Machine learning boosts drug discovery along an unprecedented scale

This file handout photo provided to AFP on November 16, 2021 by Pfizer shows the making of its experimental Covid-19 antiviral pills, Paxlovid, at its laboratory in Freiburg, Germany – Copyright AFP Dimitar DILKOFF

Increasing the capabilities of virtual screening through the use of machine learning has enabled a 10-fold time reduction in the processing of 1.56 billion drug-like molecules.

This is the outcome of a study conducted by researchers from the University of Eastern Finland, who teamed up with industry and supercomputers to carry out one of the world’s largest virtual drug screens.

The aim was to find novel drug molecules. This is common in pharmaceutical research, where scientists rely on fast computer-aided screening of large compound libraries to identify agents that can block a drug target.

Such a target can be an enzyme that enables a bacterium to withstand antibiotics or a virus to infect its host. The size of these collections of small organic molecules has grown significantly over the past years. This means libraries are growing faster than the speed of the computers needed to process them.

The complication that emerges is where the screening of a modern billion-scale compound library against only a single drug target can take several months or years. That is with conventional computers.

The study drew upon one of Finland’s powerful supercomputers, CSC – IT Center for Science Ltd. – and industrial collaborators from Orion Pharma to study the prospect of machine learning in the speed-up of giga-scale virtual screens.

Before applying artificial intelligence to accelerate the screening, the researchers first established a baseline: In a virtual screening campaign of unprecedented size, 1.56 billion drug-like molecules were evaluated against two pharmacologically relevant targets over almost six months with the help of the supercomputers, and molecular docking.

Docking is a computational technique that fits the small molecules into a binding region of the target and computes a “docking score” to express how well they fit. This way, docking scores were first determined for all 1.56 billion molecules.

The results were compared to a machine learning-boosted screen using HASTEN, a tool developed by Dr Tuomo Kalliokoski from Orion Pharma. HASTEN uses machine learning to learn the properties of molecules and how those properties affect how well the compounds score.

The research has been published in the Journal of Chemical Information and Modeling, here the researchers state their study represented the first rigorous comparison of a machine learning-boosted docking tool with a conventional docking baseline on the giga-scale.

The research is titled: “Machine Learning-Boosted Docking Enables the Efficient Structure-Based Virtual Screening of Giga-Scale Enumerated Chemical Libraries.”