def export_case_review_individuals(request, project_guid):
"""Export case review Individuals table.
Args:
project_guid (string): GUID of the project for which to export case review individual data
"""
format = request.GET.get('file_format', 'tsv')
project = Project.objects.filter(guid=project_guid)
if not project:
raise ValueError("Invalid project GUID: %s" % project_guid)
project = project[0]
individuals = Individual.objects.filter(
family__project=project,
case_review_status__regex="[\w].*").order_by('family__family_id',
'affected')
filename_prefix = "%s_case_review_individuals" % _slugify(project.name)
return export_individuals(filename_prefix,
individuals,
format,
include_case_review_columns=True,
include_phenotips_columns=True)
def export_project_individuals_handler(request, project_guid):
"""Export project Individuals table.
Args:
project_guid (string): GUID of the project for which to export individual data
"""
format = request.GET.get('file_format', 'tsv')
include_phenotypes = bool(request.GET.get('include_phenotypes'))
project = get_project_and_check_permissions(project_guid, request.user)
# get all individuals in this project
individuals = Individual.objects.filter(family__project=project).order_by(
'family__family_id', 'affected')
filename_prefix = "%s_individuals" % _slugify(project.name)
return export_individuals(
filename_prefix,
individuals,
format,
include_hpo_terms_present=include_phenotypes,
include_hpo_terms_absent=include_phenotypes,
)
def export_case_review_families(request, project_guid):
"""Export case review Families table.
Args:
project_guid (string): GUID of the project for which to export case review family data
"""
format = request.GET.get('file_format', 'tsv')
project = Project.objects.filter(guid=project_guid)
if not project:
raise ValueError("Invalid project GUID: %s" % project_guid)
project = project[0]
# get all families in this project that have at least 1 individual in case review.
families = set()
for i in Individual.objects.filter(
family__project=project,
case_review_status__regex="[\w].*").order_by('family__family_id'):
families.add(i.family)
filename_prefix = "%s_case_review_families" % _slugify(project.name)
return export_families(filename_prefix,
families,
format,
include_case_review_columns=True)
def export_case_review_individuals_handler(request, project_guid):
"""Export case review Individuals table.
Args:
project_guid (string): GUID of the project for which to export case review individual data
"""
format = request.GET.get('file_format', 'tsv')
project = get_project_and_check_permissions(project_guid, request.user)
individuals = Individual.objects.filter(
family__project=project,
case_review_status__regex="[\w].*").order_by('family__family_id',
'affected')
filename_prefix = "%s_case_review_individuals" % _slugify(project.name)
return export_individuals(
filename_prefix,
individuals,
format,
include_case_review_status=True,
include_case_review_discussion=True,
include_hpo_terms_present=True,
include_hpo_terms_absent=True,
)
def _deprecated_create_original_project(project):
"""DEPRECATED - create project in original xbrowse schema to keep things in sync.
Args:
project (object): new-style seqr project model
"""
# keep new seqr.Project model in sync with existing xbrowse_server.base.models
base_project, created = BaseProject.objects.get_or_create(
project_id=_slugify(project.name))
if created:
logger.info("Created base project %s" % base_project)
base_project.project_name = project.name
base_project.description = project.description
base_project.seqr_project = project
base_project.save()
for reference_population_id in [
"gnomad-genomes2", "gnomad-exomes2", "topmed"
]:
try:
population = ReferencePopulation.objects.get(
slug=reference_population_id)
logger.info("Adding population " + reference_population_id +
" to project " + str(project))
base_project.private_reference_populations.add(population)
except Exception as e:
logger.error("Unable to add reference population %s: %s" %
(reference_population_id, e))
return base_project
def _deprecated_create_original_project(project):
"""DEPRECATED - create project in original xbrowse schema to keep things in sync.
Args:
project (object): new-style seqr project model
"""
# keep new seqr.Project model in sync with existing xbrowse_server.base.models
base_project, created = BaseProject.objects.get_or_create(
project_id=_slugify(project.name)
)
if created:
logger.info("Created base project %s" % base_project)
base_project.project_name = project.name
base_project.genome_version = project.genome_version
base_project.description = project.description or ""
base_project.seqr_project = project
base_project.save()
for reference_population_id in ["gnomad-genomes2", "gnomad-exomes2", "topmed"]:
try:
population = ReferencePopulation.objects.get(slug=reference_population_id)
logger.info("Adding population " + reference_population_id + " to project " + str(project))
base_project.private_reference_populations.add(population)
except Exception as e:
logger.error("Unable to add reference population %s: %s" % (reference_population_id, e))
return base_project
def __init__(self, cluster_id):
"""
Args:
cluster_id (string): unique id for the dataproc cluster.
"""
# from gcloud error message: clusterName must be a match of regex '(?:[a-z](?:[-a-z0-9]{0,49}[a-z0-9])?).'
self.cluster_id = "c" + _slugify(cluster_id).replace("_", "-").lower()
def create_project(request):
"""Create a new project.
