International Life Sciences Awards 2020

GHP / 2020 International Life Sciences Awards 23 , Best Chemistry Predictive Analytics Specialists 2020 Information is a powerful thing, but only if it can be interpreted in a way that is useful. The Reaxys teamat Elsevier has built a strong reputation for their work in extraction, curation and analysis of chemistry data. This year they have taken their chemistry information software Reaxys to the next level by developing an AI solution for drug discovery - the best-in-class Predictive Retrosynthesis software that combines reaction data with deep learning algorithm to support innovation in synthetic andmedicinal chemistry. We take a look at the company, and this latest development, to find out more. New answers to old questions can reshape human knowledge. As scientists uncover the truth, it is possible to make important steps towards the most urgent of human crises. This is what the team behind Elsevier does. They do not do this work alone. The last 140 years has seen the firm partner with some of the leading minds and thinkers in the world. Together with other organizations, Elsevier has been responsible for the curation and verification of generations of accumulated scientific knowledge. Their work includes the publication of the seminal reference work, Gray’s Anatomy, but it also looks forward to what potential lies ahead. Of course, the world of today is very different to the world of Sep201056 yesterday. Elsevier’s talented team has set their sights on what technology is able to bring to the world of predictive analytics. Working in collaboration with Pending.AI (PAI), the team has developed an AI solution for drug discovery - the predictive retrosynthesis tool based on deep learning to support innovation in synthetic and medicinal chemistry. This tool started life in Elsevier’s R&D Collaboration Network and is now being integrated into Elsevier’s flagship chemistry solution, Reaxys. By combining Reaxys’ content with cutting-edge AI and machine learning technologies developed by PAI, the team is able to offer solutions unparalleled by competitors in the market. The result is the Reaxys-PAI Predictive Retrosynthesis solution, Elsevier which uses deep neural networks to create a model trained on Reaxys data. The tool has been tested and is being used by the world’s leading pharmaceutical and chemical companies. It serves their researchers need of synthesis planning by providing scientifically robust and innovative synthetic route suggestions. Companies around the globe aim to use digitization of research and AI solutions to make molecules synthesize, cheaper and to increase productivity. Digitalization is the use of information shared through systems integration and connection of siloed data, which plays an integral role in ensuring that speed, productivity and accuracy increases in novel molecule development. Reaxys boosts the digitization journey of companies by enabling the extraction, curation and harmonization of chemistry data from companies’ electronic lab notebooks and combining it with published chemistry data. The synergistic benefit of this combined dataset is unparalleled as it is utilised for: 1. Retraining of the predictive retrosynthesis model so that it leverages not only published knowledge contained in Reaxys data, but also decades of in-house research. This augmented Reaxys- PAI Predictive Retrosynthesis solution better fits the current and future research needs of the company 2. Providing a single point of access to search, retrieve and draw insights from published and in-house knowledge thereby enhancing productivity. This customized cheminformatics solution can help remove data silos and unlock the potential of learning from previously failed as well as successful experiments This innovative solution that Elsevier has been involved in is just another step along the road for the company. Their business is not just in finding solutions and settling but using these solutions to ask even more questions. It’s a way forward that drives innovation, saving researchers considerable time and money. It’s an invaluable tool to the entire scientific community. As such, the Reaxys-PAI Predictive Retrosynthesis solution sets up a whole new strategy for the company which will see the construction of AI for the next generation of scientists. These machine-learning enabled decision support tools will help to bring drugs to market for patients most in need as well as helping to find synthesis routes for novel chemical compounds that may bemore environmentally friendly. For the last 140 years, Elsevier has been an essential resource for the scientific community. By partnering with exciting start-ups such as PAI, and exploring bold new directions, it ensures that this will remain the case for years to come. Company: Elsevier Contact: Dr. Ivan Krstic Email: [email protected] Reaxys© helps boost Design-Make-Test-Analyse cycle