首页 > 最新文献

Nature computational science最新文献

英文 中文
Programmable responsive metamaterials for mechanical computing and robotics 用于机械计算和机器人技术的可编程响应超材料。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-27 DOI: 10.1038/s43588-024-00673-w
Qiguang He, Samuele Ferracin, Jordan R. Raney
Unconventional computing based on mechanical metamaterials has been of growing interest, including how such metamaterials might process information via autonomous interactions with their environment. Here we describe recent efforts to combine responsive materials with nonlinear mechanical metamaterials to achieve stimuli-responsive mechanical logic and computation. We also describe some key challenges and opportunities in the design and construction of these devices, including the lack of comprehensive computational tools, and the challenges associated with patterning multi-material mechanisms. Mechanical metamaterials have shown potential for processing information via autonomous environmental interactions. This Perspective summarizes recent efforts and challenges on integrating stimuli-responsive materials with mechanical metamaterials for mechanical computing, and explores the remaining challenges in the field.
人们对基于机械超材料的非常规计算越来越感兴趣,包括这种超材料如何通过与环境的自主互动来处理信息。在此,我们将介绍最近将响应材料与非线性机械超材料相结合,以实现刺激响应式机械逻辑和计算的努力。我们还介绍了设计和建造这些装置的一些关键挑战和机遇,包括缺乏全面的计算工具,以及与多材料机制图案化相关的挑战。
{"title":"Programmable responsive metamaterials for mechanical computing and robotics","authors":"Qiguang He, Samuele Ferracin, Jordan R. Raney","doi":"10.1038/s43588-024-00673-w","DOIUrl":"10.1038/s43588-024-00673-w","url":null,"abstract":"Unconventional computing based on mechanical metamaterials has been of growing interest, including how such metamaterials might process information via autonomous interactions with their environment. Here we describe recent efforts to combine responsive materials with nonlinear mechanical metamaterials to achieve stimuli-responsive mechanical logic and computation. We also describe some key challenges and opportunities in the design and construction of these devices, including the lack of comprehensive computational tools, and the challenges associated with patterning multi-material mechanisms. Mechanical metamaterials have shown potential for processing information via autonomous environmental interactions. This Perspective summarizes recent efforts and challenges on integrating stimuli-responsive materials with mechanical metamaterials for mechanical computing, and explores the remaining challenges in the field.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"567-573"},"PeriodicalIF":12.0,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142082867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Computational design of art-inspired metamaterials 艺术启发超材料的计算设计。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-27 DOI: 10.1038/s43588-024-00671-y
Gary P. T. Choi
In recent years, there has been a surge of interest in the design of mechanical metamaterials for different science and engineering applications. In particular, various computational approaches have been developed to facilitate the systematic design of art-inspired metamaterials including origami and kirigami metamaterials. In this Comment, we highlight the recent advances and discuss the outlook for the computational design of art-inspired metamaterials.
近年来,人们对用于不同科学和工程应用的机械超材料设计兴趣大增。特别是,人们开发了各种计算方法,以促进艺术启发超材料(包括折纸和叽里格米超材料)的系统设计。在本评论中,我们将重点介绍最新进展,并探讨艺术启发超材料计算设计的前景。
{"title":"Computational design of art-inspired metamaterials","authors":"Gary P. T. Choi","doi":"10.1038/s43588-024-00671-y","DOIUrl":"10.1038/s43588-024-00671-y","url":null,"abstract":"In recent years, there has been a surge of interest in the design of mechanical metamaterials for different science and engineering applications. In particular, various computational approaches have been developed to facilitate the systematic design of art-inspired metamaterials including origami and kirigami metamaterials. In this Comment, we highlight the recent advances and discuss the outlook for the computational design of art-inspired metamaterials.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"549-552"},"PeriodicalIF":12.0,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142082863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Synergy between photonic metamaterials and AI 光子超材料与人工智能的协同作用。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-27 DOI: 10.1038/s43588-024-00675-8
Jie Pan
Dr Yongmin Liu — professor of mechanical and industrial engineering and professor of electrical and computer engineering at Northeastern University — talks to Nature Computational Science about his career trajectory, his research on photonic metamaterials, and the synergistic effects between photonic metamaterials research and artificial intelligence (AI).
