def test_extract_xrefs():
soup = make_soup("""
<Allele>
<XRefList>
<XRef DB="UniProtKB" ID="Uniprot-1" />
<XRef DB="OMIM" ID="Omim-1" />
<XRef DB="dbSNP" ID="123" Type="rs" />
</XRefList>
</Allele>
""")
result = ClinvarVariationAnnotator._extract_xrefs(soup)
assert result['uniprot_id'] == 'Uniprot-1'
assert result['omim_id'] == 'Omim-1'
assert result['dbsnp_id'] == 'rs123'
soup = make_soup("""
<Allele>
<XRefList>
</XRefList>
</Allele>
""")
# Check it does not break when info is missing
ClinvarVariationAnnotator._extract_xrefs(soup)
def test_extract_single_gene_name():
soup = make_soup("""
<VariationReport>
<GeneList GeneCount="1">
<Gene Symbol="Gene-Symbol-1"></Gene>
<GeneList/>
</VariationReport>
""")
result = ClinvarVariationAnnotator._extract_single_gene_name(soup)
assert result == 'Gene-Symbol-1'
soup = make_soup("""
<VariationReport>
<GeneList GeneCount="2">
<Gene Symbol="Gene-Symbol-1"></Gene>
<Gene Symbol="Gene-Symbol-2"></Gene>
<GeneList/>
</VariationReport>
""")
result = ClinvarVariationAnnotator._extract_single_gene_name(soup)
assert result is None
soup = make_soup("""<VariationReport></VariationReport>""")
result = ClinvarVariationAnnotator._extract_single_gene_name(soup)
assert result is None
def clinvar_variants_for_genes(gene_list, cache, as_dataframe=False):
"""Given a list of gene symbols (e.g. AIP, BRCA2), return a list of
variants from a search in Clinvar with those gene symbols.
Requires a *cache* argument for the ClinvarVariationAnnotator annotator.
Options are: 'myqsl', 'postgres', 'dict', 'redis'.
Returns a tuple of (variant_ids, annotations).
If *as_dataframe* is set to True, it returns a pandas DataFrame with
the annotations.
"""
annotator = ClinvarVariationAnnotator(cache=cache)
variant_ids = []
variants = {}
for gene_variant_ids in clinvar_variant_ids_for_genes(gene_list):
variants_for_this_gene = annotator.annotate(gene_variant_ids)
variant_ids += gene_variant_ids
variants.update(variants_for_this_gene)
if as_dataframe:
df = pd.DataFrame(list(variants.values()))
columns = [col for col in COLUMN_ORDER if col in df.columns]
return df[columns]
else:
return (variant_ids, variants)
def annotate_items_with_clinvar(items, items_type, clinvar_vcf_path,
grouped_by_unique_item=False, cache='mysql',
proxies={}, use_web=True, use_cache=True):
"""
Given a list of *items* like rs IDs and the *items_type*, "rs_id",
return the Clinvar Variation Reports associated with them.
"""
items = list(unique_everseen(items))
# Associate each item (an RS ID, a genomic location, etc.) to a list of
# one or many ClinVar Variation IDs
variation_ids_per_item = _get_clinvar_variation_ids_from(
items, items_type, clinvar_vcf_path=clinvar_vcf_path
)
variation_ids = list(collapse(list(variation_ids_per_item.values())))
# Use the ClinVar Variation IDs to get Variation Reports
annotator_options = dict(cache=cache, proxies=proxies)
annotate_options = dict(use_web=use_web, use_cache=use_cache)
annotator = ClinvarVariationAnnotator(**annotator_options)
clinvar_variations = annotator.annotate(variation_ids, **annotate_options)
# Associate the original items with the obtained Variation Reports
if grouped_by_unique_item:
result = defaultdict(list)
for item, item_variation_ids in variation_ids_per_item.items():
result[item] = [clinvar_variations[variation_id]
for variation_id in item_variation_ids
if variation_id in clinvar_variations]
else:
result = list(clinvar_variations.values())
return result
def test_extract_allele_hgvs():
soup = make_soup("""
<Allele>
<HGVSlist>
<HGVS Assembly="GRCh37"
Change="genomic-change-37"
AccessionVersion="Accession-1">Name-37</HGVS>
<HGVS Assembly="GRCh38"
Change="genomic-change-38"
AccessionVersion="Accession-1">Name-38</HGVS>
<HGVS Type="HGVS, coding, RefSeq">accession:coding-change-1</HGVS>
<HGVS Type="HGVS, coding, RefSeq">accession:coding-change-2</HGVS>
<HGVS Type="HGVS, protein, RefSeq">accession:protein-change-1</HGVS>
</HGVSlist>
</Allele>
""")
result = ClinvarVariationAnnotator._extract_allele_hgvs(soup)
assert result['genomic_change_g37'] == 'genomic-change-37'
assert result['genomic_change_g37_accession'] == 'Accession-1'
assert result['genomic_change_g37_name'] == 'Name-37'
assert result['genomic_change_g38'] == 'genomic-change-38'
assert result['genomic_change_g38_accession'] == 'Accession-1'
assert result['genomic_change_g38_name'] == 'Name-38'
assert result['coding_changes'] == ['accession:coding-change-1',
'accession:coding-change-2']
assert result['protein_changes'] == ['accession:protein-change-1']
soup = make_soup("<Allele></Allele>")
# Test it does not break when data is missing
ClinvarVariationAnnotator._extract_allele_hgvs(soup)
def search_clinvar_variants_for_phenotypes(
pheno_terms,
clinsigs=['pathogenic', 'likely pathogenic'],
cache='dict',
):
"""
Queries ClinVar API for all the variants associated with any of the
phenotypes passed as *pheno_terms*. Optionally, pass a list of *clinsigs*
to filter the results. You may also pass a *cache* for
ClinvarVariationAnnotator to use.
