DNA sequence analysis is essential for comprehending genetic disorders; yet, the intricacy of high-performance machine learning models frequently leads to a “black box” issue that prevents clinical adoption and confidence. This study develops a paradigm that functions across the model transparency spectrum to answer the urgent need for transparency. A Decision Tree is what we use, a classifier model renowned for its inherent interpretability—and supplement it with cutting-edge post-hoc Explainable AI (XAI) methods, particularly SHAP and LIME that stand for Shapely Additive explanations and Local Interpretable Model-agnostic Explanations respectively. Our approach achieved 80% prediction accuracy while retaining transparency by using a carefully selected dataset of benign and harmful DNA variations. More significantly, the XAI integration offered precise, useful insights for both local explanations of specific forecasts and global knowledge of general model behaviour. This study tells us more about the reliability of Ai in precision medicine by showing how successfully it balances clinical interpretability and predictive performance by using interpretable models with explainability techniques.