = list(stream_utils.gene_stream_to_variant_stream(gene_stream, get_reference()))
elif search_spec.search_mode == 'allele_count':
variants = list(get_variants_allele_count(
get_datastore(),
xfamily,
search_spec.allele_count_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.genotype_quality_filter,
))
elif search_spec.search_mode == 'all_variants':
variants = list(get_variants_family(
get_datastore(),
xfamily,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.genotype_quality_filter,
))
return variants
def handle(self, *args, **options):
if settings.CUSTOM_ANNOTATOR_SETTINGS is not None:
print("Load dbNSFP.. ")
# note that you could use mall.get_custom_annotator() here too
custom_annotator = CustomAnnotator(settings.CUSTOM_ANNOTATOR_SETTINGS)
custom_annotator.load()
get_reference().load()
mall.get_annotator().load()
def gene_info(request, gene_str):
real_gene_id = get_gene_id_from_str(gene_str, get_reference())
gene = get_reference().get_gene(real_gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(real_gene_id)
gene_json = json.dumps(gene)
return render(request, 'gene_info.html', {
'gene_json': gene_json,
'gene_symbol': gene['symbol'],
})
def family_coverage_gene(request, family, gene_id):
project_id = family.project.project_id
gene = get_reference().get_gene(gene_id)
gene_structure = get_reference().get_gene_structure(gene_id)
individuals = family.get_individuals()
indiv_ids = [i.indiv_id for i in individuals]
num_individuals = len(indiv_ids)
coding_regions = []
for c in get_coding_regions_from_gene_structure(gene_id, gene_structure):
coding_region = {}
coding_region['start'] = genomeloc.get_chr_pos(c.xstart)[1]
coding_region['stop'] = genomeloc.get_chr_pos(c.xstop)[1]
coding_region['gene_id'] = c.gene_id
coding_region['size'] = c.xstop - c.xstart + 1
coding_regions.append(coding_region)
coverages = {}
for individual in individuals:
coverages[
individual.indiv_id] = get_coverage_store().get_coverage_for_gene(
str(individual.pk), gene['gene_id'])
whole_gene = Counter({'callable': 0, 'low_coverage': 0, 'poor_mapping': 0})
for coverage_spec in coverages.values():
whole_gene['callable'] += coverage_spec['gene_totals']['callable']
whole_gene['low_coverage'] += coverage_spec['gene_totals'][
'low_coverage']
whole_gene['poor_mapping'] += coverage_spec['gene_totals'][
'poor_mapping']
gene_coding_size = 0
for c in coding_regions:
gene_coding_size += c['stop'] - c['start'] + 1
totalsize = gene_coding_size * num_individuals
whole_gene['ratio_callable'] = whole_gene['callable'] / float(totalsize)
whole_gene['ratio_low_coverage'] = whole_gene['low_coverage'] / float(
totalsize)
whole_gene['ratio_poor_mapping'] = whole_gene['poor_mapping'] / float(
totalsize)
whole_gene['gene_coding_size'] = gene_coding_size
return render(
request, 'coverage/family_coverage_gene.html', {
'project': family.project,
'family': family,
'gene': gene,
'coverages_json': json.dumps(coverages),
'whole_gene_json': json.dumps(whole_gene),
'coding_regions_json': json.dumps(coding_regions),
'indiv_ids_json': json.dumps(indiv_ids),
'individuals': individuals,
'whole_gene': whole_gene,
})
def gene_info(request, gene_str):
real_gene_id = get_gene_id_from_str(gene_str, get_reference())
gene = get_reference().get_gene(real_gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(
real_gene_id)
gene_json = json.dumps(gene)
return render(request, 'gene_info.html', {
'gene_json': gene_json,
'gene_symbol': gene['symbol'],
})
def calculate_mendelian_variant_search(search_spec, xfamily):
sys.stderr.write(" mendelian_variant_search for %s - search mode: %s %s\n" % (xfamily.project_id, search_spec.search_mode, search_spec.__dict__))
variants = None
if search_spec.search_mode == 'standard_inheritance':
variants = list(get_variants_with_inheritance_mode(
get_mall(xfamily.project_id),
xfamily,
search_spec.inheritance_mode,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.quality_filter,
))
elif search_spec.search_mode == 'custom_inheritance':
variants = list(get_variants_family(
get_datastore(xfamily.project_id),
xfamily,
genotype_filter=search_spec.genotype_inheritance_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.quality_filter,
))
elif search_spec.search_mode == 'gene_burden':
gene_stream = get_genes_family(
get_datastore(xfamily.project_id),
get_reference(),
xfamily,
burden_filter=search_spec.gene_burden_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.quality_filter,
)
variants = list(stream_utils.gene_stream_to_variant_stream(gene_stream, get_reference()))
elif search_spec.search_mode == 'allele_count':
variants = list(get_variants_allele_count(
get_datastore(xfamily.project_id),
xfamily,
search_spec.allele_count_filter,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.quality_filter,
))
elif search_spec.search_mode == 'all_variants':
variants = list(get_variants_family(
get_datastore(xfamily.project_id),
xfamily,
variant_filter=search_spec.variant_filter,
quality_filter=search_spec.quality_filter,
indivs_to_consider=xfamily.indiv_id_list(),
))
return variants
def handle(self, *args, **options):
if settings.CUSTOM_ANNOTATOR_SETTINGS is not None:
print("Load dbNSFP.. ")
# note that you could use mall.get_custom_annotator() here too
custom_annotator = CustomAnnotator(
settings.CUSTOM_ANNOTATOR_SETTINGS)
custom_annotator.load()
get_reference().load()
mall.get_annotator().load()
def gene_quicklook(request, project_id, gene_id):
"""
Summary of a gene in a project
"""
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
return HttpResponse("Unauthorized")
gene_id = get_gene_id_from_str(gene_id, get_reference())
gene = get_reference().get_gene(gene_id)
sys.stderr.write(project_id + " - staring gene search for: %s %s \n" %
(gene_id, gene))
variant_filter = get_default_variant_filter(
'all_coding',
mall.get_annotator().reference_population_slugs)
num_indivs = len(
[i for i in project.get_individuals() if i.has_variant_data()])
aac_threshold = (.2 * num_indivs) + 5
rare_variants = []
for variant in project_analysis.get_variants_in_gene(
project, gene_id, variant_filter=variant_filter):
aac = get_alt_allele_count(variant)
max_af = max(variant.annotation['freqs'].values())
if aac <= aac_threshold and max_af < .01:
rare_variants.append(variant)
add_extra_info_to_variants_project(get_reference(), project, rare_variants)
knockouts = []
knockout_ids, variation = get_knockouts_in_gene(project, gene_id)
for kid in knockout_ids:
variants = variation.get_relevant_variants_for_indiv_ids([kid])
