Supervised learning techniques to predict compounds in pathway modules based on molecular properties

Machine learning algorithms provide significant indications in metabolomics to predict chemical compounds in metabolic pathways and in their modules. The modules in the metabolic pathway are sub networks of functionally related genes based on rules such as protein-protein interactions, co-regulated expression, coordinated physiological activity, and successive reaction steps. Fully functional modules are helpful to improve the diseases process, drug discover, and prediction of missing reaction. All modules in the metabolic pathway are not functional due to missing reaction steps. The structural mapping of chemical compounds with the pathway module is helpful to understand the mechanism of prediction unknown reaction step. The main purpose of this paper to predict the chemical compounds in pathway modules and their classes. We have constructed binary and multi-label classification data sets to predict pathway module and module classes, respectively. In order to identify the pathway module and its classes, we have built an ensemble Extra trees classifier to learn the molecular and atomic properties of chemical compounds. We have also experimented with different ensemble machine learning algorithm for the prediction of pathway modules. The overall prediction rate of the classifier 98.59%, indicating extra tree classifier features are more interpretable and have a high predictive performance on various tasks.