PDAC-ANN: an artificial neural network to predict pancreatic ductal adenocarcinoma based on gene expression

BMC Cancer

Table 1 Characteristics of studies used in the meta-analysis

Accession number	Study	Array platform	Differentially expressed genes		Samples
			Upregulated	Downregulated	Tumor	Normal
GSE23397	*	Affymetrix Human Exon 1.0 ST Array	4031	870	15	6
GSE28735	[29]		245	146	45	45
GSE41368	[30]		1200	462	6	6
GSE32676	[31]	Affymetrix Human Genome U133 Plus 2.0 Array	686	319	25	7
GSE71989	[32]		3052	661	13	8
GSE15471	[33]		1546	227	39	39
GSE62165	[13]	Affymetrix Human Genome U219 Array	2638	1266	118	13
GSE43795	[34]	Illumina HumanHT-12 V4.0 expression beadchip	1978	1343	6	5
GSE71729	[35]	Agilent-014850 Whole Human Genome Microarray 4x44K G4112F	285	175	145	46
GSE60979	[36]	Agilent-028004 SurePrint G3 Human GE 8x60K Microarray	1365	1336	49	12
Total					461	187
GSE16515	[37]	Affymetrix Human Genome U133 Plus 2.0 Array	–	–	36	16
GSE62452	[38]	Affymetrix Human Gene 1.0 ST Array	–	–	69	61
Total					105	77

* No publication available
- Analysis of differentially expressed genes was not applied to the validation dataset

ISSN: 1471-2407