Nature Machine Intelligence最新文献

Histopathology-based protein multiplex generation using deep learning 基于组织病理学的深度学习蛋白复合生成

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-08-04 DOI: 10.1038/s42256-025-01074-y

Sonali Andani, Boqi Chen, Joanna Ficek-Pascual, Simon Heinke, Ruben Casanova, Bernard Friedrich Hild, Bettina Sobottka, Bernd Bodenmiller, Viktor H. Koelzer, Gunnar Rätsch

{"title":"Histopathology-based protein multiplex generation using deep learning","authors":"Sonali Andani, Boqi Chen, Joanna Ficek-Pascual, Simon Heinke, Ruben Casanova, Bernard Friedrich Hild, Bettina Sobottka, Bernd Bodenmiller, Viktor H. Koelzer, Gunnar Rätsch","doi":"10.1038/s42256-025-01074-y","DOIUrl":"https://doi.org/10.1038/s42256-025-01074-y","url":null,"abstract":"Multiplexed protein imaging offers valuable insights into interactions between tumours and their surrounding tumour microenvironment, but its widespread use is limited by cost, time and tissue availability. Here we present HistoPlexer, a deep learning framework that generates spatially resolved protein multiplexes directly from standard haematoxylin and eosin (H&E) histopathology images. HistoPlexer jointly predicts multiple tumour and immune markers using a conditional generative adversarial architecture with custom loss functions designed to ensure pixel- and embedding-level similarity while mitigating slice-to-slice variations. A comprehensive evaluation of metastatic melanoma samples demonstrates that HistoPlexer-generated protein maps closely resemble real maps, as validated by expert assessment. They preserve crucial biological relationships by capturing spatial co-localization patterns among proteins. The spatial distribution of immune infiltration from HistoPlexer-generated protein multiplex enables stratification of tumours into immune subtypes. In an independent cohort, integration of HistoPlexer-derived features into predictive models enhances performance in survival prediction and immune subtype classification compared to models using H&E features alone. To assess broader applicability, we benchmarked HistoPlexer on publicly available pixel-aligned datasets from different cancer types. In all settings, HistoPlexer consistently outperformed baseline methods, demonstrating robustness across diverse tissue types and imaging conditions. By enabling whole-slide protein multiplex generation from routine H&E images, HistoPlexer offers a cost- and time-efficient approach to tumour microenvironment characterization with strong potential to advance precision oncology.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"15 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A semantic-enhanced multi-modal remote sensing foundation model for Earth observation 基于语义增强的多模态遥感对地观测基础模型

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-08-04 DOI: 10.1038/s42256-025-01078-8

Kang Wu, Yingying Zhang, Lixiang Ru, Bo Dang, Jiangwei Lao, Lei Yu, Junwei Luo, Zifan Zhu, Yue Sun, Jiahao Zhang, Qi Zhu, Jian Wang, Ming Yang, Jingdong Chen, Yongjun Zhang, Yansheng Li

{"title":"A semantic-enhanced multi-modal remote sensing foundation model for Earth observation","authors":"Kang Wu, Yingying Zhang, Lixiang Ru, Bo Dang, Jiangwei Lao, Lei Yu, Junwei Luo, Zifan Zhu, Yue Sun, Jiahao Zhang, Qi Zhu, Jian Wang, Ming Yang, Jingdong Chen, Yongjun Zhang, Yansheng Li","doi":"10.1038/s42256-025-01078-8","DOIUrl":"https://doi.org/10.1038/s42256-025-01078-8","url":null,"abstract":"Remote sensing foundation models, pretrained on massive remote sensing data, have shown impressive performance in several Earth observation (EO) tasks. These models usually use single-modal temporal data for pretraining, which is insufficient for multi-modal applications. Moreover, these models require a considerable number of samples for fine-tuning in downstream tasks, posing challenges in time-sensitive scenarios, such as rapid flood mapping. We present SkySense++, a multi-modal remote sensing foundation model for diverse EO tasks. SkySense++ has a factorized architecture to accommodate multi-modal images acquired by diverse sensors. We adopt progressive pretraining, which involves two stages, on meticulously curated datasets of 27 million multi-modal remote sensing images. The first representation-enhanced pretraining stage uses multi-granularity contrastive learning to obtain general representations. The second semantic-enhanced pretraining stage leverages masked semantic learning to learn semantically enriched representations, enabling few-shot capabilities. This ability allows the model to handle unseen tasks with minimal labelled data, alleviating the need for fine-tuning on extensive annotated data. SkySense++ demonstrates consistent improvements in classification, detection and segmentation over previous state-of-the-art models across 12 EO tasks in 7 domains: agriculture, forestry, oceanography, atmosphere, biology, land surveying and disaster management. This generalizability may lead to a new chapter of remote sensing foundation model applications for EO tasks at scale.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"31 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Type II mechanoreceptors and cuneate spiking neuronal network enable touch localization on a large-area e-skin II型机械感受器和楔形尖峰神经元网络实现了大面积电子皮肤的触觉定位

