def test_get_allele_frequencies(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
output1 = [
0.3391, 0.2859, 0.4539, 0.4274, 0.253, 0.3272, 0.392, 0.4176,
0.3196, 0.43, 0.2793, 0.5536, 0.4469, 0.4538, 0.3747, 0.2942,
0.4278
]
self.assertEqual(vcf_parser.get_allele_frequencies(input1), output1)
def test_get_rs_number(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
output1 = 'rs3795269'
input2 = {
'NM_002617.3': {
'Existing_variation': 'gene1'
},
'NM_002614.3': {
'Existing_variation': 'gene2'
}
}
output2 = 'gene1|gene2'
input3 = {
'NM_002617.3': {
'Existing_variation': ''
},
'NM_002614.3': {
'Existing_variation': 'gene2'
}
}
output3 = 'gene2'
input4 = {
'NM_002617.3': {
'gene1': ''
},
'NM_002614.3': {
'test': 'gene2'
}
}
output4 = "None"
self.assertEqual(vcf_parser.get_rs_number(input1), output1)
self.assertEqual(vcf_parser.get_rs_number(input2), output2)
self.assertEqual(vcf_parser.get_rs_number(input3), output3)
self.assertEqual(vcf_parser.get_rs_number(input4), output4)
def test_get_max_af(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
input2 = {'NM_002617.3': {'MAX_AF': '0.01'}}
input3 = {'NM_002617.3': {'MAX_AF': ''}}
input4 = {'NM_002617.3': {'NO_MAX_AF': ''}}
self.assertEqual(vcf_parser.get_max_af(input1), 0.5536)
self.assertEqual(vcf_parser.get_max_af(input2), 0.01)
self.assertEqual(vcf_parser.get_max_af(input3), 0.0)
self.assertRaises(KeyError, vcf_parser.get_max_af, input4)
def test_get_variant_genes_list(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
output1 = [('PEX10', '-1'), ('RER1', '1')]
input2 = {
'NM_002617.3': {
'SYMBOL': 'gene1',
'STRAND': '1'
},
'NM_002614.3': {
'SYMBOL': 'gene2',
'STRAND': '1'
}
}
output2 = [('gene1', '1'), ('gene2', '1')]
input3 = {
'NM_002617.3': {
'SYMBOL': '',
'STRAND': '1'
},
'NM_002614.3': {
'SYMBOL': 'gene2',
'STRAND': '1'
}
}
output3 = [('gene2', '1')]
self.assertEqual(vcf_parser.get_variant_genes_list(input1), output1)
self.assertEqual(set(vcf_parser.get_variant_genes_list(input2)),
set(output2))
self.assertEqual(set(vcf_parser.get_variant_genes_list(input3)),
set(output3))
def test_get_clin_sig(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
input2 = {'NM_002617.3': {'CLIN_SIG': 'test2'}}
input3 = {'NM_002617.3': {'CLIN_SIG': 'test3'}}
input4 = {'NM_002617.3': {'NO_CLIN_SIG': 'test4'}}
input5 = master_list[5]['transcript_data']
self.assertEqual(vcf_parser.get_clin_sig(input1), 'None')
self.assertEqual(vcf_parser.get_clin_sig(input2), 'test2')
self.assertEqual(vcf_parser.get_clin_sig(input3), 'test3')
self.assertRaises(KeyError, vcf_parser.get_clin_sig, input4)
self.assertEqual(vcf_parser.get_clin_sig(input5), 'benign')
def upload_sample_vcf(output_dir, sample_name):
"""
Takes a VEP annotated vcf and uploads the variants contained within it.
Input:
output_dir = The directory containing the pipline output.
sample_name = A string representing the unique sample name.
Output:
None - uploads and returns None.
"""
#Get the sample - if we cannot find it then raise and error.
try:
sample = Sample.objects.get(name=sample_name)
except Sample.DoesNotExist:
raise CommandError("Cannot find a sample in the DB with name " +
sample_name + ".")
already_uploaded = VariantSample.objects.filter(sample=sample)
if already_uploaded.exists():
raise CommandError("Stuff already uploaded against this sample.")
