Table 6. Predictive performance of different approaches based on the variance-stability transformation for dataset 2 and 3.

	Dataset 2			Dataset 3
Method	ER (SD)	AUC (SD)	AUPRC (SD)	ER (SD)	AUC (SD)	AUPRC (SD)
Oracle	0.04 (0.02)	0.97 (0.02)	0.97 (0.03)	0.08 (0.04)	0.96 (0.04)	0.94 (0.07)
LASSO	0.2 (0.04)	0.87 (0.04)	0.87 (0.06)	0.28 (0.07)	0.79 (0.07)	0.79 (0.1)
ZINB+GLM	0.23 (0.05)	0.84 (0.06)	0.84 (0.08)	0.36 (0.08)	0.68 (0.11)	0.68 (0.15)
TPNB+GLM	0.19 (0.05)	0.88 (0.04)	0.88 (0.06)	0.3 (0.08)	0.75 (0.09)	0.74 (0.13)
NB+GLM	0.23 (0.05)	0.84 (0.06)	0.84 (0.08)	0.36 (0.08)	0.68 (0.11)	0.68 (0.15)
ZINB+LASSO	0.23 (0.05)	0.84 (0.06)	0.84 (0.08)	0.34 (0.08)	0.69 (0.1)	0.69 (0.15)
TPNB+LASSO	0.19 (0.05)	0.88 (0.04)	0.88 (0.06)	0.3 (0.07)	0.75 (0.09)	0.75 (0.12)
NB+LASSO	0.23 (0.05)	0.84 (0.05)	0.84 (0.08)	0.34 (0.07)	0.7 (0.09)	0.7 (0.14)