def get_all_clinical_data_for_family(project_id, family_id, indiv_id):
"""
Gets phenotype and genotype data for this individual
Args:
family_id: id of family
Returns:
A JSON object as per MME spec of a patient
"""
project = get_object_or_404(Project, project_id=project_id)
#species (only human for now) till seqr starts tracking species
species = "NCBITaxon:9606"
#contact (this should be set in settings
href = settings.MME_CONTACT_HREF
if settings.MME_PATIENT_PRIMARY_DATA_OWNER[project_id]["email"] != "":
href = href + ',' + settings.MME_PATIENT_PRIMARY_DATA_OWNER[
project_id]["email"]
contact = {
"name":
settings.MME_CONTACT_NAME + ' (data owner: ' +
settings.MME_PATIENT_PRIMARY_DATA_OWNER[project_id]["PI"] + ')',
"institution":
settings.MME_CONTACT_INSTITUTION,
"href":
href
}
#genomicFeatures section
genomic_features = []
variants = []
project_tags = ProjectTag.objects.filter(project__project_id=project_id)
for project_tag in project_tags:
variant_tags = VariantTag.objects.filter(project_tag=project_tag)
for variant_tag in variant_tags:
if family_id == variant_tag.toJSON()['family']:
variant = get_datastore(project.project_id).get_single_variant(
project.project_id,
variant_tag.toJSON()['family'],
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
if variant is None:
raise ValueError(
"Variant no longer called in this family (did the callset version change?)"
)
variants.append({
"variant": variant.toJSON(),
"tag": project_tag.title,
"family": variant_tag.family.toJSON(),
"tag_name": variant_tag.toJSON()['tag']
})
#start compiling a matchmaker-esque data structure to send back
genomic_features = []
for variant in variants:
start = variant['variant']['pos']
reference_bases = variant['variant']['ref']
alternate_bases = variant['variant']['alt']
end = int(
variant['variant']['pos_end']) #int and long are unified in python
reference_name = variant['variant']['chr'].replace('chr', '')
#now we have more than 1 gene associated to these VAR postions,
#so we will associate that information to each gene symbol
for i, gene_id in enumerate(variant['variant']['gene_ids']):
genomic_feature = {}
genomic_feature['gene'] = {"id": gene_id}
genomic_feature['variant'] = {
'assembly': settings.GENOME_ASSEMBLY_NAME,
'referenceBases': reference_bases,
'alternateBases': alternate_bases,
'start': start,
'end': end,
'referenceName': reference_name
}
genomic_feature['zygosity'] = variant['variant']['genotypes'][
indiv_id]['num_alt']
gene_symbol = ""
if gene_id != "":
gene = get_reference().get_gene(gene_id)
gene_symbol = gene['symbol']
genomic_feature['auxiliary'] = {
"tag_name": variant['tag_name'],
"gene_symbol": gene_symbol
}
genomic_features.append(genomic_feature)
#Find phenotype information
indiv = Individual.objects.get(indiv_id=indiv_id, project=project)
phenotypes_entered = get_phenotypes_entered_for_individual(
project_id, indiv.phenotips_id)
#need to eventually support "FEMALE"|"MALE"|"OTHER"|"MIXED_SAMPLE"|"NOT_APPLICABLE",
#as of now PhenoTips only has M/F
sex = "NOT_APPLICABLE"
if "M" == indiv.gender:
sex = "MALE"
if "F" == indiv.gender:
sex = "FEMALE"
features = []
if phenotypes_entered.has_key('features'):
#as of now non-standard features ('nonstandard_features') without HPO
#terms cannot be sent to MME
for f in phenotypes_entered['features']:
features.append({
"id": f['id'],
"observed": f['observed'],
"label": f['label']
})
#--depracating obfuscation as per discussion on slack and green light by @dgmacarthur
#make a unique hash to represent individual in MME for MME_ID
#h = hashlib.md5()
#h.update(indiv.indiv_id)
#id=h.hexdigest()
#label=id #using ID as label
id = indiv.indiv_id
label = indiv.indiv_id
#add new patient to affected patients
affected_patient = {
"id": id,
"species": species,
"label": label,
"contact": contact,
"features": features,
"sex": sex,
