The article explores the impact of Bayesian optimization on feature selection methods, particularly in the context of high-dimensional molecular data analysis. Bayesian optimization is known for its ability to automatically configure hyper-parameters, which are crucial for the performance of many machine learning models. The study conducted extensive simulation studies to compare various feature selection methods, focusing on those that require hyper-parameter tuning. The methods evaluated included filter-based (SIS, MRMR), wrapper-based (RFE), and embedded-based (Lasso, Enet, XGBoost) approaches. The simulation results showed that feature selection methods with hyper-parameters tuned using Bayesian optimization often achieved better recall rates, especially when the number of selected features was large. Additionally, the study utilized gene expression data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) to predict brain imaging-related phenotypes. The results indicated that Bayesian-optimized feature selection methods improved the accuracy of disease risk prediction models. The authors concluded that Bayesian optimization can significantly enhance the performance of feature selection methods, particularly in high-dimensional data analysis, and provided practical suggestions for its application in big molecular data analysis.The article explores the impact of Bayesian optimization on feature selection methods, particularly in the context of high-dimensional molecular data analysis. Bayesian optimization is known for its ability to automatically configure hyper-parameters, which are crucial for the performance of many machine learning models. The study conducted extensive simulation studies to compare various feature selection methods, focusing on those that require hyper-parameter tuning. The methods evaluated included filter-based (SIS, MRMR), wrapper-based (RFE), and embedded-based (Lasso, Enet, XGBoost) approaches. The simulation results showed that feature selection methods with hyper-parameters tuned using Bayesian optimization often achieved better recall rates, especially when the number of selected features was large. Additionally, the study utilized gene expression data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) to predict brain imaging-related phenotypes. The results indicated that Bayesian-optimized feature selection methods improved the accuracy of disease risk prediction models. The authors concluded that Bayesian optimization can significantly enhance the performance of feature selection methods, particularly in high-dimensional data analysis, and provided practical suggestions for its application in big molecular data analysis.