P. A. Hunt, M. D. Segall & J. D. Tyzack, J. Comput.-Aided Mol. Des. 2018, 32, pp537–546

DOI: 10.1007/s10822-018-0107-0

This paper describes the underlying methods and validation of a model predicting the most likely Cytochrome P450 isoforms responsible for metabolism of a compound. The model makes up part of StarDrop’s P450 module.

Abstract

In the development of novel pharmaceuticals, the knowledge of how many, and which, Cytochrome P450 isoforms are involved in the phase I metabolism of a compound is important. Potential problems can arise if a compound is metabolised predominantly by a single isoform in terms of drug-drug interactions or genetic polymorphisms that would lead to variations in exposure in the general population. Combined with models of regioselectivities of metabolism by each isoform, such a model would also aid in the prediction of the metabolites likely to be formed by P450-mediated metabolism.

We describe the generation of a multi-class random forest model. It predicts which, out of a list of the 7 leading Cytochrome P450 isoforms, would be the major metabolising isoforms for a novel compound. The model has a 76% success rate with a top-1 criterion and an 88% success rate for a top-2 criterion. It shows significant enrichment over randomised models.