add_extra_info_to_variants_project(get_reference(), project, variants)
knockouts.append({
'indiv_id': kid,
'variants': [v.toJSON() for v in variants],
})
sys.stderr.write("Retrieved %s variants \n" % len(rare_variants))
return render(
request, 'project/gene_quicklook.html', {
'gene':
gene,
'gene_json':
json.dumps(gene),
'project':
project,
'rare_variants_json':
json.dumps([v.toJSON() for v in rare_variants]),
'individuals_json':
json.dumps([i.get_json_obj() for i in project.get_individuals()]),
'knockouts_json':
json.dumps(knockouts),
})
def gene_info(request, gene_id):
gene = get_reference().get_gene(gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(gene_id)
ret = {
'gene': gene,
'is_error': False,
'found_gene': gene is not None,
}
return JSONResponse(ret)
def gene_info(request, gene_id):
gene = get_reference().get_gene(gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(gene_id)
ret = {
'gene': gene,
'is_error': False,
'found_gene': gene is not None,
}
return JSONResponse(ret)
def gene_info(request, gene_str):
real_gene_id = get_gene_id_from_str(gene_str, get_reference())
gene = get_reference().get_gene(real_gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(real_gene_id)
add_notes_to_genes([gene], request.user)
gene_json = json.dumps(gene)
return render(request, 'gene_info.html', {
'gene_json': gene_json,
'gene_symbol': gene['symbol'],
'new_page_url': '/gene_info/{}'.format(gene_str),
})
def family_coverage_gene(request, family, gene_id):
project_id = family.project.project_id
gene = get_reference().get_gene(gene_id)
gene_structure = get_reference().get_gene_structure(gene_id)
individuals = family.get_individuals()
indiv_ids = [i.indiv_id for i in individuals]
num_individuals = len(indiv_ids)
coding_regions = []
for c in get_coding_regions_from_gene_structure(gene_id, gene_structure):
coding_region = {}
coding_region['start'] = genomeloc.get_chr_pos(c.xstart)[1]
coding_region['stop'] = genomeloc.get_chr_pos(c.xstop)[1]
coding_region['gene_id'] = c.gene_id
coding_region['size'] = c.xstop-c.xstart+1
coding_regions.append(coding_region)
coverages = {}
for individual in individuals:
coverages[individual.indiv_id] = get_coverage_store().get_coverage_for_gene(
str(individual.pk),
gene['gene_id']
)
whole_gene = Counter({'callable': 0, 'low_coverage': 0, 'poor_mapping': 0})
for coverage_spec in coverages.values():
whole_gene['callable'] += coverage_spec['gene_totals']['callable']
whole_gene['low_coverage'] += coverage_spec['gene_totals']['low_coverage']
whole_gene['poor_mapping'] += coverage_spec['gene_totals']['poor_mapping']
gene_coding_size = 0
for c in coding_regions:
gene_coding_size += c['stop']-c['start']+1
totalsize = gene_coding_size*num_individuals
whole_gene['ratio_callable'] = whole_gene['callable'] / float(totalsize)
whole_gene['ratio_low_coverage'] = whole_gene['low_coverage'] / float(totalsize)
whole_gene['ratio_poor_mapping'] = whole_gene['poor_mapping'] / float(totalsize)
whole_gene['gene_coding_size'] = gene_coding_size
return render(request, 'coverage/family_coverage_gene.html', {
'project': family.project,
'family': family,
'gene': gene,
'coverages_json': json.dumps(coverages),
'whole_gene_json': json.dumps(whole_gene),
'coding_regions_json': json.dumps(coding_regions),
'indiv_ids_json': json.dumps(indiv_ids),
'individuals': individuals,
'whole_gene': whole_gene,
})
def gene_info(request, gene_str):
real_gene_id = get_gene_id_from_str(gene_str, get_reference())
gene = get_reference().get_gene(real_gene_id)
gene['expression'] = get_reference().get_tissue_expression_display_values(
real_gene_id)
add_notes_to_genes([gene], request.user)
gene_json = json.dumps(gene)
return render(
request, 'gene_info.html', {
'gene_json': gene_json,
'gene_symbol': gene['symbol'],
'new_page_url': '/gene_info/{}'.format(gene_str),
})
def cohort_gene_search_variants(request):
# TODO: this view not like the others - refactor to forms
error = None
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
if not project.can_view(request.user):
return PermissionDenied
form = api_forms.CohortGeneSearchVariantsForm(request.GET)
if form.is_valid():
gene_id = form.cleaned_data['gene_id']
inheritance_mode = form.cleaned_data['inheritance_mode']
variant_filter = form.cleaned_data['variant_filter']
quality_filter = form.cleaned_data['quality_filter']
else:
error = server_utils.form_error_string(form)
if not error:
indivs_with_inheritance, gene_variation = cohort_search.get_individuals_with_inheritance_in_gene(
get_datastore(project.project_id),
get_reference(),
cohort.xcohort(),
inheritance_mode,
gene_id,
variant_filter=variant_filter,
quality_filter=quality_filter
)
relevant_variants = gene_variation.get_relevant_variants_for_indiv_ids(cohort.indiv_id_list())
api_utils.add_extra_info_to_variants_family(get_reference(), cohort, relevant_variants)
ret = {
'is_error': False,
'variants': [v.toJSON() for v in relevant_variants],
'gene_info': get_reference().get_gene(gene_id),
}
return JSONResponse(ret)
else:
ret = {
'is_error': True,
'error': error
}
return JSONResponse(ret)
def control_comparison(control_cohort, gene_id, sample_hits, sample_size, inheritance_mode, variant_filter, quality_filter):
"""
Compare the results of num_hits, total against the reference population
Return dict of 'num_hits', 'fisher_2sided_palue',
"""
cohort = get_population_datastore().get_control_cohort(control_cohort)
indivs_with_inheritance, gene_variation = get_individuals_with_inheritance_in_gene(
get_population_datastore(),
get_reference(),
cohort,
inheritance_mode,
gene_id,
variant_filter=variant_filter,
quality_filter=quality_filter
)
control_hits = len(indivs_with_inheritance)
fisher_results = fisher.pvalue(
sample_hits,
sample_size,
control_hits,
get_population_datastore().get_control_cohort_size(settings.DEFAULT_CONTROL_COHORT)
)
return {
'control_hits': control_hits,
'fisher_2sided_pvalue': fisher_results.two_tail,
}
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 causal_variants(request, project_id):
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
raise PermissionDenied
variants = get_causal_variants_for_project(project)
add_extra_info_to_variants_project(get_reference(),
project,
variants,
add_family_tags=True,
add_populations=True)
return render(
request, 'project/causal_variants.html', {
'project':
project,
'variants_json':
json.dumps([v.toJSON() for v in variants]),
'families_json':
json.dumps({
family.family_id: family.get_json_obj()
for family in project.get_families()
})
})
