This work develops photocatalytic coatings for indoor formaldehyde removal on wood surfaces. It explores coating‐induced color variations and visual perception via eye‐tracking tests. Combined with XGBoost and Bayesian optimization, an intelligent design framework is established to guide the optimization of visually friendly wood photocatalytic ...
Xinyue Wei +8 more
wiley +1 more source
CoSP: Reconfigurable Metamaterial Inverse Design via Contrastive Pretrained Large Language Model
In this work, CoSP (contrastive multi‐state pretrain), an intelligent inverse design method for reconfigurable metamaterials based on a contrastive pretrained large language model, is proposed. Numerical experiments demonstrate that CoSP can design reconfigurable metamaterial structures for multi‐state, multi‐band optical responses, showing great ...
Shujie Yang +4 more
wiley +1 more source
Systematic Multi‐Level Analyses Decode the Arthritis‐Neurodegeneration Axis With In Vivo Validation
Arthritis and neurodegeneration are usually studied as separate disorders, but this study connects them through population evidence, genetic inference, transcriptomic mapping, and mouse models. It highlights RNF40 as a context‐dependent joint‐brain candidate, induced in inflammatory joints yet functionally linked to dopamine‐neuron vulnerability ...
Jinwen Wang +7 more
wiley +1 more source
TSTScope is an interpretable AI framework that integrates single‐cell transcriptomes with TCR information through curated gene‐program constraints. By linking receptor context to functional T cell states, it reveals response‐associated tumor‐specific T cell programs in lung cancer immunotherapy cohorts and defines an MPR score associated with ...
Shiwei Cao +8 more
wiley +1 more source
Accounting for cell lineage and sex effects in the identification of cell-specific DNA methylation using a Bayesian model selection algorithm. [PDF]
White N +6 more
europepmc +1 more source
A Robotic Testing Platform for Pipelined Discovery of Resilient Dielectric Elastomer Actuators
This work introduces a robotic testing platform enabling high‐throughput, multi‐parameter lifetime mapping of linear dielectric elastomer actuators (DEAs). By integrating programmable high‐voltage control and automated electromechanical measurements, the system identifies material and operating conditions that double actuator lifetime.
Ang (Leo) Li +10 more
wiley +1 more source
Machine Learning of Temperature‐Dependent Chemical Kinetics Using Parallel Droplet Microreactors
An integrated droplet microfluidics and machine learning framework enables high‐throughput characterization of temperature‐dependent reaction kinetics. Time‐resolved measurements from thousands of droplets train Neural ODE models that accurately predict nonlinear reaction dynamics across diverse thermal environments, bridging large‐scale ...
Mamoru Saita, Yutaka Hori
wiley +1 more source
Elucidation of Genetic Interactions in the Yeast GATA-Factor Network Using Bayesian Model Selection. [PDF]
Milias-Argeitis A +5 more
europepmc +1 more source
Perceived Time Shapes Physical Fatigue Accumulation and Its Neural Oscillatory Correlates
Can a clock change the course of physical fatigue? By covertly manipulating the clock calibration during repeated isometric contractions, we showed that perceived time shapes fatigue accumulation and frontal oscillatory dynamics. While fatigue accumulation was only reduced when the clock was slowed down, frontal oscillatory dynamics followed perceived ...
Pierre‐Marie Matta +3 more
wiley +1 more source
Canonical Antibodies Adopt Distinct Binding Modes to Recognize Viral Glycan Shields
Canonical Y‐shaped antibodies recognize viral glycan shields through adaptive Fab assembly states shaped by glycan organization and somatic hypermutation. Structural analyses ofbroadly neutralizing antibodies VRC35 and VRC36 across glycoproteins of HIV‐1, influenza, SARS‐CoV‐2, and Lassa viruses reveal distinct Fab assembly states, spanning monovalent,
Jiaxuan Cheng +71 more
wiley +1 more source