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-08-04 DOI: 10.1038/s42256-025-01076-w

Ana Clara Pereira Resende da Costa, Mariangela Filosa, Alcimar Barbosa Soares, Calogero Maria Oddo

{"title":"Type II mechanoreceptors and cuneate spiking neuronal network enable touch localization on a large-area e-skin","authors":"Ana Clara Pereira Resende da Costa, Mariangela Filosa, Alcimar Barbosa Soares, Calogero Maria Oddo","doi":"10.1038/s42256-025-01076-w","DOIUrl":"https://doi.org/10.1038/s42256-025-01076-w","url":null,"abstract":"The sense of touch is essential for humans to perceive, locate and react to physical stimuli. Notwithstanding the substantial advancements in e-skin research and related applications with collaborative robots and bionic prostheses, biomimetic intelligence remains a challenge in the attempt to understand and mimic somatosensory processing schemes. In this work, we present a large-area e-skin embedded with photonic fibre Bragg gratings, capable of decoding touch localization through a bioinspired two-layered spiking neuronal network. The implemented biomimicry of slowly adapting and fast-adapting type II primary afferents, cuneate neurons with overlapping receptive fields and neuroplasticity, enable unsupervised learning in localizing tactile stimuli with an error lower than 10 mm, and two-point discrimination thresholds matching human psychophysical thresholds in the forearm. These results align with biological findings and offer a promising step towards the development of bionic systems, opening new avenues for both practical applications and scientific explorations of somatosensation.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"26 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Protein–peptide docking with a rational and accurate diffusion generative model 合理准确的扩散生成模型实现蛋白肽对接

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-08-04 DOI: 10.1038/s42256-025-01077-9

Huifeng Zhao, Odin Zhang, Dejun Jiang, Zhenxing Wu, Hongyan Du, Xiaorui Wang, Yihao Zhao, Yuansheng Huang, Jingxuan Ge, Tingjun Hou, Yu Kang

{"title":"Protein–peptide docking with a rational and accurate diffusion generative model","authors":"Huifeng Zhao, Odin Zhang, Dejun Jiang, Zhenxing Wu, Hongyan Du, Xiaorui Wang, Yihao Zhao, Yuansheng Huang, Jingxuan Ge, Tingjun Hou, Yu Kang","doi":"10.1038/s42256-025-01077-9","DOIUrl":"https://doi.org/10.1038/s42256-025-01077-9","url":null,"abstract":"Therapeutic peptides represent the forefront of drug discovery, offering potent and safe alternatives to traditional small molecules. However, their weak and context-dependent nature complicates the efficient virtual screening and structural characterization of protein–peptide patterns. Here we introduce RAPiDock, a diffusion generative model designed for rational, accurate and rapid protein–peptide docking at an all-atomic level. RAPiDock efficiently reduces the sampling space by incorporating physical constraints and uses a bi-scale graph to effectively capture multidimensional structural information while balancing efficiency. In addition, the model uses a Clebsch–Gordan tensor product-based architecture to ensure physical symmetry. RAPiDock outperforms existing tools in prediction of protein–peptide-binding patterns, achieving a 93.7% success rate at top-25 predictions (13.4% higher than AlphaFold2-Multimer), with an execution speed of 0.35 seconds per complex (~270 times faster than AlphaFold2-Multimer). Extensive experiments demonstrate RAPiDock’s remarkable ability to handle 92 types of residue including posttranslational modifications, accurately predict subtle docking patterns, successfully identify multiple potential peptide-binding sites in global docking and serve as a powerful tool for high-throughput virtual screening with structural precision. All these push the boundaries of efficient protein–peptide docking in multiple real-application scenarios.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"28 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering 用逆神经渲染实现可概括和可解释的三维跟踪

