Browse Articles
Discover research articles across all indexed journals
Proximal labeling of the Golgi secretome reveals fat body–derived humoral factors in Drosophila disc regeneration
Flexible Organic Single Crystals with Unconventional Negative Photoconductivity for Intelligent Security Systems
Unlocking Photocatalytic CO <sub>2</sub> Conversion to Ethylene Glycol by Microdroplet‐Enabled Interfacial Electric Field
ABSTRACT The electrochemical carbon dioxide (CO 2 ) reduction features attractive potentials of carbon neutrality and chemical upgrading, while the types of products from photocatalysis or electrocatalysis still remain limited. The conversion of CO 2 into ethylene glycol, an important chemical that is typically produced by thermo‐catalytic process, has not been achieved by photocatalysis or electrocatalysis. Herein, we demonstrate a microdroplet‐enhanced photocatalytic strategy with a Pd‐TiO 2 catalyst, which allows for efficient conversion of CO 2 into ethylene glycol with high selectivity. The electro‐sprayed microdroplets provide a high interfacial electric field that facilitates deep reduction of CO 2 into *CH 3 , as well as generation of reactive *OH intermediates to form CH 3 OH. The in situ formed CH 3 OH is further photo‐catalytically activated to cleave a C−H bond to *CH 2 OH, followed by the C−C coupling to produce ethylene glycol. Under 1‐sun illumination, the electro‐sprayed microdroplets and Pd‐TiO 2 catalyst exhibited a high CO 2 ‐to‐ethylene glycol conversion rate of 2985 µmol L −1 h −1 , which significantly outperformed those by only photocatalysis or only electro‐sprayed microdroplets, revealing the capability of combining multiple energy fields for tuning reaction pathways and advancing green synthesis.
Global metabolomic profiling of serum biomarkers in women with polycystic ovary syndrome
Novel drug targets in 2025
Clinical implementation of multi-cancer early detection tests: can we find a path forward?
Targeted enzyme discovery using metal-coordination mining
Correction: Colchicine inhibits vascular calcification by suppressing inflammasome activation through the enhancement of the Sirt2-PP2Ac signaling pathway
Linking device dynamics to neural network performance in ionically gated synaptic transistors
Abstract Neuromorphic computing based on artificial synapses requires devices capable of gradual, repeatable, and energy-efficient conductance modulation. Ionically gated transistors are promising candidates because their ion dynamics naturally produce synaptic behavior under low-voltage operation. However, how key device characteristics—conductance range, number of accessible conductance states N , and weight-update nonlinearity β —jointly influence neural network performance remains insufficiently understood. Here, we investigate MoS 2 -based ionically gated synaptic transistors using a combined experimental and modeling framework that links device physics to hardware-aware artificial neural network (ANN) simulations across image-classification tasks of varying complexity. We show that under fixed-amplitude pulsing, increasing N introduces a fundamental trade-off: finer weight resolution is accompanied by stronger update nonlinearity. ANN simulations further reveal that, within the nonlinearity range studied here, classification accuracy is governed by a task-dependent optimal weight resolution rather than a simply maximized number of states. To overcome the nonlinear weight updates, we employ a physics-informed transient model to develop a predictive pulse-engineering algorithm and experimentally demonstrate that it can linearize synaptic weight evolution in the same device. These linearized updates improve ANN accuracy by 1.5%–5.2% for MNIST, 4.0%–5.2% for FMNIST, and 1.2%–12% for KMNIST across the tested state numbers, establishing a quantitative link between device-level dynamics and neural network performance in ionically gated synaptic transistors.
The unglamourous AI wins that are quietly transforming drug discovery and development
Pan-RAS inhibitor daraxonrasib shows promise in pancreatic cancer
Daily briefing: What’s really happening with trust in science
The Rho guanine-nucleotide exchange factor P-Rex2 exhibits structural and regulatory features distinct from the related RhoGEF P-Rex1
Adsorption and Sulfur-Selective Photooxidation of Cysteine on Anatase TiO <sub>2</sub> (101)
Multimodal MRI-based quantification of tumoral spatial heterogeneity for preoperative upgrade prediction of ductal carcinoma in situ
Abstract This retrospective study aimed to develop a multimodal habitat-based model integrating clinicopathological factors, radiomics features, and intratumoral heterogeneity (ITH) score for predicting upstaging in patients with ductal carcinoma in situ (DCIS). A total of 389 DCIS patients from two institutions were included. Lesions were segmented on DCE-MRI, DWI, and ADC images with peritumoral expansion, and intratumoral and peritumoral habitats were identified using Gaussian mixture model (GMM) clustering. Radiomics features were extracted from each habitat, and ITH scores were calculated. Independent predictors of upstaging were identified using multivariate logistic regression, and multiple models were constructed and evaluated. Age, nuclear grade, palpable mass, and DCE-ITH score were independent risk factors for DCIS upstaging. Habitat-based models derived from intratumoral and peritumoral regions demonstrated good predictive performance, while the combined model integrating habitat features, ITH score, and clinicopathological variables achieved the best performance, with AUCs of 0.945 and 0.869 in the training and test cohorts, respectively. This multimodal combined model may facilitate individualized preoperative risk assessment and support clinical decision-making in DCIS patients.