EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ManyFold: an efficient and flexible library for training and validating protein folding models

Amelia Villegas-Morcillo iD, Louis Robinson, Arthur Flajolet, Thomas D Barrett

Bioinformatics · 2023

DOI: 10.1093/bioinformatics/btac773 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2023. Verfasst von Amelia Villegas-Morcillo und 3 weiteren Autor:innen. Beteiligte Einrichtung: InstaDeep, London W2 1AY, UK. Hauptthemen laut MEDLINE: Software, Proteins. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 8 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: TPUs from Google's TPU Research Cloud, TPUs from Google’s TPU Research Cloud.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Summary ManyFold is a flexible library for protein structure prediction with deep learning that (i) supports models that use both multiple sequence alignments (MSAs) and protein language model (pLM) embedding as inputs, (ii) allows inference of existing models (AlphaFold and OpenFold), (iii) is fully trainable, allowing for both fine-tuning and the training of new models from scratch and (iv) is written in Jax to support efficient batched operation in distributed settings. A proof-of-concept pLM-based model, pLMFold, is trained from scratch to obtain reasonable results with reduced computational overheads in comparison to AlphaFold. Availability and implementation The source code for ManyFold, the validation dataset and a small sample of training data are available at https://github.com/instadeepai/manyfold. Supplementary information Supplementary data are available at Bioinformatics online.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SoftwareProteinsProtein FoldingSequence AlignmentLanguage

Substanzen

Proteins

Autor:innen und Institutionen

  • InstaDeep, London W2 1AY, UK.
  • Department of Signal Theory, Telematics and Communications, University of Granada, Granada 18071, Spain.

ORCID-Profile: Amelia Villegas-Morcillo

Förderung

  • TPUs from Google's TPU Research Cloud
  • TPUs from Google’s TPU Research Cloud

Zitieren

APA

Villegas-Morcillo, A., Robinson, L., Flajolet, A., & Barrett, T. D. (2023). ManyFold: an efficient and flexible library for training and validating protein folding models. Bioinformatics. https://doi.org/10.1093/bioinformatics/btac773

Vancouver

Villegas-Morcillo A, Robinson L, Flajolet A, Barrett TD. ManyFold: an efficient and flexible library for training and validating protein folding models. Bioinformatics. 2023. doi:10.1093/bioinformatics/btac773

BibTeX
@article{VillegasMorcillo2023,
  author = {Villegas-Morcillo, Amelia and Robinson, Louis and Flajolet, Arthur and Barrett, Thomas D},
  title = {ManyFold: an efficient and flexible library for training and validating protein folding models},
  journal = {Bioinformatics},
  year = {2023},
  doi = {10.1093/bioinformatics/btac773}
}
Kostenlos starten →