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-08-04 DOI: 10.1038/s42256-025-01083-x

Julian Ost, Tanushree Banerjee, Mario Bijelic, Felix Heide

{"title":"Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering","authors":"Julian Ost, Tanushree Banerjee, Mario Bijelic, Felix Heide","doi":"10.1038/s42256-025-01083-x","DOIUrl":"https://doi.org/10.1038/s42256-025-01083-x","url":null,"abstract":"Today, the most successful methods for image-understanding tasks rely on feed-forward neural networks. Although this approach offers empirical accuracy, efficiency and task adaptation through fine-tuning, it also comes with fundamental disadvantages. Existing networks often struggle to generalize across different datasets, even on the same task. By design, these networks ultimately reason about high-dimensional scene features, which are challenging to analyse. This is true especially when attempting to predict three-dimensional (3D) information based on two-dimensional images. We propose to recast vision problems with RGB inputs as an inverse rendering problem by optimizing through a differentiable rendering pipeline over the latent space of pretrained 3D object representations and retrieving latents that best represent object instances in a given input image. Specifically, we solve the task of 3D multi-object tracking by optimizing an image loss over generative latent spaces that inherently disentangle shape and appearance properties. Not only do we investigate an alternative take on tracking but our method also enables us to examine the generated objects, reason about failure situations and resolve ambiguous cases. We validate the generalization and scaling capabilities of our method by learning the generative prior exclusively from synthetic data and assessing camera-based 3D tracking on two large-scale autonomous robot datasets. Both datasets are completely unseen to our method and do not require fine-tuning.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"15 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data meets prior knowledge for interpretable mechanistic inference in biology 数据满足生物学中可解释的机制推理的先验知识

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-07-22 DOI: 10.1038/s42256-025-01075-x

David Gomez-Cabrero, Jesper N. Tegnér

引用次数: 0

Emotional risks of AI companions demand attention AI同伴的情感风险需要关注

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-07-22 DOI: 10.1038/s42256-025-01093-9

引用次数: 0

Integrating multimodal cancer data using deep latent variable path modelling 利用深潜变量路径模型整合多模态癌症数据

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-07-22 DOI: 10.1038/s42256-025-01052-4

Alex Ing, Alvaro Andrades, Marco Raffaele Cosenza, Jan O. Korbel

{"title":"Integrating multimodal cancer data using deep latent variable path modelling","authors":"Alex Ing, Alvaro Andrades, Marco Raffaele Cosenza, Jan O. Korbel","doi":"10.1038/s42256-025-01052-4","DOIUrl":"https://doi.org/10.1038/s42256-025-01052-4","url":null,"abstract":"Cancers are commonly characterized by a complex pathology encompassing genetic, microscopic and macroscopic features, which can be probed individually using imaging and omics technologies. Integrating these data to obtain a full understanding of pathology remains challenging. We introduce a method called deep latent variable path modelling, which combines the representational power of deep learning with the capacity of path modelling to identify relationships between interacting elements in a complex system. To evaluate the capabilities of deep latent variable path modelling, we initially trained a model to map dependencies between single-nucleotide variant, methylation profiles, microRNA sequencing, RNA sequencing and histological data using breast cancer data from The Cancer Genome Atlas. This method exhibited superior performance in mapping associations between data types compared with classical path modelling. We additionally performed successful applications of the model to stratify single-cell data, identify synthetic lethal interactions using CRISPR–Cas9 screens derived from cell lines and detect histologic–transcriptional associations using spatial transcriptomic data. Results from each of these data types can then be understood with reference to the same holistic model of illness.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"14 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144678143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing deep learning-based field reconstruction with a differentiable learning framework 用可微学习框架增强基于深度学习的场重构

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-07-22 DOI: 10.1038/s42256-025-01063-1