东北大学机械与工业工程系教授、电气与计算机工程系教授刘勇民博士向《自然-计算科学》讲述了他的职业轨迹、他对光子超材料的研究以及光子超材料研究与人工智能(AI)之间的协同效应。
{"title":"Synergy between photonic metamaterials and AI","authors":"Jie Pan","doi":"10.1038/s43588-024-00675-8","DOIUrl":"10.1038/s43588-024-00675-8","url":null,"abstract":"Dr Yongmin Liu — professor of mechanical and industrial engineering and professor of electrical and computer engineering at Northeastern University — talks to Nature Computational Science about his career trajectory, his research on photonic metamaterials, and the synergistic effects between photonic metamaterials research and artificial intelligence (AI).","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"556-557"},"PeriodicalIF":12.0,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142082868","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
AI-recognized mitochondrial phenotype enables identification of drug targets 人工智能识别的线粒体表型有助于确定药物靶点。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-22 DOI: 10.1038/s43588-024-00682-9
Revealing a drug’s mechanism of action (MOA) is costly and time-consuming. In this study, we used deep learning to extract temporal mitochondrial phenotypic features after exposure to drugs with known MOAs using re-identification algorithms. The trained model could then predict the MOAs of unidentified substances, facilitating phenotypic screening-based drug discovery and repurposing.
揭示药物的作用机制(MOA)既费钱又费时。在这项研究中,我们利用深度学习,使用再识别算法提取暴露于已知MOA药物后的线粒体表型特征。经过训练的模型可以预测未识别物质的MOA,从而促进基于表型筛选的药物发现和再利用。
{"title":"AI-recognized mitochondrial phenotype enables identification of drug targets","authors":"","doi":"10.1038/s43588-024-00682-9","DOIUrl":"10.1038/s43588-024-00682-9","url":null,"abstract":"Revealing a drug’s mechanism of action (MOA) is costly and time-consuming. In this study, we used deep learning to extract temporal mitochondrial phenotypic features after exposure to drugs with known MOAs using re-identification algorithms. The trained model could then predict the MOAs of unidentified substances, facilitating phenotypic screening-based drug discovery and repurposing.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"563-564"},"PeriodicalIF":12.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142037931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Deep learning large-scale drug discovery and repurposing 深度学习大规模药物发现和再利用。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-21 DOI: 10.1038/s43588-024-00679-4
Min Yu, Weiming Li, Yunru Yu, Yu Zhao, Lizhi Xiao, Volker M. Lauschke, Yiyu Cheng, Xingcai Zhang, Yi Wang
Large-scale drug discovery and repurposing is challenging. Identifying the mechanism of action (MOA) is crucial, yet current approaches are costly and low-throughput. Here we present an approach for MOA identification by profiling changes in mitochondrial phenotypes. By temporally imaging mitochondrial morphology and membrane potential, we established a pipeline for monitoring time-resolved mitochondrial images, resulting in a dataset comprising 570,096 single-cell images of cells exposed to 1,068 United States Food and Drug Administration-approved drugs. A deep learning model named MitoReID, using a re-identification (ReID) framework and an Inflated 3D ResNet backbone, was developed. It achieved 76.32% Rank-1 and 65.92% mean average precision on the testing set and successfully identified the MOAs for six untrained drugs on the basis of mitochondrial phenotype. Furthermore, MitoReID identified cyclooxygenase-2 inhibition as the MOA of the natural compound epicatechin in tea, which was successfully validated in vitro. Our approach thus provides an automated and cost-effective alternative for target identification that could accelerate large-scale drug discovery and repurposing. A deep learning-based model, MitoReID, is presented for profiling changes in mitochondrial phenotypes, allowing for the identification of various drugs’ mechanism of action.
