Fig. 26

Precision-recall curves of CNN models of increasing number of layers on the three datasets of increasing size. The higher the curve, the better performance with a horizontal dashed line to represent random prediction. An ideal precision-recall curve would cross the (1,1) point. The splice data has three panels since precision-recall curves assume binary classification and the splice dataset has three classes (0, 1, 2). Each panel corresponds to prediction one class vs the other two combined. DeepRAM outperforms all models in the histone and motif discovery datasets, and behaves well on the splice data. The CNN models (Nguyen and Zeng) outperform all models on the splice data