Single-cell classification using graph convolutional networks

BMC Bioinformatics

Table 4 Median F1 of the eight scRNAseq data classifier and the four conventional classifiers tools on the four experiments (N = 1000)

Training dataset	BaronHuman + Muraro + Segerstolpe	Xin + Muraro + Segerstolpe	Xin + BaronHuman + Segerstolpe	Xin + BaronHuman + Muraro
Testing dataset	Xin	BaronHuman	Muraro	Segerstolpe
sigGCN	0.993	0.976	0.957	0.989
FC	0.976	0.965	0.945	0.990
scID	0.991	0.98	0.976	0.994
scPred	0.923	0.231	0.912	0.725
CasTLe	0.974	0.916	0.966	0.988
SingleR	0.988	0.974	0.969	0.994
scmapcluster	0.133	0.001	0.001	0
scmapcell	0.415	0.232	0.353	0.241
ACTINN	0.982	0.977	0.957	0.989
RF	0.946	0.914	0.957	0.903
SVM-linear	0.986	0.972	0.952	0.986
SVM-rbf	0.963	0.97	0.958	0.952
KNN	0.766	0.74	0.504	0.602

ISSN: 1471-2105