Browse Articles

Attention mechanism based CNN-LSTM hybrid deep learning model for atmospheric ozone concentration prediction

Spatiotemporal evolution and driving forces of landscape ecological risk in the lower reaches of the Yellow River from 2000 to 2020

Novel control strategies for electric vehicle charging stations using stochastic modeling and queueing analysis

The human autophagy-initiating complexes ULK1C and PI3KC3-C1

Methane-powered sea spiders: Diverse, epibiotic methanotrophs serve as a source of nutrition for deep-sea methane seep <i>Sericosura</i>

Methane seeps harbor uncharacterized animal–microbe symbioses with unique nutritional strategies. Three undescribed sea spider species (family Ammotheidae; genus Sericosura ) endemic to methane seeps were found along the eastern Pacific margin, from California to Alaska, hosting diverse methane- and methanol-oxidizing bacteria on their exoskeleton. δ 13 C tissue isotope values of in situ specimens corroborated methane assimilation (−45‰, on average). Live animal incubations with 13 C-labeled methane and methanol, followed by nanoscale secondary ion mass spectrometry, confirmed that carbon derived from both compounds was actively incorporated into the tissues within five days. Methano- and methylotrophs of the bacterial families Methylomonadaceae, Methylophagaceae and Methylophilaceae were abundant, based on environmental metagenomics and 16S rRNA sequencing, and fluorescence and electron microscopy confirmed dense epibiont aggregations on the sea spider exoskeleton. Egg sacs carried by the males hosted identical microbes suggesting vertical transmission. We propose that these sea spiders farm and feed on methanotrophic and methylotrophic bacteria, expanding the realm of animals known to harness C1 compounds as a carbon source. These findings advance our understanding of the biology of an understudied animal lineage, unlocking some of the unique nutritional links between the microbial and faunal food webs in the oceans.

Human neutrophil α-defensin HNP1 interacts with bacterial OmpA to promote Acinetobacter baumannii biofilm formation

Modular shape-changing tensegrity-blocks enable self-assembling robotic structures

The familiarity effect of Chinese stroke stimulus and imagery on contextual integration: Evidence from ERP correlates

The neural process of contextual integration has been examined through the phenomenology of semantic incongruence of words. The present study investigated whether the effort of contextual integration would be heightened by the increased demand of selective attention and attention orientation to unfamiliar Chinese stroke style and sequence. It also examined whether visual imagery of unfamiliar stroke style and sequence would mitigate the effort of contextual integration of unfamiliar Chinese stroke. Nineteen participants take part in two cognitive tasks: (a) imagery of Chinese strokes and (b) detection of Chinese familiar and unfamiliar stroke style. An electroencephalogram was concurrently recorded for the analysis of event-related potential (ERP). Results revealed significant differences in attention orientation and effort of contextual integration between familiar and unfamiliar Chinese strokes, as indicated by larger amplitudes of N160 (100–200 ms) &amp; P200 (260–380 ms) components. Furthermore, a larger amplitude of N400 (300–500 ms) component, signifying the neural process of integrating external from the context, was obtained when individuals viewed unfamiliar Chinese strokes. These findings suggest a cognitive effort was needed to process unfamiliar Chinese stimuli, followed by a greater mental effort required for contextual integration of the unfamiliar stimuli. Furthermore, top-down control of visual imagery would facilitate the process of contextual integration via generating internal representation. This finding provides a new insight that the effort expended in contextual integration may be associated with both attentional control and the generation of internal representation from long-term storage across visual stimuli with varying levels of stimulus familiarity. In summary, our study provided insights into the cognitive mechanisms underlying attentional control, contextual integration, and the role of visual imagery in the processing of stimuli with different levels of familiarity. Furthermore, it suggested the potential utility of the N400 component as a biomarker for assessing attention control and memory retrieval functions.

Bolt loosening evaluation method based on normalized screw root equivalent stress and loosening life curve

Ocean acidification induces changes in circadian alternative splicing profiles in a coral reef fish

Innovative deep learning classifiers for breast cancer detection through hybrid feature extraction techniques

Abstract Breast cancer remains a major cause of mortality among women, where early and accurate detection is critical to improving survival rates. This study presents a hybrid classification approach for mammogram analysis by combining handcrafted statistical features and deep learning techniques. The methodology involves preprocessing with the Shearlet Transform, segmentation using Improved Otsu thresholding and Canny edge detection, followed by feature extraction through Gray Level Co-occurrence Matrix (GLCM), Gray Level Run Length Matrix (GLRLM), and 1st-order statistical descriptors. These features are input into a 2D BiLSTM-CNN model designed to learn spatial and sequential patterns in mammogram images. Evaluated on the MIAS dataset, the proposed method achieved 97.14% accuracy, outperforming several benchmark models. The results indicate that this hybrid strategy offers improvements in classification performance and may assist radiologists in more effective breast cancer screening.

Effects of sport disciplines on offspring sex ratio in elite athletes: an observational study

Gpbar1-mediated SIRT1-PGC-1α axis maintains mitochondrial homeostasis and mitigates renal injury in obstructive jaundice

Treatment outcomes of ABVD in classical Hodgkin lymphoma patients from Thailand without procarbazine access

Physical activity enhances college students’ mental health through social adaptability and exercise behavior chain mediation

Empathy and mental health distress in informal caregivers of dementia and stroke patients: a cross-sectional study

FDA approves a c-MET-targeted ADC for lung cancer

Written in chromatin: The enduring legacy of C. David Allis

Transient infrared nanoscopy resolves the millisecond photoswitching dynamics of single lipid vesicles in water

Abstract Understanding the biophysical and biochemical properties of molecular nanocarriers under physiological conditions with minimal interference is critical for advancing photopharmacology, drug delivery, nanotheranostics and synthetic biology. However, analytical methods often struggle to combine precise chemical imaging and dynamic measurements without perturbative labeling. This challenge is exemplified by azobenzene-based photoswitchable lipids, which are intriguing reagents for controlling nanocarrier properties on fast timescales, enabling precise light-induced drug release. Here, we leverage the chemical recognition and high spatio-temporal resolution of scattering-type scanning near-field optical microscopy (s-SNOM) to demonstrate a non-destructive, label-free technique for mid-infrared imaging and spectroscopy of individual photoswitchable liposomes. Our transient nanoscopy approach enables imaging below the diffraction limit and tracks dynamics with sampling times as fast as 30 ms. We resolve photoinduced changes in shape and MIR spectral signature of individual vesicles and discover abrupt and delayed photoisomerization dynamics. Our findings highlight the method’s potential for studying complex dynamics of unlabeled nanoscale soft matter.

Deciphering coulombic loss in lithium-ion batteries and beyond

Abstract Lithium-ion batteries are pivotal for modern energy storage, yet accurately predicting their lifespan remains a critical challenge. While descriptors like coulombic efficiency are widely used to assess battery longevity, the unclear physical origins of coulombic losses cause semi-quantitative correlation with capacity, complicating battery development. Here, we combine high-precision leakage current and open-circuit-voltage measurements with charge conservation principles to explore microscopic charge consumptions at electrode-electrolyte interfaces across diverse chemistries. We demonstrate that coulombic loss arises from a synergy between local charge neutrality and global charge compensation, reconciling its quantitative correlation to capacity. Contrary to conventional assumptions equating coulombic loss with irreversible capacity loss, this framework resolves systematic overestimations and paradoxical phenomena in existing chemistries. Our findings establish physics-informed criteria for accelerated lifespan evaluation and guide rational design of long-life lithium-ion batteries and beyond.