Veit-Lorenz Heuthe, Lukas Seemann, Samuel Tovey, Clemens Bechinger
{"title":"Reservoir computing from collective dynamics of active colloidal oscillators.","authors":"Veit-Lorenz Heuthe, Lukas Seemann, Samuel Tovey, Clemens Bechinger","doi":"10.1038/s44488-026-00001-3","DOIUrl":"10.1038/s44488-026-00001-3","url":null,"abstract":"<p><p>Physical reservoir computing is a computational framework that offers an energy- and computation-efficient alternative to conventional training of neural networks. In reservoir computing, input signals are mapped into the high-dimensional dynamics of a nonlinear system, and only a simple readout layer is trained. In most physical implementations, the interactions that give rise to the dynamics cannot be tuned directly and high dimensionality is typically achieved through time-multiplexing, which can limit flexibility and efficiency. Here we introduce a reservoir composed of hundreds of hydrodynamically coupled active colloidal oscillators forming a fully parallel physical reservoir whose coupling strength and memory can be tuned in situ. The collective dynamics of the active oscillators allow accurate predictions of chaotic time series without time-multiplexing. We further demonstrate real-time detection of subtle hidden anomalies that preserve all instantaneous statistical properties of the signal. These results establish interacting active colloids as a reconfigurable platform for physical computation and model-free detection of irregularities in complex time signals.</p>","PeriodicalId":524152,"journal":{"name":"Communications AI & computing","volume":"1 1","pages":"6"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13446078/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148687706","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}
Barbora Rehák Bučková, Charlotte Fraza, Cecilie Koldbæk Lemvigh, Camilla Bärthel Flaaten, Linn Sofie Sæther, Giulia Cattarinussi, Karen Marie Sandø Ambrosen, Ole Andreas Andreassen, Lars Tjelta Westlye, Christian Beckmann, Bjørn Hylsebeck Ebdrup, Paola Dazzan, Torill Ueland, Robin Mitra, Andre Marquand
{"title":"Structured missingness in biomedical research.","authors":"Barbora Rehák Bučková, Charlotte Fraza, Cecilie Koldbæk Lemvigh, Camilla Bärthel Flaaten, Linn Sofie Sæther, Giulia Cattarinussi, Karen Marie Sandø Ambrosen, Ole Andreas Andreassen, Lars Tjelta Westlye, Christian Beckmann, Bjørn Hylsebeck Ebdrup, Paola Dazzan, Torill Ueland, Robin Mitra, Andre Marquand","doi":"10.1038/s44488-026-00025-9","DOIUrl":"https://doi.org/10.1038/s44488-026-00025-9","url":null,"abstract":"<p><p>Missing data are ubiquitous in biomedical research, and become particularly problematic when combining data from multiple clinical studies, an increasingly common practice for building models that generalise across populations. This process introduces complex structured patterns of missing data ('structured missingness'), where one study may omit variables that another collects, leaving deterministic gaps that standard imputation methods were not designed to handle. Here, we show that many popular imputation algorithms systematically fail under structured missingness, distorting the underlying statistical properties of the data in ways that conventional accuracy metrics cannot detect. Methods designed to preserve data distributions are more robust, and a hierarchical modelling approach that explicitly accounts for site-specific differences further improves performance. We introduce a multi-metric evaluation framework and a practical decision guide to support method selection. Our findings highlight the need for more principled imputation approaches in multi-site biomedical research and provide concrete tools to address this challenge.</p>","PeriodicalId":524152,"journal":{"name":"Communications AI & computing","volume":"1 1","pages":"19"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13652732/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149171470","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}
Kate A Richardson, Sokratis Trifinopoulos, Mike Williams
{"title":"Interpretable AI reveals hidden structure in nuclear masses.","authors":"Kate A Richardson, Sokratis Trifinopoulos, Mike Williams","doi":"10.1038/s44488-026-00017-9","DOIUrl":"https://doi.org/10.1038/s44488-026-00017-9","url":null,"abstract":"<p><p>Predicting nuclear masses with high precision remains a central challenge in nuclear physics, with implications ranging from stellar nucleosynthesis to determining the boundaries of the nuclear landscape. Approaches using artificial intelligence (AI) have recently achieved remarkable accuracy on this task, but their usefulness is limited by their black-box nature and by the difficulty of assessing their reliability when extrapolating beyond measured nuclei. Here, we present