Table 3.

Comparison of our method, MethCGI, Methylator, and iDNA-Methyl on the benchmark dataset.

Predictor	ACC	MCC	SN	SP
iDNA-Methyl	0.7749	0.5471	0.6125	0.9033
Methylator	0.7135	0.3327	0.5172	0.8078
MethCGI	0.7383	0.3748	0.4968	0.8542
Our Method ${}^1$	0.8017	0.5558	0.7268	0.8377

${}^1$ Feature size is 114-D feature, selected by feature selection in SVM. Our method used SVM classifier and target-jackknife cross-validation. The bold digits are the greatest values in each column.