Integrating Polyoxometalate into Dy(III)-based Single-molecule Magnets with Pentagonal Bipyramidal Symmetry.

Polyoxometalates (POMs) with various coordination fashions are versatile ligands for constructing single-ion magnets (SIMs), but enforcing POM-SIMs with a specific geometry remains a synthetic challenge. Herein, we synthesized a POM-cocrystallized DyIII-SIM [Dy(OPPh3)4(H2O)3][PW12O40]·4EtOH (1Dy) and a POM-ligated DyIII-SIM [{Dy(OPPh3)3(H2O)3}{PW12O40}]·Ph3PO·H2O (2Dy) with pentagonal bipyramidal local coordination geometry. Magnetic measurements indicate that 1Dy displays field-induced single-molecule magnet (SMM) behavior and the relaxation is dominated by under-barrier processes. 2Dy exhibits spin-lattice relaxation at a broader temperature region with a reversal barrier over 300 K. Magneto-structural analysis reveals that the enhancement of SMM behavior originated from the equatorial replacement of Ph3PO by POM, which strengthens the axial anisotropy in 2Dy. Luminescent experiments indicate that the characteristic DyIII emissions of 1Dy are covered up by the strong π-π* emission of Ph3PO at low-temperature regions. As for 2Dy, partial DyIII emission persists thanks to the antenna effect between DyIII and POM.

Textile-Based Membraneless Microfluidic Double-Inlet Hybrid Microbial-Enzymatic Biofuel Cell.

This study reports the development of a textile-based colaminar flow hybrid microbial-enzymatic biofuel cell. Shewanella MR-1 was used as a biocatalyst on the anode, and bienzymatic system catalysts based on glucose oxidase and horseradish peroxidase were applied on an air-breathing cathode to address the overpotential loss in a body-friendly way. A single-layer Y-shaped channel configuration with a double-inlet was adopted. Microchannels of biofuel cells were patterned by silk screen printing with Ecoflex to maintain the flexibility of textile substrates without harm to the human body. The electrodes were fabricated with poly(3,4-ethylenedioxythiophene):polystyrene sulfonate and a mixture of multiwalled carbon nanotubes and single-walled carbon nanotubes by screen printing. The effects of electrode materials, catalyst type, catalyst concentration, and glucose concentration in the catholyte were investigated to optimize the fuel cell performance. The peak power density (44.9 µW cm-2) and maximum current density (388.9 µA cm-2) of the optimized hybrid biofuel cell were better than those of previously reported textile- or paper-substrate microscale single microbial fuel cells. The developed biofuel cell will be a useful platform as a microscale power source that is harmless to the environment and living organisms.

A novel PDGFR inhibitor WQ-C-401 prevents pulmonary vascular remodeling in rats with monocrotaline-induced pulmonary arterial hypertension.

Platelet-derived growth factor (PDGF) is one of the most important cytokines associated with pulmonary vascular remodeling in pulmonary arterial hypertension (PAH). PDGF receptor (PDGFR) inhibition exerted therapeutic effects on PAH in clinical trials, but serious side effects warrant the withdrawal of existing drugs. In this study, a novel highly selective PDGFR inhibitor WQ-C-401 was developed, and its effects on PDGFR signaling pathway and pulmonary vascular remodeling in PAH were investigated. Cell proliferation assays and Western blot analysis of PDGFRα/ß phosphorylation showed that WQ-C-401 inhibited PDGFR-mediated cell proliferation assay and suppressed PDGFR phosphorylation in a concentration-dependent manner. DiscoverX's KinomeScanTM technology confirmed the good kinome selectivity of WQ-C-401 (S score (1) of PDGFR = (0.01)). In monocrotaline (MCT)-induced PAH rats, intragastric administration of WQ-C-401 (25, 50, 100 mg/kg/d) or imatinib (50 mg/kg/d, positive control) significantly decreased right ventricular systolic pressure (RVSP). Histological analysis demonstrated that WQ-C-401 inhibited pulmonary vascular remodeling by reducing muscularization and fibrosis, as well as alleviated right ventricular hypertrophy in MCT-treated rats. In addition, WQ-C-401 suppressed MCT-induced cell hyperproliferation and CD68+ macrophage infiltration around the pulmonary artery. In vitro, WQ-C-401 inhibited PDGF-BB-induced proliferation and migration of human pulmonary arterial smooth muscle cells (PASMCs). Moreover, Western blot analysis showed that WQ-C-401 concertration-dependently inhibited PDGF-BB-induced phosphorylation of ERK1/2 and PDGFRß Y751, decreased collagen Ⅰ synthesis and increased alpha smooth muscle actin (α-SMA) expression in PASMCs. Collectively, our results suggest that WQ-C-401 is a selective and potent PDGFR inhibitor which could be a promising drug for the therapeutics of PAH by preventing pulmonary vascular remodeling.

