Table 5.

Comparison of prediction quality measured with sensitivity and specificity for the prediction of the crystallizable and noncrystallizable proteins by the CRYSTALP2 method.

	crystallizable proteins		Noncrystallizable proteins
Dataset	sensitivity	specificity	sensitivity	specificity
TEST-RL	74.4	65.1	65.1	74.4
TEST	79.1	72.2	72.2	79.1
TEST-NEW	76.1	62.6	62.5	76.1

Results are based on training the classification model on FEAT dataset and testing on the TEST-RL, TEST, and TEST-NEW datasets, respectively.