Improved Intrusion Detection Based on Hybrid Deep Learning Models and Federated Learning.

The security of the Industrial Internet of Things (IIoT) is of vital importance, and the Network Intrusion Detection System (NIDS) plays an indispensable role in this. Although there is an increasing number of studies on the use of deep learning technology to achieve network intrusion detection, the limited local data of the device may lead to poor model performance because deep learning requires large-scale datasets for training. Some solutions propose to centralize the local datasets of devices for deep learning training, but this may involve user privacy issues. To address these challenges, this study proposes a novel federated learning (FL)-based approach aimed at improving the accuracy of network intrusion detection while ensuring data privacy protection. This research combines convolutional neural networks with attention mechanisms to develop a new deep learning intrusion detection model specifically designed for the IIoT. Additionally, variational autoencoders are incorporated to enhance data privacy protection. Furthermore, an FL framework enables multiple IIoT clients to jointly train a shared intrusion detection model without sharing their raw data. This strategy significantly improves the model's detection capability while effectively addressing data privacy and security issues. To validate the effectiveness of the proposed method, a series of experiments were conducted on a real-world Internet of Things (IoT) network intrusion dataset. The experimental results demonstrate that our model and FL approach significantly improve key performance metrics such as detection accuracy, precision, and false-positive rate (FPR) compared to traditional local training methods and existing models.

Ethereum Phishing Scam Detection Based on Data Augmentation Method and Hybrid Graph Neural Network Model.

The rapid advancement of blockchain technology has fueled the prosperity of the cryptocurrency market. Unfortunately, it has also facilitated certain criminal activities, particularly the increasing issue of phishing scams on blockchain platforms such as Ethereum. Consequently, developing an efficient phishing detection system is critical for ensuring the security and reliability of cryptocurrency transactions. However, existing methods have shortcomings in dealing with sample imbalance and effective feature extraction. To address these issues, this study proposes an Ethereum phishing scam detection method based on DA-HGNN (Data Augmentation Method and Hybrid Graph Neural Network Model), validated by real Ethereum datasets to prove its effectiveness. Initially, basic node features consisting of 11 attributes were designed. This study applied a sliding window sampling method based on node transactions for data augmentation. Since phishing nodes often initiate numerous transactions, the augmented samples tended to balance. Subsequently, the Temporal Features Extraction Module employed Conv1D (One-Dimensional Convolutional neural network) and GRU-MHA (GRU-Multi-Head Attention) models to uncover intrinsic relationships between features from the time sequences and to mine adequate local features, culminating in the extraction of temporal features. The GAE (Graph Autoencoder) concept was then leveraged, with SAGEConv (Graph SAGE Convolution) as the encoder. In the SAGEConv reconstruction module, by reconstructing the relationships between transaction graph nodes, the structural features of the nodes were learned, obtaining reconstructed node embedding representations. Ultimately, phishing fraud nodes were further identified by integrating temporal features, basic features, and embedding representations. A real Ethereum dataset was collected for evaluation, and the DA-HGNN model achieved an AUC-ROC (Area Under the Receiver Operating Characteristic Curve) of 0.994, a Recall of 0.995, and an F1-score of 0.994, outperforming existing methods and baseline models.

Qualifying Event and Recurrence of Ischemic Stroke in Symptomatic Artery Occlusion: A Post Hoc Analysis of CMOSS.

