Augmenting DMTA using predictive AI modelling at AstraZeneca

• Published in: Drug discovery today 2024-04, Vol.29 (4), p.103945-103945, Article 103945
• Main Authors: Ghiandoni, Gian Marco, Evertsson, Emma, Riley, David J., Tyrchan, Christian, Rathi, Prakash Chandra
• Format: Article
• Language: English
• Subjects: artificial intelligence; cloud computing; drug discovery; machine learning
• ISSN: 1359-6446; 1878-5832
• DOI: 10.1016/j.drudis.2024.103945

•The evolution of the design-make-test-analyse (DMTA) cycle, mainly within AstraZeneca, but also generalizing in the context of drug discovery, is presented.•The challenges faced by scientists throughout the DMTA cycle and the use of predictive modelling to improve its efficiency in drug discovery are discussed.•We present the Predictive Insight Platform (PIP), a bespoke infrastructure and collection of services for molecular predictive modelling created at AstraZeneca.•The PIP is described from a technical standpoint, aiming to inform the reader and provide suggestions to improve similar systems.•We speculate on the role of predictive modelling in the future of drug R&D. Design-Make-Test-Analyse (DMTA) is the discovery cycle through which molecules are designed, synthesised, and assayed to produce data that in turn are analysed to inform the next iteration. The process is repeated until viable drug candidates are identified, often requiring many cycles before reaching a sweet spot. The advent of artificial intelligence (AI) and cloud computing presents an opportunity to innovate drug discovery to reduce the number of cycles needed to yield a candidate. Here, we present the Predictive Insight Platform (PIP), a cloud-native modelling platform developed at AstraZeneca. The impact of PIP in each step of DMTA, as well as its architecture, integration, and usage, are discussed and used to provide insights into the future of drug discovery.