On Tuesday, Google DeepMind and Isomorphic Labs announced a significant progress update on the next iteration of AlphaFold, the deep learning model that in 2020 cracked the problem of predicting a protein’s structure from its amino acid sequence, leapfrogging other approaches in computational biology.
Now, the companies say they have developed the model to predict not just standalone protein structures, but what proteins look like in combination with several classes of molecules, including small molecules and nucleic acids — capabilities that are key for drug development.
“We really needed to make progress on protein-ligand binding and co-folding of ligands with proteins in order to help us design more effective drugs,” said Demis Hassabis, CEO of both DeepMind and Isomorphic Labs, which spun out in 2021 with the goal of applying deep learning approaches to drug discovery. The updated model can also predict protein structures after they’ve been modified. “Basically the new AlphaFold can cope with all of these different things, which is pretty amazing.”
This article is exclusive to STAT+ subscribers
Unlock this article — and get additional analysis of the technologies disrupting health care — by subscribing to STAT+.
Already have an account? Log in
Already have an account? Log in
To submit a correction request, please visit our Contact Us page.
STAT encourages you to share your voice. We welcome your commentary, criticism, and expertise on our subscriber-only platform, STAT+ Connect