BACKGROUND: The authors aimed to elucidate the relationship between latest ischemic event and the incidence of subsequent ischemic stroke in patients with symptomatic artery occlusion. METHODS AND RESULTS: We analyzed the association between qualifying event-the latest ischemic event (transient ischemic attack [TIA] or stroke)-and the incidence of ipsilateral ischemic stroke in patients with symptomatic artery occlusion treated with medical therapy alone in CMOSS (Carotid or Middle Cerebral Artery Occlusion Surgery Study). The incidence of CMOSS primary outcomes, including any stroke or death within 30 days after randomization or ipsilateral ischemic stroke between 30 days and 2 years, between the bypass surgical and medical groups, stratified by qualifying events, was also compared. Of the 165 patients treated with medical therapy alone, 75 had a TIA and 90 had a stroke as their qualifying event. The incidence of ipsilateral ischemic stroke did not significantly differ between patients with a TIA and those with a stroke as their qualifying event (13.3% versus 6.7%, P=0.17). In multivariate analysis, the qualifying event was not associated with the incidence of ipsilateral ischemic stroke. There were no significant differences in the CMOSS primary outcomes between the surgical and medical groups, regardless of the qualifying event being TIA (10.1% versus 12.2%, P=0.86) or stroke (6.7% versus 8.9%, P=0.55). CONCLUSIONS: Among patients with symptomatic artery occlusion and hemodynamic insufficiency, the risk of subsequent ipsilateral ischemic stroke does not appear to be lower in patients presenting with a TIA compared with those with a stroke. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01758614.

[Research progress of chemerin in polycystic ovary syndrome].

As a multifunctional adipokine, chemerin plays a crucial role in various pathophysiological processes through endocrine and paracrine manner. It can bind to three known receptors (ChemR23, GPR1 and CCRL2) and participate in energy metabolism, glucose and lipid metabolism, and inflammation, especially in metabolic diseases. Polycystic ovary syndrome (PCOS) is one of the most common endocrine diseases, which seriously affects the normal life of women of childbearing age. Patients with PCOS have significantly increased serum levels of chemerin and high expression of chemerin in their ovaries. More and more studies have shown that chemerin is involved in the occurrence and development of PCOS by affecting obesity, insulin resistance, hyperandrogenism, oxidative stress and inflammatory response. This article mainly reviews the production, subtypes, function and receptors of chemerin protein, summarizes and discusses the research status of chemerin protein in PCOS from the perspectives of metabolism, reproduction and inflammation, and provides theoretical basis and reference for the clinical diagnosis and treatment of PCOS.

Chlorogenic acid can improve spermatogenic dysfunction in rats with varicocele by regulating mitochondrial homeostasis and inhibiting the activation of NLRP3 inflammasomes by oxidative mitochondrial DNA and cGAS/STING pathway.

In recent years, Varicocele (VC) has been recognized as a common cause of male infertility that can be treated by surgery or drugs. How to reduce the damage of VC to testicular spermatogenic function has attracted extensive attention in recent years. Among them, overexpressed ROS and high levels of inflammation may play a key role in VC-induced testicular damage. As the key mediated innate immune pathways, cGAS-STING shaft under pathological conditions, such as in cell and tissue damage stress can be cytoplasmic DNA activation, induce the activation of NLRP3 inflammatory corpuscle, triggering downstream of the inflammatory cascade reaction. Chlorogenic acid (CGA), as a natural compound from a wide range of sources, has strong anti-inflammatory and antioxidant activities, and is a potential effective drug for the treatment of varicocele infertility. The aim of this study is to investigate the role of CGA in the spermatogenic dysfunction of the rat testis induced by VC and the potential mechanisms. The results of this study have shown that CGA gavage treatment ameliorated the pathological damage of seminiferous tubules, increased the number of sperm in the lumen, and increased the expression levels of Occludin and ZO-1, which indicated the therapeutic effect of CGA on spermatogenic dysfunction in the testis of VC rats. Meanwhile, the damage of mitochondrial structure was alleviated and the expression levels of ROS, NLRP3 and pro-inflammatory cytokines (IL-1ß, IL-6, IL-18) were significantly reduced in the testicular tissues of model rats after CGA treatment. In addition, we demonstrated for the first time the high expression status of cGAS and STING in testicular tissues of VC model rats, and this was ameliorated to varying degrees after CGA treatment. In conclusion, this study suggests that CGA can improve the spermatogenic function of the testis by reducing mitochondrial damage and inhibiting the activation of the cGAS-STING axis, inhibiting the activation of the NLRP3 inflammasome, and improving the inflammatory damage of the testis, highlighting the potential of CGA as a therapeutic agent for varicocele infertility.

Molecular engineering of thiophene- and pyrrole-fused core arylamine systems: Tuning redox properties, NIR spectral responsiveness and bacterial imaging applications.