#Find VCF file
query = output_dir + "vcfs*/" + sample_name + "*.vcf.gz"
vcf_file_path = glob.glob(query)
if len(vcf_file_path) != 1:
raise CommandError("Found more than one file for sample: " +
sample_name)
else:
vcf_file_path = vcf_file_path[0]
#validate vcf
validation_report = vcf_parser.validate_input(vcf_file_path, sample_name)
if validation_report[0] == False:
raise CommandError('Error opening vcf file: ' + validation_report[1])
if vcf_parser.vep_annotated(vcf_file_path) == False:
raise CommandError("No VEP annotations detected in vcf")
#Check we have an admin user in the database for the next stage
try:
user = User.objects.get(pk=1) # a superuser has to have been created
except:
raise CommandError(
'Could not find an appropiate user to use for downstream data entry - please create an admin with pk=1'
)
#Try and parse the vcf using the vcf_parser
try:
vcf_data = vcf_parser.create_master_list(vcf_file_path, sample_name)
except:
raise CommandError('Could not process data using vcf_parser function')
#update sample information e.g. vcf location.
sample.vcf_file = vcf_file_path
sample.save()
for variant in vcf_data:
chromosome = variant['chrom']
pos = str(variant['pos'])
ref = variant['reference']
alt = variant['alt_alleles'][0]
#hash_id = hashlib.sha256(chromosome+" "+pos+" "+ref+" "+alt).hexdigest()
hash_id = variant_utilities.get_variant_hash(chromosome, pos, ref, alt)
gene_list = vcf_parser.get_variant_genes_list(
variant['transcript_data'])
rs_number = vcf_parser.get_rs_number(variant['transcript_data'])
worst_consequence = vcf_parser.worst_consequence(
variant['transcript_data'])
worst_consequence = Consequence.objects.get(name=worst_consequence)
max_af = vcf_parser.get_max_af(variant['transcript_data'])
allele_frequencies = vcf_parser.get_allele_frequencies(
variant['transcript_data'])
clin_sig = vcf_parser.get_clin_sig(variant['transcript_data'])
af = allele_frequencies[0]
afr_af = allele_frequencies[1]
amr_af = allele_frequencies[2]
eur_af = allele_frequencies[3]
eas_af = allele_frequencies[4]
sas_af = allele_frequencies[5]
exac_af = allele_frequencies[6]
exac_adj_af = allele_frequencies[7]
exac_afr_af = allele_frequencies[8]
exac_amr_af = allele_frequencies[9]
exac_eas_af = allele_frequencies[10]
exac_fin_af = allele_frequencies[11]
exac_nfe_af = allele_frequencies[12]
exac_oth_af = allele_frequencies[13]
exac_sas_af = allele_frequencies[14]
esp_aa_af = allele_frequencies[15]
esp_ea_af = allele_frequencies[16]
#Look for a variant in the database if we have not seen it before create a new one
try:
new_variant = Variant.objects.get(variant_hash=hash_id)
except Variant.DoesNotExist:
new_variant = Variant(chromosome=chromosome,
position=pos,
ref=ref,
alt=alt,
variant_hash=hash_id,
rs_number=rs_number,
last_updated=timezone.now(),
worst_consequence=worst_consequence,
max_af=max_af,
af=af,
afr_af=afr_af,
amr_af=amr_af,
eur_af=eur_af,
eas_af=eas_af,
sas_af=sas_af,
exac_af=exac_af,
exac_adj_af=exac_adj_af,
exac_afr_af=exac_afr_af,
exac_amr_af=exac_amr_af,
exac_eas_af=exac_eas_af,
exac_fin_af=exac_fin_af,
exac_nfe_af=exac_nfe_af,
exac_oth_af=exac_oth_af,
exac_sas_af=exac_sas_af,
esp_aa_af=esp_aa_af,
esp_ea_af=esp_ea_af,
clinical_sig=clin_sig)
new_variant.save()
for gene in gene_list:
try:
gene_model = Gene.objects.get(name=gene[0])
except Gene.DoesNotExist:
gene_model = Gene(name=gene[0])
gene_model.save()
#Now create transcripts
for transcript_key in variant['transcript_data']:
