Predictive performance of mlLGPR on T1 golden datasets.

mlLGPR-L1: the mlLGPR with L1 regularizer, mlLGPR-L2: the mlLGPR with L2 regularizer, mlLGPR-EN: the mlLGPR with elastic net penalty, L2: AB: abundance features, RE: reaction evidence features, and PE: pathway evidence features. For each performance metric, ‘↓’ indicates the lower score is better while ‘↑’ indicates the higher score is better.mlLGPR-L1：采用L1正则化器（L1 regularizer）的mlLGPR模型；mlLGPR-L2：采用L2正则化器（L2 regularizer）的mlLGPR模型；mlLGPR-EN：采用弹性网络惩罚（elastic net penalty）的mlLGPR模型。其中L2类型特征包含丰度特征（abundance features，AB）、反应证据特征（reaction evidence features，RE）以及通路证据特征（pathway evidence features，PE）三类。针对每项性能指标，符号“↓”表示分值越低性能越佳，“↑”表示分值越高性能越优。