Table 1.

Description of the datasets used in this study.

		Training datasets	Validation datasets
		BLCA158	BLCA57	BLCA114
		TCGA	GSE31684	GSE32894
Platform		Illumina Hiseq-RNAseq	GPL570	GPL6974
Sample Size		158	57	114
Stage	0	0	5	0
	I	1	9	63
	II	60	12	43
	III	58	22	8
	IV	39	9	0