Key points are not available for this paper at this time.
Abstract Most gene expression studies aim to discover genes associated with specific diseases. The standard approaches based on machine learning utilize several feature selection techniques to identify significant genes that can serve as biomarkers for a given disease. In recent times, the integration of prior knowledge-based approaches for biomarker identification has shown much promise in discovering several biomarkers, thus allowing scope for an increase in translational applications. In this study, we developed a novel approach GediNET that integrates prior biological knowledge about genes associated with diseases like cancer to group genes into groups. The novelty of GediNet is that it discovers disease-disease gene associations within groups rather than disease-gene associations. These groups are later subject to a Scoring component for performing group selections rather than single feature selection. The top-ranked groups are used to train the machine learning model. The process of Grouping and Scoring using the (G-S-M) model is then applied to discover groups of disease genes or biomarkers for a specific disease. One of the outputs of the suggested tool GediNET is a list of significant groups of diseases that combine their associated genes can contribute to developing biomarkers and drugs. GediNET identifies the relationships between diseases, Disease–disease association (DDA) based machine learning, which explores novel associations of diseases that enhance knowledge of disease relationships, which could further improve approaches to disease diagnosis, prognosis, and treatment. The GediNET Knime workflow can be downloaded from: https: //github. com/malikyousef/GediNET. git or https: //kni. me/w/3kH1SQVₘMUsMTS-.
Yousef et al. (Tue,) studied this question.