Table 1. The performance of models trained with various features for methylarginine.

Training features	Sn (%)	Sp (%)	Acc (%)	MCC (%)
SPC	68.96±1.52	92.71±1.09	86.78±0.56	63.78±1.66
PWAA	60.85±0.57	95.51±0.50	86.85±0.41	62.76±1.16
ASA	51.94±0.15	99.81±0.15	87.84±0.12	66.37±0.45
VDWV	56.34±1.22	98.91±0.34	88.24±0.54	67.08±1.81
SPC+PWAA	71.66±2.14	91.87±0.60	86.82±0.75	64.40±2.55
SPC+ASA	65.69±1.22	95.46±0.69	88.01±0.68	66.42±1.93
PWAA+ASA	66.42±2.37	92.13±0.90	85.70±1.02	60.72±2.78
SPC+PWAA+ASA	74.09±1.44	93.35±1.45	88.54±0.64	68.93±1.88
SPC+PWAA+VDWV	74.54±3.56	94.31±1.17	89.37±0.83	71.69±3.79
SPC+PWAA+ASA+VDWV	80.73±1.58	92.28±1.24	89.39±1.35	72.45±1.73

The corresponding measurement was represented as the average value±standard deviation. The window size was 15 and the ratio between positive and negative samples was 1∶3.