Search

Named Entity Recognition (NER) is a crucial component of Natural Language Processing (NLP). When dealing with the high diversity and complexity of the Chinese language, existing Chinese NER models face challenges in addressing word sense ambiguity, capturing long-range dependencies, and maintaining robustness, which hinders the accuracy of entity recognition. To this end, a Chinese NER model based on multi-level representation learning is proposed. The model leverages a pre-trained word-based embedding to capture contextual information. A linear layer adjusts dimensions to fit an Extended Long Short-Term Memory (XLSTM) network, enabling the capture of long-range dependencies and contextual information, and providing deeper representations. An adaptive multi-head attention mechanism is proposed to enhance the ability to capture global dependencies and comprehend deep semantic context. Additionally, GlobalPointer with rotational position encoding integrates global information for entity category prediction. Projected Gradient Descent (PGD) is incorporated, introducing perturbations in the embedding layer of the pre-trained model to enhance stability in noisy environments. The proposed model achieves F1-scores of 96.89%, 74.89%, 72.19%, and 80.96% on the Resume, Weibo, CMeEE, and CLUENER2020 datasets, respectively, demonstrating improvements over baseline and comparison models.

The economically important seaweed Gracilariopsis lemaneiformis (Gp. lemaneiformis) is the main resource for agar production and abalone bait. Few studies have focused on the functions of glycosyltransferases (GTs) and glycoside hydrolases (GHs) involved in saccharometabolism, including that of agar. To investigate the possible functions of glycosyltransferase family 7 (GT7) and glycoside hydrolase family 16 (GH16) members in agar metabolism, analyses of these gene families were conducted using bioinformatics-based and physiological methods. In total, five GlGT7 genes and four GlGH16 genes from the Gp. lemaneiformis genome were identified and analyzed. These GT7 and GH16 members are absent in higher plants, and the majority of GlGT7s and GlGH16s have no introns. Catalytic pocket residue analysis revealed that GlGT7s may function as β-1,4-galactosyltransferases and that GlGH16s play roles as agarases. Promoter prediction and qRT-PCR experiments verified that these genes can be regulated by light and phytohormones. GlGT7s were predicted to interact with carbohydrate sulfotransferases (STs), whereas yeast two-hybrid (Y2H) assays revealed no interactions between these proteins. Under heat stress, no significant difference was found in agar content; however, the expression of GlGT7s fluctuated and that of GlGH16 increased. Low nitrogen stress significantly increased the agar content, and the expression of GlGT7s increased, whereas that of GlGH16s decreased. In three cultivars of Gp. lemaneiformis, the expression of GlGT7 genes in cultivars 981 and Lulong No. 1 was upregulated relative to that in the wild type, whereas GlGH16 expression levels were significantly decreased. Fluorescence microscopy further showed that β-1,4-galactose accumulation was consistent with increases in agar content and GlGT7 expression. Pearson correlation analysis confirmed that the expression levels of GlGT7s and GlGH16s were positively and negatively correlated with agar accumulation, respectively. Taken together, these results demonstrated that GlGT7 and GlGH16 are intimately correlated with agar metabolism; in particular, GlGT7-2, GlGT7-5, and GlGH16-4 could act as molecular markers to indicate agar yield. This study will provide a valuable basis for breeding new cultivars with high agar content.

Gracilariopsis lemaneiformis, an important commercial red macroalga, is facing significant impacts from global warming, which limits algal growth and yield in China. Polyamines (PAs), spermidine (SPD) and spermine (SPM), are ubiquitous polycations important for growth and environmental stress responses including high temperature (HT) tolerance. Spermidine synthase (SPDS) gene is one of the important genes in higher PA biosynthesis, which plays critical roles in HT stress response. Here, we isolated an SPDS gene from G. lemaneiformis and further analyzed its phylogenetic tree, subcellular localization, and gene expression patterns under stress conditions. Meanwhile, supplemented with SPD and SPM were used to study the effects of PAs on HT tolerance in G. lemaneiformis. It showed exogenous 0.5 mM SPD and SPM, respectively, remarkably improved the algal relative growth rate (RGR) compared to those in the CK treatment groups under HT conditions. In addition, they both significantly elevated the activities of antioxidant enzymes and significantly upregulated the expression of genes encoding antioxidant enzymes, triggered transcription factors (TFs) signaling, and improved the expression of genes encoding small heat shock proteins (sHSP20s) during HT stress. Moreover, exogenous PA also enhanced the expression of genes involved in pyruvate metabolism, ascorbate and aldarate metabolism, and nucleotide excision repair in G. lemaneiformis, which helped to maintain better energy supply, redox homeostasis, and genome integrity under HT stress. Taken together, these data provided valuable information for functional characterization of specific gene in endogenous PA synthesis and uncovered the importance of exogenous PAs in promoting algae adaptation to HT stress.