Returns a list with those variants. Each variant is a dictionary with lots
of data.
Usage:
```
from anotala.recipes import search_clinvar_variants_for_phenotype
phenos = ['Diabetes mellitus', 'Insulin']
variants = search_clinvar_variants_for_phenotype(phenos)
```
"""
variation_ids = _search_clinvar(pheno_terms, clinsigs)
annotator = ClinvarVariationAnnotator(cache)
annotations = annotator.annotate(variation_ids)
return annotations
def test_extract_variation_type():
soup = make_soup('<VariationReport VariationType="Var-1-Type"></VariationReport>')
result = ClinvarVariationAnnotator._extract_variation_type(soup)
assert result == 'Var-1-Type'
soup = make_soup('<VariationReport></VariationReport>')
# Just test it does not break when VariantType is missing
assert ClinvarVariationAnnotator._extract_variation_type(soup) is None
def test_extract_dbsnp_ids_from_alleles_for_haplotypes():
info = {
'variation_type': 'Haplotype',
'alleles': [
{'dbsnp_id': 'rs1'},
{'dbsnp_id': 'rs2'},
{},
]
}
ClinvarVariationAnnotator._extract_dbsnp_ids_from_alleles_for_haplotypes(info)
assert info['dbsnp_ids'] == ['rs1', 'rs2']
def test_extract_sequence_info_from_allele():
soup = make_soup("""
<Allele>
<SequenceLocation Assembly="GRCh37"
Chr="X"
Accession="NC_1"
start="1000"
stop="1000"
variantLength="1"
referenceAllele="A"
alternateAllele="G"/>
<SequenceLocation Assembly="GRCh38"
Chr="X"
Accession="NC_2"
start="2000"
stop="2000"
variantLength="1"
referenceAllele="A"
alternateAllele="G"/>
</Allele>
""")
result = ClinvarVariationAnnotator._extract_sequence_info_from_allele(soup)
assert result['start_g37'] == 1000
assert result['stop_g37'] == 1000
assert result['chrom_g37'] == 'X'
assert result['length_g37'] == 1
assert result['ref_g37'] == 'A'
assert result['alt_g37'] == 'G'
assert result['accession_g37'] == 'NC_1'
assert result['start_g38'] == 2000
assert result['stop_g38'] == 2000
assert result['chrom_g38'] == 'X'
assert result['length_g38'] == 1
assert result['ref_g38'] == 'A'
assert result['alt_g38'] == 'G'
assert result['accession_g38'] == 'NC_2'
assert result['genomic_allele'] == 'G'
soup = make_soup("""
<Allele>
<SequenceLocation Assembly="GRCh37"
innerStart="1000"
innerStop="1000" />
</Allele>
""")
result = ClinvarVariationAnnotator._extract_sequence_info_from_allele(soup)
assert result['start_g37'] == 1000
assert result['stop_g37'] == 1000
def test_extract_obervation():
soup = make_soup("""
<VariationReport VariationID="1">
<ObservationList>
<Observation VariationID="1" ObservationType="primary">
<ClinicalSignificance DateLastEvaluated="2001-01-01">
<Description>Pathogenic</Description>
</ClinicalSignificance>
<ReviewStatus>criteria provided</ReviewStatus>
<PhenotypeList>
<Phenotype Name="Pheno-1"></Phenotype>
</PhenotypeList>
</Observation>
