On Predicting lung cancer subtypes using ‘omic’ data from tumor and tumor-adjacent histologically-normal tissue

BMC Cancer

Table 1 Datasets and sample distributions

Dataset Source	Tissue type	ADC	SCC
GEO: GDS3257 (gene expression)	Tumor	58	***
GEO: GDS3257 (gene expression)	TAHN	49	***
TCGA: LUAD+LUSC (gene expression)	Tumor	32	153
TCGA: LUAD+LUSC (gene expression)	TAHN	***	***
TCGA: LUAD+LUSC (DNA methylation)	Tumor	65	132
TCGA: LUAD+LUSC (DNA methylation)	TAHN	24	27

See challenge in Background on lack of TAHN tissue availability (***). GEO gene expression platform: Affymetrix Human Genome U133A Array (22,283 features), TCGA gene expression platform: Agilent 244 K Custom Gene Expression (17,814 features). TCGA methylation platform: Illumina Infinium HumanMethylation 27 k (27,578 features)

ISSN: 1471-2407