Pub Date : 2024-12-19DOI: 10.1038/s43588-024-00745-x
Emmet A. Francis, Justin G. Laughlin, Jørgen S. Dokken, Henrik N. T. Finsberg, Christopher T. Lee, Marie E. Rognes, Padmini Rangamani
Biological cells rely on precise spatiotemporal coordination of biochemical reactions to control their functions. Such cell signaling networks have been a common focus for mathematical models, but they remain challenging to simulate, particularly in realistic cell geometries. Here we present Spatial Modeling Algorithms for Reactions and Transport (SMART), a software package that takes in high-level user specifications about cell signaling networks and then assembles and solves the associated mathematical systems. SMART uses state-of-the-art finite element analysis, via the FEniCS Project software, to efficiently and accurately resolve cell signaling events over discretized cellular and subcellular geometries. We demonstrate its application to several different biological systems, including yes-associated protein (YAP)/PDZ-binding motif (TAZ) mechanotransduction, calcium signaling in neurons and cardiomyocytes, and ATP generation in mitochondria. Throughout, we utilize experimentally derived realistic cellular geometries represented by well-conditioned tetrahedral meshes. These scenarios demonstrate the applicability, flexibility, accuracy and efficiency of SMART across a range of temporal and spatial scales. Spatial Modeling Algorithms for Reactions and Transport (SMART) is a software package that allows users to simulate spatially resolved biochemical signaling networks within realistic geometries of cells and organelles.
生物细胞依靠生化反应的精确时空协调来控制其功能。这种细胞信号网络一直是数学模型的研究重点，但它们的模拟仍然具有挑战性，尤其是在现实的细胞几何结构中。我们在此介绍反应和运输的空间建模算法（SMART），这是一个软件包，可接收用户关于细胞信号网络的高级规格，然后组装并求解相关的数学系统。通过 FEniCS 项目软件，SMART 利用最先进的有限元分析技术，高效、准确地解决离散化细胞和亚细胞几何结构上的细胞信号传导问题。我们展示了它在几个不同生物系统中的应用，包括 yes-associated protein (YAP)/PDZ-binding motif (TAZ) 机械传导、神经元和心肌细胞中的钙信号转导以及线粒体中的 ATP 生成。在整个过程中，我们利用实验得出的现实细胞几何图形，这些几何图形由条件良好的四面体网格表示。这些场景证明了 SMART 在一系列时间和空间尺度上的适用性、灵活性、准确性和效率。
{"title":"Spatial modeling algorithms for reactions and transport in biological cells","authors":"Emmet A. Francis, Justin G. Laughlin, Jørgen S. Dokken, Henrik N. T. Finsberg, Christopher T. Lee, Marie E. Rognes, Padmini Rangamani","doi":"10.1038/s43588-024-00745-x","DOIUrl":"10.1038/s43588-024-00745-x","url":null,"abstract":"Biological cells rely on precise spatiotemporal coordination of biochemical reactions to control their functions. Such cell signaling networks have been a common focus for mathematical models, but they remain challenging to simulate, particularly in realistic cell geometries. Here we present Spatial Modeling Algorithms for Reactions and Transport (SMART), a software package that takes in high-level user specifications about cell signaling networks and then assembles and solves the associated mathematical systems. SMART uses state-of-the-art finite element analysis, via the FEniCS Project software, to efficiently and accurately resolve cell signaling events over discretized cellular and subcellular geometries. We demonstrate its application to several different biological systems, including yes-associated protein (YAP)/PDZ-binding motif (TAZ) mechanotransduction, calcium signaling in neurons and cardiomyocytes, and ATP generation in mitochondria. Throughout, we utilize experimentally derived realistic cellular geometries represented by well-conditioned tetrahedral meshes. These scenarios demonstrate the applicability, flexibility, accuracy and efficiency of SMART across a range of temporal and spatial scales. Spatial Modeling Algorithms for Reactions and Transport (SMART) is a software package that allows users to simulate spatially resolved biochemical signaling networks within realistic geometries of cells and organelles.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"5 1","pages":"76-89"},"PeriodicalIF":12.0,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11774757/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142866530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-18DOI: 10.1038/s43588-024-00752-y
Muratahan Aykol, Amil Merchant, Simon Batzner, Jennifer N. Wei, Ekin Dogus Cubuk
Crystallization of amorphous precursors into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to the synthesis and development of new materials in the laboratory. Reliably predicting the outcome of such a process would enable new research directions in these areas, but has remained beyond the reach of molecular modeling or ab initio methods. Here we show that candidates for the crystallization products of amorphous precursors can be predicted in many inorganic systems by sampling the local structural motifs at the atomistic level using universal deep learning interatomic potentials. We show that this approach identifies, with high accuracy, the most likely crystal structures of the polymorphs that initially nucleate from amorphous precursors, across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides and metal alloys. This study introduces a2c, a computational method that leverages machine learning and atomistic simulations to predict the most likely crystallization products upon annealing of amorphous precursors. The a2c tool was demonstrated on a variety of materials, including oxides, nitrides and metallic glasses, and can assist researchers in discovering synthesis pathways for materials design.
