Table 3.

Individual and average AUCs and MCCs from the validation phase and the additional validation approaches applied to the T2D datasets. The standard deviation of each result was excluded to keep the table simple and avoid complexity

PRJNA325931*	9 features (REFS)		3316 features		SelectKbest (k = 9)
Classifier	AUC	MCC	AUC	MCC	AUC	MCC
AdaBoostClassifier	0.8000	0.6749	0.4500	0.0438	0.7600	0.4530
Extra trees	0.8400	0.7532	0.5000	0.0000	0.7600	0.5512
KNeighbors	0.6500	0.4033	0.5000	− 0.0428	0.6500	0.2319
MLP	0.8800	0.8064	0.5200	− 0.0083	0.8200	0.6792
Lasso CV	0.7800	0.5661	0.5000	0.1828	0.7600	0.5758
Average	0.7900	0.6407	0.4940	0.0351	0.7500	0.4982

PRJNA554535	5 of 9 features (REFS)		SelectKbest (4 of 9 features)		10-time random selection
Classifier	AUC	MCC	AUC	MCC	AUC	MCC
AdaBoostClassifier	0.8200	0.6090	0.5260	0.5800	0.8000	0.0525
Extra Trees	0.8500	0.6504	0.5310	0.6093	0.8000	0.0684
KNeighbors	0.6700	0.3840	0.5230	0.4984	0.7300	0.0374
MLP	0.7100	0.4765	0.5230	0.3952	0.6000	0.0600
Lasso CV	0.5200	− 0.0146	0.5160	− 0.0158	0.5100	0.0296
Average	0.7140	0.4210	0.5238	0.4134	0.6880	0.0496

PRJEB53017	5 of 9 features (REFS)		SelectKbest (4 of 9 features)		10-time random selection
Classifier	AUC	MCC	AUC	MCC	AUC	MCC
AdaBoostClassifier	0.6700	0.3036	0.5200	0.0425	0.5500	0.0517
Extra Trees	0.6900	0.3659	0.5230	0.2526	0.6000	0.0550
KNeighbors	0.6800	0.4124	0.4970	0.2977	0.6200	−0.0164
MLP	0.6600	0.3823	0.5270	0.0711	0.5400	0.0741
Lasso CV	0.6100	0.2505	0.5100	0.2035	0.6000	0.0189
Average	0.6620	0.3429	0.5154	0.1734	0.5820	0.0366

*Discovery dataset