Figure 5E source data files

PR_curve_values.csv — Precision-recall curve values for Joint-RPCA and RPCA classification models using either the top 5% of metabolite features or the top 5% of features across all omics, ranked by the corresponding method. The file contains the recall and precision coordinates used for plotting, lower and upper precision bounds, and summary average precision statistics for each method and dataset combination.

fold_level_predictions.csv — Fold-level classification predictions used to generate the precision-recall curves shown in Figure 5E. For each cross-validation fold, the file contains the true class labels (y_true) and predicted scores (y_score) for Joint-RPCA and RPCA models applied to either metabolomics data alone (gcms) or aggregated multi-omic features (all).