PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models

Abstract Protein-protein interactions (PPIs) refer to the phenomenon of protein binding through various types of bonds to execute biological functions. These interactions are critical for understanding biological mechanisms and drug research. Among these, the protein binding interface is a critical...

Full description

Saved in:
Bibliographic Details
Main Authors: Jun Xie, Youli Zhang, Ziyang Wang, Xiaocheng Jin, Xiaoli Lu, Shengxiang Ge, Xiaoping Min
Format: Article
Language:English
Published: BMC 2025-04-01
Series:BMC Bioinformatics
Subjects:
Online Access:https://doi.org/10.1186/s12859-025-06123-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850042466220113920
author Jun Xie
Youli Zhang
Ziyang Wang
Xiaocheng Jin
Xiaoli Lu
Shengxiang Ge
Xiaoping Min
author_facet Jun Xie
Youli Zhang
Ziyang Wang
Xiaocheng Jin
Xiaoli Lu
Shengxiang Ge
Xiaoping Min
author_sort Jun Xie
collection DOAJ
description Abstract Protein-protein interactions (PPIs) refer to the phenomenon of protein binding through various types of bonds to execute biological functions. These interactions are critical for understanding biological mechanisms and drug research. Among these, the protein binding interface is a critical region involved in protein-protein interactions, particularly the hotspot residues on it that play a key role in protein interactions. Current deep learning methods trained on large-scale data can characterize proteins to a certain extent, but they often struggle to adequately capture information about protein binding interfaces. To address this limitation, we propose the PPI-Graphomer module, which integrates pretrained features from large-scale language models and inverse folding models. This approach enhances the characterization of protein binding interfaces by defining edge relationships and interface masks on the basis of molecular interaction information. Our model outperforms existing methods across multiple benchmark datasets and demonstrates strong generalization capabilities.
format Article
id doaj-art-11532669cd364d68b60d8236b93ae5d0
institution DOAJ
issn 1471-2105
language English
publishDate 2025-04-01
publisher BMC
record_format Article
series BMC Bioinformatics
spelling doaj-art-11532669cd364d68b60d8236b93ae5d02025-08-20T02:55:32ZengBMCBMC Bioinformatics1471-21052025-04-0126111510.1186/s12859-025-06123-2PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer modelsJun Xie0Youli Zhang1Ziyang Wang2Xiaocheng Jin3Xiaoli Lu4Shengxiang Ge5Xiaoping Min6Institute of Artificial Intelligence, School of Informatic, Xiamen UniversityNational Institute of Diagnostics and Vaccine Development in Infectious Diseases, School of Public Health, Xiamen UniversityInstitute of Artificial Intelligence, School of Informatic, Xiamen UniversityNational Institute of Diagnostics and Vaccine Development in Infectious Diseases, School of Public Health, Xiamen UniversityInformation and Networking Center, Xiamen UniversityNational Institute of Diagnostics and Vaccine Development in Infectious Diseases, School of Public Health, Xiamen UniversityInstitute of Artificial Intelligence, School of Informatic, Xiamen UniversityAbstract Protein-protein interactions (PPIs) refer to the phenomenon of protein binding through various types of bonds to execute biological functions. These interactions are critical for understanding biological mechanisms and drug research. Among these, the protein binding interface is a critical region involved in protein-protein interactions, particularly the hotspot residues on it that play a key role in protein interactions. Current deep learning methods trained on large-scale data can characterize proteins to a certain extent, but they often struggle to adequately capture information about protein binding interfaces. To address this limitation, we propose the PPI-Graphomer module, which integrates pretrained features from large-scale language models and inverse folding models. This approach enhances the characterization of protein binding interfaces by defining edge relationships and interface masks on the basis of molecular interaction information. Our model outperforms existing methods across multiple benchmark datasets and demonstrates strong generalization capabilities.https://doi.org/10.1186/s12859-025-06123-2Bind affinity predictionESMGraph transformer
spellingShingle Jun Xie
Youli Zhang
Ziyang Wang
Xiaocheng Jin
Xiaoli Lu
Shengxiang Ge
Xiaoping Min
PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
BMC Bioinformatics
Bind affinity prediction
ESM
Graph transformer
title PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
title_full PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
title_fullStr PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
title_full_unstemmed PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
title_short PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models
title_sort ppi graphomer enhanced protein protein affinity prediction using pretrained and graph transformer models
topic Bind affinity prediction
ESM
Graph transformer
url https://doi.org/10.1186/s12859-025-06123-2
work_keys_str_mv AT junxie ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT youlizhang ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT ziyangwang ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT xiaochengjin ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT xiaolilu ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT shengxiangge ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels
AT xiaopingmin ppigraphomerenhancedproteinproteinaffinitypredictionusingpretrainedandgraphtransformermodels