This work introduces an optimal transportation (OT) view of generative adversarial networks (GANs). Natural datasets have intrinsic patterns, which can be summarized as the manifold distribution principle: the distribution of a class of data is close to a low-dimensional manifold. GANs mainly accomplish two tasks: manifold learning and probability distribution transformation. The latter can be carried out using the classical OT method. From the OT perspective, the generator computes the OT map, while the discriminator computes the Wasserstein distance between the generated data distribution and the real data distribution; both can be reduced to a convex geometric optimization process. Furthermore, OT theory discovers the intrinsic collaborative—instead of competitive—relation between the generator and the discriminator, and the fundamental reason for mode collapse. We also propose a novel generative model, which uses an autoencoder (AE) for manifold learning and OT map for probability distribution transformation. This AE–OT model improves the theoretical rigor and transparency, as well as the computational stability and efficiency; in particular, it eliminates the mode collapse. The experimental results validate our hypothesis, and demonstrate the advantages of our proposed model. Keywords: Generative, Adversarial, Deep learning, Optimal transportation, Mode collapse

Gracilariopsis lemaneiformis (G. lemaneiformis) is an important marine red macroalgae with high economic and ecological value all over the world. To date, global warming is a key issue that has a great impact on all living organisms, such as macroalgae. L-arginine (Arg) is a precursor of nitric oxide (NO) and polyamines (PAs), which can induce stress defense responses in land plants. However, its role in inducing algae resistance at high temperature (HT) is unclear. In this study, G. lemaneiformis thalli were treated with different concentrations of Arg to investigate its effect and the mechanism on the tolerance of G. lemaneiformis against HT stress. It turned out that exogenous Arg significantly alleviated the HT-induced oxidative damage as indicated by a markedly decrease in malondialdehyde (MDA) content. Notably, Arg remarkably improved the relative growth rate (RGR) and phycobiliprotein (PBP) contents of G. lemaneiformis at HT. Moreover, Arg significantly elevated the activities of antioxidant enzymes, such as superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), to efficiently scavenge reactive oxygen species (ROS). In addition, it also promoted the accumulation of free amino acids (AAs) as compared to those in the control treatment (CK) group under HT conditions. To investigate the mechanism of G. lemaneiformis to Arg, a transcriptome analysis was performed and revealed 1,414 and 3,825 differentially expressed genes (DEGs) in Arg-treated groups as compared to CK groups at 24 and 48 h of HT stress, respectively. Results showed that Arg significantly upregulated the expression of genes encoding antioxidant enzymes, heat shock proteins, and triggered transcription factors (TFs) signaling during HT stress. Moreover, Arg enhanced the DEGs involved in arginine and proline (Pro) metabolism, AAs biosynthesis, glycolysis, tricarboxylic acid (TCA) cycle, and oxidative phosphorylation. These results may help in understanding the role of Arg in G. lemaneiformis resistance to HT and provide a practical viewpoint for obtaining heat-tolerant G. lemaneiformis to further promote the development of the cultivated seaweed industry in the future.

Agar is widely applied across the food, pharmaceutical and biotechnology industries, owing to its various bioactive functions. To better understand the agar biosynthesis in commercial seaweed Gracilariopsis lemaneiformis, the activities of four enzymes participating in the agar biosynthesis were detected, and phosphoglucomutase (PGM) was confirmed as highly correlated with agar accumulation. Three genes of PGM (GlPGM1, GlPGM2 and GlPGM3) were identified from the G. lemaneiformis genome. The subcellular localization analysis validated that GlPGM1 was located in the chloroplast and GlPGM3 was not significantly distributed in the organelles. Both the GlPGM1 and GlPGM3 protein levels showed a remarkable consistency with the agar variations, and GlPGM3 may participate in the carbon flux between (iso)floridoside, floridean starch and agar synthesis. After treatment with the PGM inhibitor, the agar and floridean starch contents and the activities of floridean starch synthase were significantly decreased; products identified in the Calvin cycle, the pentose phosphate pathway, the Embden-Meyerhof-Parnas pathway and the tricarboxylic acid cycle were depressed; however, lipids, phenolic acids and the intermediate metabolites, fructose-1,6-phosphate were upregulated. These findings reveal the essential role of PGM in regulating the carbon flux between agar and other carbohydrates in G. lemaneiformis, providing a guide for the artificial regulation of agar accumulation.

Data quality management, especially data cleansing, has been extensively studied for many years in the areas of data management and visual analytics. In the paper, we first review and explore the relevant work from the research areas of data management, visual analytics and human-computer interaction. Then for different types of data such as multimedia data, textual data, trajectory data, and graph data, we summarize the common methods for improving data quality by leveraging data cleansing techniques at different analysis stages. Based on a thorough analysis, we propose a general visual analytics framework for interactively cleansing data. Finally, the challenges and opportunities are analyzed and discussed in the context of data and humans. MSC: 00-01, 99-00, Keywords: Data quality management, Visual analytics, Data cleansing

