Inferred global dense residue transition graphs from primary structure sequences enable protein interaction prediction via directed graph convolutional neural networks

2025-10-15

AI Summary

Accurate prediction of protein–protein interactions (PPIs) is critical for functional annotation and drug discovery. Existing methods either rely on computationally expensive 3D structure modeling or shallow sequence embeddings, incurring high resource overhead. To address this, we propose ProtGram-DirectGCN: a novel framework that first constructs the first global directed residue transition graph (an n-gram graph) directly from protein sequences; then introduces DirectGCN—a customized graph convolutional network designed for dense, heterophilic, directed graphs—incorporating path-specific transformations and learnable gating mechanisms to enable multi-path information propagation and attention-based pooling. Crucially, our approach requires no 3D structural data, drastically reducing computational demands while maintaining strong robustness and generalization under few-shot settings. On standard graph benchmarks, it matches state-of-the-art performance; in PPI prediction, it achieves stable, superior accuracy over leading sequence-based models.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Introduction Accurate prediction of protein-protein interactions (PPIs) is crucial for understanding cellular functions and advancing drug development. Existing in-silico methods use direct sequence embeddings from Protein Language Models (PLMs). Others use Graph Neural Networks (GNNs) for 3D protein structures. This study explores less computationally intensive alternatives. We introduce a novel framework for downstream PPI prediction through link prediction. Methods We introduce a two-stage graph representation learning framework, ProtGram-DirectGCN. First, we developed ProtGram. This approach models a protein's primary structure as a hierarchy of globally inferred n-gram graphs. In these graphs, residue transition probabilities define edge weights. Each edge connects a pair of residues in a directed graph. The probabilities are aggregated from a large corpus of sequences. Second, we propose DirectGCN, a custom directed graph convolutional neural network. This model features a unique convolutional layer. It processes information through separate path-specific transformations: incoming, outgoing, and undirected. A shared transformation is also applied. These paths are combined via a learnable gating mechanism. We apply DirectGCN to ProtGram graphs to learn residue-level embeddings. These embeddings are pooled via attention to generate protein-level embeddings for prediction. Results We first established the efficacy of DirectGCN on standard node classification benchmarks. Its performance matches established methods on general datasets. The model excels at complex, directed graphs with dense, heterophilic structures. When applied to PPI prediction, the full ProtGram-DirectGCN framework delivers robust predictive power. This strong performance holds even with limited training data.

Problem

• Predicting protein-protein interactions using sequence-based graphs
• Developing directed graph neural networks for residue transitions
• Overcoming computational intensity of existing protein interaction methods

Innovation

• Global residue transition graphs from sequences
• Directed graph convolutional network with path-specific transformations
• Attention-pooled residue embeddings for protein interaction prediction