Fig. 10.

Performance comparison of well performing conventional descriptor sets and learned representations for three different splits of ion-channel and protease family datasets in terms of; (a) Spearman rank correlation, and (b) median corrected MCC scores