Table 3. Area under the Curve Values for Prediction of 10 Assays of the Matrix 2 Test Set^a.

assay code	CCBM	NB	RF	SVM	single-task DNN	multitask DNN	GraphConv
A	0.85	0.84	0.91	0.92	0.91	0.91	0.90
B	0.77	0.79	0.85	0.85	0.82	0.82	0.83
C	0.64	0.71	0.73	0.72	0.69	0.67	0.72
D	0.63	0.72	0.69	0.65	0.67	0.62	0.64
E	0.81	0.82	0.86	0.84	0.84	0.85	0.85
F	0.82	0.82	0.88	0.88	0.87	0.87	0.86
G	0.73	0.79	0.84	0.84	0.81	0.79	0.82
H	0.80	0.85	0.90	0.90	0.88	0.87	0.89
I	0.80	0.85	0.89	0.89	0.88	0.85	0.89
J	0.84	0.87	0.92	0.92	0.91	0.86	0.92

^aReported are AUC values for prediction of 10 assays (codes A–J) using different machine learning methods. For each assay, best results are indicated in bold.