Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Rapid identification of pathogenic viruses remains a critical challenge. A recent study advances this frontier by demonstrating a fully integrated memristor-based hardware system that accelerates genomic analysis by a factor of 51, while reducing energy consumption to just 0.2% of that required by conventional computational methods.
We propose a computationally efficient genome-wide association study (GWAS) method, WtCoxG, for time-to-event (TTE) traits in the presence of case ascertainment— a form of oversampling bias. WtCoxG addresses case ascertainment bias by applying a weighted Cox proportional hazard model, and outperforms existing approaches when incorporating information on external allele frequencies.
This Perspective discusses that generative AI aligns with generative linguistics by showing that neural language models (NLMs) are formal generative models. Furthermore, generative linguistics offers a framework for evaluating and improving NLMs.
A benchmark — MaCBench — is developed for evaluating the scientific knowledge of vision language models (VLMs). Evaluation of leading VLMs reveals that they excel at basic scientific tasks such as equipment identification, but struggle with spatial reasoning and multistep analysis — a limitation for autonomous scientific discovery.
A recent study demonstrates the potential of using in-memory computing architecture for implementing large language models for an improved computational efficiency in both time and energy while maintaining a high accuracy.
Large language models remain largely unexplored is the design of cities. In this Perspective, the authors discuss the potential opportunities brought by these models in assisting urban planning.
An integrated platform, Digital Twin for Chemical Science (DTCS), is developed to connect first-principles theory with spectroscopic measurements through a bidirectional feedback loop. By predicting and refining chemical reaction mechanisms before, during and after experiments, DTCS enables the interpretation of spectra and supports real-time decision-making in chemical characterization.
Large language models are increasingly important in social science research. The authors provide guidance on how best to validate and use these models as rigorous tools to further scientific inference.
A recent study proposed ZeoBind, an AI-accelerated workflow enabling the discovery and experimental verification of hits within chemical spaces containing hundreds of millions of zeolites.
A recent study sought to replicate published experimental research using large language models, finding that human behavior is replicated surprisingly well overall, but deviates in important ways that could lead social scientists astray.
A recent study provides intuition and guidelines for deciding whether to incorporate cheaper, lower-fidelity experiments into a closed-loop search for molecules and materials with desired properties.
An artificial neural network-based strategy is developed to learn committor-consistent transition pathways, providing insight into rare events in biomolecular systems.
A recent study introduces a neural code conversion method that aligns brain activity across individuals without shared stimuli, using deep neural network-derived features to match stimulus content.
We developed group technical effects (GTE) as a quantitative metric for evaluating gene-level batch effects in single-cell data. It identifies highly batch-sensitive genes — the primary contributors to batch effects — that vary across datasets, and whose removal effectively mitigates the batch effects.
This Perspective highlights the potential integrations of large language models (LLMs) in chemical research and provides guidance on the effective use of LLMs as research partners, noting the ethical and performance-based challenges that must be addressed moving forward.
Enhanced sampling methods aim to simulate rare physical and chemical reactive processes involving transitions between long-lived states. Existing methods often disproportionally sample either metastable or transition states. A machine-learning approach combines the strengths of these two cases to characterize entire rare events with the same thoroughness in a single calculation.
A framework with large language models is proposed to predict disease spread in real-time by incorporating complex, multi-modal information and using a artificial intelligence–human cooperative prompt design.
Predicting how molecular changes affect brain activity is a challenge in neuroscience. We introduced a multiscale modeling approach to simulate these microscopic changes and how they impact macroscale brain activity. This approach predicted how the anesthetic action on synaptic receptors can lead to the transitions in macroscale brain activity observed empirically.
A new framework disentangles the nature of disruption in science, revealing how rare but persistent breakthroughs shake the foundations of research fields while remaining central to future work.