<Observation VariationID="2">
<ClinicalSignificance>
<Description>Benign</Description>
</ClinicalSignificance>
</Observation>
</ObservationList>
</VariationReport>
""")
result = ClinvarVariationAnnotator._extract_observation(soup)
assert result == {
'variation_id': '1',
'clinical_significances': ['Pathogenic'],
'date_last_evaluated': '2001-01-01',
'type': 'primary',
'review_status': 'criteria provided',
'phenotypes': [{'name': 'Pheno-1'}],
}
def test_parse_citation():
soup = make_soup("""
<Citation Type="Citation-Type">
<ID Source="Source-1">123</ID>
</Citation>
""")
result = ClinvarVariationAnnotator._parse_citation(soup)
assert result == {'type': 'Citation-Type', 'Source-1': '123'}
soup = make_soup("""
<Citation Type="Citation-Type">
<ID Source="PubMed">123</ID>
</Citation>
""")
result = ClinvarVariationAnnotator._parse_citation(soup)
assert result == {'type': 'Citation-Type', 'pmid': '123'}
def test_extract_genes():
soup = make_soup("""
<VariationReport>
<GeneList GeneCount="1">
<Gene GeneID="Gene-1"
Symbol="Gene-1-Symbol"
FullName="Gene 1 full name"
strand="-"
HGNCID="HGNC:123">
<OMIM>234</OMIM>
</Gene>
</GeneList>
</VariationReport>
""")
result = ClinvarVariationAnnotator._extract_genes(soup)
assert result == [
{
'entrez_id': 'Gene-1',
'symbol': 'Gene-1-Symbol',
'full_name': 'Gene 1 full name',
'hgnc_id': '123',
'strand': '-',
'omim_id': '234',
}
]
def test_associated_phenotypes():
clinical_assertions = [
{'phenotypes': [{'name': 'Pheno-1'}, {'name': 'Pheno-2'}]},
{'phenotypes': [{'name': 'Pheno-2'}, {'name': 'Pheno-3'}]},
{} # Has no phenotypes key, shouldn't break
]
result = ClinvarVariationAnnotator._associated_phenotypes(clinical_assertions)
assert result == ['Pheno-1', 'Pheno-2', 'Pheno-3']
def test_generate_clinical_summary():
assertions = [
{'clinical_significances': ['ClinSig-1']},
{'clinical_significances': ['ClinSig-1', 'ClinSig-2', 'ClinSig-3']},
{'clinical_significances': ['ClinSig-2']},
]
result = ClinvarVariationAnnotator._generate_clinical_summary(assertions)
assert result == {'ClinSig-1': 2, 'ClinSig-2': 2, 'ClinSig-3': 1}
def test_extract_allele_frequencies():
soup = make_soup("""
<Allele>
<AlleleFrequency Value="0.01" Type="Type-1" MinorAllele="A" />
<AlleleFrequency Value="0.02" Type="Type-2" MinorAllele="A" />
<AlleleFrequency Value="0.03" Type="Type-3" />
</Allele>
""")
result = ClinvarVariationAnnotator._extract_allele_frequencies(soup)
assert result == {'A': {'Type-1': 0.01, 'Type-2': 0.02}}
def test_parse_comment():
soup = make_soup("""
<Comment DataSource="Comment-Source-1" Type="Comment-type">
Some comment.
</Comment>
""")
result = ClinvarVariationAnnotator._parse_comment(soup)
assert result == {
'data_source': 'Comment-Source-1',
'type': 'Comment-type',
'text': 'Some comment.'