大规模药物发现和再利用具有挑战性。确定药物的作用机制(MOA)至关重要,但目前的方法成本高、通量低。在这里,我们提出了一种通过线粒体表型变化进行MOA鉴定的方法。通过对线粒体形态和膜电位进行时间成像,我们建立了一个用于监测时间分辨线粒体图像的管道,形成了一个由 570,096 张单细胞图像组成的数据集,这些图像是暴露于 1,068 种美国食品药品管理局批准药物的细胞的图像。利用重新识别(ReID)框架和膨胀三维 ResNet 主干网,开发了名为 MitoReID 的深度学习模型。该模型在测试集上取得了 76.32% 的 Rank-1 和 65.92% 的平均精度,并根据线粒体表型成功识别了六种未经训练药物的 MOA。此外,MitoReID还确定了茶叶中天然化合物表儿茶素的MOA为环氧化酶-2抑制,并成功地进行了体外验证。因此,我们的方法为靶点识别提供了一种自动化、经济高效的替代方法,可以加速大规模药物发现和再利用。
{"title":"Deep learning large-scale drug discovery and repurposing","authors":"Min Yu, Weiming Li, Yunru Yu, Yu Zhao, Lizhi Xiao, Volker M. Lauschke, Yiyu Cheng, Xingcai Zhang, Yi Wang","doi":"10.1038/s43588-024-00679-4","DOIUrl":"10.1038/s43588-024-00679-4","url":null,"abstract":"Large-scale drug discovery and repurposing is challenging. Identifying the mechanism of action (MOA) is crucial, yet current approaches are costly and low-throughput. Here we present an approach for MOA identification by profiling changes in mitochondrial phenotypes. By temporally imaging mitochondrial morphology and membrane potential, we established a pipeline for monitoring time-resolved mitochondrial images, resulting in a dataset comprising 570,096 single-cell images of cells exposed to 1,068 United States Food and Drug Administration-approved drugs. A deep learning model named MitoReID, using a re-identification (ReID) framework and an Inflated 3D ResNet backbone, was developed. It achieved 76.32% Rank-1 and 65.92% mean average precision on the testing set and successfully identified the MOAs for six untrained drugs on the basis of mitochondrial phenotype. Furthermore, MitoReID identified cyclooxygenase-2 inhibition as the MOA of the natural compound epicatechin in tea, which was successfully validated in vitro. Our approach thus provides an automated and cost-effective alternative for target identification that could accelerate large-scale drug discovery and repurposing. A deep learning-based model, MitoReID, is presented for profiling changes in mitochondrial phenotypes, allowing for the identification of various drugs’ mechanism of action.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"600-614"},"PeriodicalIF":12.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142019787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Bridging the gap between artificial intelligence and natural intelligence 缩小人工智能与自然智能之间的差距。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-16 DOI: 10.1038/s43588-024-00677-6
Rui-Jie Zhu, Skye Gunasekaran, Jason Eshraghian
According to a recent study, a small network consisting of four leaky integrate-and-fire neurons can reproduce the behavior of a single Hodgkin–Huxley neuron, thereby bridging the gap between endogenous and exogenous complexity.
根据最近的一项研究,由四个漏性整合-发射神经元组成的小型网络可以重现单个霍奇金-赫胥黎神经元的行为,从而缩小了内源性和外源性复杂性之间的差距。
{"title":"Bridging the gap between artificial intelligence and natural intelligence","authors":"Rui-Jie Zhu, Skye Gunasekaran, Jason Eshraghian","doi":"10.1038/s43588-024-00677-6","DOIUrl":"10.1038/s43588-024-00677-6","url":null,"abstract":"According to a recent study, a small network consisting of four leaky integrate-and-fire neurons can reproduce the behavior of a single Hodgkin–Huxley neuron, thereby bridging the gap between endogenous and exogenous complexity.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"559-560"},"PeriodicalIF":12.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141997046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Accelerating predictions of electronic transport and superconductivity 加速电子传输和超导预测。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-16 DOI: 10.1038/s43588-024-00678-5
Ting Cao
By developing a machine learning framework, a recent study substantially accelerates the calculation of electron–phonon coupling, making it computationally feasible to predict and understand a range of important physical phenomena, including electronic transport, hot-carrier relaxation, and superconductivity in complex materials.
