Nature Methods最新文献

{"title":"Scaling a foundational protein language model to 100 billion parameters.","authors":"","doi":"10.1038/s41592-025-02637-y","DOIUrl":"https://doi.org/10.1038/s41592-025-02637-y","url":null,"abstract":"","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143780653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"xTrimoPGLM: unified 100-billion-parameter pretrained transformer for deciphering the language of proteins.","authors":"Bo Chen, Xingyi Cheng, Pan Li, Yangli-Ao Geng, Jing Gong, Shen Li, Zhilei Bei, Xu Tan, Boyan Wang, Xin Zeng, Chiming Liu, Aohan Zeng, Yuxiao Dong, Jie Tang, Le Song","doi":"10.1038/s41592-025-02636-z","DOIUrl":"https://doi.org/10.1038/s41592-025-02636-z","url":null,"abstract":"<p><p>Protein language models have shown remarkable success in learning biological information from protein sequences. However, most existing models are limited by either autoencoding or autoregressive pretraining objectives, which makes them struggle to handle protein understanding and generation tasks concurrently. We propose a unified protein language model, xTrimoPGLM, to address these two types of tasks simultaneously through an innovative pretraining framework. Our key technical contribution is an exploration of the compatibility and the potential for joint optimization of the two types of objectives, which has led to a strategy for training xTrimoPGLM at an unprecedented scale of 100 billion parameters and 1 trillion training tokens. Our extensive experiments reveal that (1) xTrimoPGLM substantially outperforms other advanced baselines in 18 protein understanding benchmarks across four categories. The model also facilitates an atomic-resolution view of protein structures, leading to an advanced three-dimensional structural prediction model that surpasses existing language model-based tools. (2) xTrimoPGLM not only can generate de novo protein sequences following the principles of natural ones, but also can perform programmable generation after supervised fine-tuning on curated sequences. These results highlight the substantial capability and versatility of xTrimoPGLM in understanding and generating protein sequences, contributing to the evolving landscape of foundation models in protein science. Trained weight for the xTrimoPGLM model, and downstream datasets are available at https://huggingface.co/biomap-research .</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143780656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Life beyond labels.","authors":"Caroline Seydel","doi":"10.1038/s41592-025-02649-8","DOIUrl":"https://doi.org/10.1038/s41592-025-02649-8","url":null,"abstract":"","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143772467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The need to implement FAIR principles in biomolecular simulations.","authors":"Rommie E Amaro, Johan Åqvist, Ivet Bahar, Federica Battistini, Adam Bellaiche, Daniel Beltran, Philip C Biggin, Massimiliano Bonomi, Gregory R Bowman, Richard A Bryce, Giovanni Bussi, Paolo Carloni, David A Case, Andrea Cavalli, Chia-En A Chang, Thomas E Cheatham, Margaret S Cheung, Christophe Chipot, Lillian T Chong, Preeti Choudhary, G Andres Cisneros, Cecilia Clementi, Rosana Collepardo-Guevara, Peter Coveney, Roberto Covino, T Daniel Crawford, Matteo Dal Peraro, Bert L de Groot, Lucie Delemotte, Marco De Vivo, Jonathan W Essex, Franca Fraternali, Jiali Gao, Josep Ll Gelpí, Francesco L Gervasio, Fernando D González-Nilo, Helmut Grubmüller, Marina G Guenza, Horacio V Guzman, Sarah Harris, Teresa Head-Gordon, Rigoberto Hernandez, Adam Hospital, Niu Huang, Xuhui Huang, Gerhard Hummer, Javier Iglesias-Fernández, Jan H Jensen, Shantenu Jha, Wanting Jiao, William L Jorgensen, Shina C L Kamerlin, Syma Khalid, Charles Laughton, Michael Levitt, Vittorio Limongelli, Erik Lindahl, Kresten Lindorff-Larsen, Sharon Loverde, Magnus Lundborg, Yun L Luo, F Javier Luque, Charlotte I Lynch, Alexander D MacKerell, Alessandra Magistrato, Siewert J Marrink, Hugh Martin, J Andrew McCammon, Kenneth Merz, Vicent Moliner, Adrian J Mulholland, Sohail Murad, Athi N Naganathan, Shikha Nangia, Frank Noe, Agnes Noy, Julianna Oláh, Megan L O'Mara, Mary Jo Ondrechen, Jose N Onuchic, Alexey Onufriev, Sílvia Osuna, Giulia Palermo, Anna R Panchenko, Sergio Pantano, Carol Parish, Michele Parrinello, Alberto Perez, Tomas Perez-Acle, Juan R Perilla, B Montgomery Pettitt, Adriana Pietropaolo, Jean-Philip Piquemal, Adolfo B Poma, Matej Praprotnik, Maria J Ramos, Pengyu Ren, Nathalie Reuter, Adrian Roitberg, Edina Rosta, Carme Rovira, Benoit Roux, Ursula Rothlisberger, Karissa Y Sanbonmatsu, Tamar Schlick, Alexey K Shaytan, Carlos Simmerling, Jeremy C Smith, Yuji Sugita, Katarzyna Świderek, Makoto Taiji, Peng Tao, D Peter Tieleman, Irina G Tikhonova, Julian