Results From the World's Largest Distributed Prospective Application of Machine Learning to Small-Molecule Hit Discovery

At the American Chemical Society Fall 2020 Virtual Meeting & Expo, Atomwise CEO, Abraham Heifets, was selected to present our work and results. Learn what our team of Atoms have been working on below, and visit Atomwise at ACS Fall 2020 Virtual Meeting & Expo for other Atomwise presentations during #ACSFall2020. 

Abraham Heifets, PhD 
Atomwise

Title: Results From the World's Largest Distributed Prospective Application of Machine Learning to Small-Molecule Hit Discovery

Division: Computers in Chemistry

Abstract

We report here the results from more than 100 prospective applications of machine-learning to small-molecule hit-discovery. Physical high-throughput screens are no longer a viable option for early-stage drug discovery because synthesized-on-demand ultra-large screening libraries (ULSL), which cover tens of billions of molecules, are too big for brute-force physical screening. Effectively exploring ULSL requires extraordinarily accurate algorithms, to avoid an overwhelmingly large number of false-positives because of the magnitude of the explored space. While machine learning tools promise to revolutionize drug discovery in precisely this manner, it has been extremely challenging to evaluate their actual prospective performance. We showed that the excellent performance reported on standard medicinal chemistry benchmarks are likely due to weaknesses in the design of these benchmarks, rather than the generalizability of the algorithms. Therefore, we have been running the largest assessment of machine learning for hit discovery in history, comprising over 700 projects on 575 targets with 330 universities in 40 countries. Our first 100+ results empirically demonstrate the viability of our machine-learning technology as a replacement for high-throughput screens.

Speaker Presentation

View on-demand video

Join our team