Title:Review of Current Chemoinformatic Tools for Modeling Important Aspects of CYPsmediated Drug Metabolism. Integrating Metabolism Data with Other Biological Profiles to Enhance Drug Discovery
VOLUME: 15 ISSUE: 4
Author(s):Alejandro Speck-Planche and Maria Natalia Dias Soeiro Cordeiro
Affiliation:REQUIMTE/Department of Chemistry and Biochemistry, Faculty of Sciences, University of Porto, 4169-007 Porto, Portugal.
Keywords:Chemoinformatic tools, cytochrome P450, drug metabolism, linear discriminant analysis, mtk-QSBER, SOM, support vector
machine, quadratic indices.
Abstract:The study of the metabolism of xenobiotics by the human body is an essential stage in the complex and expensive process of
drug discovery, being one of the main causes of disapproval and/or withdrawal of drugs. Regarding this, enzymes known as cytochromes
P450 (CYPs) play a very decisive role in the biotransformation of many chemicals. For this reason, the use of chemoinformatics to predict
and /or analyze from different points of view CYPs-mediated drug metabolism, can help to reduce time and financial resources. This
work is focused on the most remarkable advances in the last 5 years of the chemoinformatics tools towards the virtual analysis of CYPsmediated
drug metabolism. First, a brief section is dedicated to the applicability of chemoinformatics in different areas associated with
drug metabolism. Then, both the models for prediction of CYPs substrates and those allowing the assessment of sites of metabolism
(SOM) are discussed. At the same time, the principal limitations of the current chemoinformatic tools are pointed out. Finally, and taking
into account that metabolism is an essential step in the whole process of designing any drug, we introduce here as a case of study, the first
multitasking model for quantitative-structure biological effect relationships (mtk-QSBER). The purpose of this model is to integrate different
types of biological profiles such as ADMET (absorption, distribution, metabolism, excretion, toxicity) profiles and antistaphylococci
activities. The mtk-QSBER model was created by employing a heterogeneous dataset of more than 66000 cases tested in
6510 different experimental conditions. The model displayed a total accuracy higher than 94%. To the best of our knowledge, this is the
first attempt to complement metabolism assays with other relevant biological data in order to speed up the discovery of efficacious antistaphylococci
agents.
