BayesQuantify: an R package utilized to refine the ACMG/AMP criteria according to the Bayesian framework

Abstract

Improving the precision and accuracy of variant classification in clinical genetic testing involves further specification and stratification of the ACMG/AMP criteria. The Bayesian framework proposed by ClinGen has provided a mathematical foundation for evidence refinement, successfully quantifying, and extending the evidence strengths of PS1, PS4, PM5, and PP3/BP4. However, existing software and tools designed for quantifying the evidence strength and establishing corresponding thresholds to refine the ACMG/AMP criteria are lacking. To address this gap, we have developed BayesQuantify, an R package that aims to provide users with a unified resource for quantifying the strength of evidence for ACMG/AMP criteria using a naive Bayes classifier. By analyzing publicly available data, we demonstrate BayesQuantify's capability to offer objective and consistent refinement of the ACMG/AMP evidence. BayesQuantify is available from GitHub at https://github.com/liusihan/BayesQuantify.