Identification and External Validation of Differentially Expressed Genes in Pancreatic Ductal Adenocarcinoma: A Public GEO Bioinformatics Study
Advances in Modern Biomedicine · 2026
DOI: 10.64229/udja9831 ↗Abstract
Background: Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis, and molecular markers that distinguish tumor tissue from non-tumor pancreatic tissue may support early diagnostic research. Public gene-expression datasets provide an accessible approach for identifying reproducible candidate genes. Methods Two public microarray datasets were analyzed. GSE62452 contained 69 PDAC tumor tissues and 61 adjacent non-tumor tissues profiled on the Affymetrix Human Gene 1.0 ST Array. Sixty tumor–adjacent tissue pairs with matching information were used for paired differential-expression analysis. GSE15471 contained 39 tumor and 39 matched normal pancreatic tissue arrays profiled on the Affymetrix U133 Plus 2.0 Array. Six technical replicate arrays were excluded, leaving 36 tumor and 36 normal arrays for external validation. Probe identifiers were mapped to gene symbols using the corresponding NCBI GEO platform annotations. Differential expression was assessed using paired mean expression differences and Benjamini–Hochberg false-discovery-rate correction. Selected genes were validated using expression differences and direction-adjusted area under the receiver-operating-characteristic curve (AUC). Results GSE62452 contained 20,003 gene symbols after annotation mapping. Ten genes met the discovery criteria of false-discovery rate
EUVIMED