EUVIMEDEUVIMED entdecken
Crossref

Evolutionary-scale prediction of atomic-level protein structure with a language model

Zeming Lin iD, Halil Akin iD, Roshan Rao iD, Brian Hie iD, Zhongkai Zhu, Wenting Lu, Nikita Smetanin, Robert Verkuil iD, Ori Kabeli iD, Yaniv Shmueli iD, Allan Dos Santos Costa iD, Maryam Fazel-Zarandi und 3 weitere

Science · 17.03.2023 · 379 (6637) · 1123-1130

DOI: 10.1126/science.ade2574 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science, erschienen am 17.03.2023. Verfasst von Zeming Lin und 14 weiteren Autor:innen. Beteiligte Einrichtung: FAIR, Meta AI, New York, NY, USA. Hauptthemen laut MEDLINE: Machine Learning, Proteins, Evolution, Molecular, Sequence Analysis, Protein. Die Arbeit wurde bisher 6271-mal zitiert (Crossref) und nennt 74 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent advances in machine learning have leveraged evolutionary information in multiple sequence alignments to predict protein structure. We demonstrate direct inference of full atomic-level protein structure from primary sequence using a large language model. As language models of protein sequences are scaled up to 15 billion parameters, an atomic-resolution picture of protein structure emerges in the learned representations. This results in an order-of-magnitude acceleration of high-resolution structure prediction, which enables large-scale structural characterization of metagenomic proteins. We apply this capability to construct the ESM Metagenomic Atlas by predicting structures for >617 million metagenomic protein sequences, including >225 million that are predicted with high confidence, which gives a view into the vast breadth and diversity of natural proteins.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningProteinsEvolution, MolecularSequence Analysis, ProteinAmino Acid SequenceProtein Conformation

Substanzen

Proteins

Autor:innen und Institutionen

  • FAIR, Meta AI, New York, NY, USA.
  • New York University, New York, NY, USA.
  • Stanford University, Palo Alto, CA, USA.
  • Massachusetts Institute of Technology, Cambridge, MA, USA.

ORCID-Profile: Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Robert Verkuil, Ori Kabeli, Yaniv Shmueli, Allan Dos Santos Costa, Tom Sercu, Salvatore Candido, Alexander Rives

Zitieren

APA

Lin, Z., Akin, H., Rao, R., Hie, B., Zhu, Z., Lu, W., Smetanin, N., Verkuil, R., Kabeli, O., Shmueli, Y., Dos Santos Costa, A., Fazel-Zarandi, M., Sercu, T., Candido, S., & Rives, A. (2023). Evolutionary-scale prediction of atomic-level protein structure with a language model. Science, 379(6637), 1123-1130. https://doi.org/10.1126/science.ade2574

Vancouver

Lin Z, Akin H, Rao R, Hie B, Zhu Z, Lu W, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science. 2023;379(6637):1123-1130. doi:10.1126/science.ade2574

BibTeX
@article{Lin2023,
  author = {Lin, Zeming and Akin, Halil and Rao, Roshan and Hie, Brian and Zhu, Zhongkai and Lu, Wenting and Smetanin, Nikita and Verkuil, Robert and Kabeli, Ori and Shmueli, Yaniv and Dos Santos Costa, Allan and Fazel-Zarandi, Maryam and Sercu, Tom and Candido, Salvatore and Rives, Alexander},
  title = {Evolutionary-scale prediction of atomic-level protein structure with a language model},
  journal = {Science},
  year = {2023},
  volume = {379},
  number = {6637},
  pages = {1123--1130},
  doi = {10.1126/science.ade2574}
}
Kostenlos starten →