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.
The authors have developed an adaptive reinforced dynamics approach, which improves the efficiency when exploring the configurational space and free energy landscape of large biomolecules, such as proteins.
A data-driven solution of partial differential equations is developed with conditional generative adversarial networks, which could be used in both forward and inverse problems.
The authors demonstrate an effective approach to lower the computing time required to accurately reach the thermodynamic limit in quantum many-body calculations. This method can be applied to solve problems in a wide range of material systems, including metals, insulators and semiconductors.
The authors present a full-scale model of the entorhinal cortex–dentate gyrus–CA3 network based on experimental data to show that fast lateral inhibition plays a key role in pattern separation.
The study shows that a memory-aware and socially coupled human movement model can reproduce urban growth patterns at the macro level, providing a bottom-up approach to understand urban growth and to reveal its connection to human mobility behavior.
A machine learning-assisted directed evolution method is developed, combining hierarchical unsupervised clustering and supervised learning, to guide protein engineering by iteratively exploring the large mutational sequence space.
Climate data are often stored at higher precision than is needed. The proposed compression automatically determines the precision from the data’s bitwise real information, removing any false information and leading to a more efficient compression.
The authors present a framework for modeling cell interactions using rigid bodies, which can used to represent cells as free moving polygons, to allow epithelial layers to smoothly interact, to model bacteria and to robustly represent membranes.
The authors propose a molecular modeling approach to simulating the galvanostatic charge–discharge process of supercapacitors under constant-potential conditions. This model can accurately predict supercapacitor dynamics when compared with experimental observations.
Combining bioinformatics data and atomistic simulations, this study develops a sequence-dependent coarse-grained model for biomolecular phase separation. This model achieves a quantitative agreement with experimental observations. Extensive benchmarks exemplify its performance.
A plurality of epidemiological models are analyzed using physics-informed neural networks to identify time-dependent parameters and data-driven fractional models. The results are reported for different geographical locations by inferring unknown parameters and unobserved dynamics.
A method for comorbidity discovery informed by each patient’s demographic and medical history is introduced. Statistics for 4,623,841 pairs of potentially comorbid medical terms are provided as a searchable web resource.
Using a statistical method for transient correlations, the waxing and waning in levels of population infection by SARS-CoV-2 are shown to respond to temperature and absolute humidity, across geographical locations and for different temporal and spatial resolutions.
An analysis of GPS pedestrian traces shows that (1) people increasingly deviate from the shortest path when the distance between origin and destination increases and that (2) chosen paths are statistically different when origin and destination are swapped. Ultimately, this can explain the observed human attitude in selecting different paths upon return trips.
A variational autoencoder-based order parameter is proposed and demonstrated on various high-entropy alloys, providing a computational technique for understanding chemical ordering in alloys. Ultimately, this has the potential to facilitate the development of rational alloy design strategies.
An algorithmic approach is developed to analyze large-scale patient safety data and remove the confounders of reporting trajectory and drug inference. Such an approach can be effectively used to investigate demographic disparities of drug safety and to identify at-risk patients during a pandemic.
The authors demonstrate how neural systems can encode cognitive functions, and use the proposed model to train robust, scalable deep neural networks that are explainable and capable of symbolic reasoning and domain generalization.
Combining human mobility data and nonlinear mathematical analysis techniques, this study offers insights into the interplay between human behavior, disease transmission and non-pharmaceutical interventions during the COVID-19 pandemic.
The authors propose Detect, a browser-based anomaly detection framework for diffusion magnetic resonance imaging tractometry data. The tool leverages normative microstructural brain features derived from healthy participants using deep autoencoders to detect anomalies at the individual level.
The authors propose EPICS, a method to predict microbial community structures by estimating effective pairwise interactions that subsume high-order interactions between species. EPICS is more efficient and applicable to larger communities than current approaches.