Overview

Variation databases systematically catalog the genetic differences that distinguish individuals and drive phenotypic diversity, including disease susceptibility. As sequencing technologies have matured, the number of known human genetic variants has grown exponentially, reaching hundreds of millions of entries. Variation databases serve two critical functions: they define the baseline spectrum of normal population variation, against which disease-associated variants can be identified, and they aggregate clinical interpretations that guide diagnostic and therapeutic decisions.

Key Concepts

dbSNP (Database of Single Nucleotide Polymorphisms) catalogs single nucleotide variants, small insertions and deletions, and microsatellite repeats. Each variant receives an rs (reference SNP) identifier. dbSNP records include allele frequency data from large population studies such as the 1000 Genomes Project and gnomAD. ClinVar focuses on the relationship between genetic variants and human health, providing clinical significance classifications (pathogenic, benign, uncertain significance) with supporting evidence. COSMIC (Catalogue of Somatic Mutations in Cancer) catalogs somatic mutations found in human cancers, including point mutations, copy number alterations, and gene fusions.

Applications

Variation databases are essential tools in genomic medicine. DNA sequencing projects use dbSNP to filter common polymorphisms from candidate pathogenic variants. Next-generation sequencing pipelines annotate variants against ClinVar to prioritize clinically actionable findings. Cancer biochemistry research relies on COSMIC to identify driver mutations, while DNA repair mechanism studies connect mutation signatures with specific repair pathway deficiencies.

Practical Protocol

To query ClinVar for a disease-associated variant, go to ncbi.nlm.nih.gov/clinvar and search by gene name, variant (e.g., “BRCA1 VUS”), or phenotype. Each record displays the clinical significance classification (Pathogenic, Likely Pathogenic, Uncertain Significance, Likely Benign, Benign) along with the review status and supporting evidence count. For a missense variant interpretation workflow: start with the variant’s genomic coordinates from your sequencing pipeline. Use the dbSNP rsID to pull population frequency data from the “Freq” table, variants with allele frequency >1% in gnomAD are unlikely to be pathogenic for rare diseases. Click through to ClinVar to assess whether the variant has been clinically reported. If the variant is novel (no rsID), use tools like SIFT or PolyPhen-2 (linked from dbSNP’s “Molecular Consequence” section) to predict the functional impact of the amino acid substitution. For a concrete example, consider a patient exome revealing a missense variant in MYH7 (p.Arg403Trp). Querying dbSNP shows rs121913652 with a rare frequency (0.002% in gnomAD). ClinVar classifies this variant as Pathogenic for familial hypertrophic cardiomyopathy, supported by functional studies demonstrating disrupted actin binding. COSMIC confirms this residue is not recurrently mutated in cancer, narrowing the differential diagnosis. This stepwise workflow, frequency filtering, clinical annotation, and functional prediction, converts raw variant calls into clinically actionable information. As clinical sequencing becomes routine, automated pipelines increasingly integrate these databases to generate real-time variant interpretation reports for physicians.