def cohort_gene_search(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
sys.stderr.write("cohort_gene_search %s %s: starting ... \n" % (project.project_id, cohort.cohort_id))
form = api_forms.CohortGeneSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.cohort_id = cohort.cohort_id
sys.stderr.write("cohort_gene_search %s %s: search spec: %s \n" % (project.project_id, cohort.cohort_id, str(search_spec.toJSON())))
genes = api_utils.calculate_cohort_gene_search(cohort, search_spec)
sys.stderr.write("cohort_gene_search %s %s: get %s genes \n" % (project.project_id, cohort.cohort_id, len(genes)))
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), genes)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
sys.stderr.write("cohort_gene_search %s %s: done adding extra info \n" % (project.project_id, cohort.cohort_id))
return JSONResponse({
'is_error': False,
'genes': genes,
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def saved_variants(request, project_id):
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
return HttpResponse('unauthorized')
variants = get_all_saved_variants_for_project(project)
if 'family' in request.GET:
requested_family_id = request.GET.get('family')
variants = filter(lambda v: v.extras['family_id'] == requested_family_id, variants)
variants = sorted(variants, key=lambda v: (v.extras['family_id'], v.xpos))
grouped_variants = itertools.groupby(variants, key=lambda v: v.extras['family_id'])
for family_id, family_variants in grouped_variants:
family = Family.objects.get(project=project, family_id=family_id)
family_variants = list(family_variants)
add_extra_info_to_variants_family(get_reference(), family, family_variants)
return render(request, 'project/saved_variants.html', {
'project': project,
'tag': None,
'variants_json': json.dumps([v.toJSON() for v in variants]),
'families_json': json.dumps({family.family_id: family.get_json_obj() for family in project.get_families()})
})
def family_coverage_gene_list(request, family, gene_list):
"""
Table of summary coverages for each gene in the gene list
"""
sample_id_list = [str(individual.pk) for individual in family.get_individuals()]
cache_key = ('family_coverage_gene_list', family.project.project_id, family.family_id, gene_list.slug)
cached_results = server_utils.get_cached_results(cache_key)
if cached_results:
gene_coverages = cached_results
else:
gene_coverages = []
for gene_id in gene_list.gene_id_list():
d = {
'gene_id': gene_id,
'gene_name': get_reference().get_gene_symbol(gene_id),
'totals': get_coverage_store().get_coverage_totals_for_gene(gene_id, sample_id_list)
}
d['coding_size'] = sum(d['totals'].values())
try:
d['percent'] = float(d['totals']['callable'])*100 / d['coding_size']
except ZeroDivisionError:
d['percent'] = 0
gene_coverages.append(d)
server_utils.save_results_cache(cache_key, gene_coverages)
return render(request, 'coverage/family_coverage_gene_list.html', {
'project': family.project,
'family': family,
'gene_list': gene_list,
'gene_coverages': gene_coverages,
})
def breakpoint_search(request, project_id, family_id):
log.info("Showing main breakpoint search page")
project = get_object_or_404(Project, project_id=project_id)
family = get_object_or_404(Family, project=project, family_id=family_id)
if not project.can_view(request.user):
raise PermissionDenied
if not family.has_data('breakpoints'):
return render(
request, 'analysis_unavailable.html',
{'reason': 'This family does not have any breakpoint data.'})
gene_lists = project.get_gene_list_map()
gene_list_json = dict([(get_reference().get_gene_symbol(g),
[gl.name for gl in gll])
for g, gll in gene_lists.iteritems()])
bam_file_paths = dict(
(ind.indiv_id,
os.path.join(settings.READ_VIZ_BAM_PATH, ind.bam_file_path))
for ind in family.get_individuals())
return render(
request, 'breakpoint_search.html', {
'project': project,
'family': family,
'gene_lists_json': json.dumps(gene_list_json),
'bam_files_json': json.dumps(bam_file_paths),
})
def fetch_project_individuals_data(project_id):
"""
Notes:
1. ONLY project-authorized user has access to this individual
"""
project = get_object_or_404(Project, project_id=project_id)
variants = get_causal_variants_for_project(project)
variants = sorted(variants, key=lambda v: (v.extras['family_id'], v.xpos))
grouped_variants = itertools.groupby(variants, key=lambda v: v.extras['family_id'])
for family_id, family_variants in grouped_variants:
family = Family.objects.get(project=project, family_id=family_id)
family_variants = list(family_variants)
add_extra_info_to_variants_family(get_reference(), family, family_variants)
family_data = [v.toJSON() for v in variants]
variant_data = {family.family_id: family.get_json_obj() for family in project.get_families()}
phenotype_entry_counts = gather_phenotype_data_for_project(project_id, variant_data)
status_description_map = {}
for abbrev, details in ANALYSIS_STATUS_CHOICES:
status_description_map[abbrev] = details[0]
families_json = json_displays.family_list(project.get_families())
family_statuses = {}
for f in families_json:
family_statuses[f['family_id']] = status_description_map[f['analysis_status']['status']]
return family_data, variant_data, phenotype_entry_counts, family_statuses
def deprecated_retrieve_saved_variants_json(project, variant_tuples,
create_if_missing):
project_id = project.deprecated_project_id
xbrowse_project = BaseProject.objects.get(project_id=project_id)
user = User.objects.filter(is_staff=True).first(
) # HGMD annotations are only returned for staff users
variants = get_variants_from_variant_tuples(xbrowse_project,
variant_tuples,
user=user)
if not create_if_missing:
variants = [
var for var in variants if not var.get_extra('created_variant')
]
add_extra_info_to_variants_project(get_reference(),
xbrowse_project,
variants,
add_populations=True)
family_guids_by_id = {
f.family_id: f.guid
for f in Family.objects.filter(project=project)
}
individual_guids_by_id = {
i.individual_id: i.guid
for i in Individual.objects.filter(family__project=project)
}
return [
_variant_details(variant.toJSON(), family_guids_by_id,
individual_guids_by_id) for variant in variants
]
def diagnostic_search(request):
project, family = utils.get_project_and_family_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
form = api_forms.DiagnosticSearchForm(family, request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = family.family_id
gene_list = form.cleaned_data['gene_list']
diagnostic_info_list = []
for gene_id in gene_list.gene_id_list():
