When to Use

• User wants to check if variants in ENCODE regulatory peaks have clinical significance in ClinVar
• User asks about "ClinVar", "pathogenic variants", "clinical significance", or "variant classification"
• User needs to annotate ENCODE-derived regulatory variants with disease associations
• User wants to find clinically relevant variants within enhancers, promoters, or open chromatin regions
• Example queries: "check ClinVar for variants in my ATAC-seq peaks", "find pathogenic variants in pancreas enhancers", "annotate regulatory variants with clinical significance"

Annotating ENCODE Regulatory Variants with ClinVar Clinical Significance

Cross-reference ENCODE functional genomic elements with ClinVar clinical variant classifications to identify pathogenic variants in regulatory regions and understand non-coding disease mechanisms.

Scientific Rationale

The question: "Do any clinically significant variants fall within my ENCODE regulatory elements, and can ENCODE data explain their pathogenic mechanism?"

ClinVar is NCBI's public archive of variant-disease associations, aggregating submissions from clinical laboratories, research groups, and expert panels. Most ClinVar annotations focus on coding variants, but a growing number of non-coding variants are being classified. ENCODE provides the functional context to explain WHY a non-coding variant is pathogenic — by showing that it disrupts an active enhancer, promoter, or insulator in disease-relevant tissue.

This bidirectional integration serves two use cases:

1. Forward: Start from ENCODE peaks, find clinically significant variants within them
2. Reverse: Start from ClinVar pathogenic variants, use ENCODE to explain their mechanism

The Non-Coding Variant Challenge

• ~90% of GWAS-associated variants are in non-coding regions (Maurano et al. 2012)
• ClinVar increasingly includes non-coding variants, but most lack mechanistic annotation
• ENCODE regulatory annotations provide the "why" behind non-coding pathogenicity
• A variant classified as VUS (variant of uncertain significance) may be reclassified with ENCODE functional evidence

Key Literature

• Landrum et al. 2018 "ClinVar: improving access to variant interpretations and supporting evidence" (Nucleic Acids Research, ~2,000 citations). Describes the ClinVar database architecture, submission standards, and the star-rating review system for variant classifications. DOI: 10.1093/nar/gkx1153
• Riggs et al. 2020 "Technical standards for the interpretation and reporting of constitutional copy-number variants: a joint consensus recommendation of the ACMG and ClinGen" (Genetics in Medicine, ~500 citations). Framework for interpreting structural variants, relevant when ENCODE elements overlap CNVs. DOI: 10.1038/s41436-019-0686-8
• Richards et al. 2015 "Standards and guidelines for the interpretation of sequence variants: ACMG/AMP joint consensus recommendation" (Genetics in Medicine, ~12,000 citations). The ACMG variant classification framework (pathogenic through benign). ENCODE functional data can provide evidence for PS3/BS3 (functional studies) criteria. DOI: 10.1038/gim.2015.30
• ENCODE Project Consortium 2020 (Nature, ~1,656 citations). Registry of 926,535 human cCREs — the functional annotation layer for interpreting non-coding ClinVar variants. DOI: 10.1038/s41586-020-2493-4

ClinVar Clinical Significance Categories

ClinVar Star Ratings

Always check star ratings. A 0-star "pathogenic" classification has very different reliability than a 3-star classification.

NCBI E-utilities API Reference

Base URL: https://eutils.ncbi.nlm.nih.gov/entrez/eutils/

No authentication required for low-volume use. Rate limit: 3 requests/second without API key, 10/second with NCBI API key.

Key Endpoints

ClinVar VCF Downloads

For bulk intersection with ENCODE peaks, download the ClinVar VCF:

• GRCh38: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh38/clinvar.vcf.gz
• GRCh37: https://ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh37/clinvar.vcf.gz

Updated monthly on the first Thursday.

Step 1: Define the Scope

Determine which direction the analysis runs:

Forward: ENCODE Peaks to ClinVar Variants

Starting from ENCODE regulatory elements, find clinically significant variants within them.

# Get ENCODE peaks for target tissue
encode_search_experiments(
    assay_title="Histone ChIP-seq",
    target="H3K27ac",
    organ="pancreas",
    biosample_type="tissue"
)

encode_list_files(
    experiment_accession="ENCSR...",
    file_format="bed",
    output_type="IDR thresholded peaks",
    assembly="GRCh38"
)

Reverse: ClinVar Variants to ENCODE Context

Starting from ClinVar pathogenic variants, determine if they overlap ENCODE regulatory elements.

import requests

# Search ClinVar for pathogenic variants in a gene
url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi"
params = {
    "db": "clinvar",
    "term": "INS[gene] AND pathogenic[clinical significance]",
    "retmax": 50,
    "retmode": "json"
}
response = requests.get(url, params=params)
result = response.json()
variant_ids = result["esearchresult"]["idlist"]

Step 2: Query ClinVar via E-utilities

Search for Variants by Gene

import requests
import time

def search_clinvar(gene_symbol, significance="pathogenic"):
    """Search ClinVar for variants in a gene with given clinical significance."""
    url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi"
    term = f"{gene_symbol}[gene] AND {significance}[clinical significance]"
    params = {
        "db": "clinvar",
        "term": term,
        "retmax": 100,
        "retmode": "json"
    }
    response = requests.get(url, params=params)
    time.sleep(0.34)  # Rate limit: 3/sec
    return response.json()["esearchresult"]["idlist"]

Get Variant Details

def get_clinvar_summary(variant_ids):
    """Get summary for ClinVar variant IDs."""
    url = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi"
    params = {
        "db": "clinvar",
        "id": ",".join(variant_ids[:20]),  # Max 20 per request
        "retmode": "json"
    }
    response = requests.get(url, params=params)
    time.sleep(0.34)
    return response.json()["result"]

Search by Genomic Region

# Se