if transcript_key == "":
try:
transcript_model = Transcript.objects.get(
name='no_transcript')
except Transcript.DoesNotExist:
transcript_model = Transcript(name='no_transcript',
canonical=False)
transcript_model.save()
else:
try:
transcript_model = Transcript.objects.get(
name=transcript_key)
except Transcript.DoesNotExist:
canonical = variant['transcript_data'][transcript_key][
'CANONICAL']
if canonical == 'YES':
canonical = True
else:
canonical = False
gene = variant['transcript_data'][transcript_key][
'SYMBOL']
if gene != "":
gene = Gene.objects.get(name=gene)
transcript_model = Transcript(name=transcript_key,
canonical=canonical,
gene=gene)
transcript_model.save()
else:
transcript_model = Transcript(name=transcript_key,
canonical=canonical)
transcript_model.save()
#now create transcriptvariant model
consequence = variant['transcript_data'][transcript_key][
'Consequence']
exon = variant['transcript_data'][transcript_key]['EXON']
intron = variant['transcript_data'][transcript_key]['INTRON']
hgvsc_t = variant['transcript_data'][transcript_key]['HGVSc']
hgvsp_t = variant['transcript_data'][transcript_key]['HGVSp']
codons = variant['transcript_data'][transcript_key]['Codons']
cdna_position = variant['transcript_data'][transcript_key][
'cDNA_position']
cds_position = variant['transcript_data'][transcript_key][
'CDS_position']
protein_position = variant['transcript_data'][transcript_key][
'Protein_position']
amino_acids = variant['transcript_data'][transcript_key][
'Amino_acids']
picked = variant['transcript_data'][transcript_key]['PICK']
if picked == '1':
picked = True
else:
picked = False
variant_transcript = VariantTranscript(
variant=new_variant,
transcript=transcript_model,
consequence=consequence,
exon=exon,
intron=intron,
hgvsc=hgvsc_t,
hgvsp=hgvsp_t,
codons=codons,
cdna_position=cdna_position,
protein_position=protein_position,
amino_acids=amino_acids,
picked=picked)
variant_transcript.save()
genotype = variant['genotype']
caller = variant['Caller']
allele_depth = variant['allele_depth']
filter_status = variant['filter_status']
total_count_forward = variant['TCF']
total_count_reverse = variant['TCR']
vafs = ":".join(str(x) for x in variant['VAFS'])
new_variant_sample = VariantSample(
variant=new_variant,
sample=sample,
genotype=genotype,
caller=caller,
allele_depth=allele_depth,
filter_status=filter_status,
total_count_forward=total_count_forward,
total_count_reverse=total_count_reverse,
vafs=vafs)
new_variant_sample.save()
return None
def test_worst_consequence(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
input1 = master_list[0]['transcript_data']
input2 = {
'NM_002617.3': {
'Consequence': 'transcript_ablation'
},
'NM_002614.3': {
'Consequence': 'splice_acceptor_variant'
}
}
input3 = {
'NM_002617.3': {
'Consequence': 'transcript_ablation'
},
'NM_002614.3': {
'Consequence': 'splice_acceptor_variant&splice_donor_variant'
},
'NM_002684.3': {
'Consequence': 'inframe_deletion'
}
}
input4 = {
'NM_002617.3': {
'Consequence': 'stop_gained'
},
'NM_002614.3': {
'Consequence': 'splice_acceptor_variant'
}
}
input5 = {
'NM_002617.3': {
'Consequence': 'frameshift_variant'
},
'NM_002614.3': {
'Consequence': 'splice_acceptor_variant'
}
}
input6 = {
'NM_002617.3': {
'Consequence':
'intron_variant&missense_variant&non_coding_transcript_variant'
},
'NM_002614.3': {
'Consequence': 'frameshift_variant'
}
}
input7 = {
'NM_002617.3': {