The thiophene- and pyrrole-fused heterocyclic compounds have garnered significant interest for their distinctive electron-rich characteristics and notable optoelectronic properties. However, the construction of high-performance systems within this class is of great challenge. Herein, we develop a series of novel dithieno[3,2-b:2',3'-d] pyrrole (DTP) and tetrathieno[3,2-b:2',3'-d] pyrrole (TTP) bridged arylamine compounds (DTP-C4, DTP-C12, DTP-C4-Fc, TTP-C4-OMe, TTP-C4, and TTP-C12) with varying carbon chain lengths. The pertinent experimental results reveal that this series of compounds undergo completely reversible multistep redox processes. Notably, TTP-bridged compounds TTP-C4 and TTP-C12 exhibit impressive multistep near-infrared (NIR) absorption alterations with notable color changes and electroluminescent behaviors, which are mainly attributed to the charge transfer transitions from terminal arylamine units to central bridges, as supported by theoretical calculations. Additionally, compound DTP-C4 demonstrates the ability to visually identify gram-positive and gram-negative bacteria. Therefore, this work suggests the promising electroresponsive nature of compounds TTP-C4 and TTP-C12, positioning them as excellent materials for various applications. It also provides a facile approach to constructing high-performance multifunctional luminescent materials, particularly those with strong and long-wavelength NIR absorption capabilities.

Optimizing biochar addition strategies in combined processes: Comprehensive assessment of earthworm growth, lignocellulose degradation and vermicompost quality.

The sustainable management of agricultural waste is essential for curtailing environmental contamination. To address the shortcomings of single treatment methods, this study evaluated the feasibility of combining membrane-covered composting (MC) with vermicomposting. Based on this, the integrated effects of different biochar addition strategies on the combined process were investigated. The aim was to improve the efficiency of vermicomposting while eliminating the negative effects of biochar on earthworms. Addition of biochar before membrane-covered composting increased total earthworm biomass by 25.6 - 31.4 % and reproduction rate by 13.4 - 23.9 %. Specifically, the electrical conductivity (EC) (1061.0 - 1112.0 uS/cm) of the vermicompost was significantly reduced, while the total nutrient content (42.3 - 42.6 mg/g) and germination index (GI) (103.9 - 108.4 %) were maximized. Additionally, reductions in the carbon-to-nitrogen ratio and volatile content were observed. Overall, combination process is a promising approach to improve the quality of vermicomposting. The study's results offer a novel perspective on the value-added treatment of agricultural waste.

MSGM: An Advanced Deep Multi-Size Guiding Matching Network for Whole Slide Histopathology Images Addressing Staining Variation and Low Visibility Challenges.

Matching whole slide histopathology images to provide comprehensive information on homologous tissues is beneficial for cancer diagnosis. However, the challenge arises with the Giga-pixel whole slide images (WSIs) when aiming for high-accuracy matching. Learning-based methods are difficult to generalize well with large-size WSIs, necessitating the integration of traditional matching methods to enhance accuracy as the size increases. In this paper, we propose a multi-size guiding matching method applicable high-accuracy requirements. Specifically, we design learning multiscale texture to train deep descriptors, called TDescNet, that trains 64 ×64×256 and 256 ×256×128 size convolution layer as C64 and C256 descriptors to overcome staining variation and low visibility challenges. Furthermore, we develop the 3D-ring descriptor using sparse keypoints to support the description of large-size WSIs. Finally, we employ C64, C256, and 3D-ring descriptors to progressively guide refined local matching, utilizing geometric consistency to identify correct matching results. Experiments show that when matching WSIs of size 4096×4096 pixels, our average matching error is 123.48 [Formula: see text] and the success rate is 93.02 % in 43 cases. Notably, our method achieves an average improvement of 65.52 [Formula: see text] in matching accuracy compared to recent state-of-the-art methods, with enhancements ranging from 36.27 [Formula: see text] to 131.66 [Formula: see text]. Therefore, we achieve high-fidelity whole-slice image matching, and overcome staining variation and low visibility challenges, enabling assistance in comprehensive cancer diagnosis through matched WSIs.

