biostudies-literatureUnknownStewart RNHGRI NIH HHSWellcome TrustNIGMS NIH HHSDisruption of protein-protein interactions (PPIs) is a major mechanism of a variant's deleterious effect. Computational tools are needed to assess such variants at scale, yet existing predictors rarely consider loss of specific interactions, particularly when variants perturb binding interfaces without significantly affecting protein stability. To address this problem, we present MutPred-PPI, a graph attention network that predicts interaction-specific (edgetic) effects of missense variants by operating on AlphaFold3-based protein complex contact graphs with protein language model embeddings imposed upon nodes. We systematically evaluated our model with stringent group cross-validation as well as benchmark data recently collected within the IGVF Consortium. MutPred-PPI outperformed all basbioRxiv : the preprint server for biology2025.12.20.695738https://www.ebi.ac.uk/biostudies/studies/S-EPMC12767517biostudies-literaturePredicting interaction-specific protein-protein interaction perturbations by missense variants with MutPred-PPI.PMC12767517U01 HG012022UM1 HG011989R01 GM145937Calderwood MACooper DNCoppin GTixhon MMort MRadivojac PStewart RVidal MSpirohn-Fitzgerald KLaval FHao TfalsePredicting interaction-specific protein-protein interaction perturbations by missense variants with MutPred-PPI.Disruption of protein-protein interactions (PPIs) is a major mechanism of a variant's deleterious effect. Computational tools are needed to assess such variants at scale, yet existing predictors rarely consider loss of specific interactions, particularly when variants perturb binding interfaces without significantly affecting protein stability. To address this problem, we present MutPred-PPI, a graph attention network that predicts interaction-specific (edgetic) effects of missense variants by operating on AlphaFold3-based protein complex contact graphs with protein language model embeddings imposed upon nodes. We systematically evaluated our model with stringent group cross-validation as well as benchmark data recently collected within the IGVF Consortium. MutPred-PPI outperformed all bas2025-01-01T00:00:00Z2025 Dec2026-06-03T03:17:11.691Z2026-06-03T03:10:48.863ZS-EPMC127675174149760010.64898/2025.12.20.695738