ICML 2019 Expo Workshop

July 12, 2020

Expo 2019 Schedule »

Machine Learning for Drug discovery

Sponsor: Insilico Medicine, Inc

Organizers:
Alex Zhebrak (Insilico Medicine)

Presenters:
Alex Zhebrak (Insilico Medicine)

Abstract:

Recent advancements in machine learning methods for chemistry and biology are rapidly transforming the healthcare industry with multiple applications across the field ranging from de-novo drug discovery to genetic variant analysis. Scientific progress in building reliable machine learning architectures shifts the focus from conventional computational approaches towards intelligent analysis and inference with modern statistical models. Growing attention towards applications of machine learning to molecular biology and medicinal chemistry will be potentially disruptive for the healthcare field including pharmaceutical and clinical industry. Novel machine learning techniques applied to the relevant data can bring improvements in developing effective and safe drugs, efficient treatment process and reliable diagnostics.

The main focus of this workshop is applications of machine learning to problems in drug discovery and challenges involving molecules such as molecular representations, predictive models for molecular properties, generative models for drug discovery and biological data, graph representations for molecules and more. While some of the machine learning methods and approaches are applicable right away, many of them require deep domain expertise and understanding of the underlying processes. This workshop is an excellent opportunity for interdisciplinary communication between researchers and industry professionals in the fields of machine learning, chemistry, molecular biology, and computer science. We aim to create a collaborative environment for sharing the experience and recent advancements that will foster the progress in the scientific community. The workshop will feature several invited talks from prominent researchers in the field. The open discussion will focus on current state-of-the-art techniques with their possible applications in industry and will outline the challenges and promising research directions.