BackgroundPancreatic adenocarcinoma(PAAD) is a kind of terribly malignant tumor of the digestive tract with high mortality globally, however, the pathogenic molecular mechanisms of which remain unclear currently. Therefore, it is very important to explore the underlying mechanisms and screen for novel prognostic markers and treatment targets. We performed a weighted gene co-expression network analysis(WGCNA) to identify hub genes associated with clinical parameters of interest using The Cancer Genome Atlas(TCGA) data. In addition, Gene ontology(GO) enrichments and survival analyses were conducted.MethodsThe 14 datasets including gene expression profiles and clinical phenotypes for PAAD patients in TCGA were analyzed by WGCNA in R. After constructing the scale-free network of gene co-expression, modules with clinical significance were distinguished, and hub genes were validated, which also key genes regulated relevant traits. GO enrichments were performed on the modules. Furthermore, the significance of hub genes was confirmed via survival analysis.ResultsWe discovered 12 candidate genes in total which had significant effects on the overall survival of PAAD patients. The expression of DUSP26, TCEAL2, CELF3, CDK5R2, NAP1L2, INA had strong and positive influences on the pathological T stage, while the expression of CEP55, HJURP, KIF23, CDCA5, SAC3D1, ATP6L had great negative impacts on the histological stage. Meanwhile, we also found that the tumor primary location and age of onset were key factors affecting the prognosis of PAAD.ConclusionThese candidate genes probably play important roles in proliferation, differentiation and metastasis of PAAD cells, which may be acted as novel prognostic markers and effective therapeutic targets. Besides, most of them are first reported in PAAD and deserved in-depth research.