Deep learning models for unbiased sequence-based PPI prediction plateau at an accuracy of 0.65
Open Access
- 1 July 2025
- journal article
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (Supplement) , i590-i598
Abstract
Motivation: As most proteins interact with other proteins to perform their respective functions, methods to computationally predict these interactions have been developed. However, flawed evaluation schemes and data leakage in test sets have obscured the fact that sequence-based protein–protein interaction (PPI) prediction is still an open problem. Recently, methods achieving better-than-random performance on leakage-reduced PPI data have been proposed. Results: Here, we show that the use of ESM-2 protein embeddings explains this performance gain irrespective of model architecture. We compared the performance of models with varying complexity, per-protein, and per-token embeddings, as well as the influence of self- or cross-attention, where all models plateaued at an accuracy of 0.65. Moreover, we show that the tested sequence-based models cannot implicitly learn a contact map as an intermediate layer. These results imply that other input types, such as structure, might be necessary for producing reliable PPI predictions. Availability and implementation: All code for models and execution of the models is available at https://github.com/daisybio/PPI_prediction_study. Python version 3.8.18 and PyTorch version 2.1.1 were used for this study. The environment containing the versions of all other packages used can be found in the GitHub repository. The used data are available at https://doi.org/10.6084/m9.figshare.21591618.v3.Keywords
Funding Information
- German Federal Ministry of Education and Research (031L0309A)
- Klaus Tschira Stiftung (00.003.2024)
- German Federal Ministry of Education and Research (031L0305A, DROP2AI)
- Deutsche Forschungsgemeinschaft
- German Research Foundation (422216132)
This publication has 21 references indexed in Scilit:
- TUnA: an uncertainty-aware transformer model for sequence-based protein–protein interaction predictionBriefings in Bioinformatics, 2024
- Democratizing protein language models with parameter-efficient fine-tuningProceedings of the National Academy of Sciences, 2024
- ProteinCLIP: enhancing protein language models with natural languagebioRxiv, 2024
- Cracking the black box of deep sequence-based protein–protein interaction predictionBriefings in Bioinformatics, 2024
- D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein-protein interactionsCell Systems, 2021
- Highly accurate protein structure prediction with AlphaFoldNature, 2021
- HIPPIE v2.0: enhancing meaningfulness and reliability of protein–protein interaction networksNucleic Acids Research, 2016
- Current Experimental Methods for Characterizing Protein–Protein InteractionsChemMedChem, 2016
- An evaluation of in vitro protein–protein interaction techniques: Assessing contaminating background proteinsProteomics, 2006
- The Protein Data BankNucleic Acids Research, 2000