Partnership aims to boost predictive metabolism models using machine learning for faster, greener R&D

Drug discovery software and AI specialist Optibrium has partnered with Tallinn University of Technology (TalTech) to co-supervise a new PhD position as part of the European Commission-funded INNOCHEMBIO programme. The research project will focus on developing next-generation predictive models for drug metabolism, with the goal of improving accuracy, speed, and sustainability in pharmaceutical R&D.

The PhD research will explore how machine learning interatomic potentials (MLIPs) can replace traditional physics-based simulations, which are often too computationally expensive for routine use. By developing MLIPs specifically tailored to drug-like molecules and metabolism via Cytochrome P450 enzymes—a group responsible for metabolising up to 80% of small molecule drugs—the project aims to reduce reliance on laboratory testing while supporting more efficient compound design.

The resulting models will be integrated into Optibrium’s StarDrop platform, helping drug discovery teams make better-informed decisions earlier in development. This integration is expected to deliver practical benefits for pharmaceutical researchers, supporting the industry’s broader move toward green chemistry and more sustainable innovation pathways.

The PhD project is part of the wider Marie Skłodowska-Curie Actions (MSCA) COFUND initiative. INNOCHEMBIO will fund 15 PhD positions in total, with the first call for applications now open. The programme is designed to train the next generation of experts in sustainable chemistry and biotechnology, while fostering industry-academic collaborations that ensure research has immediate relevance and application.

Mario Öeren, principal scientist at Optibrium and senior researcher at TalTech, said: “Collaborations like this offer PhD candidates hands-on experience with real-world industry challenges. The models we develop will directly enhance our ability to support pharmaceutical R&D, improving the predictive power and efficiency of early-stage decision-making.”

Matthew Segall, CEO at Optibrium, added: “We’re proud to support the next generation of scientists and contribute to the global effort to make drug discovery more sustainable. Faster and more accurate predictive models mean teams can operate more cost-effectively while reducing waste and conserving valuable resources.”

The partnership reflects growing interest across the sector in AI-driven discovery tools that support sustainability goals and reduce development timelines. By combining machine learning with advanced modelling techniques, projects like this help lay the groundwork for a new era of data-driven, environmentally responsible drug development.