What is ClinVar?

ClinVar was launched in 2013 to address a critical gap in clinical genetics: the lack of a centralized, publicly accessible repository of variant interpretations. Before ClinVar, genetic testing laboratories maintained proprietary databases, leading to inconsistent variant classifications across labs. ClinVar aggregates submissions from hundreds of sources, including major clinical labs like GeneDx, Invitae, and Ambry Genetics, as well as expert curation panels like ClinGen. As of 2024, the database contains over 2.5 million submitted variant interpretations covering more than 1.2 million unique variants.

The five-tier classification system follows guidelines established by the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) in 2015. A 'pathogenic' classification means there is strong evidence that the variant causes disease. 'Likely pathogenic' indicates greater than 90% certainty. 'Variant of uncertain significance' (VUS) means the evidence is insufficient or conflicting — this is the most challenging category for clinicians, as it provides no actionable guidance. The 'likely benign' and 'benign' classifications indicate the variant is not expected to cause disease. Conflicts between submitters are flagged, and star ratings (0-4 stars) indicate the level of review.

PepFold queries ClinVar as the first step in its pharmacogenomic pipeline. When a user submits an rsID, the system retrieves the variant's clinical significance, associated conditions, review status, and molecular consequence from ClinVar's API. This annotation determines how the variant is handled downstream: pathogenic variants in drug-metabolizing enzymes trigger different peptide design strategies than risk-factor variants in disease-associated proteins. The ClinVar data feeds directly into PepFold's variant annotation layer, ensuring every peptide candidate is designed with full clinical context.