A risk stratification model for lung cancer based on gene coexpression network
Risk stratification model for lung cancer with gene expression profile is of great interest. Instead of the previously reported models based on individual prognostic genes, we aimed to develop a novel system-level risk stratification model for lung adenocarcinoma based on gene coexpression network. Using multiple microarray datasets obtained from lung adenocarcinoma, gene coexpression network analysis was performed to identify survival-related network modules. Representative genes of these network modules were selected and then, risk stratification model was constructed exploiting deep learning algorithm. The model was validated in two independent test cohorts. Survival analysis using univariate and multivariate Cox regression was performed using the output of the model to evaluate whether the model could predict patients overall survival independent of clinicopathological variables. Five network modules were significantly associated with patients survival. Considering prognostic significance and representativeness, genes of the two survival-related modules were selected for input data of the risk stratification model. The output of the model was significantly associated with patients overall survival in the two independent test sets as well as training set (p < 0.00001, p < 0.0001 and p = 0.02 for training set, test set 1 and 2, respectively). In multivariate analyses, the model was associated with patients prognosis independent of other clinical and pathological features. Our study presents a new perspective on incorporating gene coexpression networks into the gene expression signature, and the clinical application of deep learning in genomic data science for prognosis prediction.