Table 3.

Comparison of proposed method with Lasso, elastic net, adaptive Lasso, using three experimental mixtures containing 6, 7, and 20 metabolites, respectively; and a library size of 61, 101, and 200 metabolites, respectively. Performance was evaluated based on average accuracy, sensitivity, and specificity

Met.	Metrics	Proposed method	Lasso	Elastic Net	Adaptive Lasso	Chenomx	Bayesil	ASICS
Library size 61
	Accuracy	1.00	0.72	0.62	0.75	0.80	0.64	0.80
6	Sensitivity	1.00	1.00	1.00	1.00	0.67	1.00	0.57
	Specificity	1.00	0.69	0.58	0.73	0.82	0.60	0.81
	Accuracy	1.00	0.77	0.79	0.64	0.89	0.64	0.83
7	Sensitivity	1.00	1.00	1.00	1.00	0.71	1.00	0.33
	Specificity	1.00	0.74	0.76	0.59	0.91	0.59	0.86
20	Accuracy	0.93	0.67	0.69	0.70	0.74	0.75	0.56
	Sensitivity	1.00	1.00	1.00	1.00	0.80	0.95	0.45
	Specificity	0.90	0.51	0.54	0.56	0.71	0.66	0.59
Library size 101
	Accuracy	0.94	0.79	0.88	0.72	0.87	0.30	0.80
6	Sensitivity	1.00	1.00	1.00	1.00	0.67	1.00	0.57
	Specificity	0.94	0.78	0.87	0.71	0.88	0.25	0.81
	Accuracy	0.97	0.90	0.84	0.87	0.93	0.42	0.83
7	Sensitivity	1.00	1.00	1.00	1.00	0.71	1.00	0.33
	Specificity	0.97	0.89	0.83	0.86	0.95	0.37	0.86
	Accuracy	0.88	0.72	0.73	0.72	0.69	0.33	0.56
20	Sensitivity	1.00	1.00	1.00	1.00	0.80	0.95	0.45
	Specificity	0.85	0.65	0.67	0.65	0.67	0.16	0.59
Library size 200
	Accuracy	0.94	0.86	0.86	0.83	0.92	0.30	0.80
6	Sensitivity	1.00	1.00	1.00	1.00	0.67	1.00	0.57
	Specificity	0.93	0.85	0.85	0.82	0.93	0.25	0.81
	Accuracy	0.98	0.86	0.86	0.92	0.96	0.42	0.83
7	Sensitivity	1.00	1.00	1.00	1.00	0.71	1.00	0.33
	Specificity	0.97	0.85	0.85	0.92	0.97	0.37	0.86
	Accuracy	0.86	0.74	0.75	0.74	0.76	0.33	0.56
20	Sensitivity	1.00	1.00	1.00	1.00	0.80	0.95	0.45
	Specificity	0.84	0.71	0.72	0.71	0.75	0.16	0.59