We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


Optibrium’s Quantum Mechanics and Machine Learning Methods Predict Routes of Drug Metabolism

A computer chip that resembles a brain.
Credit: iStock.
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 1 minute

Optibrium, announced the publication of its peer-reviewed study in Xenobiotica, ‘Predicting routes of phase I and II metabolism based on quantum mechanics and machine learning’1. In the paper, the team demonstrate a new method that better determines the routes of metabolism and metabolites in early drug discovery.

Unexpected metabolism can lead to late-stage drug candidate failure, or even the withdrawal of approved drugs. Early in silico prediction of the dominant routes of metabolism is therefore vital to improve a drug’s chance of success.

The paper first describes the development and validation of Optibrium’s WhichEnzymeTM model, which accurately predicts the enzyme families most likely to metabolise a drug candidate. The team then combine this new model with Optibrium’s previously published models. These include regioselectivity models for key Phase I and Phase II drug metabolising enzymes, which use quantum mechanical simulations with machine learning methods to predict sites of metabolism and the resulting metabolites. Additionally, the WhichP450 model, which predicts the Cytochrome P450 isoform(s) responsible for a compound’s metabolism.

Want more breaking news?

Subscribe to Technology Networks’ daily newsletter, delivering breaking science news straight to your inbox every day.

Subscribe for FREE

Based on the combined model outputs, Optibrium showcase a new method to determine the most likely routes of metabolism and metabolites to be observed experimentally. The paper demonstrates that this method delivers high sensitivity in identifying experimentally reported metabolites, as well as higher precision than other methods for predicting in vivo metabolite profiles. It enables researchers to identify compounds with greater metabolic stability and better safety profiles, and underpins Optibrium’s recently launched StarDrop Metabolism module.

“Our latest study is the result of six years of focused research, delivering a practical model that allows users to predict metabolic pathways for a wide range of drug-like compounds. Thanks to our carefully curated datasets and our signature reactivity-accessibility approach, we have managed to build accurate isoform-specific regioselectivity models for the important Phase I and II enzyme families” Dr Mario Öeren, Principal Scientist, Optibrium said. “Using that very same data, we've trained models that predict the likely enzyme families and isoforms which metabolise a compound. And then by combining these models, we've trained and validated a ‘model of models’ that predicts the full metabolism pathway for a given compound.”