PRING: Rethinking Protein-Protein Interaction Prediction from Pairs to Graphs

2025-07-07

AI Summary

Existing PPI prediction methods predominantly focus on pairwise interaction classification, neglecting their capacity to reconstruct biologically plausible protein–protein interaction (PPI) networks. To address this gap, we introduce PRING—the first graph-level PPI prediction benchmark—comprising a high-quality, multi-species dataset of 21,484 proteins and 186,818 interactions. PRING establishes a novel dual-dimensional evaluation framework integrating topological structure and functional semantics, enabling assessment across network reconstruction, module detection, and disease mechanism inference. We systematically evaluate four major method classes—sequence similarity, naïve sequence models, protein language models, and structure-based predictors—under strict data-leakage mitigation protocols. Results reveal substantial limitations in current models’ ability to recover both the structural integrity and functional coherence of ground-truth PPI networks. PRING provides a reproducible, multi-level evaluation standard for PPI model development, advancing accurate functional annotation and mechanistic interpretation.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Deep learning-based computational methods have achieved promising results in predicting protein-protein interactions (PPIs). However, existing benchmarks predominantly focus on isolated pairwise evaluations, overlooking a model's capability to reconstruct biologically meaningful PPI networks, which is crucial for biology research. To address this gap, we introduce PRING, the first comprehensive benchmark that evaluates protein-protein interaction prediction from a graph-level perspective. PRING curates a high-quality, multi-species PPI network dataset comprising 21,484 proteins and 186,818 interactions, with well-designed strategies to address both data redundancy and leakage. Building on this golden-standard dataset, we establish two complementary evaluation paradigms: (1) topology-oriented tasks, which assess intra and cross-species PPI network construction, and (2) function-oriented tasks, including protein complex pathway prediction, GO module analysis, and essential protein justification. These evaluations not only reflect the model's capability to understand the network topology but also facilitate protein function annotation, biological module detection, and even disease mechanism analysis. Extensive experiments on four representative model categories, consisting of sequence similarity-based, naive sequence-based, protein language model-based, and structure-based approaches, demonstrate that current PPI models have potential limitations in recovering both structural and functional properties of PPI networks, highlighting the gap in supporting real-world biological applications. We believe PRING provides a reliable platform to guide the development of more effective PPI prediction models for the community. The dataset and source code of PRING are available at https://github.com/SophieSarceau/PRING.

Problem

• Evaluates PPI prediction from graph-level, not just pairs
• Assesses network topology and functional biological tasks
• Identifies limitations in current PPI models for real-world use

Innovation

• PRING introduces graph-level PPI prediction benchmark
• Uses multi-species dataset with redundancy control
• Combines topology and function-oriented evaluation paradigms