}
def test_extract_allele_basic_info():
soup = make_soup("""
<Allele AlleleID="Allele-1">
<Name>Allele-Name</Name>
<VariantType>Var-Type</VariantType>
</Allele>
""")
result = ClinvarVariationAnnotator._extract_allele_basic_info(soup)
assert result['name'] == 'Allele-Name'
assert result['allele_id'] == 'Allele-1'
assert result['variant_type'] == 'Var-Type'
def test_extract_molecular_consequences():
soup = make_soup("""
<Allele>
<MolecularConsequenceList>
<MolecularConsequence HGVS="change-1"
Function="type-1" />
</MolecularConsequenceList>
</Allele>
""")
results = ClinvarVariationAnnotator._extract_molecular_consequences(soup)
result = results[0]
assert result['hgvs'] == 'change-1'
assert result['function'] == 'type-1'
def test_annotations_by_id():
variations_xml = """
<ClinvarResult-Set>
<VariationReport VariationID="Var-1"></VariationReport>
<VariationReport VariationID="Var-2"></VariationReport>
</ClinvarResult-Set>
"""
result = ClinvarVariationAnnotator._annotations_by_id(
ids=['Var-1', 'Var-2'],
xml=variations_xml,
)
result = dict(result)
assert len(result) == 2
assert result['Var-1'].startswith('<VariationReport VariationID="Var-1"')
assert result['Var-2'].startswith('<VariationReport VariationID="Var-2"')
def test_parse_phenotype_list(monkeypatch):
phenotype_list = make_soup("""
<PhenotypeList>
<Phenotype Name="Pheno-1"></Phenotype>
<Phenotype Name="Pheno-2">
<XRefList>
<XRef ID="MIM-2" DB="OMIM" />
</XRefList>
</Phenotype>
</PhenotypeList>
""")
result = ClinvarVariationAnnotator._parse_phenotype_list(phenotype_list)
assert len(result) == 2
assert result[0] == {'name': 'Pheno-1'}
assert result[1] == {'name': 'Pheno-2', 'omim_id': 'MIM-2',
'incidental': False}
def test_extract_clinical_assertions():
soup = make_soup("""
<VariationReport>
<ClinicalAssertionList>
<GermlineList>
<Germline SubmitterName="Submitter-1"
DateLastSubmitted="2000-01-01">
<PhenotypeList>
<Phenotype Name="Pheno-1">
<XRefList>
<XRef ID="MIM-1" DB="OMIM" />
</XRefList>
</Phenotype>
</PhenotypeList>
<ClinicalSignificance>
<Description>ClinSig-1</Description>
<Citation Type="general">
<ID Source="PubMed">123</ID>
</Citation
<Citation Type="general">
<ID Source="PubMed">234</ID>
</Citation
<Comment DataSource="Comment-Source-1" Type="Comment-type">
Some comment about the variant.
</Comment>
<Method>Method-1</Method>
</ClinicalSignificance>
</Germline>
</GermlineList>
<SomaticList>
<Somatic SubmitterName="Submitter-2"
DateLastSubmitted="2000-01-02">
<PhenotypeList>
<Phenotype Name="Pheno-2">
<XRefList>
<XRef Type="MIM" ID="MIM-2" DB="OMIM"/>
</XRefList>
</Phenotype>
</PhenotypeList>
<ClinicalSignificance>
<Description>ClinSig-2</Description>
<Method>Method-2</Method>
</ClinicalSignificance>
</Somatic>
</SomaticList>
</ClinicalAssertionList>
</VariationReport>
""")
results = ClinvarVariationAnnotator._extract_clinical_assertions(soup)
assert len(results) == 2
germline = results[0]
assert germline == {
'clinical_significances': ['ClinSig-1'],
'clinical_significance_detail': {
'description': 'ClinSig-1',
'citations': [{'type': 'general', 'pmid': '123'},
{'type': 'general', 'pmid': '234'}],
'method': 'Method-1',
'comments': [{'data_source': 'Comment-Source-1',
'type': 'Comment-type',
'text': 'Some comment about the variant.'}]
},
'date_last_submitted': '2000-01-01',
'phenotypes': [{'name': 'Pheno-1', 'omim_id': 'MIM-1',
'incidental': False}],
'submitter_name': 'Submitter-1',
'type': 'germline',
}
somatic = results[1]
assert somatic == {
'clinical_significances': ['ClinSig-2'],
'clinical_significance_detail': {
'description': 'ClinSig-2',
'citations': [],
'comments': [],
'method': 'Method-2',
},
'date_last_submitted': '2000-01-02',
'phenotypes': [{'name': 'Pheno-2', 'omim_id': 'MIM-2',
'incidental': False}],
'submitter_name': 'Submitter-2',
'type': 'somatic',
}
soup = make_soup('<VariationReport><GermlineList><Germline></Germline>'
'</GermlineList></VariationReport>')
# Make sure it does not break on missing data
ClinvarVariationAnnotator._extract_clinical_assertions(soup)
def test_extract_variation_id():
soup = make_soup('<VariationReport VariationID="Var-1"></VariationReport>')
result = ClinvarVariationAnnotator._extract_variation_id(soup)
assert result == 'Var-1'
def test_extract_variation_name():
soup = make_soup('<VariationReport VariationName="Var-1-Name"></VariationReport>')
result = ClinvarVariationAnnotator._extract_variation_name(soup)
assert result == 'Var-1-Name'