Prioritize and classify variants for clinical significance using ClinVar annotations, population frequencies, and ACMG classification guidelines.
# InterVar for ACMG classification
pip install intervar
# ClinVar database
wget ftp://ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh38/clinvar.vcf.gzAnnotated VCF
├── ClinVar lookup (known clinical assertions)
├── Population frequency (gnomAD)
├── Computational predictions (SIFT, PolyPhen, CADD)
├── ACMG classification
└── Prioritized variant list
bcftools annotate \
-a clinvar.vcf.gz \
-c INFO/CLNSIG,INFO/CLNDN \
input.vcf.gz -Oz -o with_clinvar.vcf.gzfrom cyvcf2 import VCF
for v in VCF('annotated.vcf.gz'):
clnsig = v.INFO.get('CLNSIG', '')
if 'Pathogenic' in clnsig or 'Likely_pathogenic' in clnsig:
print(f'{v.CHROM}:{v.POS} {v.REF}>{v.ALT[0]} - {clnsig}')The 5-tier system for variant pathogenicity:
| Class | Meaning | Action |
|---|---|---|
| Pathogenic | Disease-causing | Report, clinical action |
| Likely pathogenic | Probably disease-causing | Report, consider action |
| VUS | Uncertain significance | Report with caution |
| Likely benign | Probably not disease-causing | May report |
| Benign | Not disease-causing | Usually not reported |
def prioritize_variant(v):
score = 0
# ClinVar pathogenic
if 'Pathogenic' in v.INFO.get('CLNSIG', ''):
score += 10
# Rare in gnomAD
af = v.INFO.get('gnomAD_AF', 1.0)
if af < 0.001:
score += 5
# High CADD score
cadd = v.INFO.get('CADD_PHRED', 0)
if cadd > 20:
score += 3
# Protein-altering
if 'missense' in v.INFO.get('Consequence', ''):
score += 2
return score| Database | Purpose |
|---|---|
| ClinVar | Clinical variant assertions |
| gnomAD | Population allele frequencies |
| OMIM | Disease-gene associations |
| CADD | Deleteriousness scores |
- Always include variant evidence and classification rationale
- VUS variants may be reclassified with new evidence
- Consider family segregation data when available
- Follow laboratory/institutional reporting guidelines
"Annotate my VCF with ClinVar clinical significance"
"Filter for pathogenic and likely pathogenic variants"
"Prioritize variants by clinical relevance"
"Apply ACMG classification criteria to my variants"
- Always include variant evidence and classification rationale
- VUS variants may be reclassified with new evidence
- Consider family segregation data when available
- Follow laboratory/institutional reporting guidelines