Identification of Human Cell Cycle Phase Markers Based on Single-Cell RNA-Seq Data by Using Machine Learning Methods

Table 1

Performance of optimal classifiers corresponding to the optimal feature subsets based on different classification algorithms and feature ranking algorithms.

Feature ranking algorithms	Classification algorithm	Number of features	ACC	MCC
mRMR	Random forest	195	0.906	0.859
	Support vector machine	124	0.884	0.826
	k-nearest neighbor	37	0.869	0.803
	Decision tree	158	0.836	0.753
MCFS	Random forest	52	0.906	0.859
	Support vector machine	526	0.918	0.878
	k-nearest neighbor	76	0.881	0.822
	Decision tree	479	0.842	0.762
SHAP by LightGBM	Random forest	87	0.917	0.875
	Support vector machine	222	0.917	0.875
	k-nearest neighbor	18	0.866	0.800
	Decision tree	103	0.856	0.784