Likelihood ratios to assess genetic evidence for clinical significance of uncertain variants: hereditary hemorrhagic telangiectasia as a model

Abstract

Clinical laboratories performing gene sequencing discover previously unreported and/or uncharacterized variants. Often these are missense or intronic mutations in which the contribution to disease cannot be predicted, and consequently these mutations are reported as variants of uncertain significance. Follow-up to assess family concordance is recommended by the American College of Medical Genetics to provide genetic evidence for clinical significance. Although family concordance studies show whether a variant segregates with disease in the family, the strength of evidence varies depending on the number and degree of relatedness of family members available for testing. We investigated a statistical model which accounts for the pedigree, inheritance patterns, and penetrance to determine the likelihood of a variant being a causative or deleterious mutation. We used hereditary hemorrhagic telangiectasia (HHT) as a model for an autosomal dominant disease. Pedigree data were transferred to MLINK, and a Bayesian analysis was calculated to determine the likelihood that a variant is causative of disease. In applying this analysis to HHT pedigrees we found Bayes Factors of variants showing odds in favor of causality ranging from approximately 4:1 to over 400:1. These numbers provide an objective measure of the strength of genetic evidence. Other parameters such as amino acid severity predictions, ortholog and paralog comparisons and functional assays can be included in the analysis to increase the evidence of causality.