TABLE 4.

AUCs for the performance of the multi-omics based predictive models.

Cohort	Classifiers	AUC	Accuracy	Sensitivity	Specificity	PPV	NPV	F1 score
Training	LR	0.932 (0.892–0.972)	0.888	0.784	0.946	0.889	0.888	0.833
	RF	0.949 (0.914–0.984)	0.902	0.902	0.902	0.836	0.943	0.868
	XGBoost	0.957 (0.922–0.992)	0.923	0.922	0.924	0.870	0.955	0.895
	SVM	0.910 (0.865–0.956)	0.832	0.843	0.826	0.729	0.905	0.782
	MLP	0.952 (0.921–0.984)	0.888	0.922	0.870	0.797	0.952	0.855
	KNN	0.892 (0.839–0.945)	0.790	0.882	0.739	0.652	0.919	0.750
	DT	0.776 (0.700–0.852)	0.769	0.725	0.793	0.661	0.839	0.692
	LightGBM	0.998 (0.994–1.000)	0.986	0.980	0.989	0.980	0.989	0.980
Test	LR	0.810 (0.704–0.915)	0.661	0.480	0.784	0.600	0.690	0.533
	RF	0.706 (0.568–0.844)	0.661	0.560	0.730	0.583	0.711	0.571
	XGBoost	0.695 (0.561–0.829)	0.613	0.480	0.703	0.522	0.667	0.500
	SVM	0.799 (0.691–0.907)	0.677	0.720	0.649	0.581	0.774	0.643
	MLP	0.802 (0.693–0.911)	0.726	0.760	0.703	0.633	0.812	0.691
	KNN	0.750 (0.630–0.871)	0.597	0.720	0.514	0.500	0.731	0.590
	DT	0.549 (0.417–0.680)	0.548	0.400	0.649	0.435	0.615	0.417
	LightGBM	0.726 (0.594–0.859)	0.694	0.560	0.784	0.636	0.725	0.596

SVM: support vector machine, LR: logistic regression, RF: random forest, KNN: k-nearest neighbor, AUC: area under the curve, NPV: negative predictive value, PPV: positive predictive value.