通过开发一种机器学习框架,最近的一项研究大大加快了电子-声子耦合的计算速度,使预测和理解一系列重要物理现象(包括复杂材料中的电子传输、热载流子弛豫和超导性)在计算上变得可行。
{"title":"Accelerating predictions of electronic transport and superconductivity","authors":"Ting Cao","doi":"10.1038/s43588-024-00678-5","DOIUrl":"10.1038/s43588-024-00678-5","url":null,"abstract":"By developing a machine learning framework, a recent study substantially accelerates the calculation of electron–phonon coupling, making it computationally feasible to predict and understand a range of important physical phenomena, including electronic transport, hot-carrier relaxation, and superconductivity in complex materials.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"561-562"},"PeriodicalIF":12.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141997045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Network model with internal complexity bridges artificial intelligence and neuroscience 具有内部复杂性的网络模型是人工智能和神经科学的桥梁。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-16 DOI: 10.1038/s43588-024-00674-9
Linxuan He, Yunhui Xu, Weihua He, Yihan Lin, Yang Tian, Yujie Wu, Wenhui Wang, Ziyang Zhang, Junwei Han, Yonghong Tian, Bo Xu, Guoqi Li
Artificial intelligence (AI) researchers currently believe that the main approach to building more general model problems is the big AI model, where existing neural networks are becoming deeper, larger and wider. We term this the big model with external complexity approach. In this work we argue that there is another approach called small model with internal complexity, which can be used to find a suitable path of incorporating rich properties into neurons to construct larger and more efficient AI models. We uncover that one has to increase the scale of the network externally to stimulate the same dynamical properties. To illustrate this, we build a Hodgkin–Huxley (HH) network with rich internal complexity, where each neuron is an HH model, and prove that the dynamical properties and performance of the HH network can be equivalent to a bigger leaky integrate-and-fire (LIF) network, where each neuron is a LIF neuron with simple internal complexity. This study shows that by enhancing internal complexity of neurons in a Hodgkin–Huxley network, similar performance to larger, simpler networks can be achieved, suggesting an alternative path for powerful AI systems by focusing on neuron complexity.
人工智能(AI)研究人员目前认为,建立更通用模型问题的主要方法是大人工智能模型,即现有的神经网络变得更深、更大、更广。我们称之为外部复杂性大模型方法。在这项工作中,我们认为还有另一种方法,称为具有内部复杂性的小模型,可以用来找到一条合适的路径,将丰富的属性融入神经元,从而构建更大、更高效的人工智能模型。我们发现,要激发同样的动态特性,必须从外部扩大网络的规模。为了说明这一点,我们构建了一个具有丰富内部复杂性的霍奇金-赫胥黎(HH)网络,其中每个神经元都是一个 HH 模型,并证明 HH 网络的动态特性和性能可以等同于一个更大的泄漏积分发射(LIF)网络,其中每个神经元都是一个具有简单内部复杂性的 LIF 神经元。
{"title":"Network model with internal complexity bridges artificial intelligence and neuroscience","authors":"Linxuan He, Yunhui Xu, Weihua He, Yihan Lin, Yang Tian, Yujie Wu, Wenhui Wang, Ziyang Zhang, Junwei Han, Yonghong Tian, Bo Xu, Guoqi Li","doi":"10.1038/s43588-024-00674-9","DOIUrl":"10.1038/s43588-024-00674-9","url":null,"abstract":"Artificial intelligence (AI) researchers currently believe that the main approach to building more general model problems is the big AI model, where existing neural networks are becoming deeper, larger and wider. We term this the big model with external complexity approach. In this work we argue that there is another approach called small model with internal complexity, which can be used to find a suitable path of incorporating rich properties into neurons to construct larger and more efficient AI models. We uncover that one has to increase the scale of the network externally to stimulate the same dynamical properties. To illustrate this, we build a Hodgkin–Huxley (HH) network with rich internal complexity, where each neuron is an HH model, and prove that the dynamical properties and performance of the HH network can be equivalent to a bigger leaky integrate-and-fire (LIF) network, where each neuron is a LIF neuron with simple internal complexity. This study shows that by enhancing internal complexity of neurons in a Hodgkin–Huxley network, similar performance to larger, simpler networks can be achieved, suggesting an alternative path for powerful AI systems by focusing on neuron complexity.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"584-599"},"PeriodicalIF":12.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141997048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Enhancing structured light with optimized metasurfaces 利用优化的元表面增强结构光。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-16 DOI: 10.1038/s43588-024-00684-7
Jie Pan
