BMC Bioinformatics

Table 2 Characteristics of the TCGA dataset

From: Assessment of deep learning and transfer learning for cancer prediction based on gene expression data

Disease	Size	Cancer	Non-cancer	Prior
BRCA	1214	1101	113	0.91
KIRC	610	538	72	0.88
LUAD	592	533	59	0.90
UCEC	574	551	23	0.96
THCA	560	502	58	0.89
LUSC	551	502	49	0.91
PRAD	550	498	52	0.90
HNSC	544	500	44	0.92
LGG	510	510	0	1
OV	374	374	0	1
LIHC	371	371	0	1
Total	6450	5980	470	0.927

The columns represent respectively the type of tissues (Disease), the numbers of samples (Size), cancer samples (Cancer), non-cancer samples (Non-cancer) and the proportion of the majority class (Prior). This dataset contains only patient data

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com