KGOT: Unified Knowledge Graph and Optimal Transport Pseudo-Labeling for Molecule-Protein Interaction Prediction

TL;DR

KGOT tackles MPI prediction under data scarcity by unifying a large multimodal knowledge graph with an optimal-transport–based pseudo-labeling framework. A pretrained Uni-Mol backbone extracts 3D-aware molecule/protein embeddings, while molecular similarity regularizes the transport-based pseudo-labeling that expands training signals beyond sparse labels. The model is trained with a multi-objective loss that combines KG structure and pseudo-label alignment, achieving state-of-the-art zero-shot performance on DUD-E and LIT-PCBA and enhancing KG link prediction across multiple embedding schemes. This work demonstrates that integrating diverse biological knowledge with OT-guided pseudo-labels provides a scalable, robust approach for data-limited biological prediction tasks and sets a paradigm for multimodal, knowledge-driven foundation models in biomedicine.

Abstract

Predicting molecule-protein interactions (MPIs) is a fundamental task in computational biology, with crucial applications in drug discovery and molecular function annotation. However, existing MPI models face two major challenges. First, the scarcity of labeled molecule-protein pairs significantly limits model performance, as available datasets capture only a small fraction of biological relevant interactions. Second, most methods rely solely on molecular and protein features, ignoring broader biological context such as genes, metabolic pathways, and functional annotations that could provide essential complementary information. To address these limitations, our framework first aggregates diverse biological datasets, including molecular, protein, genes and pathway-level interactions, and then develop an optimal transport-based approach to generate high-quality pseudo-labels for unlabeled molecule-protein pairs, leveraging the underlying distribution of known interactions to guide label assignment. By treating pseudo-labeling as a mechanism for bridging disparate biological modalities, our approach enables the effective use of heterogeneous data to enhance MPI prediction. We evaluate our framework on multiple MPI datasets including virtual screening tasks and protein retrieval tasks, demonstrating substantial improvements over state-of-the-art methods in prediction accuracies and zero shot ability across unseen interactions. Beyond MPI prediction, our approach provides a new paradigm for leveraging diverse biological data sources to tackle problems traditionally constrained by single- or bi-modal learning, paving the way for future advances in computational biology and drug discovery.

Figures (1)

• Figure 1: Overview of KGOT.(A)Knowledge integration(B)Supervised score learning(C)Pseudo-label generation(D)KG augmentation & link prediction