Research group Group Bryant

We aim to create a universal molecular framework, where any molecule can be predicted with AI. With this technology, the design of new molecules will be possible at the click of a button – for any application.

Using the latest AI, it is possible to predict the structure of proteins from sequence information. We have taken this one step further to enable the structure prediction of interacting proteins. With reinforcement learning, we can assemble these interactions into large complexes consisting of up to 30 proteins. Although the problem of protein structure is within our grasp, the interaction partners of proteins are difficult to study. Indeed, not only proteins but many nucleic acids and other biological molecules interact with proteins and so do pharmaceuticals. Using advanced deep learning, we are developing methods to illuminate these interactions. From the trained models, we can extract insights that can be used to design new molecules for different biotechnological applications.

Group members

This research group has no members.

Research projects

There are no research project connections.

Publications

EvoBind: in silico directed evolution of peptide binders with AlphaFold - Binder design BioRxiv preprint this version posted July 23, 2022 Patrick Bryant & Arne Elofsson

Peptide binder design with inverse folding and protein structure prediction - Binder design Nature, 25 October 2023 Patrick Bryant & Arne Elofsson

Structure prediction of protein-ligand complexes from sequence information with Umol - Structure prediction of protein-ligand complexes BioRxiv, posted November 05, 2023 Patrick Bryant, Atharva Kelkar, Andrea Guljas, Cecilia Clementi,Frank Noé

Structure prediction of alternative protein conformations - Structure prediction of alternative protein conformations BioRxiv, posted September 25, 2023 Patrick Bryant

Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile - Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile Patrick Bryant, Frank Noé bioRxiv 2023.07.04.547638;

Predicting the structure of large protein complexes using AlphaFold and Monte Carlo tree search - Bryant, P., Pozzati, G., Zhu, W. et al. Predicting the structure of large protein complexes using AlphaFold and Monte Carlo tree search. Nat Commun 13, 6028 (2022).

Improved prediction of protein-protein interactions using AlphaFold2 - Nature, March, 10, 2022. Bryant, P., Pozzati, G. & Elofsson, A. Improved prediction of protein-protein interactions using AlphaFold2. Nat Commun 13, 1265 (2022).

Towards a structurally resolved human protein interaction network - Burke, D.F., Bryant, P., Barrio-Hernandez, I. et al. Towards a structurally resolved human protein interaction network. Nat Struct Mol Biol 30, 216–225 (2023).

A structural biology community assessment of AlphaFold2 applications - Nature, Publishes: November, 07, 2023 Akdel, M., Pires, D.E.V., Pardo, E.P. et al. A structural biology community assessment of AlphaFold2 applications. Nat Struct Mol Biol 29, 1056–1067 (2022).

News

2024-05-31

Department of Molecular Biosciences, The Wenner-Gren Institute

One step closer to the goal of AI-based drug discovery

One step closer to the goal of AI-based drug discovery: a new AI system can predict the structure of protein-ligand complexes directly from sequence information Abstract: Protein-ligand docking is an established tool in drug discovery and development to narrow down potential therapeutics for experimental testing. However, a high-quality protein structure is required and often the protein is treated as fully or partially rigid. Here we develop an AI system that can predict the fully flexible all-atom structure of protein-ligand complexes directly from sequence information. We find that classical docking methods are still superior, but depend upon having crystal structures of the target protein. In addition to predicting flexible all-atom structures, predicted confidence metrics (plDDT) can be used to select accurate predictions as well as to distinguish between strong and weak binders. The advances presented here suggest that the goal of AI-based drug discovery is one step closer, but there is still a way to go to grasp the complexity of protein-ligand interactions fully. Umol is available at: Github Read the full article here

Calendar

No events available.