diagnostic_info = get_gene_diangostic_info(family, gene_id, search_spec.variant_filter)
add_extra_info_to_variants_family(get_reference(), family, diagnostic_info._variants)
diagnostic_info_list.append(diagnostic_info)
return JSONResponse({
'is_error': False,
'gene_diagnostic_info_list': [d.toJSON() for d in diagnostic_info_list],
'gene_list_info': gene_list.toJSON(details=True),
'data_summary': family.get_data_summary(),
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def _set_saved_variant_json(new_variant_tag_or_note, source_variant_tag_or_note, new_family):
if new_family is None:
return
project_id = new_family.project.deprecated_project_id
project = Project.objects.get(project_id=project_id)
try:
variant_info = get_datastore(project).get_single_variant(
project_id,
new_family.family_id,
source_variant_tag_or_note.xpos,
source_variant_tag_or_note.ref,
source_variant_tag_or_note.alt)
except Exception as e:
logger.error("Unable to retrieve variant annotations for %s %s: %s" % (
new_family, source_variant_tag_or_note, e))
return
if variant_info:
add_extra_info_to_variants_project(get_reference(), project, [variant_info], add_family_tags=True,
add_populations=True)
variant_json = variant_info.toJSON()
new_variant_tag_or_note.saved_variant_json = json.dumps(variant_json)
new_variant_tag_or_note.save()
def combine_mendelian_families_variants(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
form = api_forms.CombineMendelianFamiliesVariantsForm(request.GET)
if form.is_valid():
variants_grouped = get_variants_by_family_for_gene(
get_mall(project.project_id),
[f.xfamily() for f in form.cleaned_data['families']],
form.cleaned_data['inheritance_mode'],
form.cleaned_data['gene_id'],
variant_filter=form.cleaned_data['variant_filter'],
quality_filter=form.cleaned_data['quality_filter']
)
variants_by_family = []
for family in form.cleaned_data['families']:
variants = variants_grouped[(family.project.project_id, family.family_id)]
add_extra_info_to_variants_family(get_reference(), family, variants)
variants_by_family.append({
'project_id': family.project.project_id,
'family_id': family.family_id,
'family_name': str(family),
'variants': [v.toJSON() for v in variants],
})
return JSONResponse({
'is_error': False,
'variants_by_family': variants_by_family,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def combine_mendelian_families(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
if not project.can_view(request.user):
return PermissionDenied
form = api_forms.CombineMendelianFamiliesForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_group_id = family_group.slug
genes = api_utils.calculate_combine_mendelian_families(family_group, search_spec)
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), genes)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
return JSONResponse({
'is_error': False,
'genes': genes,
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def cohort_variant_search(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
if not project.can_view(request.user):
return HttpResponse('unauthorized')
form = api_forms.CohortVariantSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = cohort.cohort_id
variants = api_utils.calculate_mendelian_variant_search(search_spec, cohort.xfamily())
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), [v.toJSON() 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_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def family_variant_view(request, project_id, family_id):
project = get_object_or_404(Project, project_id=project_id)
family = get_object_or_404(Family, project=project, family_id=family_id)
if not project.can_view(request.user):
raise PermissionDenied
try:
xpos = int(request.GET.get('xpos'))
ref = request.GET.get('ref')
alt = request.GET.get('alt')
except:
return HttpResponse('Invalid View')
variant = get_datastore(project).get_single_variant(
project_id, family_id, xpos, ref, alt)
add_extra_info_to_variants_project(get_reference(),
project, [variant],
add_family_tags=True,
add_populations=True)
return render(
request, 'family/family_variant_view.html', {
'project': project,
'family': family,
'variant_json': json.dumps(variant.toJSON()),
})
def cohort_variant_search(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
if not project.can_view(request.user):
return PermissionDenied
form = api_forms.CohortVariantSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = cohort.cohort_id
sys.stderr.write("cohort_variant_search - starting: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
variants = api_utils.calculate_mendelian_variant_search(search_spec, cohort.xfamily())
list_of_variants = [v.toJSON() for v in variants]
sys.stderr.write("cohort_variant_search - done calculate_mendelian_variant_search: %s %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id, len(list_of_variants)))
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), list_of_variants)
sys.stderr.write("cohort_variant_search - done save_results_for_spec: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
api_utils.add_extra_info_to_variants_cohort(get_reference(), cohort, variants)
sys.stderr.write("cohort_variant_search - done add_extra_info_to_variants_cohort: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def variants_with_tag(request, project_id, tag):
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
return HttpResponse('unauthorized')
project_tag = get_object_or_404(ProjectTag, project=project, tag=tag)
variants = get_variants_by_tag(project, tag)
variants = sorted(variants, key=lambda v: v.extras['family_id'])
grouped_variants = itertools.groupby(variants,
key=lambda v: v.extras['family_id'])
for family_id, family_variants in grouped_variants:
family = Family.objects.get(project=project, family_id=family_id)
family_variants = list(family_variants)
add_extra_info_to_variants_family(get_reference(), family,
family_variants)
return render(
request, 'project/saved_variants.html', {
'project':
project,
'variants_json':
json.dumps([v.toJSON() for v in variants]),
'families_json':
json.dumps({
family.family_id: family.get_json_obj()
for family in project.get_families()
})
})
def combine_mendelian_families_variants(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
form = api_forms.CombineMendelianFamiliesVariantsForm(request.GET)
if form.is_valid():
variants_grouped = get_variants_by_family_for_gene(
get_mall(project.project_id),
[f.xfamily() for f in form.cleaned_data['families']],
form.cleaned_data['inheritance_mode'],