Xu Liu, Wei Peng, Xiaoya Zhang, Xiaoyu Zhao, Weien Zhou, Wen Yao, Xiaoqian Chen

{"title":"Enhancing deep learning-based field reconstruction with a differentiable learning framework","authors":"Xu Liu, Wei Peng, Xiaoya Zhang, Xiaoyu Zhao, Weien Zhou, Wen Yao, Xiaoqian Chen","doi":"10.1038/s42256-025-01063-1","DOIUrl":"https://doi.org/10.1038/s42256-025-01063-1","url":null,"abstract":"Achieving accurate reconstructions of complex high-dimensional fields from sparse sensors remains a long-standing challenge. Frequently, reconstruction performance is mainly constrained by models and placement. The placement of sparse and prohibitive experimental sensors restricts information quality, resulting in formidable reconstruction tasks. Despite deep learning-based models having made strides, they typically lack the ability to co-optimize sensor placement. The joint optimization of high-dimensional neural network parameters versus low-dimensional sensor placement further poses significant difficulties. Here we present a general bilevel differentiable learning framework that effectively integrates models with sensor placement optimization (DSPO), enabling the dynamical search for the placement and accurate global field reconstruction. Within this framework, models are complemented with a differentiable operator to achieve the differentiability of placement. A gradient-based optimizer further empowers models by dynamically updating placement. The alternating optimization strategy is adopted to efficiently solve the joint optimization. We demonstrate the efficiency and generalizability of the DSPO on baseline models across various scenarios, including periodic and acyclic physical fields, regular and irregular grid datasets, and noisy and noiseless observations. Our results show that the DSPO significantly improves the reconstruction accuracy of models and robustness and advances baseline models comparable with the state-of-the-art performance. Our framework provides a new and general paradigm for the practical use of neural networks and placement optimization techniques for real-world applications.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"37 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144678141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Designing metamaterials with programmable nonlinear responses and geometric constraints in graph space 在图空间中设计具有可编程非线性响应和几何约束的超材料

IF 23.8 1区计算机科学

Nature Machine Intelligence Pub Date : 2025-07-22 DOI: 10.1038/s42256-025-01067-x

Marco Maurizi, Derek Xu, Yu-Tong Wang, Desheng Yao, David Hahn, Mourad Oudich, Anish Satpati, Mathieu Bauchy, Wei Wang, Yizhou Sun, Yun Jing, Xiaoyu Rayne Zheng

{"title":"Designing metamaterials with programmable nonlinear responses and geometric constraints in graph space","authors":"Marco Maurizi, Derek Xu, Yu-Tong Wang, Desheng Yao, David Hahn, Mourad Oudich, Anish Satpati, Mathieu Bauchy, Wei Wang, Yizhou Sun, Yun Jing, Xiaoyu Rayne Zheng","doi":"10.1038/s42256-025-01067-x","DOIUrl":"https://doi.org/10.1038/s42256-025-01067-x","url":null,"abstract":"Advances in data-driven design and additive manufacturing have substantially accelerated the development of truss metamaterials—three-dimensional truss networks—offering exceptional mechanical properties at a fraction of the weight of conventional solids. While existing design approaches can generate metamaterials with target linear properties, such as elasticity, they struggle to capture complex nonlinear behaviours and to incorporate geometric and manufacturing constraints—including defects—crucial for engineering applications. Here we present GraphMetaMat, an autoregressive graph-based framework capable of designing three-dimensional truss metamaterials with programmable nonlinear responses, originating from hard-to-capture physics such as buckling, frictional contact and wave propagation, along with arbitrary geometric constraints and defect tolerance. Integrating graph neural networks, physics biases, imitation learning, reinforcement learning and tree search, we show that GraphMetaMat can target stress–strain curves across four orders of magnitude and vibration transmission responses with varying attenuation gaps, unattainable by previous methods. We further demonstrate the use of GraphMetaMat for the inverse design of novel material topologies with tailorable high-energy absorption and vibration damping that outperform existing polymeric foams and phononic crystals, potentially suitable for protective equipment and electric vehicles. This work sets the stage for the automatic design of manufacturable, defect-tolerant materials with on-demand functionalities.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":"14 1","pages":""},"PeriodicalIF":23.8,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144677427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0