Dynamic Single Cell Transcriptomics Defines Kidney FGF23/KL Bioactivity and Novel Segment-Specific Inflammatory Targets.

FGF23 via its coreceptor αKlotho (KL) provides critical control of phosphate metabolism, which is altered in rare and very common syndromes, however the spatial-temporal mechanisms dictating renal FGF23 functions remain poorly understood. Thus, developing approaches to modify specific FGF23-dictated pathways has proven problematic. Herein, wild type mice were injected with rFGF23 for 1, 4 and 12h and renal FGF23 bioactivity was determined at single cell resolution. Computational analysis identified distinct epithelial, endothelial, stromal, and immune cell clusters, with differential expressional analysis uniquely tracking FGF23 bioactivity at each time point. FGF23 actions were sex independent but critically relied upon constitutive KL expression mapped within proximal tubule (S1-S3) and distal tubule (DCT/CNT) cell sub-populations. Temporal KL-dependent FGF23 responses drove unique and transient cellular identities, including genes in key MAPK- and vitamin D-metabolic pathways via early- (AP-1-related) and late-phase (EIF2 signaling) transcriptional regulons. Combining ATACseq/RNAseq data from a cell line stably expressing KL with the in vivo scRNAseq pinpointed genomic accessibility changes in MAPK-dependent genes, including the identification of FGF23-dependent EGR1 distal enhancers. Finally, we isolated unexpected crosstalk between FGF23-mediated MAPK signaling and pro-inflammatory TNF receptor activation via NF-κB, which blocked FGF23 bioactivity in vitro and in vivo . Collectively, our findings have uncovered novel pathways at the single cell level that likely influence FGF23-dependent disease mechanisms. Translational statement: Inflammation and elevated FGF23 in chronic kidney disease (CKD) are both associated with poor patient outcomes and mortality. However, the links between these manifestations and the effects of inflammation on FGF23-mediated mineral metabolism within specific nephron segments remain unclear. Herein, we isolated an inflammatory pathway driven by TNF/NF-κB associated with regulating FGF23 bioactivity. The findings from this study could be important in designing future therapeutic approaches for chronic mineral diseases, including potential combination therapies or early intervention strategies. We also suggest that further studies could explore these pathways at the single cell level in CKD models, as well as test translation of our findings to interactions of chronic inflammation and elevated FGF23 in human CKD kidney datasets.

Prediction of miRNAs and diseases association based on sparse autoencoder and MLP.

Introduction: MicroRNAs (miRNAs) are small and non-coding RNA molecules which have multiple important regulatory roles within cells. With the deepening research on miRNAs, more and more researches show that the abnormal expression of miRNAs is closely related to various diseases. The relationship between miRNAs and diseases is crucial for discovering the pathogenesis of diseases and exploring new treatment methods. Methods: Therefore, we propose a new sparse autoencoder and MLP method (SPALP) to predict the association between miRNAs and diseases. In this study, we adopt advanced deep learning technologies, including sparse autoencoder and multi-layer perceptron (MLP), to improve the accuracy of predicting miRNA-disease associations. Firstly, the SPALP model uses a sparse autoencoder to perform feature learning and extract the initial features of miRNAs and diseases separately, obtaining the latent features of miRNAs and diseases. Then, the latent features combine miRNAs functional similarity data with diseases semantic similarity data to construct comprehensive miRNAs-diseases datasets. Subsequently, the MLP model can predict the unknown association among miRNAs and diseases. Result: To verify the performance of our model, we set up several comparative experiments. The experimental results show that, compared with traditional methods and other deep learning prediction methods, our method has significantly improved the accuracy of predicting miRNAs-disease associations, with 94.61% accuracy and 0.9859 AUC value. Finally, we conducted case study of SPALP model. We predicted the top 30 miRNAs that might be related to Lupus Erythematosus, Ecute Myeloid Leukemia, Cardiovascular, Stroke, Diabetes Mellitus five elderly diseases and validated that 27, 29, 29, 30, and 30 of the top 30 are indeed associated. Discussion: The SPALP approach introduced in this study is adept at forecasting the links between miRNAs and diseases, addressing the complexities of analyzing extensive bioinformatics datasets and enriching the comprehension contribution to disease progression of miRNAs.

