VariantMatcher : Phenotypic and Genomic Data Sharing to Facilitate Variant Classification and Disease Gene Discovery

VariantMatcher is a web-based platform developed to share variant-level genomic data and phenotypic information. Here, we describe the experience of a clinical laboratory in Brazil, DASA, utilizing VariantMatcher to enhance variant classification in a clinical setting. We retrospectively analyzed 3025 molecular genetic test results to identify variants classified as variants of uncertain significance (VUS). Of 2302 reported variants, 1679 were classified as VUS. Next, we selected the 542 VUS not described in gnomAD to investigate their presence in VariantMatcher. Of these 542, 80 variants were present in VariantMatcher. Sixty-three of them were identified in individuals with a phenotypic overlap. Further investigation of the phenotypic features of individuals with the same variants led to the reclassification of 20 (32%) of the 63 variants. Notably, most of the reclassifications helped rule out the queried variant as causative for the phenotype being investigated, particularly for autosomal or X-linked dominant diseases with early onset and complete penetrance. Our experience supports the benefits of data sharing in the clinical setting, which can improve variant classification accuracy and provide more precise guidance for informed clinical decision-making.