{"title":"Predicting emergence of crystals from amorphous precursors with deep learning potentials","authors":"Muratahan Aykol, Amil Merchant, Simon Batzner, Jennifer N. Wei, Ekin Dogus Cubuk","doi":"10.1038/s43588-024-00752-y","DOIUrl":"10.1038/s43588-024-00752-y","url":null,"abstract":"Crystallization of amorphous precursors into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to the synthesis and development of new materials in the laboratory. Reliably predicting the outcome of such a process would enable new research directions in these areas, but has remained beyond the reach of molecular modeling or ab initio methods. Here we show that candidates for the crystallization products of amorphous precursors can be predicted in many inorganic systems by sampling the local structural motifs at the atomistic level using universal deep learning interatomic potentials. We show that this approach identifies, with high accuracy, the most likely crystal structures of the polymorphs that initially nucleate from amorphous precursors, across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides and metal alloys. This study introduces a2c, a computational method that leverages machine learning and atomistic simulations to predict the most likely crystallization products upon annealing of amorphous precursors. The a2c tool was demonstrated on a variety of materials, including oxides, nitrides and metallic glasses, and can assist researchers in discovering synthesis pathways for materials design.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"5 2","pages":"105-111"},"PeriodicalIF":12.0,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00752-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142857211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-16DOI: 10.1038/s43588-024-00738-w
Mengdi Zhao, Ning Wang, Xinrui Jiang, Xiaoyang Ma, Haixin Ma, Gan He, Kai Du, Lei Ma, Tiejun Huang
The behavior of an organism is influenced by the complex interplay between its brain, body and environment. Existing data-driven models focus on either the brain or the body–environment. Here we present BAAIWorm, an integrative data-driven model of Caenorhabditis elegans, which consists of two submodels: the brain model and the body–environment model. The brain model was built by multicompartment models with realistic morphology, connectome and neural population dynamics based on experimental data. Simultaneously, the body–environment model used a lifelike body and a three-dimensional physical environment. Through the closed-loop interaction between the two submodels, BAAIWorm reproduced the realistic zigzag movement toward attractors observed in C. elegans. Leveraging this model, we investigated the impact of neural system structure on both neural activities and behaviors. Consequently, BAAIWorm can enhance our understanding of how the brain controls the body to interact with its surrounding environment. BAAIWorm is an integrative data-driven model of C. elegans that simulates interactions between the brain, body and environment. The biophysically detailed neuronal model is capable of replicating the zigzag movement observed in this species.