Universal inverse modeling of point spread functions for SMLM localization and microscope characterization.

The point spread function (PSF) of a microscope describes the image of a point emitter. Knowing the accurate PSF model is essential for various imaging tasks, including single-molecule localization, aberration correction and deconvolution. Here we present universal inverse modeling of point spread functions (uiPSF), a toolbox to infer accurate PSF models from microscopy data, using either image stacks of fluorescent beads or directly images of blinking fluorophores, the raw data in single-molecule localization microscopy (SMLM). Our modular framework is applicable to a variety of microscope modalities and the PSF model incorporates system- or sample-specific characteristics, for example, the bead size, field- and depth- dependent aberrations, and transformations among channels. We demonstrate its application in single or multiple channels or large field-of-view SMLM systems, 4Pi-SMLM, and lattice light-sheet microscopes using either bead data or single-molecule blinking data.

Insight into Interfacial Heat Transfer of ß-Ga2O3/Diamond Heterostructures via the Machine Learning Potential.

ß-Ga2O3 is an ultrawide-band gap semiconductor with excellent potential for high-power and ultraviolet optoelectronic device applications. Low thermal conductivity is one of the major obstacles to enable the full performance of ß-Ga2O3-based devices. A promising solution for this problem is to integrate ß-Ga2O3 with a diamond heat sink. However, the thermal properties of the ß-Ga2O3/diamond heterostructures after the interfacial bonding have not been studied extensively, which are influenced by the crystal orientations and interfacial atoms for the ß-Ga2O3 and diamond interfaces. In this work, molecular dynamics simulations based on machine learning potential have been adopted to investigate the crystal-orientation-dependent and interfacial-atom-dependent thermal boundary resistance (TBR) of the ß-Ga2O3/diamond heterostructure after interfacial bonding. The differences in TBR at different interfaces are explained in detail through the explorations of thermal conductivity value, thermal conductivity spectra, vibration density of states, and interfacial structures. Based on the above explorations, a further understanding of the influence of different crystal orientations and interfacial atoms on the ß-Ga2O3/diamond heterostructure was achieved. Finally, insightful optimization strategies have been proposed in the study, which could pave the way for better thermal design and management of ß-Ga2O3/diamond heterostructures according to guidance in the selection of the crystal orientations and interfacial atoms of the ß-Ga2O3 and diamond interfaces.

Axial de-scanning using remote focusing in the detection arm of light-sheet microscopy.

Rapid, high-resolution volumetric imaging without moving heavy objectives or disturbing delicate samples remains challenging. Pupil-matched remote focusing offers a promising solution for high NA systems, but the fluorescence signal's incoherent and unpolarized nature complicates its application. Thus, remote focusing is mainly used in the illumination arm with polarized laser light to improve optical coupling. Here, we introduce a novel optical design that can de-scan the axial focus movement in the detection arm of a microscope. Our method splits the fluorescence signal into S and P-polarized light, lets them pass through the remote focusing module separately, and combines them with the camera. This allows us to use only one focusing element to perform aberration-free, multi-color, volumetric imaging without (a) compromising the fluorescent signal and (b) needing to perform sample/detection-objective translation. We demonstrate the capabilities of this scheme by acquiring fast dual-color 4D (3D space + time) image stacks with an axial range of 70 µm and camera-limited acquisition speed. Owing to its general nature, we believe this technique will find its application in many other microscopy techniques that currently use an adjustable Z-stage to carry out volumetric imaging, such as confocal, 2-photon, and light sheet variants.

Bizarre parosteal osteochondromatous proliferation (Nora lesion) involving the spine: a case report and systematic review.

