ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment

Abstract Multi-species acute toxicity assessment forms the basis for chemical classification, labelling and risk management. Existing deep learning methods struggle with diverse experimental conditions, imbalanced data, and scarce target data, hindering their ability to reveal endpoint associations...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiang Lu, Lianlian Wu, Ruijiang Li, Mengxuan Wan, Jun Yang, Peng Zan, Hui Bai, Song He, Xiaochen Bo
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-60989-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849402095703162880
author Jiang Lu
Lianlian Wu
Ruijiang Li
Mengxuan Wan
Jun Yang
Peng Zan
Hui Bai
Song He
Xiaochen Bo
author_facet Jiang Lu
Lianlian Wu
Ruijiang Li
Mengxuan Wan
Jun Yang
Peng Zan
Hui Bai
Song He
Xiaochen Bo
author_sort Jiang Lu
collection DOAJ
description Abstract Multi-species acute toxicity assessment forms the basis for chemical classification, labelling and risk management. Existing deep learning methods struggle with diverse experimental conditions, imbalanced data, and scarce target data, hindering their ability to reveal endpoint associations and accurately predict data-scarce endpoints. Here we propose a machine learning paradigm, Adjoint Correlation Learning, for multi-condition acute toxicity assessment (ToxACoL) to address these challenges. ToxACoL models endpoint associations via graph topology and achieves knowledge transfer via graph convolution. The adjoint correlation mechanism encodes compounds and endpoints synchronously, yielding endpoint-aware and task-focused representations. Comprehensive analyses demonstrate that ToxACoL yields 43%-87% improvements for data-scarce human endpoints, while reducing training data by 70% to 80%. Visualization of the learned top-level representation interprets structural alert mechanisms. Filled-in toxicity values highlight potential for extrapolating animal results to humans. Finally, we deploy ToxACoL as a free web platform for rapid prediction of multi-condition acute toxicities.
format Article
id doaj-art-f3231c70bb19475caabc47f925d321d6
institution Kabale University
issn 2041-1723
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-f3231c70bb19475caabc47f925d321d62025-08-20T03:37:37ZengNature PortfolioNature Communications2041-17232025-07-0116111910.1038/s41467-025-60989-7ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessmentJiang Lu0Lianlian Wu1Ruijiang Li2Mengxuan Wan3Jun Yang4Peng Zan5Hui Bai6Song He7Xiaochen Bo8Academy of Medical Engineering and Translational Medicine, Tianjin UniversityAcademy of Medical Engineering and Translational Medicine, Tianjin UniversityDepartment of Advanced & Interdisciplinary Biotechnology, Academy of Military Medical SciencesShanghai Key Laboratory of Power Station Automation Technology, School of Mechatronics Engineering and Automation, Shanghai UniversityDepartment of Cell Biology, School of Life Sciences, Central South UniversityShanghai Key Laboratory of Power Station Automation Technology, School of Mechatronics Engineering and Automation, Shanghai UniversityClinical Translational Research Center, Beijing Tsinghua Changgung Hospital, Tsinghua UniversityDepartment of Advanced & Interdisciplinary Biotechnology, Academy of Military Medical SciencesAcademy of Medical Engineering and Translational Medicine, Tianjin UniversityAbstract Multi-species acute toxicity assessment forms the basis for chemical classification, labelling and risk management. Existing deep learning methods struggle with diverse experimental conditions, imbalanced data, and scarce target data, hindering their ability to reveal endpoint associations and accurately predict data-scarce endpoints. Here we propose a machine learning paradigm, Adjoint Correlation Learning, for multi-condition acute toxicity assessment (ToxACoL) to address these challenges. ToxACoL models endpoint associations via graph topology and achieves knowledge transfer via graph convolution. The adjoint correlation mechanism encodes compounds and endpoints synchronously, yielding endpoint-aware and task-focused representations. Comprehensive analyses demonstrate that ToxACoL yields 43%-87% improvements for data-scarce human endpoints, while reducing training data by 70% to 80%. Visualization of the learned top-level representation interprets structural alert mechanisms. Filled-in toxicity values highlight potential for extrapolating animal results to humans. Finally, we deploy ToxACoL as a free web platform for rapid prediction of multi-condition acute toxicities.https://doi.org/10.1038/s41467-025-60989-7
spellingShingle Jiang Lu
Lianlian Wu
Ruijiang Li
Mengxuan Wan
Jun Yang
Peng Zan
Hui Bai
Song He
Xiaochen Bo
ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
Nature Communications
title ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
title_full ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
title_fullStr ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
title_full_unstemmed ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
title_short ToxACoL: an endpoint-aware and task-focused compound representation learning paradigm for acute toxicity assessment
title_sort toxacol an endpoint aware and task focused compound representation learning paradigm for acute toxicity assessment
url https://doi.org/10.1038/s41467-025-60989-7
work_keys_str_mv AT jianglu toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT lianlianwu toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT ruijiangli toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT mengxuanwan toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT junyang toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT pengzan toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT huibai toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT songhe toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment
AT xiaochenbo toxacolanendpointawareandtaskfocusedcompoundrepresentationlearningparadigmforacutetoxicityassessment