Fig. 1From: Identification of common oncogenic and early developmental pathways in the ovarian carcinomas controlling by distinct prognostically significant microRNA subsetsThe workflow of the identification of a prognostic signature. The workflow includes the steps of our method of feature selection, cross validation, construction of the prognostic models and regulatory networks. This pipeline processes the miRNA and patient survival data sets, using the updated data-driven grouping (DDg) method, called DDSS. It further performs cluster analysis, statistically weighted voting grouping (SWVg) to select the high-confidence prognostic features (expressed miRNAs) providing the patient stratification two or more disease development risk subgroups. Finally, the work flow links these results with pathway data. DDg and SWVg are the both prognosis and feature selection and patient’s risk classification methods in [32, 34] (see Additional file 1: Methods)Back to article page