{"title":"An integrative data-driven model simulating C. elegans brain, body and environment interactions","authors":"Mengdi Zhao, Ning Wang, Xinrui Jiang, Xiaoyang Ma, Haixin Ma, Gan He, Kai Du, Lei Ma, Tiejun Huang","doi":"10.1038/s43588-024-00738-w","DOIUrl":"10.1038/s43588-024-00738-w","url":null,"abstract":"The behavior of an organism is influenced by the complex interplay between its brain, body and environment. Existing data-driven models focus on either the brain or the body–environment. Here we present BAAIWorm, an integrative data-driven model of Caenorhabditis elegans, which consists of two submodels: the brain model and the body–environment model. The brain model was built by multicompartment models with realistic morphology, connectome and neural population dynamics based on experimental data. Simultaneously, the body–environment model used a lifelike body and a three-dimensional physical environment. Through the closed-loop interaction between the two submodels, BAAIWorm reproduced the realistic zigzag movement toward attractors observed in C. elegans. Leveraging this model, we investigated the impact of neural system structure on both neural activities and behaviors. Consequently, BAAIWorm can enhance our understanding of how the brain controls the body to interact with its surrounding environment. BAAIWorm is an integrative data-driven model of C. elegans that simulates interactions between the brain, body and environment. The biophysically detailed neuronal model is capable of replicating the zigzag movement observed in this species.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 12","pages":"978-990"},"PeriodicalIF":12.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00738-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142840495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-16DOI: 10.1038/s43588-024-00740-2
We created an open-source model that simulates Caenorhabditis elegans in a closed-loop system, by integrating simulations of its brain, its physical body, and its environment. BAAIWorm replicated C. elegans locomotive behaviors, and synthetic perturbations of synaptic connections impacted neural control of movement and affected the embodied motor behavior.
{"title":"A simulated C. elegans with biophysically detailed neurons and muscle dynamics","authors":"","doi":"10.1038/s43588-024-00740-2","DOIUrl":"10.1038/s43588-024-00740-2","url":null,"abstract":"We created an open-source model that simulates Caenorhabditis elegans in a closed-loop system, by integrating simulations of its brain, its physical body, and its environment. BAAIWorm replicated C. elegans locomotive behaviors, and synthetic perturbations of synaptic connections impacted neural control of movement and affected the embodied motor behavior.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 12","pages":"888-889"},"PeriodicalIF":12.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142840494","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}
Pub Date : 2024-12-12DOI: 10.1038/s43588-024-00741-1
Tiancheng Hu, Yara Kyrychenko, Steve Rathje, Nigel Collier, Sander van der Linden, Jon Roozenbeek
Social identity biases, particularly the tendency to favor one’s own group (ingroup solidarity) and derogate other groups (outgroup hostility), are deeply rooted in human psychology and social behavior. However, it is unknown if such biases are also present in artificial intelligence systems. Here we show that large language models (LLMs) exhibit patterns of social identity bias, similarly to humans. By administering sentence completion prompts to 77 different LLMs (for instance, ‘We are…’), we demonstrate that nearly all base models and some instruction-tuned and preference-tuned models display clear ingroup favoritism and outgroup derogation. These biases manifest both in controlled experimental settings and in naturalistic human–LLM conversations. However, we find that careful curation of training data and specialized fine-tuning can substantially reduce bias levels. These findings have important implications for developing more equitable artificial intelligence systems and highlight the urgent need to understand how human–LLM interactions might reinforce existing social biases. Researchers show that large language models exhibit social identity biases similar to humans, having favoritism toward ingroups and hostility toward outgroups. These biases persist across models, training data and real-world human–LLM conversations.