form.cleaned_data['gene_id'],
variant_filter=form.cleaned_data['variant_filter'],
quality_filter=form.cleaned_data['quality_filter']
)
variants_by_family = []
for family in form.cleaned_data['families']:
variants = variants_grouped[(family.project.project_id, family.family_id)]
add_extra_info_to_variants_family(get_reference(), family, variants)
variants_by_family.append({
'project_id': family.project.project_id,
'family_id': family.family_id,
'family_name': str(family),
'variants': [v.toJSON() for v in variants],
})
return JSONResponse({
'is_error': False,
'variants_by_family': variants_by_family,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def get_variants_in_gene(family_group,
gene_id,
variant_filter=None,
quality_filter=None):
"""
"""
variants_by_family = []
for family in family_group.get_families():
variant_list = list(
get_mall(
family.project.project_id).variant_store.get_variants_in_gene(
family.project.project_id,
family.family_id,
gene_id,
variant_filter=variant_filter))
variant_list = search_utils.filter_gene_variants_by_variant_filter(
variant_list, gene_id, variant_filter)
add_extra_info_to_variants_family(get_reference(), family,
variant_list)
variants_by_family.append({
'variants': [v.toJSON() for v in variant_list],
'family_id':
family.family_id,
'project_id':
family.project.project_id,
'family_name':
str(family),
})
return variants_by_family
def family_coverage_gene_list(request, family, gene_list):
"""
Table of summary coverages for each gene in the gene list
"""
sample_id_list = [str(individual.pk) for individual in family.get_individuals()]
cache_key = ('family_coverage_gene_list', family.project.project_id, family.family_id, gene_list.slug)
cached_results = server_utils.get_cached_results(cache_key)
if cached_results:
gene_coverages = cached_results
else:
gene_coverages = []
for gene_id in gene_list.gene_id_list():
d = {
'gene_id': gene_id,
'gene_name': get_reference().get_gene_symbol(gene_id),
'totals': get_coverage_store().get_coverage_totals_for_gene(gene_id, sample_id_list)
}
d['coding_size'] = sum(d['totals'].values())
try:
d['percent'] = float(d['totals']['callable'])*100 / d['coding_size']
except ZeroDivisionError:
d['percent'] = 0
gene_coverages.append(d)
server_utils.save_results_cache(cache_key, gene_coverages)
return render(request, 'coverage/family_coverage_gene_list.html', {
'project': family.project,
'family': family,
'gene_list': gene_list,
'gene_coverages': gene_coverages,
})
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):
return HttpResponse('unauthorized')
search_hash = request.GET.get('search_hash')
search_spec, genes = cache_utils.get_cached_results(project.project_id, search_hash)
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
def cohort_variant_search(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
if not project.can_view(request.user):
return PermissionDenied
form = api_forms.CohortVariantSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = cohort.cohort_id
sys.stderr.write("cohort_variant_search - starting: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
variants = api_utils.calculate_mendelian_variant_search(search_spec, cohort.xfamily())
list_of_variants = [v.toJSON() for v in variants]
sys.stderr.write("cohort_variant_search - done calculate_mendelian_variant_search: %s %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id, len(list_of_variants)))
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), list_of_variants)
sys.stderr.write("cohort_variant_search - done save_results_for_spec: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
api_utils.add_extra_info_to_variants_cohort(get_reference(), cohort, variants)
sys.stderr.write("cohort_variant_search - done add_extra_info_to_variants_cohort: %s %s\n" % (json.dumps(search_spec.toJSON()), cohort.xfamily().family_id))
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def mendelian_variant_search(request):
# TODO: how about we move project getter into the form, and just test for authX here?
# esp because error should be described in json, not just 404
project, family = get_project_and_family_for_user(request.user, request.GET)
form = api_forms.MendelianVariantSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = family.family_id
variants = api_utils.calculate_mendelian_variant_search(search_spec, family.xfamily())
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), [v.toJSON() for v in variants])
add_extra_info_to_variants_family(get_reference(), family, variants)
return_type = request.GET.get('return_type', 'json')
if return_type == 'json':
return JSONResponse({
'is_error': False,
'variants': [v.toJSON() for v in variants],
'search_hash': search_hash,
})
elif return_type == 'csv':
return ''
else:
return HttpResponse("Return type not implemented")
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def saved_variants(request, project_id, family_id):
project = get_object_or_404(Project, project_id=project_id)
family = get_object_or_404(Family, project=project, family_id=family_id)
if not project.can_view(request.user):
raise PermissionDenied
variants, couldntfind = get_saved_variants_for_family(family)
# TODO: first this shouldnt be in API - base should never depend on api
# TODO: also this should have better naming
add_extra_info_to_variants_project(get_reference(),
project,
variants,
add_family_tags=True,
add_populations=True)
return render(
request, 'family/saved_family_variants.html', {
'project':
project,
'family':
family,
'variants_json':
json.dumps([v.toJSON() for v in variants]),
'new_page_url':
'/project/{0}/family_page/{1}'.format(
family.seqr_family.project.guid, family.seqr_family.guid)
if family.seqr_family else None,
})
def combine_mendelian_families(request):
project, family_group = utils.get_project_and_family_group_for_user(request.user, request.GET)
if not project.can_view(request.user):
return PermissionDenied
form = api_forms.CombineMendelianFamiliesForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_group_id = family_group.slug
genes = api_utils.calculate_combine_mendelian_families(family_group, search_spec)
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), genes)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
return JSONResponse({
'is_error': False,
'genes': genes,
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def diagnostic_search(request):
project, family = utils.get_project_and_family_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
