A graph-embedded deep feedforward network for disease outcome classification and feature selection using gene expression data

Gene expression data represents a unique challenge in predictive model building, because of the small number of samples $(n)$ compared to the huge amount of features $(p)$. This "$n<