• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

基于超图的持久同调(HPC)在药物设计中的分子表示。

Hypergraph-based persistent cohomology (HPC) for molecular representations in drug design.

机构信息

Division of Mathematical Sciences, School of Physical and Mathematical Sciences, Nanyang Technological University, 637371, Singapore.

School of Mathematical Science and LPMC, Nankai University, 300071, Tianjin, China.

出版信息

Brief Bioinform. 2021 Sep 2;22(5). doi: 10.1093/bib/bbaa411.

DOI:10.1093/bib/bbaa411
PMID:33480394
Abstract

Artificial intelligence (AI) based drug design has demonstrated great potential to fundamentally change the pharmaceutical industries. Currently, a key issue in AI-based drug design is efficient transferable molecular descriptors or fingerprints. Here, we present hypergraph-based molecular topological representation, hypergraph-based (weighted) persistent cohomology (HPC/HWPC) and HPC/HWPC-based molecular fingerprints for machine learning models in drug design. Molecular structures and their atomic interactions are highly complicated and pose great challenges for efficient mathematical representations. We develop the first hypergraph-based topological framework to characterize detailed molecular structures and interactions at atomic level. Inspired by the elegant path complex model, hypergraph-based embedded homology and persistent homology have been proposed recently. Based on them, we construct HPC/HWPC, and use them to generate molecular descriptors for learning models in protein-ligand binding affinity prediction, one of the key step in drug design. Our models are tested on three most commonly-used databases, including PDBbind-v2007, PDBbind-v2013 and PDBbind-v2016, and outperform all existing machine learning models with traditional molecular descriptors. Our HPC/HWPC models have demonstrated great potential in AI-based drug design.

摘要

人工智能(AI)药物设计已被证明具有从根本上改变制药行业的巨大潜力。目前,基于 AI 的药物设计中的一个关键问题是高效的可转移分子描述符或指纹。在这里,我们提出了基于超图的分子拓扑表示、基于超图的(加权)持久同调(HPC/HWPC)和基于 HPC/HWPC 的分子指纹,用于药物设计中的机器学习模型。分子结构及其原子相互作用非常复杂,给有效的数学表示带来了巨大挑战。我们开发了第一个基于超图的拓扑框架,以在原子水平上描述详细的分子结构和相互作用。受优雅路径复形模型的启发,最近提出了基于超图的嵌入同调与持久同调。在此基础上,我们构建了 HPC/HWPC,并使用它们生成分子描述符,用于学习模型,以预测蛋白质-配体结合亲和力,这是药物设计的关键步骤之一。我们的模型在三个最常用的数据库上进行了测试,包括 PDBbind-v2007、PDBbind-v2013 和 PDBbind-v2016,并且优于所有使用传统分子描述符的现有机器学习模型。我们的 HPC/HWPC 模型在基于 AI 的药物设计中具有巨大的潜力。

