Table 1.

Summary of 20 TCGA cohorts that we used in our two sets of experiments, namely, E1 and E2

Cohort	Disease name	Stage I	Stage II	Stage III	Stage IV	Early (E1)	Late (E1)	Total (E1)	Early (E2)	Late (E2)	Total (E2)
ACC	Adrenocortical carcinoma	9	37	16	15	—	—	—	46	31	77
BLCA	Bladder urothelial carcinoma	2	130	140	134	—	—	—	132	274	406
BRCA	Breast invasive carcinoma	181	619	247	20	181	886	1067	800	267	1067
COAD	Colon adenocarcinoma	75	176	128	64	75	368	443	251	192	443
ESCA	Esophageal carcinoma	16	69	49	8	16	126	142	85	57	142
HNSC	Head and neck squamous cell carcinoma	25	70	78	259	25	407	429	95	337	429
KICH	Kidney chromophobe	20	25	14	6	20	45	65	45	20	65
KIRC	Kidney renal clear cell carcinoma	265	57	123	82	265	262	527	322	205	527
KIRP	Kidney renal papillary cell carcinoma	172	21	51	15	172	87	259	193	66	259
LIHC	Liver hepatocellular carcinoma	171	86	85	5	171	176	347	257	90	347
LUAD	Lung adenocarcinoma	274	121	84	26	274	231	505	395	110	505
LUSC	Lung squamous cell carcinoma	244	162	84	7	244	253	497	406	91	497
MESO	Mesothelioma	10	16	44	16	—	—	—	26	60	86
PAAD	Pancreatic adenocarcinoma	21	146	3	4	21	153	174	—	—	—
READ	Rectum adenocarcinoma	30	51	51	24	30	126	156	81	75	156
SKCM	Skin cutaneous melanoma	2	66	27	3	—	—	—	68	30	98
STAD	Stomach adenocarcinoma	53	111	150	38	53	299	352	164	188	352
TGCT	Testicular germ cell tumours	55	12	14	0	55	26	81	—	—	—
THCA	Thyroid carcinoma	281	52	112	55	281	219	500	333	167	500
UVM	Uveal melanoma	0	39	36	4	—	—	—	39	40	79
					Total	1883	3664	5547	3738	2300	6038

Note: For each cohort, we report TCGA cohort code, disease name and number of samples in each stage together with details about the numbers of early-stage, late-stage and total samples in experiments E1 and E2. We included 5547 and 6038 patients in total for our two sets of experiments E1 and E2, respectively.