Interactive model analysis, the process of understanding, diagnosing, and refining a machine learning model with the help of interactive visualization, is very important for users to efficiently solve real-world artificial intelligence and data mining problems. Dramatic advances in big data analytics have led to a wide variety of interactive model analysis tasks. In this paper, we present a comprehensive analysis and interpretation of this rapidly developing area. Specifically, we classify the relevant work into three categories: understanding, diagnosis, and refinement. Each category is exemplified by recent influential work. Possible future research opportunities are also explored and discussed. Keywords: Interactive model analysis, Interactive visualization, Machine learning, Understanding, Diagnosis, Refinement

CCR4-Not complex is a multifunctional regulator that plays important roles in multiple cellular processes in eukaryotes. In the present study, the biological function of FonNot2, a core subunit of the CCR4-Not complex, was explored in Fusarium oxysporum f. sp. niveum (Fon), the causal agent of watermelon wilt disease. FonNot2 was expressed at higher levels in conidia and germinating conidia and during infection in Fon-inoculated watermelon roots than in mycelia. Targeted disruption of FonNot2 resulted in retarded vegetative growth, reduced conidia production, abnormal conidial morphology, and reduced virulence on watermelon. Scanning electron microscopy observation of infection behaviors and qRT-PCR analysis of in planta fungal growth revealed that the ΔFonNot2 mutant was defective in the ability to penetrate watermelon roots and showed reduced fungal biomass in root and stem of the inoculated plants. Phenotypic and biochemical analyses indicated that the ΔFonNot2 mutant displayed hypersensitivity to cell wall perturbing agents (e.g. Congo Red and Calcofluor White) and oxidative stress (e.g. H2O2 and paraquat), and decreased fusaric acid content, and reduced ROS production during spore germination. Our data demonstrate that FonNot2 plays critical roles in regulating vegetable growth, conidiogenesis and conidia morphology, and virulence on watermelon via modulating cell wall integrity, oxidative stress response, ROS production and FA biosynthesis through the regulation of transcription of genes involved in multiple pathways.

Trehalose and its metabolism have been demonstrated to play important roles in control of plant growth, development and stress responses. However, direct genetic evidence supporting the functions of trehalose and its metabolism in defense response against pathogens is lacking. In the present study, genome-wide characterization of putative trehalose-related genes identified 11 SlTPSs for trehalose-6-phosphate synthase, 8 SlTPPs for trehalose-6-phosphate phosphatase and one SlTRE1 for trehalase in tomato genome. Nine SlTPSs, 4 SlTPPs and SlTRE1 were selected for functional analyses to explore their involvement in tomato disease resistance. Some selected SlTPSs, SlTPPs and SlTRE1 responded with distinct expression induction patterns to Botrytis cinerea and Pseudomonas syringae pv. tomato (Pst) DC3000 as well as to defense signaling hormones (e.g. salicylic acid, jasmonic acid and a precursor of ethylene). Virus-induced gene silencing-mediated silencing of SlTPS3, SlTPS4 or SlTPS7 led to deregulation of ROS accumulation and attenuated the expression of defense-related genes upon pathogen infection and thus deteriorated the resistance against B. cinerea or Pst DC3000. By contrast, silencing of SlTPS5 or SlTPP2 led to an increased expression of the defense-related genes upon pathogen infection and conferred an increased resistance against Pst DC3000. Silencing of SlTPS3, SlTPS4, SlTPS5, SlTPS7 or SlTPP2 affected trehalose level in tomato plants with or without infection of B. cinerea or Pst DC3000. These results demonstrate that SlTPS3, SlTPS4, SlTPS5, SlTPS7 and SlTPP2 play roles in resistance against B. cinerea and Pst DC3000, implying the importance of trehalose and tis metabolism in regulation of defense response against pathogens in tomato.

Biotic and abiotic stresses are major unfavorable factors that affect crop productivity worldwide. NAC proteins comprise a large family of transcription factors that play important roles in plant growth and development as well as in responses to biotic and abiotic stresses. In a virus-induced gene silencing-based screening to identify genes that are involved in defense response against Botrytis cinerea, we identified a tomato NAC gene SlSRN1 (Solanum lycopersicum Stress-related NAC1). SlSRN1 is a plasma membrane-localized protein with transactivation activity in yeast. Expression of SlSRN1 was significantly induced by infection with B. cinerea or Pseudomonas syringae pv. tomato (Pst) DC3000, leading to 6-8 folds higher than that in the mock-inoculated plants. Expression of SlSRN1 was also induced by salicylic acid, jasmonic acid and 1-amino cyclopropane-1-carboxylic acid and by drought stress. Silencing of SlSRN1 resulted in increased severity of diseases caused by B. cinerea and Pst DC3000. However, silencing of SlSRN1 resulted in increased tolerance against oxidative and drought stresses. Furthermore, silencing of SlSRN1 accelerated accumulation of reactive oxygen species but attenuated expression of defense genes after infection by B. cinerea. Our results demonstrate that SlSRN1 is a positive regulator of defense response against B. cinerea and Pst DC3000 but is a negative regulator for oxidative and drought stress response in tomato.