Topology-driven negative sampling enhances generalizability in protein–protein interaction prediction
Open Access
- 7 April 2025
- journal article
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (5)
Abstract
Motivation: Unraveling the human interactome to uncover disease-specific patterns and discover drug targets hinges on accurate protein–protein interaction (PPI) predictions. However, challenges persist in machine learning (ML) models due to a scarcity of quality hard negative samples, shortcut learning, and limited generalizability to novel proteins. Results: In this study, we introduce a novel approach for strategic sampling of protein–protein noninteractions (PPNIs) by leveraging higher-order network characteristics that capture the inherent complementarity-driven mechanisms of PPIs. Next, we introduce Unsupervised Pre-training of Node Attributes tuned for PPI (UPNA-PPI), a high throughput sequence-to-function ML pipeline, integrating unsupervised pre-training in protein representation learning with Topological PPNI (TPPNI) samples, capable of efficiently screening billions of interactions. By using our TPPNI in training the UPNA-PPI model, we improve PPI prediction generalizability and interpretability, particularly in identifying potential binding sites locations on amino acid sequences, strengthening the prioritization of screening assays and facilitating the transferability of ML predictions across protein families and homodimers. UPNA-PPI establishes the foundation for a fundamental negative sampling methodology in graph machine learning by integrating insights from network topology. Availability and implementation: Code and UPNA-PPI predictions are freely available at https://github.com/alxndgb/UPNA-PPI.Keywords
All Related Versions
Funding Information
- Department of Bioinformatics and Data Sciences at Alexion AstraZeneca Rare Disease, Boston, USA
This publication has 109 references indexed in Scilit:
- Learning to predict protein–protein interactions from protein sequencesBioinformatics, 2003
- Convergent Mechanisms for Recognition of Divergent Cytokines by the Shared Signaling Receptor gp130Molecular Cell, 2003
- The yeast two‐hybrid system for identifying protein–protein interactionsThe Journal of Pathology, 2002
- The average distances in random graphs with given expected degreesProceedings of the National Academy of Sciences, 2002
- Protein–protein interactions: mechanisms and modification by drugsJournal of Molecular Recognition, 2002
- Specificity and Stability in Topology of Protein NetworksScience, 2002
- The Molecular Architecture of Odor and Pheromone Sensing in MammalsCell, 2000
- Long Short-Term MemoryNeural Computation, 1997
- The SWISS-PROT protein sequence data bank and its new supplement TREMBLNucleic Acids Research, 1996
- A simple method for displaying the hydropathic character of a proteinJournal of Molecular Biology, 1982