form = api_forms.DiagnosticSearchForm(family, request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.family_id = family.family_id
gene_list = form.cleaned_data['gene_list']
diagnostic_info_list = []
for gene_id in gene_list.gene_id_list():
diagnostic_info = get_gene_diangostic_info(family, gene_id, search_spec.variant_filter)
add_extra_info_to_variants_family(get_reference(), family, diagnostic_info._variants)
diagnostic_info_list.append(diagnostic_info)
return JSONResponse({
'is_error': False,
'gene_diagnostic_info_list': [d.toJSON() for d in diagnostic_info_list],
'gene_list_info': gene_list.toJSON(details=True),
'data_summary': family.get_data_summary(),
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def get_knockouts_in_gene(project, gene_id, quality_filter=None):
"""
Get all the variants in a gene, but filter out quality_filter genotypes
"""
indiv_id_list = [i.indiv_id for i in project.get_individuals()]
# filter out variants > 0.01 AF in any of the reference populations
reference_populations = mall.get_annotator().reference_population_slugs
variant_filter = get_default_variant_filter('moderate_impact',
reference_populations)
variant_list = get_project_datastore(
project.project_id).get_project_variants_in_gene(
project.project_id,
gene_id,
variant_filter=variant_filter,
)
variant_list = search_utils.filter_gene_variants_by_variant_filter(
variant_list, gene_id, variant_filter)
variation = CohortGeneVariation(
get_reference(),
gene_id,
variant_list,
indiv_id_list,
quality_filter={},
)
knockouts = get_individuals_with_inheritance('recessive', variation,
indiv_id_list)
return knockouts, variation
def causal_variants(request, project_id):
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
raise PermissionDenied
variants = get_causal_variants_for_project(project)
variants = sorted(variants, key=lambda v: (v.extras['family_id'], v.xpos))
grouped_variants = itertools.groupby(variants,
key=lambda v: v.extras['family_id'])
for family_id, family_variants in grouped_variants:
family = Family.objects.get(project=project, family_id=family_id)
family_variants = list(family_variants)
add_extra_info_to_variants_family(get_reference(), family,
family_variants)
return render(
request, 'project/causal_variants.html', {
'project':
project,
'variants_json':
json.dumps([v.toJSON() for v in variants]),
'families_json':
json.dumps({
family.family_id: family.get_json_obj()
for family in project.get_families()
})
})
def cohort_gene_search(request):
project, cohort = get_project_and_cohort_for_user(request.user, request.GET)
sys.stderr.write("cohort_gene_search %s %s: starting ... \n" % (project.project_id, cohort.cohort_id))
form = api_forms.CohortGeneSearchForm(request.GET)
if form.is_valid():
search_spec = form.cleaned_data['search_spec']
search_spec.cohort_id = cohort.cohort_id
sys.stderr.write("cohort_gene_search %s %s: search spec: %s \n" % (project.project_id, cohort.cohort_id, str(search_spec.toJSON())))
genes = api_utils.calculate_cohort_gene_search(cohort, search_spec)
sys.stderr.write("cohort_gene_search %s %s: get %s genes \n" % (project.project_id, cohort.cohort_id, len(genes)))
search_hash = cache_utils.save_results_for_spec(project.project_id, search_spec.toJSON(), genes)
api_utils.add_extra_info_to_genes(project, get_reference(), genes)
sys.stderr.write("cohort_gene_search %s %s: done adding extra info \n" % (project.project_id, cohort.cohort_id))
return JSONResponse({
'is_error': False,
'genes': genes,
'search_hash': search_hash,
})
else:
return JSONResponse({
'is_error': True,
'error': server_utils.form_error_string(form)
})
def breakpoint_search(request, project_id, family_id):
log.info("Showing main breakpoint search page")
project = get_object_or_404(Project, project_id=project_id)
family = get_object_or_404(Family, project=project, family_id=family_id)
if not project.can_view(request.user):
raise PermissionDenied
if not family.has_data('breakpoints'):
return render(request, 'analysis_unavailable.html', {
'reason': 'This family does not have any breakpoint data.'
})
gene_lists = project.get_gene_list_map()
gene_list_json = dict([ (get_reference().get_gene_symbol(g),[ gl.name for gl in gll ]) for g,gll in gene_lists.iteritems() ])
bam_file_paths = dict((ind.indiv_id, os.path.join(settings.READ_VIZ_BAM_PATH,ind.bam_file_path)) for ind in family.get_individuals())
return render(request, 'breakpoint_search.html', {
'project': project,
'family': family,
'gene_lists_json': json.dumps(gene_list_json),
'bam_files_json': json.dumps(bam_file_paths),
})
def cohort_gene_search_variants(request):
# TODO: this view not like the others - refactor to forms
error = None
project, cohort = get_project_and_cohort_for_user(request.user,
request.GET)
if not project.can_view(request.user):
return HttpResponse('unauthorized')
form = api_forms.CohortGeneSearchVariantsForm(request.GET)
if form.is_valid():
gene_id = form.cleaned_data['gene_id']
inheritance_mode = form.cleaned_data['inheritance_mode']
variant_filter = form.cleaned_data['variant_filter']
quality_filter = form.cleaned_data['quality_filter']
else:
error = server_utils.form_error_string(form)
if not error:
indivs_with_inheritance, gene_variation = cohort_search.get_individuals_with_inheritance_in_gene(
get_datastore(project.project_id),
get_reference(),
cohort.xcohort(),
inheritance_mode,
gene_id,
variant_filter=variant_filter,
quality_filter=quality_filter)
relevant_variants = gene_variation.get_relevant_variants_for_indiv_ids(
cohort.indiv_id_list())
api_utils.add_extra_info_to_variants_family(get_reference(), cohort,
relevant_variants)
ret = {
'is_error': False,
'variants': [v.toJSON() for v in relevant_variants],
'gene_info': get_reference().get_gene(gene_id),
}
return JSONResponse(ret)
else:
ret = {'is_error': True, 'error': error}
return JSONResponse(ret)
def add_family_search_flag(request):
error = None
for key in [
'project_id', 'family_id', 'xpos', 'ref', 'alt', 'note',
'flag_type', 'flag_inheritance_mode'
]:
if request.GET.get(key, None) == None:
error = "%s is requred" % key
if not error:
project = get_object_or_404(Project,
project_id=request.GET.get('project_id'))
family = get_object_or_404(Family,
project=project,
family_id=request.GET.get('family_id'))
if not project.can_edit(request.user):
raise PermissionDenied
if not error:
xpos = int(request.GET['xpos'])
ref = request.GET.get('ref')
alt = request.GET['alt']
note = request.GET.get('note')
flag_type = request.GET.get('flag_type')
flag_inheritance_mode = request.GET.get('flag_inheritance_mode')
