Predictomes, a classifier-curated database of AlphaFold-modeled protein-protein interactions
Molecular Cell · 2025-03 · 85 (6) · 1216-1232.e5
DOI: 10.1016/j.molcel.2025.01.034 ↗Kurz gefasst
Fachartikel in der Zeitschrift Molecular Cell, 2025-03. Verfasst von Ernst W Schmid und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biological Chemistry & Molecular Pharmacology, Blavatnik Institute, Harvard Medical School, Boston, MA 02115, USA. Hauptthemen laut MEDLINE: Protein Interaction Mapping, Databases, Protein, Proteome, Computational Biology. Die Arbeit wurde bisher 129-mal zitiert (Crossref) und nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Howard Hughes Medical Institute, NHLBI NIH HHS, National Science Foundation.
Abstract (Originalsprache)
Protein-protein interactions (PPIs) are ubiquitous in biology, yet a comprehensive structural characterization of the PPIs underlying cellular processes is lacking. AlphaFold-Multimer (AF-M) has the potential to fill this knowledge gap, but standard AF-M confidence metrics do not reliably separate relevant PPIs from an abundance of false positive predictions. To address this limitation, we used machine learning on curated datasets to train a structure prediction and omics-informed classifier (SPOC) that effectively separates true and false AF-M predictions of PPIs, including in proteome-wide screens. We applied SPOC to an all-by-all matrix of nearly 300 human genome maintenance proteins, generating ∼40,000 predictions that can be viewed at predictomes.org , where users can also score their own predictions with SPOC. High-confidence PPIs discovered using our approach enable hypothesis generation in genome maintenance. Our results provide a framework for interpreting large-scale AF-M screens and help lay the foundation for a proteome-wide structural interactome. • SPOC classifier identifies functional AlphaFold-Multimer structure predictions • Proteome-wide in silico screening with AlphaFold-Multimer enabled by SPOC classifier • Curated database of 40,000 binary structure predictions in genome maintenance • User-friendly website predictomes.org to browse predictions and build hypotheses Schmid and Walter train a classifier that discerns functionally relevant structure predictions in proteome-wide protein-protein interaction (PPI) screens using AlphaFold-Multimer, and they use this confidence metric to curate a database of 40,000 predicted interactions among ∼300 genome maintenance proteins.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Biological Chemistry & Molecular Pharmacology, Blavatnik Institute, Harvard Medical School, Boston, MA 02115, USA.
- Department of Biological Chemistry & Molecular Pharmacology, Blavatnik Institute, Harvard Medical School, Boston, MA 02115, USA; Howard Hughes Medical Institute, Boston, MA 02115, USA. Electronic address: johannes_walte…
Förderung
- Howard Hughes Medical Institute
- NHLBI NIH HHS
- National Science Foundation
- National Institutes of Health
- National Heart Lung and Blood Institute
- American Cancer Society
Zitieren
APA
Schmid, E. W., & Walter, J. C. (2025). Predictomes, a classifier-curated database of AlphaFold-modeled protein-protein interactions. Molecular Cell, 85(6), 1216-1232.e5. https://doi.org/10.1016/j.molcel.2025.01.034
Vancouver
Schmid EW, Walter JC. Predictomes, a classifier-curated database of AlphaFold-modeled protein-protein interactions. Molecular Cell. 2025;85(6):1216-1232.e5. doi:10.1016/j.molcel.2025.01.034
BibTeX
@article{Schmid2025,
author = {Schmid, Ernst W and Walter, Johannes C},
title = {Predictomes, a classifier-curated database of AlphaFold-modeled protein-protein interactions},
journal = {Molecular Cell},
year = {2025},
volume = {85},
number = {6},
pages = {1216--1232.e5},
doi = {10.1016/j.molcel.2025.01.034}
}
EUVIMED