A collaboration between IBM Research and Arctoris is investigating the application of AI and automation to accelerate closed loop molecule discovery.
IBM Research has developed RXN for Chemistry, an online platform leveraging state-of-the-art Natural Language Processing (NLP) architectures to automate synthetic chemistry. Representing chemical reactions via SMILES (Simplified Molecular Input Line Entry System), the system is able to perform highly accurate reaction predictions using its powerful AI. Optimised synthetic routes are then used as input for RoboRXN, an automated platform for molecule synthesis.
Oxford-based technology company Arctoris has developed Ulysses, an end-to-end automated platform for drug discovery research. The platform ensures accuracy, precision, and reproducibility by leveraging robotic experiment execution and digital data capture technologies across cell and molecular biology and biochemistry/biophysics. Experiments conducted with Ulysses generate more than 100 times more datapoints per assay compared to industry standard, leading to deeper insights and accelerated progress compared to manual methods.
The two platforms are now being combined for the first time in a research collaboration that will see new small molecule inhibitors for undisclosed targets being designed, made, tested, and analysed (DMTA) in an autonomous, closed loop approach. Concretely, IBM Research will design and synthesize novel chemical matter (Design, Make), to be profiled and evaluated by Arctoris (Test, Analyze), with the resulting data informing the subsequent iteration of the DMTA cycle.
Thomas A. Fleming, Arctoris co-founder and COO, explained: “The future of drug discovery is computational, with AI and robotics paving the way for better treatments to reach patients sooner. We are excited about partnering with IBM Research on a world-first closed loop drug discovery project bringing together two leaders in the field of AI and robotics-powered drug discovery. This collaboration will showcase how the combination of our unique technology platforms will lead to accelerated research based on better data enabling better decisions.”
Dr Teodoro Laino, Distinguished Scientist at IBM Research Europe – Zurich, said: “This collaboration is a great example of the enablement that AI, Cloud and Automation can have in the space of material design. The integration between the two complementary technologies reveals how it is more and more important in R&D to turn great research into great viable products.”
Project co-ordinator Dr Matteo Manica, Research Scientist at IBM Research Europe – Zurich, added: “This is a unique opportunity to quantify the impact of AI and automation technologies in accelerating scientific discovery. In our collaboration, we demonstrate a pipeline to perform iterative design cycles where generative models suggest candidates that are synthesized with RoboRXN and screened with Ulysses. The data produced by Ulysses will then be used to establish a feedback loop to retrain the generative AI and improve the proposed leads in a completely data-driven fashion.”