"genomicFeatures": genomic_features
}
#map to put into mongo
time_stamp = datetime.datetime.fromtimestamp(
time.time()).strftime('%Y_%m_%d_%H_%M_%S')
detailed_id_map = {
"generated_on": time_stamp,
"project_id": project_id,
"family_id": family_id,
"individual_id": indiv.indiv_id,
"mme_id": id,
"individuals_used_for_phenotypes": affected_patient
}
return detailed_id_map, affected_patient
def get_all_clinical_data_for_family(project_id,family_id,indiv_id):
"""
Gets phenotype and genotype data for this individual
Args:
family_id: id of family
Returns:
A JSON object as per MME spec of a patient
"""
project = get_object_or_404(Project, project_id=project_id)
#species (only human for now) till seqr starts tracking species
species="NCBITaxon:9606"
contact={
"name": project.mme_primary_data_owner,
"institution" : project.mme_contact_institution,
"href" : project.mme_contact_url
}
#genomicFeatures section
genomic_features=[]
variants=[]
project_tags = ProjectTag.objects.filter(project__project_id=project_id)
for project_tag in project_tags:
variant_tags = VariantTag.objects.filter(project_tag=project_tag)
for variant_tag in variant_tags:
if variant_tag.family is not None and family_id == variant_tag.family.family_id:
variant = get_datastore(project).get_single_variant(
project.project_id,
variant_tag.family.family_id,
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
if variant is None:
logging.info("Variant no longer called in this family (did the callset version change?)")
continue
variants.append({"variant": variant.toJSON(),
"tag": project_tag.title,
"family": variant_tag.family.toJSON(),
"tag_name": variant_tag.project_tag.tag,
})
current_genome_assembly = find_genome_assembly(project)
#start compiling a matchmaker-esque data structure to send back
genomic_features=[]
for variant in variants:
start = variant['variant']['pos']
reference_bases = variant['variant']['ref']
alternate_bases = variant['variant']['alt']
end = int(variant['variant']['pos_end']) #int and long are unified in python
reference_name = variant['variant']['chr'].replace('chr','')
#now we have more than 1 gene associated to these VAR postions,
#so we will associate that information to each gene symbol
for i,gene_id in enumerate(variant['variant']['gene_ids']):
genomic_feature = {}
genomic_feature['gene'] ={"id": gene_id }
genomic_feature['variant']={
'assembly':current_genome_assembly,
'referenceBases':reference_bases,
'alternateBases':alternate_bases,
'start':start,
'end':end,
'referenceName':reference_name
}
genomic_feature['zygosity'] = variant['variant']['genotypes'][indiv_id]['num_alt']
gene_symbol=""
if gene_id != "":
gene = get_reference().get_gene(gene_id)
if gene:
gene_symbol = gene['symbol']
genomic_feature['auxiliary']={
"tag_name":variant['tag_name'],
"gene_symbol":gene_symbol
}
genomic_features.append(genomic_feature)
#Find phenotype information
indiv = Individual.objects.get(indiv_id=indiv_id, project=project)
phenotypes_entered = get_phenotypes_entered_for_individual(project_id,indiv.phenotips_id)
#need to eventually support "FEMALE"|"MALE"|"OTHER"|"MIXED_SAMPLE"|"NOT_APPLICABLE",
#as of now PhenoTips only has M/F
sex="NOT_APPLICABLE"
if "M" == indiv.gender:
sex="MALE"
if "F" == indiv.gender:
sex="FEMALE"
features=[]
if phenotypes_entered.has_key('features'):
#as of now non-standard features ('nonstandard_features') without HPO
#terms cannot be sent to MME
for f in phenotypes_entered['features']:
features.append({
"id":f['id'],
"observed":f['observed'],
"label":f['label']})
id=indiv.indiv_id
label=indiv.indiv_id
#add new patient to affected patients
affected_patient={"id":id,
"species":species,
"label":label,
"contact":contact,
"features":features,
"sex":sex,
"genomicFeatures":genomic_features
}
#map to put into mongo
