Better, faster, and cheaper drug discovery with machine learning by AstraZeneca

AstraZeneca is finding new and innovative ways to use AI to help solve some of the biggest challenges facing the pharmaceutical industry today. To become better, faster, and cheaper in drug discovery and development, we believe in our AI approaches at AstraZeneca to transform R&D. Please join our session to get an overview of some of the real use-cases where AI is having a genuine impact across the R&D value chain:
* Machine learning to predict compound properties to minimize the number of compounds made and tested
* Methods to identify and improve the safety profile of new drugs as well as reduce the costs and time to bring these to the clinic
* AI approaches for discovering patients responding better to treatment
* Designing Molecules using Recurrent Neural Networks and Reinforcement Learning
Johanna is a safety bioinformatician in the Data Science and AI group within Drug Safety and Metabolism at AstraZeneca. Drug design is a multiparameter optimization problem that requires a fine balance between potency, ADME, and safety. Data science and artificial intelligence are potential methods to both improve the safety profile of new drugs as well as reduce the costs and time to bring these to the clinic, and in her talk, Johanna will exemplify its application within drug safety.
Jesper has a PhD in medicine from the Karolinska Institute, his research was focused on systems biology approaches for coronary artery disease. He joined AstraZeneca in 2011 and is now a Principal Biomedical Informatics Scientist in the Advanced Analytic Centre. In this role, he can both support clinical drug projects directly and work on projects with a more strategic focus. One of Jesper’s current focus areas includes the application of wearable sensors in clinical trials including how analytical techniques which can be applied to understand biological phenomenon and link to established endpoints.
Emma Evertsson is a computational chemist with a PhD in theoretical chemistry from Lund University. At AstraZeneca, she is involved in drug design for the treatment of respiratory diseases. In addition, she is responsible for the computational platform providing predicted property values for real and virtual molecules from machine learning models. This platform is intended to reduce the number of experiments and to accelerate the drug development process.
Esben Jannik Bjerrum wants to battle diseases by inventing novel medicines. To accomplish this, he develops AI and ML-based methods and applies them to drug discovery projects. He obtained a PhD in computational chemistry from the Danish University of Pharmaceutical Sciences in 2008. He has since worked in the interface between IT, drug discovery, and research, both in industry, as a post.doc., and as an independent consultant. Over the years, he has gotten increasingly involved in machine learning and lastly artificial neural networks and deep learning for chemical applications. He joined AstraZeneca in late 2018.
David holds a PhD in Mathematical Statistics from the Chalmers University of Technology in Sweden. He joined AstraZeneca in Gothenburg in 2003 and is now a Statistical Science Director in the Statistical Innovation group within the Advanced Analytic Centre. In that position, he is both involved in directly supporting late phase clinical drug projects and methodological longer-term work (e.g. relating to biomarkers/subgroups, personalized medicine, machine learning, and safety). He is particularly interested in computationally intensive statistics, machine learning, visualisation, and Bayesian approaches, and he is a keen R user. The lead of PSI Special Interest Group on Subgroup Analysis since 2018.
John Kumar is an engineer and data scientist at AstraZeneca. His work is focused on the problem of training and serving machine learning models at scale with portable solutions using a combination of on-premises and commercial cloud resources.
Robin joined AstraZeneca 15 years ago working in IT across R&D incl. late-stage development areas Clinical and Regulatory, and the past 5 years within Early Science. Robin has held various positions incl. developer, architect, project manager, and, now, IT lead for global labs, early science, and Gothenburg at AstraZeneca.
Talk by Johanna Sagemark, Jesper Havsol, Emma Evertsson, Esben Jannik Bjerrum, David Svensson, John Kumar, and Robin Brouwer at the 2019 GAIA Conference held at Lindholmen Conference Centre on 9th of April 2019.

[@chunyuenlau56]

This is so fascinating 👏. The fact that you can individualize treatment as opposed to one size fits all is an absolute game-changer. Well done 👏.

[@ratangaonkar107]

Good topic

[@keqiaoli4617]

May I have a copy of the slides? Thank you