AI-informed AdaptiveFlow redefines large-scale cloud computing for drug discovery
Researchers today announced AdaptiveFlow, an AI-informed platform that can virtually screen billions of drug-like molecules with a 1,000-fold reduction in computational costs over existing methods. Developed and validated by scientists from St. Jude Children's Research Hospital, University of Pavia, Dana Farber Cancer Institute and Harvard Medical School, AdaptiveFlow allows prohibitively…
AdaptiveFlow, an AI-driven platform for large-scale virtual drug screening, has been developed by researchers at St. Jude Children's Research Hospital, the University of Pavia, Dana Farber Cancer Institute, and Harvard Medical School. This platform enables the virtual screening of billions of drug-like molecules with a 1,000-fold reduction in computational costs compared to existing methods. AdaptiveFlow was published in Nature Biotechnology and is now available as an open-source tool.
The platform's framework demonstrated linear scaling up to 5.6 million CPUs, allowing for the screening of billions of molecules without a loss of efficiency. It was used to identify potent inhibitors for cancer-related targets, including PARP1, a well-known anticancer target, and FSP1, an emerging target involved in cell death and cancer cell survival.
AdaptiveFlow achieves its efficiency through an 18-dimensional grid that represents molecular properties, and a machine-learning classification model that prioritizes chemically diverse molecules for deeper screening. This approach leads to the identification of compounds with predicted binding affinities comparable to those of approved drugs. By democratizing access to ultra-large molecule libraries, AdaptiveFlow aims to accelerate the development of more effective therapeutics for challenging targets.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
