Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile
PLOS Computational Biology · 25.07.2024 · 20 (7) · e1012253
DOI: 10.1371/journal.pcbi.1012253 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 25.07.2024. Verfasst von Patrick Bryant und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics and Informatics, Freie Universität Berlin, Germany. Hauptthemen laut MEDLINE: Computational Biology, Proteins. Die Arbeit wurde bisher 45-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Commission, MATH+ excellence cluster, Wallenberg Data Driven Life Science Program.
Abstract (Originalsprache)
Structure prediction of protein complexes has improved significantly with AlphaFold2 and AlphaFold-multimer (AFM), but only 60% of dimers are accurately predicted. Here, we learn a bias to the MSA representation that improves the predictions by performing gradient descent through the AFM network. We demonstrate the performance on seven difficult targets from CASP15 and increase the average MMscore to 0.76 compared to 0.63 with AFM. We evaluate the procedure on 487 protein complexes where AFM fails and obtain an increased success rate (MMscore>0.75) of 33% on these difficult targets. Our protocol, AFProfile, provides a way to direct predictions towards a defined target function guided by the MSA. We expect gradient descent over the MSA to be useful for different tasks.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Mathematics and Informatics, Freie Universität Berlin, Germany.
- The Department of Molecular Biosciences, The Wenner-Gren Institute, Stockholm University, Stockholm, Sweden.
- Science for Life Laboratory, Solna, Sweden.
- Microsoft Research AI4Science, Berlin, Germany.
ORCID-Profile: Patrick Bryant
Förderung
- European Commission
- MATH+ excellence cluster
- Wallenberg Data Driven Life Science Program
- FP7 Ideas: European Research Council
- Berlin Mathematics Research Center MATH+
- Deutsche Forschungsgemeinschaft
- SciLifeLab & Wallenberg Data Driven Life Science Program
- Zentrum für Informationsdienste und Hochleistungsrechnen, Technische Universität Dresden
Zitieren
APA
Bryant, P., & Noé, F. (2024). Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile. PLOS Computational Biology, 20(7), e1012253. https://doi.org/10.1371/journal.pcbi.1012253
Vancouver
Bryant P, Noé F. Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile. PLOS Computational Biology. 2024;20(7):e1012253. doi:10.1371/journal.pcbi.1012253
BibTeX
@article{Bryant2024,
author = {Bryant, Patrick and Noé, Frank},
title = {Improved protein complex prediction with AlphaFold-multimer by denoising the MSA profile},
journal = {PLOS Computational Biology},
year = {2024},
volume = {20},
number = {7},
pages = {e1012253},
doi = {10.1371/journal.pcbi.1012253}
}
EUVIMED