{"title":"Generative language models exhibit social identity biases","authors":"Tiancheng Hu, Yara Kyrychenko, Steve Rathje, Nigel Collier, Sander van der Linden, Jon Roozenbeek","doi":"10.1038/s43588-024-00741-1","DOIUrl":"10.1038/s43588-024-00741-1","url":null,"abstract":"Social identity biases, particularly the tendency to favor one’s own group (ingroup solidarity) and derogate other groups (outgroup hostility), are deeply rooted in human psychology and social behavior. However, it is unknown if such biases are also present in artificial intelligence systems. Here we show that large language models (LLMs) exhibit patterns of social identity bias, similarly to humans. By administering sentence completion prompts to 77 different LLMs (for instance, ‘We are…’), we demonstrate that nearly all base models and some instruction-tuned and preference-tuned models display clear ingroup favoritism and outgroup derogation. These biases manifest both in controlled experimental settings and in naturalistic human–LLM conversations. However, we find that careful curation of training data and specialized fine-tuning can substantially reduce bias levels. These findings have important implications for developing more equitable artificial intelligence systems and highlight the urgent need to understand how human–LLM interactions might reinforce existing social biases. Researchers show that large language models exhibit social identity biases similar to humans, having favoritism toward ingroups and hostility toward outgroups. These biases persist across models, training data and real-world human–LLM conversations.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"5 1","pages":"65-75"},"PeriodicalIF":12.0,"publicationDate":"2024-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11774750/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142820326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-10DOI: 10.1038/s43588-024-00735-z
Filip Milisav, Vincent Bazinet, Richard F. Betzel, Bratislav Misic
Scientific discovery in connectomics relies on network null models. The prominence of network features is conventionally evaluated against null distributions estimated using randomized networks. Modern imaging technologies provide an increasingly rich array of biologically meaningful edge weights. Despite the prevalence of weighted graph analysis in connectomics, randomization models that only preserve binary node degree remain most widely used. Here we propose a simulated annealing procedure for generating randomized networks that preserve weighted degree (strength) sequences. We show that the procedure outperforms other rewiring algorithms and generalizes to multiple network formats, including directed and signed networks, as well as diverse real-world networks. Throughout, we use morphospace representation to assess the sampling behavior of the algorithm and the variability of the resulting ensemble. Finally, we show that accurate strength preservation yields different inferences about brain network organization. Collectively, this work provides a simple but powerful method to analyze richly detailed next-generation connectomics datasets. This study proposes an algorithm for generating randomized networks that preserve the weighted degree sequence. The procedure outperforms standard rewiring algorithms and extends to multiple network types, including directed and signed networks.
{"title":"A simulated annealing algorithm for randomizing weighted networks","authors":"Filip Milisav, Vincent Bazinet, Richard F. Betzel, Bratislav Misic","doi":"10.1038/s43588-024-00735-z","DOIUrl":"10.1038/s43588-024-00735-z","url":null,"abstract":"Scientific discovery in connectomics relies on network null models. The prominence of network features is conventionally evaluated against null distributions estimated using randomized networks. Modern imaging technologies provide an increasingly rich array of biologically meaningful edge weights. Despite the prevalence of weighted graph analysis in connectomics, randomization models that only preserve binary node degree remain most widely used. Here we propose a simulated annealing procedure for generating randomized networks that preserve weighted degree (strength) sequences. We show that the procedure outperforms other rewiring algorithms and generalizes to multiple network formats, including directed and signed networks, as well as diverse real-world networks. Throughout, we use morphospace representation to assess the sampling behavior of the algorithm and the variability of the resulting ensemble. Finally, we show that accurate strength preservation yields different inferences about brain network organization. Collectively, this work provides a simple but powerful method to analyze richly detailed next-generation connectomics datasets. This study proposes an algorithm for generating randomized networks that preserve the weighted degree sequence. The procedure outperforms standard rewiring algorithms and extends to multiple network types, including directed and signed networks.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"5 1","pages":"48-64"},"PeriodicalIF":12.0,"publicationDate":"2024-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11774763/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142807370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-09DOI: 10.1038/s43588-024-00732-2
Minglang Yin, Nicolas Charon, Ryan Brody, Lu Lu, Natalia Trayanova, Mauro Maggioni
Solving partial differential equations (PDEs) using numerical methods is a ubiquitous task in engineering and medicine. However, the computational costs can be prohibitively high when many-query evaluations of PDE solutions on multiple geometries are needed. Here we aim to address this challenge by introducing Diffeomorphic Mapping Operator Learning (DIMON), a generic artificial intelligence framework that learns geometry-dependent solution operators of different types of PDE on a variety of geometries. We present several examples to demonstrate the performance, efficiency and scalability of the framework in learning both static and time-dependent PDEs on parameterized and non-parameterized domains; these include solving the Laplace equations, reaction–diffusion equations and a system of multiscale PDEs that characterize the electrical propagation on thousands of personalized heart digital twins. DIMON can reduce the computational costs of solution approximations on multiple geometries from hours to seconds with substantially less computational resources. This work presents an artificial intelligence framework to learn geometry-dependent solution operators of partial differential equations (PDEs). The framework enables scalable and fast approximations of PDE solutions on a variety of 3D geometries.
