MassiveFold: unveiling AlphaFold’s hidden potential with optimized and parallelized massive sampling
Massive sampling in AlphaFold enables access to increased structural diversity. In combination with its efficient confidence ranking, this unlocks elevated modeling capabilities for monomeric structures and foremost for protein assemblies. However, the approach struggles with GPU cost and data storage. Here we introduce MassiveFold, an optimized and customizable version of AlphaFold that runs predictions in parallel, reducing the computing time from several months to hours. MassiveFold is scalable and able to run on anything from a single computer to a large GPU infrastructure, where it can fully benefit from all the computing nodes.
Authors
- Björn Wallner (ORCID: https://orcid.org/0000-0002-3772-8279)
- Claudio Mirabello (ORCID: https://orcid.org/0000-0001-7868-034X)
- Marc F. Lensink (ORCID: https://orcid.org/0000-0003-3957-9470)
- Thibaut Véry
- Guillaume Brysbaert (ORCID: https://orcid.org/0000-0002-6807-6621)
- Nessim Raouraoua
- Christophe Blanchet
Institutions
- Linköping University (SE)
- Centre National de la Recherche Scientifique (FR)
- Inserm (FR)
- Université de Lille (FR)
- Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR)
- Université Paris-Saclay (FR)
- Science for Life Laboratory (SE)
- Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement (FR)
- Unité de Glycobiologie Structurale et Fonctionnelle (FR)
- Institut Français de Bioinformatique (FR)
Publication Details
- Journal
- Nature Computational Science
- Published
- 2024-11-11
- DOI
- https://doi.org/10.1038/s43588-024-00714-4
- Citations
- 48
- Primary Topic
- Protein Structure and Dynamics
- Type
- article
- Field-Weighted Citation Impact
- 11.22
Funders
- Institut Français de Bioinformatique
- Agence Nationale de la Recherche
- Knut och Alice Wallenbergs Stiftelse