Tirado-Rives, Iñaki Tuñón, Marc W van der Kamp, David van der Spoel, Sameer Velankar, Gregory A Voth, Rebecca Wade, Ariel Warshel, Valerie Vaissier Welborn, Stacey D Wetmore, Travis J Wheeler, Chung F Wong, Lee-Wei Yang, Martin Zacharias, Modesto Orozco","doi":"10.1038/s41592-025-02635-0","DOIUrl":"https://doi.org/10.1038/s41592-025-02635-0","url":null,"abstract":"","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143772471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sharing data from the Human Tumor Atlas Network through standards, infrastructure and community engagement.","authors":"Ino de Bruijn, Milen Nikolov, Clarisse Lau, Ashley Clayton, David L Gibbs, Elvira Mitraka, Dar'ya Pozhidayeva, Alex Lash, Selcuk Onur Sumer, Jennifer Altreuter, Kristen Anton, Mialy DeFelice, Xiang Li, Aaron Lisman, William J R Longabaugh, Jeremy Muhlich, Sandro Santagata, Subhiksha Nandakumar, Peter K Sorger, Christine Suver, Xengie Doan, Justin Guinney, Nikolaus Schultz, Adam J Taylor, Vésteinn Thorsson, Ethan Cerami, James A Eddy","doi":"10.1038/s41592-025-02643-0","DOIUrl":"https://doi.org/10.1038/s41592-025-02643-0","url":null,"abstract":"<p><p>Data from the first phase of the Human Tumor Atlas Network (HTAN) are now available, comprising 8,425 biospecimens from 2,042 research participants profiled with more than 20 molecular assays. The data were generated to study the evolution from precancerous to advanced disease. The HTAN Data Coordinating Center (DCC) has enabled their dissemination and effective reuse. We describe the diverse datasets, how to access them, data standards, underlying infrastructure and governance approaches, and our methods to sustain community engagement. HTAN data can be accessed through the HTAN Portal, explored in visualization tools-including CellxGene, Minerva and cBioPortal-and analyzed in the cloud through the NCI Cancer Research Data Commons. Infrastructure was developed to enable data ingestion and dissemination through the Synapse platform. The HTAN DCC's flexible and modular approach to sharing complex cancer research data offers valuable insights to other data-coordination efforts and researchers looking to leverage HTAN data.</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143753445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Denoising Search doubles the number of metabolite and exposome annotations in human plasma using an Orbitrap Astral mass spectrometer.","authors":"Fanzhou Kong, Tong Shen, Yuanyue Li, Amer Bashar, Susan S Bird, Oliver Fiehn","doi":"10.1038/s41592-025-02646-x","DOIUrl":"10.1038/s41592-025-02646-x","url":null,"abstract":"<p><p>Chemical exposures may affect human metabolism and contribute to the etiology of neurodegenerative disorders such as Alzheimer's disease. Identifying these small metabolites involves matching experimental spectra to reference spectra in databases. However, environmental chemicals or physiologically active metabolites are usually present at low concentrations in human specimens. The presence of noise ions can substantially degrade spectral quality, leading to false negatives and reduced identification rates. In response to this challenge, the Spectral Denoising algorithm removes both chemical and electronic noise. Spectral Denoising outperformed alternative methods in benchmarking studies on 240 tested metabolites. It improved high confident compound identifications at an average 35-fold lower concentrations than previously achievable. Spectral Denoising proved highly robust against varying levels of both chemical and electronic noise even with a greater than 150-fold higher intensity of noise ions than true fragment ions. For human plasma samples from patients with Alzheimer's disease that were analyzed on the Orbitrap Astral mass spectrometer, Denoising Search detected 2.5-fold more annotated compounds compared to the Exploris 240 Orbitrap instrument, including drug metabolites, household and industrial chemicals, and pesticides.</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143742981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enabling global image data sharing in the life sciences.","authors":"Peter Bajcsy, Sreenivas Bhattiprolu, Katy Börner, Beth A Cimini, Lucy Collinson, Jan Ellenberg, Reto Fiolka, Maryellen Giger, Wojtek Goscinski, Matthew Hartley, Nathan Hotaling, Rick Horwitz, Florian Jug, Isabel Kemmer, Anna Kreshuk, Emma Lundberg, Aastha Mathur, Kedar Narayan, Shuichi Onami, Anne L Plant, Fred Prior, Jason R Swedlow, Adam Taylor, Antje Keppler","doi":"10.1038/s41592-024-02585-z","DOIUrl":"https://doi.org/10.1038/s41592-024-02585-z","url":null,"abstract":"<p><p>Despite the importance of imaging in biological and medical research, a large body of informative and precious image data never sees the light of day. To ensure scientific