details

Query FAVOR (GenoHub) API to get variant functional prediction scores.
API: GET https://api.genohub.org/v1/rsids/{rs_id}
Note: Returns a JSON list; use first element.
Args:
    rs_id (str): dbSNP rsID (e.g. "rs7412")
Return:
    Gene name, exonic category, functional prediction scores
    (CADD, SIFT, PolyPhen2, MutationTaster, GERP, etc.)

import requests

rs_id = "rs7412"
url = f"https://api.genohub.org/v1/rsids/{rs_id}"
resp = requests.get(url, timeout=30).json()
d = resp[0] if isinstance(resp, list) else resp

print(f"[FAVOR] variant: {d.get('variant_vcf')}")
print(f"[FAVOR] 基因: {d.get('genecode_comprehensive_info')}")
print(f"[FAVOR] 区域: {d.get('genecode_comprehensive_category')}")
print(f"[FAVOR] 外显子变异类别: {d.get('genecode_comprehensive_exonic_category')}")
print(f"[FAVOR] CADD phred: {d.get('cadd_phred')} (>20=有害)")
print(f"[FAVOR] SIFT: {d.get('sift_cat')} (val={d.get('sift_val')})")
print(f"[FAVOR] PolyPhen2 HDIV: {d.get('polyphen2_hdiv_score')}")
print(f"[FAVOR] PolyPhen2 HVAR: {d.get('polyphen2_hvar_score')}")
print(f"[FAVOR] MutationTaster: {d.get('mutation_taster_score')}")
print(f"[FAVOR] MutationAssessor: {d.get('mutation_assessor_score')}")
print(f"[FAVOR] MetaSVM pred: {d.get('metasvm_pred')}")
print(f"[FAVOR] GERP_N: {d.get('gerp_n')}, GERP_S: {d.get('gerp_s')}")
print(f"[FAVOR] BRAVO AF: {d.get('bravo_af')}")

technical

github

SpectrAI-Initiative/InnoClaw

stars

374

license

Apache-2.0

contributors

16

last commit

2026-04-20T01:27:21Z

file

.claude/skills/variant-functional-prediction/SKILL.md

related

• InnoClaw/variant-population-frequency— Query gnomAD for variant allele frequency across populations. Uses FAVOR to convert rsID→variant_id first, then queries gnomAD.
• InnoClaw/comprehensive-variant-annotation— Given an rsID, query multiple databases (dbSNP, FAVOR, GWAS Catalog, ClinVar, gnomAD, PharmGKB, ClinGen) for comprehensive annotation. Use when user asks a general question about a variant without specifying which aspect.
• InnoClaw/protein_function_annotation— Protein Function Annotation Pipeline - Annotate protein function: UniProt metadata, InterPro domains, functional prediction, and GO enrichment. Use this skill for proteomics tasks involving query uniprot query interpro predict protein function get functional enrichment. Combines 4 tools from 2 SCP server(s).
• InnoClaw/go_term_analysis— Gene Ontology Analysis - Analyze GO terms: ChEMBL GO slim, STRING functional enrichment, STRING annotation, and Ensembl ontology. Use this skill for functional genomics tasks involving get go slim by id get functional enrichment get functional annotation get ontology name. Combines 4 tools from 3 SCP server(s).
• InnoClaw/gene_therapy_target— Gene Therapy Target Analysis - Analyze gene therapy target: gene info, variant pathogenicity, protein structure, and clinical evidence. Use this skill for gene therapy tasks involving get gene metadata by gene name get vep hgvs Protein structure prediction ESMFold clinvar search. Combines 4 tools from 4 SCP server(s).