Navigating protein landscapes with a machine-learned transferable coarse-grained model
Artikel i vetenskaplig tidskrift, 2025
The most popular and universally predictive protein simulation models employ all-atom molecular dynamics, but they come at extreme computational cost. The development of a universal, computationally efficient coarse-grained (CG) model with similar prediction performance has been a long-standing challenge. By combining recent deep-learning methods with a large and diverse training set of all-atom protein simulations, we here develop a bottom-up CG force field with chemical transferability, which can be used for extrapolative molecular dynamics on new sequences not used during model parameterization. We demonstrate that the model successfully predicts metastable states of folded, unfolded and intermediate structures, the fluctuations of intrinsically disordered proteins and relative folding free energies of protein mutants, while being several orders of magnitude faster than an all-atom model. This showcases the feasibility of a universal and computationally efficient machine-learned CG model for proteins.
Författare
Nicholas E. Charron
Rice University
Zuse Institute Berlin (ZIB)
Freie Universität Berlin
Klara Bonneau
Freie Universität Berlin
Aldo S. Pasos-Trejo
Freie Universität Berlin
Andrea Guljas
Freie Universität Berlin
Yaoyi Chen
Freie Universität Berlin
Felix Musil
Freie Universität Berlin
Jacopo Venturin
Freie Universität Berlin
Daria Gusew
Freie Universität Berlin
Iryna Zaporozhets
Rice University
Freie Universität Berlin
Andreas Kraemer
Freie Universität Berlin
Clark Templeton
Freie Universität Berlin
Atharva Kelkar
Freie Universität Berlin
Aleksander E. P. Durumeric
Freie Universität Berlin
Simon Olsson
Chalmers, Data- och informationsteknik, Data Science och AI
Göteborgs universitet
Adria Perez
Acellera
Universitat Pompeu Fabra
Maciej Majewski
Acellera
Universitat Pompeu Fabra
Brooke E. Husic
Princeton University
Ankit Patel
Rice University
Baylor College of Medicine
Gianni De Fabritiis
Acellera
Universitat Pompeu Fabra
Institucio Catalana de Recerca i Estudis Avancats
Frank Noe
Freie Universität Berlin
Microsoft Corporation
Rice University
Cecilia Clementi
Rice University
Freie Universität Berlin
Nature Chemistry
1755-4330 (ISSN) 1755-4349 (eISSN)
Vol. 17 8 1284-1292Ämneskategorier (SSIF 2025)
Teoretisk kemi
Bioinformatik (beräkningsbiologi)
Biofysik
DOI
10.1038/s41557-025-01874-0
PubMed
40681718