相似文献

1
Hypergraph-based persistent cohomology (HPC) for molecular representations in drug design.基于超图的持久同调(HPC)在药物设计中的分子表示。
Brief Bioinform. 2021 Sep 2;22(5). doi: 10.1093/bib/bbaa411.
2
Persistent spectral hypergraph based machine learning (PSH-ML) for protein-ligand binding affinity prediction.基于持久谱超图的机器学习(PSH-ML)用于蛋白质-配体结合亲和力预测。
Brief Bioinform. 2021 Sep 2;22(5). doi: 10.1093/bib/bbab127.
3
Molecular persistent spectral image (Mol-PSI) representation for machine learning models in drug design.用于药物设计中机器学习模型的分子持久谱图像(Mol-PSI)表示。
Brief Bioinform. 2022 Jan 17;23(1). doi: 10.1093/bib/bbab527.
4
Persistent Path-Spectral (PPS) Based Machine Learning for Protein-Ligand Binding Affinity Prediction.基于持续路径谱(PPS)的机器学习的蛋白质-配体结合亲和力预测。
J Chem Inf Model. 2023 Feb 13;63(3):1066-1075. doi: 10.1021/acs.jcim.2c01251. Epub 2023 Jan 16.
5
Forman persistent Ricci curvature (FPRC)-based machine learning models for protein-ligand binding affinity prediction.基于 Forman 持续 Ricci 曲率 (FPRC) 的蛋白质-配体结合亲和力预测机器学习模型。
Brief Bioinform. 2021 Nov 5;22(6). doi: 10.1093/bib/bbab136.
6
Dowker complex based machine learning (DCML) models for protein-ligand binding affinity prediction.基于 Dowker 复合物的机器学习 (DCML) 模型用于预测蛋白质-配体结合亲和力。
PLoS Comput Biol. 2022 Apr 6;18(4):e1009943. doi: 10.1371/journal.pcbi.1009943. eCollection 2022 Apr.
7
Persistent Homology for RNA Data Analysis.RNA 数据分析中的持久同调。
Methods Mol Biol. 2023;2627:211-229. doi: 10.1007/978-1-0716-2974-1_12.
8
Ollivier Persistent Ricci Curvature-Based Machine Learning for the Protein-Ligand Binding Affinity Prediction.基于奥利维尔持续 Ricci 曲率的机器学习在蛋白质-配体结合亲和力预测中的应用。
J Chem Inf Model. 2021 Apr 26;61(4):1617-1626. doi: 10.1021/acs.jcim.0c01415. Epub 2021 Mar 16.
9
Persistent Cohomology for Data With Multicomponent Heterogeneous Information.具有多组分异构信息的数据的持久上同调
SIAM J Math Data Sci. 2020;2(2):396-418. doi: 10.1137/19m1272226. Epub 2020 May 19.
10
Persistent spectral-based machine learning (PerSpect ML) for protein-ligand binding affinity prediction.用于蛋白质-配体结合亲和力预测的基于持久光谱的机器学习(PerSpect ML)。
Sci Adv. 2021 May 7;7(19). doi: 10.1126/sciadv.abc5329. Print 2021 May.

引用本文的文献

1
Predicting Affinity Through Homology (PATH): Interpretable binding affinity prediction with persistent homology.通过同源性预测亲和力(PATH):利用持久同源性进行可解释的结合亲和力预测。
PLoS Comput Biol. 2025 Jun 27;21(6):e1013216. doi: 10.1371/journal.pcbi.1013216. eCollection 2025 Jun.
2
Persistent Dirac for molecular representation.分子表示的持久狄拉克态。
Sci Rep. 2023 Jul 11;13(1):11183. doi: 10.1038/s41598-023-37853-z.
3
SS-GNN: A Simple-Structured Graph Neural Network for Affinity Prediction.SS-GNN:一种用于亲和力预测的结构简单的图神经网络。
ACS Omega. 2023 Jun 15;8(25):22496-22507. doi: 10.1021/acsomega.3c00085. eCollection 2023 Jun 27.
4
ASPECTS OF TOPOLOGICAL APPROACHES FOR DATA SCIENCE.数据科学的拓扑方法的各个方面。
Found Data Sci. 2022 Jun;4(2):165-216. doi: 10.3934/fods.2022002.
5
Path Topology in Molecular and Materials Sciences.分子和材料科学中的路径拓扑。
J Phys Chem Lett. 2023 Feb 2;14(4):954-964. doi: 10.1021/acs.jpclett.2c03706. Epub 2023 Jan 23.
6
Improved compound-protein interaction site and binding affinity prediction using self-supervised protein embeddings.利用自监督蛋白质嵌入提高化合物-蛋白质相互作用位点和结合亲和力预测。
BMC Bioinformatics. 2022 Dec 16;23(1):543. doi: 10.1186/s12859-022-05107-w.
7
Biomolecular Topology: Modelling and Analysis.生物分子拓扑学:建模与分析
Acta Math Sin Engl Ser. 2022;38(10):1901-1938. doi: 10.1007/s10114-022-2326-5. Epub 2022 Oct 15.
8
HOMOTOPY CONTINUATION FOR THE SPECTRA OF PERSISTENT LAPLACIANS.持久拉普拉斯算子谱的同伦延拓
Found Data Sci. 2021 Dec;3(4):677-700. doi: 10.3934/fods.2021017.
9
Dowker complex based machine learning (DCML) models for protein-ligand binding affinity prediction.基于 Dowker 复合物的机器学习 (DCML) 模型用于预测蛋白质-配体结合亲和力。
PLoS Comput Biol. 2022 Apr 6;18(4):e1009943. doi: 10.1371/journal.pcbi.1009943. eCollection 2022 Apr.