# todo: more validation - is variant valid?
flag = FamilySearchFlag(
user=request.user,
family=family,
xpos=int(request.GET['xpos']),
ref=ref,
alt=alt,
note=note,
flag_type=flag_type,
suggested_inheritance=flag_inheritance_mode,
date_saved=timezone.now(),
)
if not error:
flag.save()
variant = get_datastore(project.project_id).get_single_variant(
family.project.project_id, family.family_id, xpos, ref, alt)
api_utils.add_extra_info_to_variant(get_reference(), family, variant)
ret = {
'is_error': False,
'variant': variant.toJSON(),
}
else:
ret = {
'is_error': True,
'error': error,
}
return JSONResponse(ret)
def handle(self, *args, **options):
#genomicFeatures section
self.all_gene_lists = defaultdict(set)
self.gene_to_gene_lists = defaultdict(set)
for gene_list in GeneList.objects.all():
print('gene list: [%s]' % gene_list.name)
self.all_gene_lists[gene_list.name] = set(
g.gene_id for g in gene_list.genelistitem_set.all())
for g in gene_list.genelistitem_set.all():
self.gene_to_gene_lists[g.gene_id].add(gene_list.name)
print("starting... ")
gene_to_projects = defaultdict(set)
gene_to_variants = defaultdict(set)
gene_to_families = defaultdict(set)
gene_to_variant_tags = defaultdict(set)
Key = namedtuple('Key', 'gene_id, gene_name')
project_ids = defaultdict(int)
for variant_tag in tqdm(VariantTag.objects.filter(), unit=' variants'):
project_tag = variant_tag.project_tag
project_id = project_tag.project.project_id
project_ids[project_id] += 1
tag_name = project_tag.tag.lower()
variant = get_datastore(project_id).get_single_variant(
project_id,
variant_tag.family.family_id,
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
# print(gene_to_projects)
if variant is None:
#print("Variant %s no longer called in this family (did the callset version change?)" % (variant_tag.toJSON()))
continue
#print(project_id,variant.toJSON()['gene_ids'])
if variant.gene_ids is not None:
for gene_id in variant.gene_ids:
gene_name = get_reference().get_gene_symbol(gene_id)
key = Key._make([gene_id, gene_name])
gene_to_variants[key].add(
"%s-%s-%s-%s" %
(variant.chr, variant.pos, variant.ref, variant.alt))
if variant_tag.family:
gene_to_families[key].add(variant_tag.family.family_id)
gene_to_variant_tags[key].add(tag_name)
gene_to_projects[key].add(project_id.lower())
if len(gene_to_projects) % 50 == 0:
self.print_out(gene_to_projects, gene_to_families,
gene_to_variants, gene_to_variant_tags)
self.print_out(gene_to_projects, gene_to_families, gene_to_variants,
gene_to_variant_tags)
def family_gene_lookup(request):
project, family = utils.get_project_and_family_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
gene_id = request.GET.get('gene_id')
if not get_reference().is_valid_gene_id(gene_id):
return JSONResponse({
'is_error': True,
'error': 'Invalid gene',
})
family_gene_data = get_gene_diangostic_info(family, gene_id)
add_extra_info_to_variants_family(get_reference(), family, family_gene_data._variants)
return JSONResponse({
'is_error': False,
'family_gene_data': family_gene_data.toJSON(),
'data_summary': family.get_data_summary(),
'gene': get_reference().get_gene(gene_id),
})
def family_gene_lookup(request):
project, family = utils.get_project_and_family_for_user(request.user, request.GET)
if not project.can_view(request.user):
raise PermissionDenied
gene_id = request.GET.get('gene_id')
if not get_reference().is_valid_gene_id(gene_id):
return JSONResponse({
'is_error': True,
'error': 'Invalid gene',
})
family_gene_data = get_gene_diangostic_info(family, gene_id)
add_extra_info_to_variants_family(get_reference(), family, family_gene_data._variants)
return JSONResponse({
'is_error': False,
'family_gene_data': family_gene_data.toJSON(),
'data_summary': family.get_data_summary(),
'gene': get_reference().get_gene(gene_id),
})
def _deprecated_retrieve_saved_variants_json(project, variant_tuples, create_if_missing):
project_id = project.deprecated_project_id
xbrowse_project = BaseProject.objects.get(project_id=project_id)
user = User.objects.filter(is_staff=True).first() # HGMD annotations are only returned for staff users
variants = get_variants_from_variant_tuples(xbrowse_project, variant_tuples, user=user)
if not create_if_missing:
variants = [var for var in variants if not var.get_extra('created_variant')]
add_extra_info_to_variants_project(get_reference(), xbrowse_project, variants, add_populations=True)
return [variant.toJSON() for variant in variants]
def _deprecated_retrieve_saved_variants_json(project, variant_tuples, create_if_missing):
project_id = project.deprecated_project_id
xbrowse_project = BaseProject.objects.get(project_id=project_id)
user = User.objects.filter(is_staff=True).first() # HGMD annotations are only returned for staff users
variants = get_variants_from_variant_tuples(xbrowse_project, variant_tuples, user=user)
if not create_if_missing:
variants = [var for var in variants if not var.get_extra('created_variant')]
add_extra_info_to_variants_project(get_reference(), xbrowse_project, variants, add_populations=True)
return [variant.toJSON() for variant in variants]
def clean(self):
cleaned_data = super(CombineMendelianFamiliesVariantsForm, self).clean()
parse_variant_filter(cleaned_data)
parse_quality_filter(cleaned_data)
parse_family_tuple_list(cleaned_data)
if not get_reference().is_valid_gene_id(cleaned_data['gene_id']):
raise forms.ValidationError("{} is not a valid gene ID".format(cleaned_data['gene_id']))
return cleaned_data
def handle(self, *args, **options):
#genomicFeatures section
self.all_gene_lists = defaultdict(set)
self.gene_to_gene_lists = defaultdict(set)
for gene_list in GeneList.objects.all():
print('gene list: [%s]' % gene_list.name)
self.all_gene_lists[gene_list.name] = set(g.gene_id for g in gene_list.genelistitem_set.all())
for g in gene_list.genelistitem_set.all():
self.gene_to_gene_lists[g.gene_id].add(gene_list.name)
print("starting... ")
gene_to_projects = defaultdict(set)
gene_to_variants = defaultdict(set)
gene_to_families = defaultdict(set)
gene_to_variant_tags = defaultdict(set)
