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.
kmindex is a tool able to index thousands of environmental metagenomes and perform sequence searches in a fraction of a second, thus enabling real-time queries on complex genomic datasets.
A mathematical framework that allows computing the input–output function of neurons with active dendrites reveals how dendrites readily and potently control the response variability, a result that is experimentally confirmed.
A graph-based contrastive learning framework, LACL, is proposed for geometric domain-agnostic prediction of molecular properties to alleviate the need for molecular geometry relaxation, enabling large-scale inference scenarios.
Designing accessible, interoperable and reusable software for applying deep learning to the study of gene regulation has been a challenge in genomics research. EUGENe is a toolkit that addresses this gap and streamlines end-to-end analyses.
A deep-learning model, DetaNet, is proposed to efficiently and precisely predict molecular scalars, vectorial and tensorial properties, as well as the infrared, Raman, ultraviolet–visible and nuclear magnetic resonance spectra.
This study develops a programmable quantum processor, named Abacus, and applies it to Gaussian boson sampling tasks targeting drug discovery challenges.
The reasoning capabilities of OpenAI’s generative pre-trained transformer family were tested using semantic illusions and cognitive reflection tests that are typically used in human studies. While early models were prone to human-like cognitive errors, ChatGPT decisively outperformed humans, avoiding the cognitive traps embedded in the tasks.
SAHMI performs in silico denoising of microbial signals from existing host genomic sequencing data to select for present microbes, as well as filter out contaminants and false-positive misclassifications.
A platform for single-cell meta-analysis of inflammatory bowel disease, named scIBD, enables identification of rare or less-characterized cell types and the dissection of the commonalities and differences between ulcerative colitis and Crohn’s disease.
A microscopic moiré spin model that enables the description of moiré magnetic exchange interactions via a sliding-mapping method is proposed. The twist-angle and substrate-influenced magnetic phase diagram addresses disagreements between theories and experiments.
A deep learning ab initio method for studying magnetic materials is developed, reducing the computational cost and opening opportunities to predict the electronic properties of magnetic superstructures, such as magnetic skyrmions.
A method to compute the quantum harmonic free energy contributions in large materials and biomolecular simulations at a reasonable cost is proposed, making quantum mechanical estimates of thermodynamic quantities possible for complex systems.
The vulnerability of quantum machine learning models against adversarial noises, together with a defense strategy way out of this dilemma, is demonstrated experimentally with a programmable superconducting quantum processor.
A computational method is introduced for mutational intratumor heterogeneity inference from noisy genotype matrices derived from single-cell sequencing data. The proposed method is shown to be accurate and faster than available alternatives.
A systematic framework is introduced to calculate the effective carrier lifetime in semiconductor crystals under realistic conditions that are comparable with experiments. It helps explain the discrepancy between the calculated and experimental lifetimes in hybrid perovskites.
Accurate structural brain connectivity estimation is key to uncovering brain–behavior relationships. ReAl-LiFE, a GPU-accelerated approach, is applied for fast and reliable evaluation of individualized brain connectomes at scale.
A dynamic probabilistic algorithm that integrates many variables over time for forecasting severe acute graft-versus-host disease is proposed to improve healthcare decisions for individual patients on a case-by-case basis.
Scallop2 enables a more accurate assembly of transcripts at both single-cell resolution and sample level through a suite of algorithms that leverage the multi-end and paired-end information in Smart-seq3 and Illumina RNA-seq data.
mm2-fast is an accelerated version of minimap2, a popular software for long-read data analysis. mm2-fast introduces high-performance parallel computing techniques to reduce the overall runtime of minimap2.