{"title":"A scalable framework for learning the geometry-dependent solution operators of partial differential equations","authors":"Minglang Yin, Nicolas Charon, Ryan Brody, Lu Lu, Natalia Trayanova, Mauro Maggioni","doi":"10.1038/s43588-024-00732-2","DOIUrl":"10.1038/s43588-024-00732-2","url":null,"abstract":"Solving partial differential equations (PDEs) using numerical methods is a ubiquitous task in engineering and medicine. However, the computational costs can be prohibitively high when many-query evaluations of PDE solutions on multiple geometries are needed. Here we aim to address this challenge by introducing Diffeomorphic Mapping Operator Learning (DIMON), a generic artificial intelligence framework that learns geometry-dependent solution operators of different types of PDE on a variety of geometries. We present several examples to demonstrate the performance, efficiency and scalability of the framework in learning both static and time-dependent PDEs on parameterized and non-parameterized domains; these include solving the Laplace equations, reaction–diffusion equations and a system of multiscale PDEs that characterize the electrical propagation on thousands of personalized heart digital twins. DIMON can reduce the computational costs of solution approximations on multiple geometries from hours to seconds with substantially less computational resources. This work presents an artificial intelligence framework to learn geometry-dependent solution operators of partial differential equations (PDEs). The framework enables scalable and fast approximations of PDE solutions on a variety of 3D geometries.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 12","pages":"928-940"},"PeriodicalIF":12.0,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00732-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142803066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-12-09DOI: 10.1038/s43588-024-00737-x
Arne Schneuing, Charles Harris, Yuanqi Du, Kieran Didi, Arian Jamasb, Ilia Igashov, Weitao Du, Carla Gomes, Tom L. Blundell, Pietro Lio, Max Welling, Michael Bronstein, Bruno Correia
Structure-based drug design (SBDD) aims to design small-molecule ligands that bind with high affinity and specificity to pre-determined protein targets. Generative SBDD methods leverage structural data of drugs with their protein targets to propose new drug candidates. However, most existing methods focus exclusively on bottom-up de novo design of compounds or tackle other drug development challenges with task-specific models. The latter requires curation of suitable datasets, careful engineering of the models and retraining from scratch for each task. Here we show how a single pretrained diffusion model can be applied to a broader range of problems, such as off-the-shelf property optimization, explicit negative design and partial molecular design with inpainting. We formulate SBDD as a three-dimensional conditional generation problem and present DiffSBDD, an SE(3)-equivariant diffusion model that generates novel ligands conditioned on protein pockets. Furthermore, we show how additional constraints can be used to improve the generated drug candidates according to a variety of computational metrics. This work applies diffusion models to conditional molecule generation and shows how they can be used to tackle various structure-based drug design problems
{"title":"Structure-based drug design with equivariant diffusion models","authors":"Arne Schneuing, Charles Harris, Yuanqi Du, Kieran Didi, Arian Jamasb, Ilia Igashov, Weitao Du, Carla Gomes, Tom L. Blundell, Pietro Lio, Max Welling, Michael Bronstein, Bruno Correia","doi":"10.1038/s43588-024-00737-x","DOIUrl":"10.1038/s43588-024-00737-x","url":null,"abstract":"Structure-based drug design (SBDD) aims to design small-molecule ligands that bind with high affinity and specificity to pre-determined protein targets. Generative SBDD methods leverage structural data of drugs with their protein targets to propose new drug candidates. However, most existing methods focus exclusively on bottom-up de novo design of compounds or tackle other drug development challenges with task-specific models. The latter requires curation of suitable datasets, careful engineering of the models and retraining from scratch for each task. Here we show how a single pretrained diffusion model can be applied to a broader range of problems, such as off-the-shelf property optimization, explicit negative design and partial molecular design with inpainting. We formulate SBDD as a three-dimensional conditional generation problem and present DiffSBDD, an SE(3)-equivariant diffusion model that generates novel ligands conditioned on protein pockets. Furthermore, we show how additional constraints can be used to improve the generated drug candidates according to a variety of computational metrics. This work applies diffusion models to conditional molecule generation and shows how they can be used to tackle various structure-based drug design problems","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 12","pages":"899-909"},"PeriodicalIF":12.0,"publicationDate":"2024-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00737-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142803183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：