A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery

Abstract Discovering therapeutic molecules requires the integration of both phenotype‐based drug discovery (PDD) and target‐based drug discovery (TDD). However, this integration remains challenging due to the inherent heterogeneity, noise, and bias present in biomedical data. In this study, Knowledg...

Full description

Saved in:
Bibliographic Details
Main Authors: Qing Ye, Yundian Zeng, Linlong Jiang, Yu Kang, Peichen Pan, Jiming Chen, Yafeng Deng, Haitao Zhao, Shibo He, Tingjun Hou, Chang‐Yu Hsieh
Format: Article
Language:English
Published: Wiley 2025-04-01
Series:Advanced Science
Subjects:
Online Access:https://doi.org/10.1002/advs.202412402
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850155641123897344
author Qing Ye
Yundian Zeng
Linlong Jiang
Yu Kang
Peichen Pan
Jiming Chen
Yafeng Deng
Haitao Zhao
Shibo He
Tingjun Hou
Chang‐Yu Hsieh
author_facet Qing Ye
Yundian Zeng
Linlong Jiang
Yu Kang
Peichen Pan
Jiming Chen
Yafeng Deng
Haitao Zhao
Shibo He
Tingjun Hou
Chang‐Yu Hsieh
author_sort Qing Ye
collection DOAJ
description Abstract Discovering therapeutic molecules requires the integration of both phenotype‐based drug discovery (PDD) and target‐based drug discovery (TDD). However, this integration remains challenging due to the inherent heterogeneity, noise, and bias present in biomedical data. In this study, Knowledge‐Guided Drug Relational Predictor (KGDRP), a graph representation learning approach is developed that effectively integrates multimodal biomedical data, including network data containing biological system information, gene expression data, and sequence data that incorporates chemical molecular structures, all within a heterogeneous graph (HG) structure. By incorporating biomedical HG (BioHG) into a heterogeneous graph neural network (HGNN)‐based architecture, KGDRP exhibits a remarkable 12% improvement compared to previous methods in real‐world screening scenarios. Notably, the biology‐informed representation, derived from KGDRP, significantly enhance target prioritization by 26% in drug target discovery. Furthermore, zero‐shot evaluation on COVID‐19 exhibited a notably higher success rate in identifying diverse potential drugs. The utilization of BioHG facilitates a unique KGDRP‐based analysis of cell‐target‐drug interactions, thereby enabling the elucidation of drug mechanisms. Overall, KGDRP provides a robust infrastructure for the seamlessly integration of multimodal data and biomedical networks, effectively accelerating PDD, guiding therapeutic target discovery, and ultimately expediting therapeutic molecule discovery.
format Article
id doaj-art-be9a8130eee44faeba61b025066cece1
institution OA Journals
issn 2198-3844
language English
publishDate 2025-04-01
publisher Wiley
record_format Article
series Advanced Science
spelling doaj-art-be9a8130eee44faeba61b025066cece12025-08-20T02:24:50ZengWileyAdvanced Science2198-38442025-04-011216n/an/a10.1002/advs.202412402A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug DiscoveryQing Ye0Yundian Zeng1Linlong Jiang2Yu Kang3Peichen Pan4Jiming Chen5Yafeng Deng6Haitao Zhao7Shibo He8Tingjun Hou9Chang‐Yu Hsieh10College of Control Science and Engineering Zhejiang University Hangzhou Zhejiang 310027 ChinaCollege of Control Science and Engineering Zhejiang University Hangzhou Zhejiang 310027 ChinaCollege of Pharmaceutical Sciences Zhejiang University Hangzhou Zhejiang 310058 ChinaCollege of Pharmaceutical Sciences Zhejiang University Hangzhou Zhejiang 310058 ChinaCollege of Pharmaceutical Sciences Zhejiang University Hangzhou Zhejiang 310058 ChinaCollege of Control Science and Engineering Zhejiang University Hangzhou Zhejiang 310027 ChinaCarbonSilicon AI Technology Co., Ltd Hangzhou Zhejiang 310018 ChinaCenter for Intelligent and Biomimetic Systems Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen Guangdong 440305 ChinaCollege of Control Science and Engineering Zhejiang University Hangzhou Zhejiang 310027 ChinaCollege of Pharmaceutical Sciences Zhejiang University Hangzhou Zhejiang 310058 ChinaCollege of Pharmaceutical Sciences Zhejiang University Hangzhou Zhejiang 310058 ChinaAbstract Discovering therapeutic molecules requires the integration of both phenotype‐based drug discovery (PDD) and target‐based drug discovery (TDD). However, this integration remains challenging due to the inherent heterogeneity, noise, and bias present in biomedical data. In this study, Knowledge‐Guided Drug Relational Predictor (KGDRP), a graph representation learning approach is developed that effectively integrates multimodal biomedical data, including network data containing biological system information, gene expression data, and sequence data that incorporates chemical molecular structures, all within a heterogeneous graph (HG) structure. By incorporating biomedical HG (BioHG) into a heterogeneous graph neural network (HGNN)‐based architecture, KGDRP exhibits a remarkable 12% improvement compared to previous methods in real‐world screening scenarios. Notably, the biology‐informed representation, derived from KGDRP, significantly enhance target prioritization by 26% in drug target discovery. Furthermore, zero‐shot evaluation on COVID‐19 exhibited a notably higher success rate in identifying diverse potential drugs. The utilization of BioHG facilitates a unique KGDRP‐based analysis of cell‐target‐drug interactions, thereby enabling the elucidation of drug mechanisms. Overall, KGDRP provides a robust infrastructure for the seamlessly integration of multimodal data and biomedical networks, effectively accelerating PDD, guiding therapeutic target discovery, and ultimately expediting therapeutic molecule discovery.https://doi.org/10.1002/advs.202412402biological networksdrug target discoverygraph representation learningphenotypic screeningtranscriptomics
spellingShingle Qing Ye
Yundian Zeng
Linlong Jiang
Yu Kang
Peichen Pan
Jiming Chen
Yafeng Deng
Haitao Zhao
Shibo He
Tingjun Hou
Chang‐Yu Hsieh
A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
Advanced Science
biological networks
drug target discovery
graph representation learning
phenotypic screening
transcriptomics
title A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
title_full A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
title_fullStr A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
title_full_unstemmed A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
title_short A Knowledge‐Guided Graph Learning Approach Bridging Phenotype‐ and Target‐Based Drug Discovery
title_sort knowledge guided graph learning approach bridging phenotype and target based drug discovery
topic biological networks
drug target discovery
graph representation learning
phenotypic screening
transcriptomics
url https://doi.org/10.1002/advs.202412402
work_keys_str_mv AT qingye aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yundianzeng aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT linlongjiang aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yukang aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT peichenpan aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT jimingchen aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yafengdeng aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT haitaozhao aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT shibohe aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT tingjunhou aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT changyuhsieh aknowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT qingye knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yundianzeng knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT linlongjiang knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yukang knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT peichenpan knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT jimingchen knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT yafengdeng knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT haitaozhao knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT shibohe knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT tingjunhou knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery
AT changyuhsieh knowledgeguidedgraphlearningapproachbridgingphenotypeandtargetbaseddrugdiscovery