Classification and survival prediction in diffuse large B-cell lymphoma by gene expression profiling

Abstract

We present a novel framework for developing a risk model for class prediction from high-dimensional gene expression data; we define a new model that relies on several already known classification methods. We make use of the model for a survival analysis of tumor and immune subtype from Diffuse Large B-cell Lymphoma patients. Experimental analyses show good level of accuracy in the detection of Cell-of-Origin of diseases.