gene_to_variant_and_families = defaultdict(lambda: defaultdict(set))
Key = namedtuple('Key', 'gene_id, gene_name')
project_ids = defaultdict(int)
for variant_tag in tqdm(VariantTag.objects.filter(), unit=' variants'):
project_tag = variant_tag.project_tag
project_id = project_tag.project.project_id
project_ids[project_id] += 1
tag_name = project_tag.tag.lower()
variant = get_datastore(project_tag.project).get_single_variant(
project_id,
variant_tag.family.family_id,
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
# print(gene_to_projects)
if variant is None:
#print("Variant %s no longer called in this family (did the callset version change?)" % (variant_tag.toJSON()))
continue
#print(project_id,variant.toJSON()['gene_ids'])
if variant.gene_ids is not None:
for gene_id in variant.gene_ids:
gene_name = get_reference().get_gene_symbol(gene_id)
key = Key._make([gene_id, gene_name])
variant_id = "%s-%s-%s-%s" % (variant.chr, variant.pos, variant.ref, variant.alt)
gene_to_variants[key].add(variant_id)
if variant_tag.family:
gene_to_families[key].add(variant_tag.family)
gene_to_variant_tags[key].add(tag_name)
gene_to_projects[key].add(project_id.lower())
gene_to_variant_and_families[key][variant_id].add(variant_tag.family.family_id)
if len(gene_to_projects) % 50 == 0:
self.print_out(gene_to_projects, gene_to_families, gene_to_variants, gene_to_variant_tags, gene_to_variant_and_families)
self.print_out(gene_to_projects, gene_to_families, gene_to_variants, gene_to_variant_tags, gene_to_variant_and_families)
def calculate_combine_mendelian_families(family_group, search_spec, user=None):
"""
Calculate search results from the params in search_spec
Should be called after cache is checked - this does all the computation
Returns (is_error, genes) tuple
"""
xfamilygroup = family_group.xfamilygroup()
genes = []
for gene_id, family_id_list in get_families_by_gene(
get_mall(family_group.project),
xfamilygroup,
search_spec.inheritance_mode,
search_spec.variant_filter,
search_spec.quality_filter,
user=user,
):
xgene = get_reference().get_gene(gene_id)
if xgene is None:
continue
try:
start_pos, end_pos = get_reference().get_gene_bounds(gene_id)
chr, start = genomeloc.get_chr_pos(start_pos)
end = genomeloc.get_chr_pos(end_pos)[1]
except KeyError:
chr, start, end = None, None, None
gene = {
'gene_info': xgene,
'gene_id': gene_id,
'gene_name': xgene['symbol'],
'chr': chr,
'start': start,
'end': end,
'family_id_list': family_id_list,
}
genes.append(gene)
return genes
def add_gene_databases_to_variants(variants):
"""
Adds variant['gene_databases'] - a list of *gene* databases that this variant's gene(s) are seen in
"""
for variant in variants:
variant.set_extra('in_disease_gene_db', False)
for gene_id in variant.coding_gene_ids:
gene = get_reference().get_gene(gene_id)
# TODO: should be part of reference cache
if gene and (len(gene['phenotype_info']['orphanet_phenotypes']) or len(gene['phenotype_info']['mim_phenotypes'])):
variant.set_extra('in_disease_gene_db', True)
def add_family_search_flag(request):
# TODO: this view not like the others - refactor to forms
error = None
for key in ['project_id', 'family_id', 'xpos', 'ref', 'alt', 'note', 'flag_type', 'flag_inheritance_mode']:
if request.GET.get(key, None) == None:
error = "%s is requred" % key
if not error:
project = get_object_or_404(Project, project_id=request.GET.get('project_id'))
family = get_object_or_404(Family, project=project, family_id=request.GET.get('family_id'))
if not project.can_edit(request.user):
return PermissionDenied
if not error:
xpos = int(request.GET['xpos'])
ref=request.GET.get('ref')
alt=request.GET['alt']
note=request.GET.get('note')
flag_type=request.GET.get('flag_type')
flag_inheritance_mode=request.GET.get('flag_inheritance_mode')
# todo: more validation - is variant valid?
flag = FamilySearchFlag(user=request.user,
family=family,
xpos=int(request.GET['xpos']),
ref=ref,
alt=alt,
note=note,
flag_type=flag_type,
suggested_inheritance=flag_inheritance_mode,
date_saved = datetime.datetime.now(),
)
if not error:
flag.save()
variant = get_datastore(project.project_id).get_single_variant(family.project.project_id, family.family_id,
xpos, ref, alt )
api_utils.add_extra_info_to_variant(get_reference(), family, variant)
ret = {
'is_error': False,
'variant': variant.toJSON(),
}
else:
ret = {
'is_error': True,
'error': error,
}
return JSONResponse(ret)
def gene_quicklook(request, project_id, gene_id):
"""
Summary of a gene in a project
"""
project = get_object_or_404(Project, project_id=project_id)
if not project.can_view(request.user):
return HttpResponse("Unauthorized")
gene_id = get_gene_id_from_str(gene_id, get_reference())
gene = get_reference().get_gene(gene_id)
sys.stderr.write(project_id + " - staring gene search for: %s %s \n" % (gene_id, gene))
variant_filter = get_default_variant_filter('all_coding', mall.get_annotator().reference_population_slugs)
num_indivs = len([i for i in project.get_individuals() if i.has_variant_data()])
aac_threshold = (.2 * num_indivs) + 5
rare_variants = []
for variant in project_analysis.get_variants_in_gene(project, gene_id, variant_filter=variant_filter):
aac = get_alt_allele_count(variant)
max_af = max(variant.annotation['freqs'].values())
if aac <= aac_threshold and max_af < .01:
rare_variants.append(variant)
add_extra_info_to_variants_project(get_reference(), project, rare_variants)
knockouts = []
knockout_ids, variation = get_knockouts_in_gene(project, gene_id)
for kid in knockout_ids:
variants = variation.get_relevant_variants_for_indiv_ids([kid])
add_extra_info_to_variants_project(get_reference(), project, variants)
knockouts.append({
'indiv_id': kid,
'variants': [v.toJSON() for v in variants],
})
sys.stderr.write("Retrieved %s variants \n" % len(rare_variants))
return render(request, 'project/gene_quicklook.html', {
'gene': gene,
'gene_json': json.dumps(gene),
'project': project,
'rare_variants_json': json.dumps([v.toJSON() for v in rare_variants]),
'individuals_json': json.dumps([i.get_json_obj() for i in project.get_individuals()]),
'knockouts_json': json.dumps(knockouts),
})