def update_xbrowse_vcfffiles(project, sample_type, elasticsearch_index, dataset_path, matched_sample_id_to_sample_record):
base_project = find_matching_xbrowse_model(project)
get_datastore(base_project).bust_project_cache(base_project.project_id)
clear_project_results_cache(base_project.project_id)
vcf_file = VCFFile.objects.filter(
project=base_project,
dataset_type=Sample.DATASET_TYPE_VARIANT_CALLS,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index).order_by('-pk').first()
if not vcf_file:
vcf_file = VCFFile.objects.create(
project=base_project,
dataset_type=Sample.DATASET_TYPE_VARIANT_CALLS,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index,
)
logging.info("Created vcf file: " + str(vcf_file.__dict__))
vcf_file.file_path = dataset_path
vcf_file.loaded_date = matched_sample_id_to_sample_record.values()[0].loaded_date
vcf_file.save()
base_individuals = BaseIndividual.objects.filter(
seqr_individual_id__in=[s.individual_id for s in matched_sample_id_to_sample_record.values()]
)
for base_indiv in base_individuals:
base_indiv.vcf_files.add(vcf_file)
def _deprecated_update_vcfffiles(project, sample_type, elasticsearch_index, dataset_path, matched_sample_id_to_sample_record):
base_project = BaseProject.objects.get(seqr_project=project)
get_datastore(base_project).bust_project_cache(base_project.project_id)
clear_project_results_cache(base_project.project_id)
vcf_file = VCFFile.objects.filter(
project=base_project,
dataset_type=Sample.DATASET_TYPE_VARIANT_CALLS,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index).order_by('-pk').first()
if not vcf_file:
vcf_file = VCFFile.objects.create(
project=base_project,
dataset_type=Sample.DATASET_TYPE_VARIANT_CALLS,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index,
)
logger.info("Created vcf file: " + str(vcf_file.__dict__))
vcf_file.file_path = dataset_path
vcf_file.loaded_date = matched_sample_id_to_sample_record.values()[0].loaded_date
vcf_file.save()
for indiv in [s.individual for s in matched_sample_id_to_sample_record.values()]:
for base_indiv in BaseIndividual.objects.filter(seqr_individual=indiv).only('id'):
base_indiv.vcf_files.add(vcf_file)
def add_dataset_handler(request, project_guid):
"""Create or update samples for the given dataset
Args:
request: Django request object
project_guid (string): GUID of the project that should be updated
HTTP POST
Request body - should contain the following json structure:
{
'sampleType': <"WGS", "WES", or "RNA"> (required)
'datasetType': <"VARIANTS", or "ALIGN"> (required)
'elasticsearchIndex': <String>
'datasetPath': <String>
'datasetName': <String>
'ignoreExtraSamplesInCallset': <Boolean>
'mappingFile': { 'uploadedFileId': <Id for temporary uploaded file> }
}
Response body - will contain the following structure:
"""
logger.info("add_dataset_handler: " + str(request))
project = get_project_and_check_permissions(project_guid,
request.user,
permission_level=CAN_EDIT)
request_json = json.loads(request.body)
logger.info("add_dataset_handler: received %s" % pformat(request_json))
required_fields = ['sampleType', 'datasetType']
if any(field not in request_json for field in required_fields):
raise ValueError("request must contain fields: {}".format(
', '.join(required_fields)))
sample_type = request_json['sampleType']
dataset_type = request_json['datasetType']
elasticsearch_index = request_json.get('elasticsearchIndex')
if elasticsearch_index:
elasticsearch_index = elasticsearch_index.strip()
dataset_path = request_json.get('datasetPath')
if dataset_path:
dataset_path = dataset_path.strip()
dataset_name = request_json.get('datasetName')
if dataset_name:
dataset_name = dataset_name.strip()
ignore_extra_samples_in_callset = request_json.get(
'ignoreExtraSamplesInCallset')
ignore_missing_family_members = request_json.get(
'ignoreMissingFamilyMembers')
mapping_file_id = request_json.get('mappingFile', {}).get('uploadedFileId')
mapping_file_path = request_json.get('mappingFilePath')
try:
updated_samples, created_sample_ids = add_dataset(
project=project,
sample_type=sample_type,
dataset_type=dataset_type,
elasticsearch_index=elasticsearch_index,
dataset_path=dataset_path,
dataset_name=dataset_name,
max_edit_distance=0,
ignore_extra_samples_in_callset=ignore_extra_samples_in_callset,
ignore_missing_family_members=ignore_missing_family_members,
mapping_file_path=mapping_file_path,
mapping_file_id=mapping_file_id,
)
# update VCFFile records
if updated_samples:
if dataset_type == Sample.DATASET_TYPE_VARIANT_CALLS:
base_project = BaseProject.objects.get(seqr_project=project)
get_datastore(base_project).bust_project_cache(
base_project.project_id)
clear_project_results_cache(base_project.project_id)
vcf_file = VCFFile.objects.filter(
project=base_project,
dataset_type=dataset_type,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index).order_by(
'-pk').first()
if not vcf_file:
vcf_file = VCFFile.objects.create(
project=base_project,
dataset_type=dataset_type,
sample_type=sample_type,
elasticsearch_index=elasticsearch_index,
)
logger.info("Created vcf file: " + str(vcf_file.__dict__))
vcf_file.file_path = dataset_path or "{}.vcf.gz".format(
elasticsearch_index
) # legacy VCFFile model requires non-empty vcf path
vcf_file.loaded_date = iter(updated_samples).next().loaded_date
vcf_file.save()
for indiv in [s.individual for s in updated_samples]:
for base_indiv in BaseIndividual.objects.filter(
seqr_individual=indiv).only('id'):
base_indiv.vcf_files.add(vcf_file)
elif dataset_type == Sample.DATASET_TYPE_READ_ALIGNMENTS:
for sample in updated_samples:
for base_indiv in BaseIndividual.objects.filter(
seqr_individual=sample.individual).only('id'):
base_indiv.bam_file_path = sample.dataset_file_path
base_indiv.save()
updated_sample_json = get_json_for_samples(updated_samples,
project_guid=project_guid)
response = {
'samplesByGuid': {s['sampleGuid']: s
for s in updated_sample_json}
}
updated_individuals = {
s['individualGuid']
for s in updated_sample_json if s['sampleId'] in created_sample_ids
}
if updated_individuals:
individuals = Individual.objects.filter(
guid__in=updated_individuals).prefetch_related(
'sample_set', 'family').only('guid')
response['individualsByGuid'] = {
ind.guid: {
'sampleGuids':
[s.guid for s in ind.sample_set.only('guid').all()]
}
for ind in individuals
}
for ind in individuals:
family = ind.family
if family.analysis_status == Family.ANALYSIS_STATUS_WAITING_FOR_DATA:
update_seqr_model(family,
analysis_status=Family.
ANALYSIS_STATUS_ANALYSIS_IN_PROGRESS)
return create_json_response(response)
except Exception as e:
traceback.print_exc()
return create_json_response({'errors': [e.message or str(e)]},
status=400)