Bizarre parosteal osteochondromatous proliferation (BPOP), also termed Nora lesion, is a rare, benign tumor most often located in the hands and feet. We herein present the second reported case of BPOP affecting the spine, an uncommon location. One year after surgical excision, the patient was pain-free and showed no evidence of recurrence. We reviewed a total of 323 cases of BPOP among 101 articles, providing the first systematic update on the latest knowledge of BPOP. The age of patients with BPOP ranges from 3 months to 87 years, peaking in the second and third decades of life. The hands are the most common location of BPOP (58.39%), followed by the feet (20.81%). Imaging features play a key role in the diagnosis of BPOP, but histopathologic diagnosis remains the gold standard. Differential diagnosis of BPOP should be based on the epidemiologic and clinical features as well as clinical examination findings. Surgical resection is the most extensively used treatment for BPOP. Recurrence is common (37.44%) and can be treated with re-excision. This article can deepen our understanding of BPOP and will be helpful for the diagnosis and treatment of BPOP in clinical practice.

External fields effectively switch the spin channels of transition metal-doped ß-phase tellurene from first principles.

Non-volatile magnetic random-access memories have proposed the need for spin channel switching. However, this presents a challenge as each spin channel reacts differently to the external field. Tellurene is a semiconductor with a tunable bandgap, excellent stability, and high carrier concentration, but its lack of magnetic properties has hindered its application in spintronics. In this work, the influence of an external field on transition metal (TM)-doped ß-tellurene is systematically analysed from first principles. First, the active-learning moment-tensor-potential (MTP) is used to verify the thermal stability of the V-doped system with the MTP proving to be 900 times faster than the traditional method. Subsequently, under biaxial strain ranging from -2% to 10%, the V-doped system undergoes a gradual transition from a magnetic semiconductor to a spin-gapless semiconductor, and further to a half-metal and magnetic metal. The band structure can be maintained under an electric field. By examining the magnetic anisotropy energy, the lattice changes profoundly impact the electromagnetic properties, particularly with the TMs being sensitive to strain. Moreover, the band structure is reflected in the spin resolution current of the magnetic tunnel junction. This work investigates the response of doped ß-Te to external fields, revealing its potential applications in spintronics.

Potential of body mass index as a screening indicator for extracranial-intracranial bypass surgery in patients with symptomatic artery occlusion: a post-hoc analysis of the CMOSS trial.

BACKGROUND: To investigate the association between body mass index (BMI) and the incidence of ischemic stroke in patients with symptomatic artery occlusion, and further to evaluate the utility of BMI as a screening tool for identifying candidates for extracranial-intracranial bypass surgery. MATERIALS AND METHODS: We analyzed the relationship between BMI and the occurrence of ipsilateral ischemic stroke (IIS) among patients receiving only medical management in the Carotid or Middle cerebral artery Occlusion Surgery Study (CMOSS). Additionally, we compared the primary endpoint of CMOSS-stroke or death within 30 days, or IIS after 30 days up to two years-among patients with varying BMIs who underwent either surgery or medical treatment. RESULTS: Of the 165 patients who treated medically only, 16 (9.7%) suffered an IIS within two years. BMI was independently associated with the incidence of IIS (hazard ratio: 1.16 per kg/m2; 95% confidence interval: 1.06-1.27). The optimal BMI cutoff for predicting IIS was 24.5 kg/m2. Patients with BMI ≥24.5 kg/m2 experienced a higher incidence of IIS compared to those with BMI <24.5 kg/m2 (17.4% vs. 0.0%, P<0.01). The incidence of the CMOSS primary endpoint was significantly different between the surgical and medical groups for patients with BMI ≥24.5 kg/m2 (5.3% vs. 19.8%, P<0.01) and those with BMI <24.5 kg/m2 (10.6% vs. 1.4%; P=0.02). Surgical intervention was independently associated with a reduced rate of the CMOSS primary endpoint in patients with BMI ≥24.5 kg/m2. CONCLUSION: Data from the CMOSS trial indicate that patients with BMI ≥24.5 kg/m2 are at a higher risk of IIS when treated medically only and appear to derive greater benefit from bypass surgery compared to those with lower BMIs. Given the small sample size and the inherent limitations of retrospective analyses, further large-scale, prospective studies are necessary to confirm these findings.

