Machine Learning for Interpretation of DNA Variants of Maturity-Onset Diabetes of the Young Genes Based on ACMG Criteria

Background: Maturity-onset diabetes of the young (MODY) is a group of dominantly inherited monogenic diabetes, with HNF4A-MODY, GCK-MODY and HNF1A-MODY being the three most common genes responsible. Molecular diagnosis of MODY is important for precise treatment. While a DNA variant causing MODY can be assessed by the criteria of the American College of Medical Genetics and Genomics (ACMG) guidelines, gene-specific assessment of disease-causing mutations is important to differentiate between the MODY subtypes. As the ACMG criteria were not originally designed for machine learning algorithms, they are not true independent variables. Methods: In this study, we applied machine learning models for interpretation of DNA variants in MODY genes defined by the ACMG criteria based on Human Gene Mutation Database (HGMD) and ClinVar. Results: The results show highly predictive abilities with accuracy over 95%, suggest that this model could serve as a fast, gene-specific method for physicians or genetic counselors assisting with diagnosis and reporting, especially when confronted by contradictory ACMG criteria. Also, the weight of the ACMG criteria shows gene specificity which advocates for the application of machine learning methods with the ACMG criteria to capture the most relevant information for each disease-related variant. Conclusion: Our results highlight the need for different weights of the ACMG criteria in relation with different MODY genes for accurate functional classification. For proof of principle, we applied the ACMG criteria as feature vectors in a machine learning model obtaining precision-based result.