Fig. 1

Flow chart display. Flow chart showing the SNP selection, model training, and performance evaluation processes. A total of 633 subjects were recruited for the current study. The data were preprocessed and randomly divided into a training set (393 participants) and a test set (240 participants). k-fold cross-validation was used in the training set, and performance evaluation indexes such as AU-ROC and AU-PRC were adopted to judge the average predictive performance of each model