Search

Word segmentation is a critical task in natural language processing for southeast Asian Abugida languages, including Thai, Burmese, and Khmer. Existing approaches demonstrate that models using fixed-length windowed context inputs can achieve high segmentation accuracy; however, they often rely on low-level character features or language-specific preprocessing. Character-based methods can limit feature learning, while language-specific features add complexity due to specialized preprocessing requirements. This paper introduces UnifiedCut, which is a neural model that leverages multiple n-grams within a windowed multi-head attention mechanism. This design captures segmentation features from local contexts and multi-perspective n-gram inputs, enhancing generalization and recall, particularly for out-of-vocabulary words. Compared to CNN- and RNN-based approaches, UnifiedCut's multi-head attention enables finer-grained feature extraction and greater parallelism, resulting in a faster, more scalable solution. Comprehensive experiments on public datasets for Thai, Burmese, and Khmer show that UnifiedCutachieves state-of-the-art performance in word segmentation. [ABSTRACT FROM AUTHOR]

Superoxide dismutase (SOD), a critical enzyme within the plant antioxidant defense system, serves as the primary shield against reactive oxygen species (ROS) under adverse environmental conditions. However, the characterization of GbSODs in ginkgo remains incomplete. In our study, a total of eight GbSODs were identified, which were unevenly distributed across 10 chromosomes and predominantly localized on the plasma membrane. Our phylogenetic analysis revealed that GbSODs from ginkgo and other species form three distinct groups, each characterized by specific binding domains and supported by relatively high bootstrap values. Our promoter analysis indicated that all GbSODs contained multiple cis-elements related to various abiotic stress responses. Our systemic analysis revealed that only one pair of GbSODs underwent segmental duplications. The transcriptome analysis showed the variable expression of GbSODs across different tissues, which were significantly influenced by drought, ultraviolet (UV), and salt stress, with GbSOD8 showing a particularly notable induction. Transgenic plants that overexpressed GbSOD8 exhibited enhanced antioxidant enzyme activities, elevated proline levels, and reduced malondialdehyde (MDA) content, collectively contributing to their improved salt tolerance compared to wild-type plants. Our findings expand the comprehensive understanding of the GbSOD gene family and provide a solid foundation for further elucidating the molecular mechanisms of GbSODs under salt stress. [ABSTRACT FROM AUTHOR]

Portulaca oleracea L. (purslane) is a food and a traditional drug worldwide. It exhibits anti-inflammatory, anti-oxidative, anti-tumor, and anti-diabetic bioactivities; but its activity on diabetic-associated endothelial dysfunction is unknown. This study aimed to investigate the effect of purslane on endothelial function and the underlying mechanisms. Male C57BL/6 mice had 14-week ad libitum access to a high-fat rodent diet containing 60% kcal% fat to induce obesity and diabetes whereas purslane extract (200 mg/kg/day) was administered during the last 4 weeks via intragastric gavage. Primary rat aortic endothelial cells and isolated mouse aortas were cultured with a risk factor, high glucose or tunicamycin, together with purslane extract. By ESI-QTOF-MS/MS, flavonoids and their glycoside products were identified in the purslane extract. Exposure to high glucose or tunicamycin impaired acetylcholine-induced endothelium-dependent relaxations in aortas and induced endoplasmic reticulum (ER) stress and oxidative stress with the downregulation of 5′ AMP-activated protein kinase (AMPK)/ endothelial nitric oxide synthase (eNOS) signaling. Co-incubation with purslane significantly ameliorated these impairments. The effects of purslane were abolished by Compound C (AMPK inhibitor). Four-week purslane treatment ameliorated aortic relaxations, ER stress, and oxidative stress in diabetic obese mice. This study supported that purslane protected endothelial function, and inhibited ER stress and oxidative stress in vasculature through AMPK/eNOS activation, revealing its therapeutic potential against vascular complications in diabetes. [ABSTRACT FROM AUTHOR]

