Table 4.

Evaluation metrics for the test sets from SVM and RF classification using the nine gene multi-omics signature.

	Minimum		1st Quantile		Median		Mean		3rd Quantile		Maximum
Model	SVM	RF	SVM	RF	SVM	RF	SVM	RF	SVM	RF	SVM	RF
Accuracy	0.8654	0.8365	0.9519	0.9423	0.9615	0.9519	0.9624	0.9548	0.9808	0.9712	1.0000	1.0000
Sensitivity	0.7500	0.6923	0.9038	0.9231	0.9423	0.9423	0.9338	0.9373	0.9615	0.9615	1.0000	1.0000
Specificity	0.8846	0.7500	0.9808	0.9615	1.0000	0.9808	0.9910	0.9723	1.0000	0.9808	1.0000	1.0000

Accuracy, sensitivity, and specificity are reported for 5000 runs. This signature shows high predictive ability, with the mean accuracies being 96% and 95% in the test set, respectively, for SVM and RF.