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 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 has_data(self, data_key):
"""
Is data for data_key available and fully loaded for this family?
"""
if data_key == 'variation':
return self.has_variant_data() and self.get_data_status(
) == 'loaded'
elif data_key == 'breakpoints':
log.info("checking for breakpoints")
return self.has_breakpoint_data()
elif data_key == 'exome_coverage':
if self.has_coverage_data() is False:
return False
sample_ids = [
indiv.get_coverage_store_id()
for indiv in self.get_individuals()
]
statuses = set(
get_coverage_store().get_sample_statuses(sample_ids).values())
# must be at least one fully loaded
if 'loaded' not in statuses:
return False
# "there are no statuses other than loaded and None"
return len(statuses.union({'loaded', None})) == 2
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 clean_project(project_id):
"""
Clear data for this project from all the xbrowse resources:
- datastore
- coverage store
- cnv store
Does not remove any of the project's data links - so no data is lost, but everything must be rebuilt
"""
project = Project.objects.get(project_id=project_id)
individuals = project.get_individuals()
# datastore
get_mall(project_id).variant_store.delete_project(project_id)
# coverage store
for individual in individuals:
get_coverage_store().remove_sample(individual.get_coverage_store_id())
# cnv store
for individual in individuals:
get_cnv_store().remove_sample(individual.get_coverage_store_id())
def clean_project(project_id):
"""
Clear data for this project from all the xbrowse resources:
- datastore
- coverage store
- cnv store
Does not remove any of the project's data links - so no data is lost, but everything must be rebuilt
"""
project = Project.objects.get(project_id=project_id)
individuals = project.get_individuals()
# datastore
get_mall(project_id).variant_store.delete_project(project_id)
# coverage store
for individual in individuals:
get_coverage_store().remove_sample(individual.get_coverage_store_id())
# cnv store
for individual in individuals:
get_cnv_store().remove_sample(individual.get_coverage_store_id())
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 has_data(self, data_key):
"""
Is data for data_key available and fully loaded for this family?
"""
if data_key == 'variation':
return self.has_variant_data() and self.get_data_status() == 'loaded'
elif data_key == 'exome_coverage':
if self.has_coverage_data() is False:
return False
sample_ids = [indiv.get_coverage_store_id() for indiv in self.get_individuals()]
statuses = set(get_coverage_store().get_sample_statuses(sample_ids).values())
# must be at least one fully loaded
if 'loaded' not in statuses:
return False
# "there are no statuses other than loaded and None"
return len(statuses.union({'loaded', None})) == 2
def get_gene_diangostic_info(family, gene_id, variant_filter=None):
diagnostic_info = GeneDiagnosticInfo(gene_id)
diagnostic_info._gene_phenotype_summary = get_gene_phenotype_summary(
get_reference(), gene_id)
diagnostic_info._gene_sequencing_summary = get_gene_sequencing_summary(
get_coverage_store(), family, gene_id)
diagnostic_info._variants = get_diagnostic_search_variants_in_family(
get_mall(family.project).variant_store, family, gene_id,
variant_filter)
diagnostic_info._cnvs = get_diagnostic_search_cnvs_in_family(
get_cnv_store(),
family,
gene_id,
)
return diagnostic_info
def get_gene_diangostic_info(family, gene_id, variant_filter=None):
diagnostic_info = GeneDiagnosticInfo(gene_id)
diagnostic_info._gene_phenotype_summary = get_gene_phenotype_summary(get_reference(), gene_id)
diagnostic_info._gene_sequencing_summary = get_gene_sequencing_summary(get_coverage_store(), family, gene_id)
diagnostic_info._variants = get_diagnostic_search_variants_in_family(
get_mall(family.project).variant_store,
family,
gene_id,
variant_filter
)
diagnostic_info._cnvs = get_diagnostic_search_cnvs_in_family(
get_cnv_store(),
family,
gene_id,
)
return diagnostic_info
def load_coverage_for_individuals(individuals):
for individual in individuals:
if individual.coverage_file:
get_coverage_store().add_sample(
individual.get_coverage_store_id(),
gzip.open(individual.coverage_file))
def load_coverage_for_individuals(individuals):
for individual in individuals:
if individual.coverage_file:
get_coverage_store().add_sample(individual.get_coverage_store_id(), gzip.open(individual.coverage_file))