CasANGCL: pre -training and fine-tuning model based on cascaded attention network and graph contrastive learning for molecular property prediction

doi:10.1093/bib/bbac566

	CasANGCL: pre -training and fine-tuning model based on cascaded attention network and graph contrastive learning for molecular property prediction
	Zheng, Zixi 1; Tan, Yanyan 1; Wang, Hong 2; Yu, Shengpeng 1; Liu, Tianyu3 ; Liang, Cheng1
	2023
发表期刊	BRIEFINGS IN BIOINFORMATICS
卷号	24 期号:1
摘要	Motivation: Molecular property prediction is a significant requirement in AI -driven drug design and discovery, aiming to predict the molecular property information (e.g. toxicity) based on the mined biomolecular knowledge. Although graph neural networks have been proven powerful in predicting molecular property, unbalanced labeled data and poor generalization capability for new -synthesized molecules are always key issues that hinder further improvement of molecular encoding performance. Results: We propose a novel self-supervised representation learning scheme based on a Cascaded Attention Network and Graph Contrastive Learning (CasANGCL). We design a new graph network variant, designated as cascaded attention network, to encode local-global molecular representations. We construct a two -stage contrast predictor framework to tackle the label imbalance problem of training molecular samples, which is an integrated end -to -end learning scheme. Moreover, we utilize the information -flow scheme for training our network, which explicitly captures the edge information in the node/graph representations and obtains more fine-grained knowledge. Our model achieves an 81.9% ROC-AUC average performance on 661 tasks from seven challenging benchmarks, showing better portability and generalizations. Further visualization studies indicate our model's better representation capacity and provide interpretability.
关键词	molecular representation cascaded attention network graph contrastive learning self-supervised learning molecular property prediction
DOI	10.1093/bib/bbac566
收录类别	SCIE
ISSN	1467-5463
语种	英语
WOS研究方向	Biochemistry & Molecular Biology ; Mathematical & Computational Biology
WOS类目	Biochemical Research Methods ; Mathematical & Computational Biology
WOS记录号	WOS:000905625700001
出版者	OXFORD UNIV PRESS
原始文献类型	Article
EISSN	1477-4054
引用统计	被引频次：6[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/35775
专题	工商管理学院国际经济与贸易学院
通讯作者	Wang, Hong
作者单位	1.Shangdong Normal Univ, Sch Informat Sci & Engn, Jinan, Peoples R China; 2.Chinese Acad Sci, Computing Inst, Beijing, Peoples R China; 3.Lanzhou Univ Finance & Econ, Dept Elect Engn, Lanzhou, Peoples R China
推荐引用方式 GB/T 7714	Zheng, Zixi,Tan, Yanyan,Wang, Hong,et al. CasANGCL: pre -training and fine-tuning model based on cascaded attention network and graph contrastive learning for molecular property prediction[J]. BRIEFINGS IN BIOINFORMATICS,2023,24(1).
APA	Zheng, Zixi,Tan, Yanyan,Wang, Hong,Yu, Shengpeng,Liu, Tianyu,&Liang, Cheng.(2023).CasANGCL: pre -training and fine-tuning model based on cascaded attention network and graph contrastive learning for molecular property prediction.BRIEFINGS IN BIOINFORMATICS,24(1).
MLA	Zheng, Zixi,et al."CasANGCL: pre -training and fine-tuning model based on cascaded attention network and graph contrastive learning for molecular property prediction".BRIEFINGS IN BIOINFORMATICS 24.1(2023).