rigor as well as the reuse of data for scientific discovery, image data need to be made FAIR (findable, accessible, interoperable and reusable). Image data experts are working together globally to agree on common data formats, metadata, ontologies and supporting tools toward image data FAIRification. With this Perspective, we call on public funders to join these efforts to support their national scientists. What researchers most urgently need are openly accessible resources for image data storage that are operated under long-term commitments by their funders. Although existing resources in Australia, Japan and Europe are already collaborating to enable global image data sharing, these efforts will fall short unless more countries invest in operating and federating their own open data resources. This will allow us to harvest the enormous potential of existing image data, preventing substantial loss of unrealized value from past investments in imaging acquisition infrastructure.</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143743075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncalled4 improves nanopore DNA and RNA modification detection via fast and accurate signal alignment.","authors":"Sam Kovaka, Paul W Hook, Katharine M Jenike, Vikram Shivakumar, Luke B Morina, Roham Razaghi, Winston Timp, Michael C Schatz","doi":"10.1038/s41592-025-02631-4","DOIUrl":"10.1038/s41592-025-02631-4","url":null,"abstract":"<p><p>Nanopore signal analysis enables detection of nucleotide modifications from native DNA and RNA sequencing, providing both accurate genetic or transcriptomic and epigenetic information without additional library preparation. At present, only a limited set of modifications can be directly basecalled (for example, 5-methylcytosine), while most others require exploratory methods that often begin with alignment of nanopore signal to a nucleotide reference. We present Uncalled4, a toolkit for nanopore signal alignment, analysis and visualization. Uncalled4 features an efficient banded signal alignment algorithm, BAM signal alignment file format, statistics for comparing signal alignment methods and a reproducible de novo training method for k-mer-based pore models, revealing potential errors in Oxford Nanopore Technologies' state-of-the-art DNA model. We apply Uncalled4 to RNA 6-methyladenine (m6A) detection in seven human cell lines, identifying 26% more modifications than Nanopolish using m6Anet, including in several genes where m6A has known implications in cancer. Uncalled4 is available open source at github.com/skovaka/uncalled4 .</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143743080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Atomic context-conditioned protein sequence design using LigandMPNN.","authors":"Justas Dauparas, Gyu Rie Lee, Robert Pecoraro, Linna An, Ivan Anishchenko, Cameron Glasscock, David Baker","doi":"10.1038/s41592-025-02626-1","DOIUrl":"https://doi.org/10.1038/s41592-025-02626-1","url":null,"abstract":"<p><p>Protein sequence design in the context of small molecules, nucleotides and metals is critical to enzyme and small-molecule binder and sensor design, but current state-of-the-art deep-learning-based sequence design methods are unable to model nonprotein atoms and molecules. Here we describe a deep-learning-based protein sequence design method called LigandMPNN that explicitly models all nonprotein components of biomolecular systems. LigandMPNN significantly outperforms Rosetta and ProteinMPNN on native backbone sequence recovery for residues interacting with small molecules (63.3% versus 50.4% and 50.5%), nucleotides (50.5% versus 35.2% and 34.0%) and metals (77.5% versus 36.0% and 40.6%). LigandMPNN generates not only sequences but also sidechain conformations to allow detailed evaluation of binding interactions. LigandMPNN has been used to design over 100 experimentally validated small-molecule and DNA-binding proteins with high affinity and high structural accuracy (as indicated by four X-ray crystal structures), and redesign of Rosetta small-molecule binder designs has increased binding affinity by as much as 100-fold. We anticipate that LigandMPNN will be widely useful for designing new binding proteins, sensors and enzymes.</p>","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143742961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simultaneous single-cell DNA damage profiling and RNA sequencing by Paired-Damage-seq.","authors":"","doi":"10.1038/s41592-025-02634-1","DOIUrl":"https://doi.org/10.1038/s41592-025-02634-1","url":null,"abstract":"","PeriodicalId":18981,"journal":{"name":"Nature Methods","volume":" ","pages":""},"PeriodicalIF":36.1,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143701109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}