Table 2.

Summary of ROC Curve Results

	Area			Threshold value			Accuracy
	VL3	VLXT	VSL2	VL3	VLXT	VSL2	VL3	VLXT	VSL2
TM-alpha	0.740 ± 0.002	0.768 ± 0.004	0.826 ± 0.003	0.18 ± 0	0.16 ± 0	0.31 ± 0	67.7% ± 0.5%	70.8% ± 0.5%	75.1% ± 0.5%
TM-beta	0.713 ± 0.009	0.705 ± 0.010	0.796 ± 0.009	0.24 ± 0	0.38 ± 0.01	0.49 ± 0.01	64.1% ± 1.3%	67.7% ± 0.6%	72.5% ± 0.8%
PDD	0.729 ± 0.003	0.712 ± 0.004	0.762 ± 0.003	0.28 ± 0.01	0.33 ± 0	0.43 ± 0.01	66.0% ± 0.4%	64.9% ± 0.5%	69.1% ± 0.3%

The areas under the ROC curves (Area), the threshold values determined from the breakeven points specified as values on the X-axis (Threshold Values) and accuracies calculated using the indicated threshold values (Accuracy) for data obtained using the VL3, VLXT, and VSL2 predictors as applied to sequences from the TM-alpha, TM-beta and PDD datasets.