'Consequence': 'transcript_amplification'
},
'NM_002614.3': {
'Consequence': 'inframe_insertion'
}
}
input8 = {
'NM_002617.3': {
'Consequence': 'inframe_insertion'
},
'NM_002614.3': {
'Consequence':
'incomplete_terminal_codon_variant&incomplete_terminal_codon_variant'
}
}
input9 = {
'NM_002617.3': {
'Consequence': 'feature_elongation'
},
'NM_002614.3': {
'Consequence': 'intergenic_variant'
}
}
input10 = {
'NM_002617.3': {
'Consequence': 'feature_truncation'
},
'NM_002614.3': {
'Consequence': 'intergenic_variant'
}
}
input11 = {
'NM_002617.3': {
'Consequence': 'stop_lost'
},
'NM_002614.3': {
'Consequence': 'stop_lost'
}
}
self.assertEqual(vcf_parser.worst_consequence(input1),
'intron_variant')
self.assertEqual(vcf_parser.worst_consequence(input2),
'transcript_ablation')
self.assertEqual(vcf_parser.worst_consequence(input3),
'transcript_ablation')
self.assertEqual(vcf_parser.worst_consequence(input4),
'splice_acceptor_variant')
self.assertEqual(vcf_parser.worst_consequence(input5),
'splice_acceptor_variant')
self.assertEqual(vcf_parser.worst_consequence(input6),
'frameshift_variant')
self.assertEqual(vcf_parser.worst_consequence(input7),
'transcript_amplification')
self.assertEqual(vcf_parser.worst_consequence(input8),
'inframe_insertion')
self.assertEqual(vcf_parser.worst_consequence(input9),
'feature_elongation')
self.assertEqual(vcf_parser.worst_consequence(input10),
'feature_truncation')
self.assertEqual(vcf_parser.worst_consequence(input11), 'stop_lost')
def test_create_master_list(self):
master_list = vcf_parser.create_master_list(
'VariantDatabase/tests/test_files/vcfs/vep_annotated_test_vcf.vcf',
'WS61594_14000835')
self.assertEqual(master_list[0]['reference'], 'C')
self.assertEqual(master_list[0]['genotype'], '0/1')
self.assertEqual(master_list[0]['alt_alleles'], ('A', ))
self.assertEqual(master_list[0]['filter_status'], '.')
self.assertEqual(
master_list[0]['hash_id'],
'd360384c2a1df84a02bc9b2f19ee584ed837d600081450beab17762532ce18ba')
self.assertEqual(master_list[0]['allele_depth'], '124:124')
self.assertEqual(
master_list[0]['transcript_data']['NM_002617.3'], {
'MAX_AF_POPS': 'ExAC_FIN',
'TSL': '',
'APPRIS': '',
'ExAC_AF': '0.392',
'ExAC_NFE_AF': '0.4469',
'AMR_AF': '0.4539',
'SYMBOL': 'PEX10',
'AFR_AF': '0.2859',
'ExAC_EAS_AF': '0.2793',
'Feature': 'NM_002617.3',
'Codons': '',
'MOTIF_NAME': '',
'DOMAINS': '',
'SIFT': '',
'VARIANT_CLASS': 'SNV',
'EA_AF': '0.4278',
'CDS_position': '',
'CCDS': '',
'Allele': 'A',
'PolyPhen': '',
'AA_AF': '0.2942',
'MOTIF_SCORE_CHANGE': '',
'IMPACT': 'MODIFIER',
'HGVSp': '',
'ENSP': 'NP_002608.1',
'MAX_AF': '0.5536',
'INTRON': '4/5',
'ExAC_AFR_AF': '0.3196',
'Existing_variation': 'rs3795269',
'HGVSc': 'NM_002617.3:c.776+33G>T',
'MOTIF_POS': '',
'HIGH_INF_POS': '',
'ExAC_FIN_AF': '0.5536',
'PICK': '',
'GENE_PHENO': '',
'ExAC_SAS_AF': '0.3747',
'UNIPARC': '',
'cDNA_position': '',
'PUBMED': '',
'EAS_AF': '0.253',
'Feature_type': 'Transcript',
'AF': '0.3391',
'ExAC_Adj_AF': '0.4176',
'ExAC_OTH_AF': '0.4538',
'HGNC_ID': '',
'SAS_AF': '0.3272',
'SWISSPROT': '',
'FLAGS': '',
'Consequence': 'intron_variant',
'Protein_position': '',
'Gene': '5192',
'STRAND': '-1',
'EUR_AF': '0.4274',
'DISTANCE': '',
'PHENO': '',
'SYMBOL_SOURCE': '',
'Amino_acids': '',
'ExAC_AMR_AF': '0.43',
'TREMBL': '',
'CLIN_SIG': '',
'REFSEQ_MATCH': '',
'HGVS_OFFSET': '',
'BIOTYPE': 'protein_coding',
'EXON': '',
'SOMATIC': '',
'CANONICAL': ''
})
self.assertEqual(len(master_list), 281)