{"title":"Enhancing structured light with optimized metasurfaces","authors":"Jie Pan","doi":"10.1038/s43588-024-00684-7","DOIUrl":"10.1038/s43588-024-00684-7","url":null,"abstract":"","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"558-558"},"PeriodicalIF":12.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141997047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Accelerating the calculation of electron–phonon coupling strength with machine learning 利用机器学习加速电子-声子耦合强度的计算。
IF 12 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2024-08-08 DOI: 10.1038/s43588-024-00668-7
Yang Zhong, Shixu Liu, Binhua Zhang, Zhiguo Tao, Yuting Sun, Weibin Chu, Xin-Gao Gong, Ji-Hui Yang, Hongjun Xiang
The calculation of electron–phonon couplings (EPCs) is essential for understanding various fundamental physical properties, including electrical transport, optical and superconducting behaviors in materials. However, obtaining EPCs through fully first-principles methods is notably challenging, particularly for large systems or when employing advanced functionals. Here we introduce a machine learning framework to accelerate EPC calculations by utilizing atomic orbital-based Hamiltonian matrices and gradients predicted by an equivariant graph neural network. We demonstrate that our method not only yields EPC values in close agreement with first-principles results but also enhances calculation efficiency by several orders of magnitude. Application to GaAs using the Heyd–Scuseria–Ernzerhof functional reveals the necessity of advanced functionals for accurate carrier mobility predictions, while for the large Kagome crystal CsV3Sb5, our framework reproduces the experimentally observed double domes in pressure-induced superconducting phase diagrams. This machine learning framework offers a powerful and efficient tool for the investigation of diverse EPC-related phenomena in complex materials. A machine learning framework is proposed to accurately predict electron–phonon coupling (EPC) strengths while reducing computational costs compared with first-principles methods. This approach facilitates EPC calculations with advanced functionals, allowing the accurate determination of real-world material properties such as carrier mobility and superconductivity.
计算电子-声子耦合(EPCs)对于理解各种基本物理特性,包括材料中的电传输、光学和超导行为至关重要。然而,通过完全第一性原理方法获得 EPCs 具有明显的挑战性,尤其是对于大型系统或采用高级函数时。在这里,我们引入了一种机器学习框架,利用基于原子轨道的哈密顿矩阵和等变图神经网络预测的梯度来加速 EPC 计算。我们证明,我们的方法不仅能得到与第一原理结果接近的 EPC 值,还能将计算效率提高几个数量级。使用海德-斯库瑟里亚-恩泽霍夫(Heyd-Scuseria-Ernzerhof)函数对砷化镓的应用揭示了高级函数对准确预测载流子迁移率的必要性,而对于大型卡戈米晶体 CsV3Sb5,我们的框架再现了实验观察到的压力诱导超导相图中的双圆顶。这一机器学习框架为研究复杂材料中与 EPC 相关的各种现象提供了强大而高效的工具。
{"title":"Accelerating the calculation of electron–phonon coupling strength with machine learning","authors":"Yang Zhong, Shixu Liu, Binhua Zhang, Zhiguo Tao, Yuting Sun, Weibin Chu, Xin-Gao Gong, Ji-Hui Yang, Hongjun Xiang","doi":"10.1038/s43588-024-00668-7","DOIUrl":"10.1038/s43588-024-00668-7","url":null,"abstract":"The calculation of electron–phonon couplings (EPCs) is essential for understanding various fundamental physical properties, including electrical transport, optical and superconducting behaviors in materials. However, obtaining EPCs through fully first-principles methods is notably challenging, particularly for large systems or when employing advanced functionals. Here we introduce a machine learning framework to accelerate EPC calculations by utilizing atomic orbital-based Hamiltonian matrices and gradients predicted by an equivariant graph neural network. We demonstrate that our method not only yields EPC values in close agreement with first-principles results but also enhances calculation efficiency by several orders of magnitude. Application to GaAs using the Heyd–Scuseria–Ernzerhof functional reveals the necessity of advanced functionals for accurate carrier mobility predictions, while for the large Kagome crystal CsV3Sb5, our framework reproduces the experimentally observed double domes in pressure-induced superconducting phase diagrams. This machine learning framework offers a powerful and efficient tool for the investigation of diverse EPC-related phenomena in complex materials. A machine learning framework is proposed to accurately predict electron–phonon coupling (EPC) strengths while reducing computational costs compared with first-principles methods. This approach facilitates EPC calculations with advanced functionals, allowing the accurate determination of real-world material properties such as carrier mobility and superconductivity.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 8","pages":"615-625"},"PeriodicalIF":12.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141908581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
Nature computational science
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1