Dried tangerine peel (Citri reticulatae Pericarpium, CRP; Chenpi in Chinese) possesses medicine and food homology with hypolipidemic, anti-inflammatory and antioxidant activities. This study aimed to explore the protective effect of CRP extract on endothelial function and inflammation in type 2 diabetic rats and the related mechanisms. Type 2 diabetes mellitus was induced by high-fat diet (HFD)/streptozotocin (STZ) in male Sprague Dawley rats, and CRP extract was orally administered at 400 mg/kg/day for 4 weeks. Rat and mouse aortas were treated with high glucose and CRP extract ex vivo. The data showed that the ethanolic extract of CRP normalized blood pressure and the plasma lipid profile as well as the plasma levels of liver enzymes in diabetic rats. Impaired endothelium-dependent relaxations in aortas, carotid arteries and renal arteries were improved. CRP extract suppressed vascular inflammatory markers and induced AMPK activation in aortas of diabetic rats. Exposure to high glucose impaired vasodilation in aortas of rats and mice, and this impairment was prevented by co-incubation with CRP extract. In conclusion, our findings suggest that CRP extract protects endothelial function by inhibiting the vascular inflammatory state on activation of AMPK in diabetic rats. [ABSTRACT FROM AUTHOR]

Jatrorrhizine (JAT) is one of the major bioactive protoberberine alkaloids found in rhizoma coptidis, which has hypoglycemic and hypolipidemic potential. This study aimed to evaluate the vasoprotective effects of JAT in diabetes and obesity and the underlying mechanism involved. Mouse aortas, carotid arteries and human umbilical cord vein endothelial cells (HUVECs) were treated with risk factors (high glucose or tunicamycin) with and without JAT ex vivo and in vitro. Furthermore, aortas were obtained from mice with chronic treatment: (1) control; (2) diet-induced obese (DIO) mice fed a high-fat diet (45% kcal% fat) for 15 weeks; and (3) DIO mice orally administered JAT at 50 mg/kg/day for the last 5 weeks. High glucose or endoplasmic reticulum (ER) stress inducer tunicamycin impaired acetylcholine-induced endothelium-dependent relaxations (EDRs) in mouse aortas, induced oxidative stress in carotid arteries and HUVECs, downregulated phosphorylations of Akt at Ser473 and eNOS at Ser1177 and enhanced ER stress in mouse aortas and HUVECs, and these impairments were reversed by cotreatment with JAT. JAT increased NO release in high-glucose-treated mouse aortas and HUVECs. In addition, chronic JAT treatment restored endothelial function with EDRs comparable to the control, increased Akt/eNOS phosphorylation, and attenuated ER stress and oxidative stress in aortas from DIO mice. Blood pressure, glucose sensitivity, fatty liver and its morphological change, as well as plasma levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) and plasma lipid profile, were also normalized by JAT treatment. Collectively, our data may be the first to reveal the vasoprotective effect of JAT that ameliorates endothelial dysfunction in diabetes and obesity through enhancement of the Akt/eNOS pathway and NO bioavailability, as well as suppression of ER stress and oxidative stress. [ABSTRACT FROM AUTHOR]

The accurate forecasting of urban taxi demands, which is a hot topic in intelligent transportation research, is challenging due to the complicated spatial-temporal dependencies, the dynamic nature, and the uncertainty of traffic. To make full use of the global and local correlations between traffic flows on road sections, this paper presents a deep learning model based on a graph convolutional network, long short-term memory (LSTM), and multitask learning. First, an undirected graph model was formed by considering the spatial pattern distribution of taxi trips on road networks. Then, LSTMs were used to extract the temporal features of traffic flows. Finally, the model was trained using a multitask learning strategy to improve the model's generalizability. In the experiments, the efficiency and accuracy were verified with real-world taxi trajectory data. The experimental results showed that the model could effectively forecast the short-term taxi demands on the traffic network level and outperform state-of-the-art traffic prediction methods. [ABSTRACT FROM AUTHOR]