Design, synthesis, and anti-oomycete activity of 3-acyloxymaltol/ethyl maltol derivatives.

Twenty 3-acyloxymaltol/ethyl maltol derivatives (7a-j and 8a-j) were synthesized and evaluated in vitro for their anti-oomycete activity against Phytophthora capsici, respectively. Among all of twenty derivatives, more than half of the compounds 7f, 7h, 8a-h and 8j had anti-oomycete activity higher than the positive control zoxamide (EC50 = 22.23 mg/L), and the EC50 values of 18.66, 20.32, 12.80, 16.18, 10.59, 14.98, 16.80, 10.36, 15.32, 12.64, and 13.59 mg/L, respectively. Especially, compounds 8c and 8f exhibited the best anti-oomycete activity against P. capsici with EC50 values of 10.59 and 10.36 mg/L, respectively. Overall, hydroxyl group of maltol/ethyl maltol is important active modification site.

Dams alter the control pattern of watershed land use to riverine nutrient distribution: Comparison of three major rivers under different hydropower development levels in Southwestern China.

Land use plays a critical role in managing water quality in a watershed, as it governs the import and distribution of nutrients. In addition to the land use, some rivers in Southwest China are encountering a new environmental stressor of damming, which is being driven by the national strategy of hydropower development. However, the coupling effect of land use and dams on nutrients remains poorly understood, challenging the effective management of riverine water quality. Therefore, this study examined the nutrients in the Nu, Yarlung Tsangpo (YT), and Lancang (LC) Rivers, which have no dam, 1 dam, and 11 dams, respectively, during different regulatory periods (spring and fall) to identify variations in nutrient control patterns influenced by land use and dams. The findings suggested that an increase in hydropower development contributed to a notable shift in nutrient patterns from land use regulation towards dam regulation and coupling effects. Land use dominated the nutrient variations of the Nu (27.4 %-32.8 %) and low hydropower development YT (25.2 %-30.9 %) Rivers during both seasons, but the primary contributors to the nutrient variations of the high hydropower development LC River were dams (17.9 %-41.6 %) and coupling effects (16.5 %-29.0 %). Dams transform nutrient levels and compositions through internal reservoir cycling, decoupling land use and nutrients. Partial least-squares structural equation model analysis further suggested that the coupling effects of the LC River were seasonal-specific, which was primarily attributed to hydrological variations that affected their interactions. During spring, the reservoir underwent a drainage mode characterized by high-level nutrients in the bottom water. Combined with the import of riverine nutrients, it exacerbated the increase of nutrients (synergistic effect). In contrast, the reservoir transitioned into a storage mode where it intercepted nutrients from the upstream and watershed during the fall, leading to a reduction in the previously observed increasing trend and an increase in nutrient variability (antagonism effect).

[Progress in the application of artificial intelligence-assisted molecular modification of enzymes].

Natural enzymes are often difficult to meet the needs of application and research in terms of activity, enantiomer selectivity or thermal stability. Therefore, it is an important task of enzyme engineering to explore efficient molecular modification technologies to improve the properties of such enzymes. The molecular modification technologies of enzymes mainly include rational design, directed evolution, and artificial intelligence-assisted design. Directed evolution and rational design are experiment-driven molecular modification approaches of enzymes and have been successfully applied to enzyme engineering. However, due to the huge space sizes of protein sequences and the lack of experimental data, the current modification methods still face major challenges. With the development of next-generation sequencing, high-throughput screening, protein databases, and artificial intelligence (AI), data-driven enzyme engineering is emerging as a promising solution to these challenges. The AI-assisted statistical learning method has been used to establish a model for predicting the sequence/structure-properties of enzymes in a data-driven manner. Excellent mutant enzymes can be selected according to the prediction results, which greatly improve the efficiency of molecular modification. Considering the application requirements of molecular modification of enzymes, this paper reviews the data acquisition methods and application examples of AI-assisted molecular modification of enzymes, with focuses on the convolutional neural network method for predicting protein thermostability, aiming to provide reference for researchers in this field.