time_stamp = datetime.datetime.fromtimestamp(time.time()).strftime('%Y_%m_%d_%H_%M_%S')
detailed_id_map={"generated_on": time_stamp,
"project_id":project_id,
"family_id":family_id,
"individual_id":indiv.indiv_id,
"mme_id":id,
"individuals_used_for_phenotypes":affected_patient}
return detailed_id_map,affected_patient
def get_all_clinical_data_for_family(project_id,family_id,indiv_id):
"""
Gets phenotype and genotype data for this individual
Args:
family_id: id of family
Returns:
A JSON object as per MME spec of a patient
"""
project = get_object_or_404(Project, project_id=project_id)
seqr_project = project.seqr_project
#species (only human for now) till seqr starts tracking species
species="NCBITaxon:9606"
contact={
"name":seqr_project.mme_primary_data_owner,
"institution" : seqr_project.mme_contact_institution,
"href" : seqr_project.mme_contact_url
}
#genomicFeatures section
genomic_features=[]
variants=[]
project_tags = ProjectTag.objects.filter(project__project_id=project_id)
for project_tag in project_tags:
variant_tags = VariantTag.objects.filter(project_tag=project_tag)
for variant_tag in variant_tags:
if family_id == variant_tag.toJSON()['family']:
variant = get_datastore(project.project_id).get_single_variant(
project.project_id,
variant_tag.toJSON()['family'],
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
if variant is None:
logging.info("Variant no longer called in this family (did the callset version change?)")
continue
variants.append({"variant":variant.toJSON(),
"tag":project_tag.title,
"family":variant_tag.family.toJSON(),
"tag_name":variant_tag.toJSON()['tag']
})
current_genome_assembly = find_genome_assembly(seqr_project)
#start compiling a matchmaker-esque data structure to send back
genomic_features=[]
for variant in variants:
start = variant['variant']['pos']
reference_bases = variant['variant']['ref']
alternate_bases = variant['variant']['alt']
end = int(variant['variant']['pos_end']) #int and long are unified in python
reference_name = variant['variant']['chr'].replace('chr','')
#now we have more than 1 gene associated to these VAR postions,
#so we will associate that information to each gene symbol
for i,gene_id in enumerate(variant['variant']['gene_ids']):
genomic_feature = {}
genomic_feature['gene'] ={"id": gene_id }
genomic_feature['variant']={
'assembly':current_genome_assembly,
'referenceBases':reference_bases,
'alternateBases':alternate_bases,
'start':start,
'end':end,
'referenceName':reference_name
}
genomic_feature['zygosity'] = variant['variant']['genotypes'][indiv_id]['num_alt']
gene_symbol=""
if gene_id != "":
gene = get_reference().get_gene(gene_id)
gene_symbol = gene['symbol']
genomic_feature['auxiliary']={
"tag_name":variant['tag_name'],
"gene_symbol":gene_symbol
}
genomic_features.append(genomic_feature)
#Find phenotype information
indiv = Individual.objects.get(indiv_id=indiv_id,project=project)
phenotypes_entered = get_phenotypes_entered_for_individual(project_id,indiv.phenotips_id)
#need to eventually support "FEMALE"|"MALE"|"OTHER"|"MIXED_SAMPLE"|"NOT_APPLICABLE",
#as of now PhenoTips only has M/F
sex="NOT_APPLICABLE"
if "M" == indiv.gender:
sex="MALE"
if "F" == indiv.gender:
sex="FEMALE"
features=[]
if phenotypes_entered.has_key('features'):
#as of now non-standard features ('nonstandard_features') without HPO
#terms cannot be sent to MME
for f in phenotypes_entered['features']:
features.append({
"id":f['id'],
"observed":f['observed'],
"label":f['label']})
id=indiv.indiv_id
label=indiv.indiv_id
#add new patient to affected patients
affected_patient={"id":id,
"species":species,
"label":label,
"contact":contact,
"features":features,
"sex":sex,
"genomicFeatures":genomic_features
}
#map to put into mongo
time_stamp = datetime.datetime.fromtimestamp(time.time()).strftime('%Y_%m_%d_%H_%M_%S')
detailed_id_map={"generated_on": time_stamp,
