A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors

Abstract Studying the molecular properties of drugs and their interactions with human targets aids in better understanding the clinical performance of drugs and guides drug development. In computer‐aided drug discovery, it is crucial to utilize effective molecular feature representations for predict...

Full description

Saved in:
Bibliographic Details
Main Authors: Xin Yang, Yang Wang, Ye Lin, Mingxuan Zhang, Ou Liu, Jianwei Shuai, Qi Zhao
Format: Article
Language:English
Published: Wiley 2025-04-01
Series:Advanced Science
Subjects:
Online Access:https://doi.org/10.1002/advs.202412987
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850272920890245120
author Xin Yang
Yang Wang
Ye Lin
Mingxuan Zhang
Ou Liu
Jianwei Shuai
Qi Zhao
author_facet Xin Yang
Yang Wang
Ye Lin
Mingxuan Zhang
Ou Liu
Jianwei Shuai
Qi Zhao
author_sort Xin Yang
collection DOAJ
description Abstract Studying the molecular properties of drugs and their interactions with human targets aids in better understanding the clinical performance of drugs and guides drug development. In computer‐aided drug discovery, it is crucial to utilize effective molecular feature representations for predicting molecular properties and designing ligands with high binding affinity to targets. However, designing an effective multi‐task and self‐supervised strategy remains a significant challenge for the pretraining framework. In this study, a multi‐task self‐supervised deep learning framework is proposed, MTSSMol, which utilizes ≈10 million unlabeled drug‐like molecules for pretraining to identify potential inhibitors of fibroblast growth factor receptor 1 (FGFR1). During the pretraining of MTSSMol, molecular representations are learned through a graph neural networks (GNNs) encoder. A multi‐task self‐supervised pretraining strategy is proposed to fully capture the structural and chemical knowledge of molecules. Extensive computational tests on 27 datasets demonstrate that MTSSMol exhibits exceptional performance in predicting molecular properties across different domains. Moreover, MTSSMol's capability is validated to identify potential inhibitors of FGFR1 through molecular docking using RoseTTAFold All‐Atom (RFAA) and molecular dynamics simulations. Overall, MTSSMol provides an effective algorithmic framework for enhancing molecular representation learning and identifying potential drug candidates, offering a valuable tool to accelerate drug discovery processes. All of the codes are freely available online at https:// github.com/zhaoqi106/MTSSMol.
format Article
id doaj-art-df9d9137e32d439ebc25f431357c8ff8
institution OA Journals
issn 2198-3844
language English
publishDate 2025-04-01
publisher Wiley
record_format Article
series Advanced Science
spelling doaj-art-df9d9137e32d439ebc25f431357c8ff82025-08-20T01:51:39ZengWileyAdvanced Science2198-38442025-04-011213n/an/a10.1002/advs.202412987A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 InhibitorsXin Yang0Yang Wang1Ye Lin2Mingxuan Zhang3Ou Liu4Jianwei Shuai5Qi Zhao6School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan Liaoning 114051 P. R. ChinaWenzhou Key Laboratory of Biomedical Imaging Center of Biomedical Physics Wenzhou Institute University of Chinese Academy of Sciences Wenzhou Zhejiang 325001 P. R. ChinaCollege of Computer Science and Technology Jilin University Changchun Jilin 130012 P. R. ChinaSchool of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan Liaoning 114051 P. R. ChinaOujiang Laboratory (Zhejiang Lab for Regenerative Medicine, Vision, and Brain Health) Wenzhou Institute University of Chinese Academy of Sciences Wenzhou Zhejiang 325001 P. R. ChinaOujiang Laboratory (Zhejiang Lab for Regenerative Medicine, Vision, and Brain Health) Wenzhou Institute University of Chinese Academy of Sciences Wenzhou Zhejiang 325001 P. R. ChinaSchool of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan Liaoning 114051 P. R. ChinaAbstract Studying the molecular properties of drugs and their interactions with human targets aids in better understanding the clinical performance of drugs and guides drug development. In computer‐aided drug discovery, it is crucial to utilize effective molecular feature representations for predicting molecular properties and designing ligands with high binding affinity to targets. However, designing an effective multi‐task and self‐supervised strategy remains a significant challenge for the pretraining framework. In this study, a multi‐task self‐supervised deep learning framework is proposed, MTSSMol, which utilizes ≈10 million unlabeled drug‐like molecules for pretraining to identify potential inhibitors of fibroblast growth factor receptor 1 (FGFR1). During the pretraining of MTSSMol, molecular representations are learned through a graph neural networks (GNNs) encoder. A multi‐task self‐supervised pretraining strategy is proposed to fully capture the structural and chemical knowledge of molecules. Extensive computational tests on 27 datasets demonstrate that MTSSMol exhibits exceptional performance in predicting molecular properties across different domains. Moreover, MTSSMol's capability is validated to identify potential inhibitors of FGFR1 through molecular docking using RoseTTAFold All‐Atom (RFAA) and molecular dynamics simulations. Overall, MTSSMol provides an effective algorithmic framework for enhancing molecular representation learning and identifying potential drug candidates, offering a valuable tool to accelerate drug discovery processes. All of the codes are freely available online at https:// github.com/zhaoqi106/MTSSMol.https://doi.org/10.1002/advs.202412987FGFR1graph neural networksmolecular propertiesmulti‐task strategypretraining framework
spellingShingle Xin Yang
Yang Wang
Ye Lin
Mingxuan Zhang
Ou Liu
Jianwei Shuai
Qi Zhao
A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
Advanced Science
FGFR1
graph neural networks
molecular properties
multi‐task strategy
pretraining framework
title A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
title_full A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
title_fullStr A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
title_full_unstemmed A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
title_short A Multi‐Task Self‐Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors
title_sort multi task self supervised strategy for predicting molecular properties and fgfr1 inhibitors
topic FGFR1
graph neural networks
molecular properties
multi‐task strategy
pretraining framework
url https://doi.org/10.1002/advs.202412987
work_keys_str_mv AT xinyang amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT yangwang amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT yelin amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT mingxuanzhang amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT ouliu amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT jianweishuai amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT qizhao amultitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT xinyang multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT yangwang multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT yelin multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT mingxuanzhang multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT ouliu multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT jianweishuai multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors
AT qizhao multitaskselfsupervisedstrategyforpredictingmolecularpropertiesandfgfr1inhibitors