Background: The urgent need to identify new pharmacological targets for diabetes treatment and prevention has been amplified by the disease's extensive impact on individuals and healthcare systems. A better understanding of the biological underpinnings of diabetes is essential for developing therapeutic strategies that target these biological processes. Current genetic-based predictive methods fail to reliably foresee diabetes.

Objectives: Our study aims to pinpoint key epigenetic factors that predispose individuals to diabetes. These factors will inform the development of an advanced predictive model that estimates diabetes risk from genetic profiles, utilizing state-of-the-art statistical and data mining methods.

Methodology: For enhanced feature selection, we used a recursive feature removal with a cross-validation approach based on support vector machines (SVMs). Building on this, we created six machine learning models to assess their performance: logistic regression, k-Nearest Neighbors (k-NN), Naive Bayes, Random Forest, Gradient Boosting, and Multilayer Perceptron Neural Network.

Findings: The Gradient Boosting Classifier excelled, achieving a median recall of 92.17% and outstanding metrics such as area under the receiver operating characteristic curve (AUC) with a median of 68%, alongside median accuracy and precision scores of 76%. Through our machine learning analysis, we identified 31 genes significantly associated with diabetes traits, highlighting their potential as biomarkers and targets for diabetes management strategies.

Conclusion: The Gradient Boosting Classifier and Multilayer Perceptron Neural Network stood out for their ability to predict diabetes outcomes. We urge that future studies use larger cohorts and a broader range of predictive variables to improve these models' predictive powers.