Motivation: Despite near-experimental accuracy on single-chain predictions, there is still scope for improvement among multimeric predictions. Methods like AlphaFold-Multimer and FoldDock can accurately model dimers. However, how well these methods fare on larger complexes is still unclear. Further, evaluation methods of the quality of multimeric complexes are not well established.

Results: We analysed the performance of AlphaFold-Multimer on a homology-reduced dataset of homo- and heteromeric protein complexes. We highlight the differences between the pairwise and multi-interface evaluation of chains within a multimer. We describe why certain complexes perform well on one metric (e.g. TM-score) but poorly on another (e.g. DockQ). We propose a new score, Predicted DockQ version 2 (pDockQ2), to estimate the quality of each interface in a multimer. Finally, we modelled protein complexes (from CORUM) and identified two highly confident structures that do not have sequence homology to any existing structures.

Availability and implementation: All scripts, models, and data used to perform the analysis in this study are freely available at https://gitlab.com/ElofssonLab/afm-benchmark.

Cellular functions are governed by molecular machines that assemble through protein-protein interactions. Their atomic details are critical to studying their molecular mechanisms. However, fewer than 5% of hundreds of thousands of human protein interactions have been structurally characterized. Here we test the potential and limitations of recent progress in deep-learning methods using AlphaFold2 to predict structures for 65,484 human protein interactions. We show that experiments can orthogonally confirm higher-confidence models. We identify 3,137 high-confidence models, of which 1,371 have no homology to a known structure. We identify interface residues harboring disease mutations, suggesting potential mechanisms for pathogenic variants. Groups of interface phosphorylation sites show patterns of co-regulation across conditions, suggestive of coordinated tuning of multiple protein interactions as signaling responses. Finally, we provide examples of how the predicted binary complexes can be used to build larger assemblies helping to expand our understanding of human cell biology.

Motivation: In the last decade, de novo protein structure prediction accuracy for individual proteins has improved significantly by utilising deep learning (DL) methods for harvesting the co-evolution information from large multiple sequence alignments (MSAs). The same approach can, in principle, also be used to extract information about evolutionary-based contacts across protein-protein interfaces. However, most earlier studies have not used the latest DL methods for inter-chain contact distance prediction. This article introduces a fold-and-dock method based on predicted residue-residue distances with trRosetta.

Results: The method can simultaneously predict the tertiary and quaternary structure of a protein pair, even when the structures of the monomers are not known. The straightforward application of this method to a standard dataset for protein-protein docking yielded limited success. However, using alternative methods for generating MSAs allowed us to dock accurately significantly more proteins. We also introduced a novel scoring function, PconsDock, that accurately separates 98% of correctly and incorrectly folded and docked proteins. The average performance of the method is comparable to the use of traditional, template-based or ab initio shape-complementarity-only docking methods. Moreover, the results of conventional and fold-and-dock approaches are complementary, and thus a combined docking pipeline could increase overall docking success significantly. This methodology contributed to the best model for one of the CASP14 oligomeric targets, H1065.

Scoring docking solutions is a difficult task, and many methods have been developed for this purpose. In docking, only a handful of the hundreds of thousands of models generated by docking algorithms are acceptable, causing difficulties when developing scoring functions. Today's best scoring functions can significantly increase the number of top-ranked models but still fail for most targets. Here, we examine the possibility of utilizing predicted interface residues to score docking models generated during the scan stage of a docking algorithm. Many methods have been developed to infer the regions of a protein surface that interact with another protein, but most have not been benchmarked using docking algorithms. This study systematically tests different interface prediction methods for scoring >300.000 low-resolution rigid-body template free docking decoys. Overall we find that contact-based interface prediction by BIPSPI is the best method to score docking solutions, with >12% of first ranked docking models being acceptable. Additional experiments indicated precision as a high-importance metric when estimating interface prediction quality, focusing on docking constraints production. Finally, we discussed several limitations for adopting interface predictions as constraints in a docking protocol.