Abstract
BACKGROUND: Multiple different evidence types as well as gene-specific variant classification guidelines need to be considered during the classification of variants, making the process complex. Therefore, tools that support variant classification by experts are urgently needed. METHODS: We present HerediVar a web application and HerediClassify a variant classification algorithm. The performance of HerediClassify was validated and compared to other variant classification tools. HerediClassify implements 19/28 variant classification criteria by the American College of Medical Genetics and gene-specific recommendations for ATM, BRCA1, BRCA2, CDH1, PALB2, PTEN, and TP53. RESULTS: HerediVar offers modular annotation services and allows for collaboration in the classification of variants. On the validation dataset, HerediClassify shows an average F1-Score of 93% across all criteria. HerediClassify outperforms other automated variant classification tools like vaRHC and Cancer SIGVAR. CONCLUSION: In HerediVar and HerediClassify we present a powerful solution to support variant classification in HBOC. Through their modular design, HerediVar and HerediClassify are easily extendable to other use cases and human genetic diagnostics as a whole.