Pharma R&D Today
Ideas and Insight supporting all stages of Drug Discovery & Development
Accelerating synthetic chemistry with a predictive retrosynthesis solution
Posted on August 12th, 2021 by Ann-Marie Roche in AI & Data
In most aspects of life, we think of “endless possibilities” as a good thing. But in science, the wealth of possibilities can be daunting. Pharmaceutical researchers typically start with a disease that they want to tackle, find a target, and then try to identify compounds which interact with that target. There may be around 200-300 drug candidates per project, and synthesis planning and execution usually takes four to eight weeks per drug candidate. This process requires many chemists and a huge, costly effort to proceed.
However, we are finding that data-driven approaches can accelerate and improve the time-consuming process of drug discovery. The right combination of high-quality data and machine learning technology helps researchers focus on the most promising of those endless possibilities.
For chemists looking to make a novel substance, predictive retrosynthesis is an attractive new technology to help generate ideas and do synthesis planning and design. Reaxys has partnered with Pending.AI to create a predictive retrosynthesis tool that integrates each company’s expertise – data in the Reaxys chemistry reaction knowledge base and deep learning from Pending.AI – to support the search for synthetic routes to take to the lab.
In this recent webinar, scientists behind the tool went into greater depth about the processes that it performs to predict new synthesis routes, saving users time and giving them confidence that they are making informed decisions.
They also showed how the tool, built with researcher needs in mind, allows users to visualize the following:
- all the published routes together with the predicted routes in one view
- commercially available starting materials needed for synthesis
- the number of steps in the route
- an interactive and graphical route overview to assess the complexity
- literature references of published routes that informed predicted routes
The retrosynthesis tool also enables users to export the routes in multiple formats (PDF, DOC, RD, SD) for easy sharing and adding to electronic lab notebooks.
Ultimately, the Reaxys retrosynthesis tool is powerful because it brings together three crucial elements:
- A foundation of high-quality and well-connected data via Reaxys
- Smart algorithms from Predictive.AI
- An excellent user experience
To learn more about how the tool works, including a helpful demo on how to use it, watch the webinar here.
R&D Solutions for Pharma & Life SciencesWe're happy to discuss your needs and show you how Elsevier's Solution can help.
Director, Corporate Markets Marketing, Elsevier
- At the leading edge of an innovation hotspot: predictive reaction modeling
- Technology and empathy help make remote working a success for Elsevier
- Confronting the Problem of AI Bias
- An Update on Global Availability of COVID-19 Vaccines from Editor of The Lancet
- AI Disease Modeling Supports Precision Medicine for Cancer