Variant Interpretation
Overview
Variant interpretation is the process of determining whether a genetic variant identified by laboratory testing is clinically significant. This chapter covers the ACMG/AMP variant classification framework and the genomic databases used to support interpretation. As genetic testing becomes broader in scope, from single-gene tests to exomes and genomes, the ability to interpret variants accurately has become one of the most critical skills in clinical genetics.
The ACMG/AMP framework classifies sequence variants into five categories: pathogenic, likely pathogenic, variant of uncertain significance (VUS), likely benign, and benign. The evidence types that support each classification (population frequency, computational predictions, functional data, segregation, and de novo status) underpin both interpreting laboratory reports and explaining a VUS result to a patient.
Genomic databases are the practical tools that support variant interpretation. Knowing which database to consult for population frequency (gnomAD), disease-variant associations (ClinVar), or gene-disease validity (ClinGen) is a daily clinical skill. These resources are used routinely in practice, and fluency with their content, strengths, and limitations is part of competent variant interpretation.
Key Concepts
- Five-tier classification: pathogenic, likely pathogenic, VUS, likely benign, benign and what each means clinically
- Evidence categories: population data, computational/predictive data, functional data, segregation data, de novo data, allelic data
- VUS counseling: how to communicate uncertainty without causing unnecessary anxiety or false reassurance
- Reclassification: why variants may be reclassified over time and the implications for patients and families
- Database literacy: knowing the purpose and limitations of ClinVar, gnomAD, ClinGen, OMIM, and other key resources
Classification Criteria
ACMG variant classification is the 2015 ACMG/AMP framework for interpreting sequence variants. The specific evidence criteria (PVS1, PS1-PS4, PM1-PM6, PP1-PP5, and their benign counterparts) combine into a final five-tier classification. Applying them well requires understanding their nuances: incomplete penetrance, populations underrepresented in reference databases, and the distinction between somatic and germline testing all change how the rules are weighted.
Interpretation Resources
Genomic databases are the practical tools that feed evidence into the ACMG framework. The major ones cover distinct needs: population frequency (gnomAD), variant-disease associations (ClinVar), gene-disease validity curation (ClinGen), and phenotype (OMIM, HPO). What each database contains, how its data is curated, and where it falls short are all relevant to critically reading a laboratory report and to independent variant assessment.
Classification criteria define the rules; variant types define what is being classified; databases provide the evidence. The three layers fit together.
Variant Types
The leaves in this section walk through the variant types you will see on real reports and how each is described, evaluated, and counseled. Sequence variants (SNV, indel, duplication) covers the small-scale changes that ACMG/AMP rules were primarily written for. Structural variants and segregation addresses larger rearrangements (deletions, duplications, inversions, translocations) and how family-segregation evidence supports their classification. Somatic tumor variants (MSI, MRD) covers somatic interpretation, which uses a parallel framework (AMP/ASCO/CAP) tuned to therapy selection and disease monitoring rather than germline disease causation.