Pathogenic Genetic Variant Classification

XGBoost and a support-vector machine classify small germline DNA variants as pathogenic or benign. The training labels come from ClinVar and the population frequencies from dbSNP. Each score is calibrated and placed in one of three zones (likely pathogenic, uncertain, likely benign). It comes with exact TreeSHAP and Shapley explanations and with the ACMG/AMP criteria the data touch. A matching MATLAB R2026a version covers classification and explainability.

Category: Bioinformatics - Genomics and Sequence Analysis (Machine Learning, Explainable AI, MATLAB, Python)

Bioinformatics Project B03 from ProTechSC Academy, with graphical abstract, video demonstration and project guidance.