Fig. 4
From: How to approach machine learning-based prediction of drug/compound–target interactions
Histogram plots displaying bioactivity distributions of transferase, ion channel, and membrane receptor families based on train (green bars) and test (orange bars) samples of; (a) random-split, (b) dissimilar-compound-split, and (c) fully-dissimilar-split datasets, along with their median values shown as vertical dashed lines