HTTP POST
Request body - should contain json params:
name: Project name
description: Project description
Response body - will be json with the following structure, representing the created project:
{
'projectsByGuid': { <projectGuid1> : { ... <project key-value pairs> ... } }
}
"""
request_json = json.loads(request.body)
if 'form' not in request_json:
return create_json_response(
{},
status=400,
reason="Invalid request: 'form' not in request_json")
form_data = request_json['form']
name = form_data.get('name')
if not name:
return create_json_response({},
status=400,
reason="'Name' cannot be blank")
description = form_data.get('description')
project = Project.objects.create(created_by=request.user,
name=name,
description=description)
# keep new seqr.Project model in sync with existing xbrowse_server.base.models - TODO remove this code after transition to new schema is finished
try:
base_project, created = BaseProject.objects.get_or_create(
project_id=_slugify(name))
if created:
print("Created project %s" % base_project)
base_project.project_name = name
base_project.description = description
base_project.save()
project.deprecated_project_id = base_project.project_id
project.save()
except:
raise
projects_by_guid = {
project.guid: _get_json_for_project(project, request.user)
}
return create_json_response({
'projectsByGuid': projects_by_guid,
})
def _get_json_for_locus_lists(project):
result = []
for locus_list in get_objects_for_group(project.can_view_group, CAN_VIEW, LocusList):
result.append({
'locusListGuid': locus_list.guid,
'createdDate': locus_list.created_date,
'name': locus_list.name,
'deprecatedGeneListId': _slugify(locus_list.name),
'description': locus_list.description,
'numEntries': LocusListEntry.objects.filter(parent=locus_list).count(),
})
return sorted(result, key=lambda locus_list: locus_list['createdDate'])
def _enable_phenotips_for_project(project):
"""Creates 2 users in PhenoTips for this project (one that will be view-only and one that'll
have edit permissions for patients in the project).
"""
project.is_phenotips_enabled = True
project.phenotips_user_id = _slugify(project.name)
# view-only user
username, password = _get_phenotips_uname_and_pwd_for_project(project.phenotips_user_id, read_only=True)
create_phenotips_user(username, password)
# user with edit permissions
username, password = _get_phenotips_uname_and_pwd_for_project(project.phenotips_user_id, read_only=False)
create_phenotips_user(username, password)
project.save()
def export_project_families_handler(request, project_guid):
"""Export project Families table.
Args:
project_guid (string): GUID of the project for which to export family data
"""
format = request.GET.get('file_format', 'tsv')
project = _get_project_and_check_permissions(project_guid, request.user)
# get all families in this project
families = Family.objects.filter(project=project).order_by('family_id')
filename_prefix = "%s_families" % _slugify(project.name)
return export_families(filename_prefix, families, format, include_case_review_columns=False)
def _deprecated_create_original_project(project):
"""DEPRECATED - create project in original xbrowse schema to keep things in sync.
Args:
project (object): new-style seqr project model
"""
# keep new seqr.Project model in sync with existing xbrowse_server.base.models
base_project, created = BaseProject.objects.get_or_create(
project_id=_slugify(project.name))
if created:
logger.info("Created base project %s" % base_project)
base_project.project_name = project.name
base_project.description = project.description
base_project.save()
return base_project
def export_case_review_families(request, project_guid):
"""Export case review Families table.
Args:
project_guid (string): GUID of the project for which to export case review family data
"""
format = request.GET.get('file_format', 'tsv')
project = _get_project_and_check_permissions(project_guid, request.user)
# get all families in this project that have at least 1 individual in case review.
families = set()
for i in Individual.objects.filter(
family__project=project,
case_review_status__regex="[\w].*").order_by('family__family_id'):
families.add(i.family)
filename_prefix = "%s_case_review_families" % _slugify(project.name)
return export_families(filename_prefix,
families,
format,
include_case_review_columns=True)
def export_to_elasticsearch(self, vds_path, dataset_id, dataset_type,
genome_version):
"""Export the dataset to elasticsearch. Assumes the dataproc cluster already exists.
Args:
vds_path (string):
dataset_id (string):
dataset_type (string):
genome_version (string): "37" or "38"
"""
#elasticsearch_host_ip = self._get_k8s_resource_name("pods", labels={'name': 'elasticsearch'}, json_path=".items[0].status.hostIP")
#solr_node_name = self._get_k8s_resource_name("pods", labels={'name': 'solr'}, json_path=".items[0].spec.nodeName")
ELASTICSEARCH_NODE_IP = subprocess.check_output(
"kubectl get pods -l name=elasticsearch -o jsonpath='{.items[0].status.hostIP}'",
shell=True)
index_name = _slugify(dataset_id).lower()
script_path = os.path.join(
BASE_DIR, "seqr/pipelines/hail/export_variants_to_ES.py")
script_args = [
"--host",
ELASTICSEARCH_NODE_IP,
"--port",
ELASTICSEARCH_PORT,
"--genome-version",
genome_version,
"--index",
index_name,
"--index-type",
dataset_type,
vds_path,
]
self.hail_runner.run_hail(script_path, *script_args)
