Abstract
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.