Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 46 min ago
Peer-reviewedCC BY-NCBriefings in bioinformatics

EvoGraph: synergizing evolutionary semantics and tail-aware graph contrastive learning for protein-protein interaction prediction.

  • Yilou Zhang1,2
  • Yuan Zhao1,2
  • Jue Zhou3
  • Haoyi Wang1,2
  • Chengkun Wu1,2
  1. 1College of Computer Science and Technology, National University of Defense Technology, No. 137 Yanwachi Street, Changsha 410073, Hunan, P. R. China.
  2. 2National Key Laboratory of Parallel and Distributed Computing, National University of Defense Technology, No. 137 Yanwachi Street, Changsha 410073, Hunan, P. R. China.
  3. 3Department of Oncology, Xiangya Hospital, Central South University, No. 87 Xiangya Street, Changsha 410008, Hunan, P. R. China.
Read full textPDFDOICite

Abstract

Accurate prediction of multilabel protein-protein interactions is pivotal for deciphering molecular mechanisms and accelerating drug discovery. However, existing methods are often hindered by the inherent long-tailed distribution of interaction types, leading to poor generalization on rare but biologically critical categories. To address this challenge, we propose EvoGraph, a novel framework that synergizes deep evolutionary semantics with tail-aware graph representation learning. Specifically, EvoGraph first leverages the pretrained protein language model to establish a semantic foundation, compensating for sparse signals in rare classes. Driven by the critical need to disentangle complex interaction types and mitigate head-class dominance, we integrate two core mechanisms. To resolve the structural heterogeneity of diverse interaction types, we design a multi-view graph neural network encoder that aggregates type-specific topological contexts. Simultaneously, to counter label imbalance, we introduce a tail-aware contrastive learning module, which regularizes the latent space by clustering minority samples, effectively preventing them from being overshadowed by head classes. Finally, these features are synthesized via a hybrid fusion strategy and optimized with an asymmetric loss. Extensive experiments on four benchmark datasets demonstrate that EvoGraph significantly outperforms state-of-the-art baselines, achieving substantial improvements in identifying tail interaction types without compromising overall performance. The implementation of EvoGraph is available at https://github.com/AlphaZYL/EvoGraph.

Related

  1. Katalyst: knowledge-guided semantic alignment via contrastive learning for enzyme turnover prediction
    Ruilin Li, Senyu Tang, Jiaqi Deng +4
    BioinformaticsPeer-reviewedCC BY 4.0Full textMolecular biology & gene regulationCrossref
  2. GeoEPred: A multimodal structure-aware geometric deep learning framework for Gram-negative bacterial secreted effector prediction with sequence semantics
    Shouzhen Song, Hua Shi, Hongfeng Wu +5
    PLOS Computational BiologyPeer-reviewedCC BY 4.0Full textMolecular biology & gene regulationGenetics & genomicsPLOS
  3. Enhancing Linear B-Cell Epitope Prediction via Evolutionary Scale Modeling (ESM-2) and Hybrid Feature Integration
    Prasun Pal, Sai Jahnavi Pasumarthi, Sandipan Chakraborty
    ChemRxivPreprintCC BY-NC-NDFull textMolecular biology & gene regulationChemRxiv
  4. DeepEnzyKat: Turnover number prediction through enzyme-substrate interaction modeling and adaptive expert ensembles
    Guangran Zhang, Yujing Zhao, Yongwen Zhang +6
    ChemRxivPreprintCC BYMolecular biology & gene regulationChemRxiv
  5. Establishing standardized evaluation protocols for peptide-target interaction prediction: the PTI-TAPE consensus framework eDelphi
    William J. Waldock, Guangshun Wang, Ara Darzi +2
    Frontiers in bioinformaticsPeer-reviewedCC BYFull textMolecular biology & gene regulationEurope PMC

Cite

BibTeXRIS
BibTeX
@article{zhang2026evograph,
  title = {{EvoGraph: synergizing evolutionary semantics and tail-aware graph contrastive learning for protein-protein interaction prediction.}},
  author = {Yilou Zhang and Yuan Zhao and Jue Zhou and Haoyi Wang and Chengkun Wu},
  journal = {Briefings in bioinformatics},
  year = {2026},
  month = sep,
  doi = {10.1093/bib/bbag521},
  url = {https://europepmc.org/article/MED/42960995},
  note = {CC BY-NC},
}
RIS
TY  - JOUR
TI  - EvoGraph: synergizing evolutionary semantics and tail-aware graph contrastive learning for protein-protein interaction prediction.
AU  - Yilou Zhang
AU  - Yuan Zhao
AU  - Jue Zhou
AU  - Haoyi Wang
AU  - Chengkun Wu
PY  - 2026
DA  - 2026/09/01
JO  - Briefings in bioinformatics
DO  - 10.1093/bib/bbag521
UR  - https://europepmc.org/article/MED/42960995
AB  - Accurate prediction of multilabel protein-protein interactions is pivotal for deciphering molecular mechanisms and accelerating drug discovery. However, existing methods are often hindered by the inherent long-tailed distribution of interaction types, leading to poor generalization on rare but biologically critical categories. To address this challenge, we propose EvoGraph, a novel framework that synergizes deep evolutionary semantics with tail-aware graph representation learning. Specifically, EvoGraph first leverages the pretrained protein language model to establish a semantic foundation, compensating for sparse signals in rare classes. Driven by the critical need to disentangle complex interaction types and mitigate head-class dominance, we integrate two core mechanisms. To resolve the structural heterogeneity of diverse interaction types, we design a multi-view graph neural network encoder that aggregates type-specific topological contexts. Simultaneously, to counter label imbalance, we introduce a tail-aware contrastive learning module, which regularizes the latent space by clustering minority samples, effectively preventing them from being overshadowed by head classes. Finally, these features are synthesized via a hybrid fusion strategy and optimized with an asymmetric loss. Extensive experiments on four benchmark datasets demonstrate that EvoGraph significantly outperforms state-of-the-art baselines, achieving substantial improvements in identifying tail interaction types without compromising overall performance. The implementation of EvoGraph is available at https://github.com/AlphaZYL/EvoGraph.
SN  - 1467-5463
SN  - 1477-4054
ER  -