an interpretable AI model that matches the best existing predictive precision while revealing the physical structure underlying nuclear binding energies. The model learns internal representations that form a double helix, in which links between proton and neutron embeddings correspond to the most stable nuclei within each isotopic chain. Analysis of its learned features shows that the AI autonomously recovers established macroscopic components such as the liquid-drop model and rediscovers a long-forgotten physical principle-localized Jaffe factorization-that accounts for much of its performance gains. These findings demonstrate that interpretable AI can both reproduce and extend nuclear models, uncovering latent physical symmetries directly from data.</p>","PeriodicalId":524152,"journal":{"name":"Communications AI & computing","volume":"1 1","pages":"20"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13686942/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149281215","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}
António D Brito, Dominik Alwardt, Beatriz de P Mariz, Sérgio R Filipe, Mariana G Pinho, Bruno M Saraiva, Ricardo Henriques
{"title":"mAIcrobe: an open-source framework for high-throughput bacterial image analysis.","authors":"António D Brito, Dominik Alwardt, Beatriz de P Mariz, Sérgio R Filipe, Mariana G Pinho, Bruno M Saraiva, Ricardo Henriques","doi":"10.1038/s44488-026-00022-y","DOIUrl":"10.1038/s44488-026-00022-y","url":null,"abstract":"<p><p>Microscopy of bacterial cells is crucial for studying bacterial growth, cell division, or responses to antibiotics. However, analyzing these images can be challenging because bacteria vary widely in shape and behaviour. Many existing tools require specialised computational expertise, which can limit their accessibility. Here we present mAIcrobe, an open-source image analysis framework that makes advanced bacterial microscopy analysis more accessible by combining deep learning-based segmentation methods, including StarDist, CellPose, and U-Net, with quantitative morphological profiling and a flexible neural network-based classification model. mAIcrobe can analyse a wide range of bacterial species, from spherical <i>Staphylococcus aureus</i> to rod-shaped <i>Escherichia coli</i>, across different microscopy modalities, within a single environment. We demonstrate the utility of mAIcrobe by using it to identify antibiotic-induced changes in <i>E. coli</i> and cell cycle defects in <i>S. aureus</i> DnaA mutants. The framework is designed to be modular and extensible, with Jupyter notebooks provided to facilitate the development of custom models, thereby making AI-driven image analysis more accessible to the microbiology community.</p>","PeriodicalId":524152,"journal":{"name":"Communications AI & computing","volume":"1 1","pages":"16"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13577909/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148993689","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}
Christian Rominger, Andreas R Schwerdtfeger, Malay Gaherwar Singh, Dimitri Khudyakow, Elizabeth A M Michels, Fabian Wolf, Jakob Nikolas Kather, Magdalena Katharina Wekenborg
{"title":"Quantifying the human visual exposome with vision language models.","authors":"Christian Rominger, Andreas R Schwerdtfeger, Malay Gaherwar Singh, Dimitri Khudyakow, Elizabeth A M Michels, Fabian Wolf, Jakob Nikolas Kather, Magdalena Katharina Wekenborg","doi":"10.1038/s44488-026-00023-x","DOIUrl":"https://doi.org/10.1038/s44488-026-00023-x","url":null,"abstract":"<p><p>The visual environment is a fundamental yet unquantified determinant of mental health. While the concept of the environmental exposome is well-established, current methods rely on coarse geospatial proxies or biased self-reports, failing to capture the first-person visual context of daily life. Here we show that coupling ecological momentary assessment with vision-language models can quantify the semantic richness of human visual experience. Across 2,674 participant-generated photographs, model-derived estimates of \"greenness\" robustly predicted momentary affect and chronic stress, consistent with established benchmarks. We then developed a semi-autonomous large language model-based pipeline that mined over seven million scientific publications to extract nearly 1,000 environmental features empirically linked to mental health. When applied to real-world imagery, up to 33% of model-extracted context ratings significantly correlated with affect and stress.</p>","PeriodicalId":524152,"journal":{"name":"Communications AI & computing","volume":"1 1","pages":"25"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13739718/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149373693","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}