Table 3.

The performance of the two models on the training set and our final selected model on the test set, evaluated using area under the curve (AUC), specificity and, sensitivity

	LOPO CV				Independent Testing
	Ensemble Model		Penalized Logistic Model		Penalized Logistic Model
	Sample-level	In silico mixing	Sample-level	In silico mixing	In vitro mixing
AUC	0.90 [0.87–0.93]	0.93 [0.88–0.98]	0.87 [0.83–0.91]	0.91 [0.85–0.97]	0.87 [0.76–0.98]
Specificity	0.92 [0.86–0.96]	0.95 [0.82–0.99]	0.91 [0.85–0.95]	0.95 [0.82–0.99]	0.88 [0.70–0.98]
Sensitivity	0.73 [0.67–0.79]	0.79 [0.66–0.89]	0.71 [0.64–0.77]	0.72 [0.58–0.83]	0.70 [0.47–0.87]