HiHPO: Multimodal Hierarchical Graph Learning for Predicting Missing Protein-Phenotype Associations.

Hancheng Liu W. Zhai Shaojun Wang Shanfeng Zhu

Aug 2026 · IEEE journal of biomedical and health informatics · Vol PP, pp. 1-11 · 0 citations

Medicine

TL;DR

HiHPO is proposed, a multimodal, hierarchy-aware graph contrastive learning framework for predicting protein-phenotype associations that consistently outperforms state-of-the-art methods and preserves semantic relationships among phenotypes and improves generalization to fine-grained and sparsely annotated terms.

Abstract

Understanding protein-phenotype associations is essential for elucidating disease mechanisms and supporting phenotype-driven diagnosis. Although the Human Phenotype Ontology (HPO) provides a standardized framework for phenotypic description, protein-HPO annotations remain incomplete and continuously evolving, posing challenges for robust computational prediction. Existing methods often fail to fully exploit hierarchical phenotype semantics and multimodal biological context, particularly under sparse annotation settings. We propose HiHPO, a multimodal, hierarchy-aware graph contrastive learning framework for predicting protein-phenotype associations. HiHPO integrates complementary biological information from protein-protein interaction networks, gene expression profiles, and protein language model embeddings, while explicitly incorporating HPO hierarchical structure into contrastive representation learning. This design enables the model to preserve semantic relationships among phenotypes and improve generalization to fine-grained and sparsely annotated terms. Extensive evaluations on both random and temporal validation splits demonstrate that HiHPO consistently outperforms state-of-the-art methods, with pronounced advantages on deep HPO terms and newly curated annotations. Additional analyses confirm the contribution of each modality and the robustness of the framework across varying annotation densities. These results highlight the potential of hierarchy-aware multimodal learning for advancing protein-phenotype association prediction and disease-related biomedical research. Code and data are available at https://github.com/ZhuLab-Fudan/HiHPO.

DOI PubMed Semantic Scholar

View source

Similar papers

Preprint Aug 2026

Interpreting Latent Protein Language Model Features with Geometric Annotations

This work introduces an automated and scalable method for interpreting SAE features in ESM-2 by using geometrically inspired features of the protein $\text{C}_\alpha$ backbone, providing a robust method of annotating proteins activated within SAE neurons at a residue level, providing a bridge between mechanistic interp...

S. Setlur, Djordje Mihajlovic, Darrick Lee · 0 citations

Open access Sep 2026

Integrating multi-modal biological knowledge via contrastive dual-view graph learning for phosphorylation site-disease association prediction

CDVGL-PDA’s strong predictive performance across balanced, imbalanced, and low-similarity datasets is demonstrated, while case studies highlight its ability to uncover potential PDAs, illustrating its promise for advancing disease mechanism research and therapeutic target discovery.

Xiang Chen, Li-Jun Quan, Ye-Xuan Mao et al. · Bioinformatics · 0 citations

Open access Sep 2026

LLM-H2G: biomedical semantic-enhanced hypergraph contrastive learning for herb–disease association prediction

Herb–disease association prediction is central to computational traditional medicine, but existing graph and hypergraph methods mainly rely on observed topology and underuse biomedical textual semantics, especially in heterogeneous or sparse association networks. We propose LLM-H2G, a biomedical semantic...

Jun Zhang, Hengchuang Yin, Chao Wu et al. · BMC Bioinformatics · 0 citations

Conference Open access Sep 2026

Structure-Aware Contrastive Learning for Biomedical Embeddings: Bridging the Gap Between HPO and Clinical Literature

A new embedding adaptation procedure is defined whose fine-tuning approach is guided by a novel "Disease-Overlap" similarity measure, which prioritizes clinical co-occurrence of phenotypes over taxonomic distance, and optimizes the embedding space using AnglE Loss to mitigate gradient saturation.

Jose L. Mellina Andreu, Alejandro Cisterna García, Juan A. Botía · Proceedings of the Thirty-Fi... · 0 citations

#graph neural networks Sep 2026

Retrieval-Augmented Residual Graph Neural Network for Protein-Protein Interaction Site Prediction.

RGLLA-PPIS, a novel multimodal prediction model that integrates retrieval-augmented learning and residual GNNs for PPIS identification, outperforms several state-of-the-art baselines in both accuracy and robustness and demonstrates its potential to guide real-world protein engineering tasks.

Jia Mi, Ya-Wen Liu, Chong Chu et al. · IEEE Transactions on Neural... · 0 citations

Open access Aug 2026

RGCNMDA: Pathway-Bridged Relational Graph Learning with Adaptive Multi-View Fusion for Human miRNA-Disease Association Prediction.

RGCNMDA is a leakage-controlled multi-view framework that integrates global latent structure, local profiles and similarities, and pathway context that supports the robustness of leakage-controlled multi-view learning across standard and cold-start evaluation settings.

Chao Hou, Mohamed Kone, Yang Xiang et al. · Bioinformatics · 0 citations