"project_id":project_id,
"family_id":family_id,
"individual_id":indiv.indiv_id,
"mme_id":id,
"individuals_used_for_phenotypes":affected_patient}
return detailed_id_map,affected_patient
def get_all_clinical_data_for_family(project_id,family_id):
"""
Gets phenotype and genotype data for this individual
Args:
family_id: id of family
Returns:
A JSON object as per MME spec of a patient
"""
project = get_object_or_404(Project, project_id=project_id)
#species (only human for now) till seqr starts tracking species
species="NCBITaxon:9606"
#contact (this should be set in settings
contact={
"name":settings.MME_CONTACT_NAME,
"institution" : settings.MME_CONTACT_INSTITUTION,
"href" : settings.MME_CONTACT_HREF
}
#genomicFeatures section
genomic_features=[]
#family_data,variant_data,_,_ = fetch_project_individuals_data(project_id)
#variants,phenotype_entry_counts = fetch_project_individuals_data(project_id)
#--
variants=[]
project_tags = ProjectTag.objects.filter(project__project_id='1kg')
for project_tag in project_tags:
variant_tags = VariantTag.objects.filter(project_tag=project_tag)
for variant_tag in variant_tags:
variant = get_datastore(project.project_id).get_single_variant(
project.project_id,
variant_tag.toJSON()['family'],
variant_tag.xpos,
variant_tag.ref,
variant_tag.alt,
)
if variant is None:
raise ValueError("Variant no longer called in this family (did the callset version change?)")
variants.append({"variant":variant.toJSON(),
"tag":project_tag.title,
"family":variant_tag.family.toJSON()})
for variant in variants:
start = variant['variant']['pos']
reference_bases = variant['variant']['ref']
alternate_bases = variant['variant']['alt']
end = int(variant['variant']['pos_end']) #int and long are unified in python
reference_name = variant['variant']['chr'].replace('chr','')
#now we have more than 1 gene associated to these VAR postions,
#so we will associate that information to each gene symbol
genomic_features=[]
for i,gene_id in enumerate(variant['variant']['gene_ids']):
genomic_feature = {}
genomic_feature['gene'] ={"id": gene_id }
genomic_feature['variant']={
'assembly':settings.GENOME_ASSEMBLY_NAME,
'referenceBases':reference_bases,
'alternateBases':alternate_bases,
'start':start,
'end':end,
'referenceName':reference_name
}
genomic_features.append(genomic_feature)
#all affected patients
affected_patients=[]
detailed_id_map=[]
id_map={}
#--find individuals in this family
family = Family.objects.get(project=project, family_id=family_id)
for indiv in family.get_individuals():
if indiv.affected_status_display() == 'Affected':
phenotypes_entered = get_phenotypes_entered_for_individual(project_id,indiv.indiv_id)
#need to eventually support "FEMALE"|"MALE"|"OTHER"|"MIXED_SAMPLE"|"NOT_APPLICABLE",
#as of now PhenoTips only has M/F
sex="FEMALE"
if "M" == indiv.gender:
sex="MALE"
features=[]
if phenotypes_entered.has_key('features'):
#as of now non-standard features ('nonstandard_features') without HPO
#terms cannot be sent to MME
for f in phenotypes_entered['features']:
features.append({
"id":f['id'],
"observed":f['observed']})
#make a unique hash to represent individual in MME for MME_ID
h = hashlib.md5()
h.update(indiv.indiv_id)
id=h.hexdigest()
label=id #using ID as label
id_map[id]=indiv.indiv_id
#add new patient to affected patients
affected_patients.append({
"patient":
{
"id":id,
"species":species,
"label":label,
"contact":contact,
"features":features,
"sex":sex,
"genomicFeatures":genomic_features
}
})
#map to put into mongo
time_stamp = datetime.datetime.fromtimestamp(time.time()).strftime('%Y_%m_%d_%H_%M_%S')
detailed_id_map.append({"generated_on": time_stamp,
"project_id":project_id,
"family_id":family_id,
"individual_id":indiv.indiv_id,
"mme_id":id,
"individuals_used_for_phenotypes":affected_patients})
return detailed_id_map,affected_patients,id_map