Cabin air dynamics: Unraveling the patterns and drivers of volatile organic compound distribution in vehicles.

Volatile organic compounds (VOCs) are ubiquitous in vehicle cabin environments, which can significantly impact the health of drivers and passengers, whereas quick and intelligent prediction methods are lacking. In this study, we firstly analyzed the variations of environmental parameters, VOC levels and potential sources inside a new car during 7 summer workdays, indicating that formaldehyde had the highest concentration and about one third of the measurements exceeded the standard limit for in-cabin air quality. Feature importance analysis reveals that the most important factor affecting in-cabin VOC emission behaviors is the material surface temperature rather than the air temperature. By introducing the attention mechanism and ensemble strategy, we present an LSTM-A-E deep learning model to predict the concentrations of 12 observed typical VOCs, together with other five deep learning models for comparison. By comparing the prediction-observation discrepancies and five evaluation metrics, the LSTM-A-E model demonstrates better performance, which is more consistent with field measurements. Extension of the developed model for predicting the 10-day VOC concentrations in a realistic residence further illustrates its excellent environmental adaptation. This study probes the not-well-explored in-cabin VOC dynamics via observation and deep learning approaches, facilitating rapid prediction and exposure assessment of VOCs in the vehicle micro-environment.

Disulfidptosis-Related LncRNA Signatures for Prognostic Prediction in Kidney Renal Clear Cell Carcinoma.

INTRODUCTION BACKGROUND: Disulfidptosis is a prevalent apoptotic mechanism, intrinsically linked to cancer prognosis. However, the specific involvement of disulfidptosis-related long non-coding RNA (DRLncRNAs) in Kidney renal clear cell carcinoma (KIRC) remains incompletely understood. This study aims to elucidate the potential prognostic significance of disulfidptosis-related LncRNAs in KIRC. MATERIALS AND METHODS: Expression profiles and clinical data of KIRC patients were retrieved from the TCGA database to discern differentially expressed DRLncRNAs correlated with overall survival. Cox univariate analysis, Lasso Regression, and Cox multivariate analysis were used to construct a clinical prediction model. RESULTS: Six signatures, namely FAM83C.AS1, AC136475.2, AC121338.2, AC026401.3, AC254562.3, and AC000050.2, were established to evaluate overall survival (OS) in the context of Kidney renal clear cell carcinoma (KIRC) in this study. Survival analysis and ROC curves demonstrated the strong predictive performance of the associated signature. The nomogram exhibited accurate prognostic predictions for overall patient survival, offering substantial clinical utility. Gene set enrichment analysis revealed that risk signals were enriched in various immune-related pathways. Furthermore, the risk features exhibited significant correlations with immune cells, immune function, immune cell infiltration, and immune checkpoints. CONCLUSION: This study has unveiled, for the first time, six disulfdptosis-related LncRNA signatures, laying a solid foundation for enhanced and precise prognostic predictions in KIRC.

Exposure to nanopollutants (nZnO) enhances the negative effects of hypoxia and delays recovery of the mussels' immune system.

Aquatic environments face escalating challenges from multiple stressors like hypoxia and nanoparticle exposure, with impact of these combined stressors on mussel immunity being poorly understood. We investigated the individual and combined effects of short-term and long-term hypoxia and exposure to zinc oxide nanoparticles (nZnO) on immune system of the mussels (Mytilus edulis). Hemocyte functional traits (mortality, adhesion capacity, phagocytosis, lysosomal abundance, and oxidative burst), and transcript levels of immune-related genes involved in pathogen recognition (the Toll-like receptors, the complement system components, and the adaptor proteins MyD88) were assessed. Short-term hypoxia minimally affected hemocyte parameters, while prolonged exposure led to immunosuppression, impacting hemocyte abundance, viability, phagocytosis, and defensin gene expression. Under normoxia, nZnO stimulated immune responses of mussel hemocytes. However, combined nZnO and hypoxia induced more pronounced and rapid immunosuppression than hypoxia alone, indicating a synergistic interaction. nZnO exposure hindered immune parameter recovery during post-hypoxic reoxygenation, suggesting persistent impact. Opposing trends were observed in pathogen-sensing and pathogen-elimination mechanisms, with a positive correlation between pathogen-recognition system activation and hemocyte mortality. These findings underscore a complex relationship and potential conflict between pathogen-recognition ability, immune function, and cell survival in mussel hemocytes under hypoxia and nanopollutant stress, and emphasize the importance of considering multiple stressors in assessing the vulnerability and adaptability of mussel immune system under complex environmental conditions of anthropogenically modified coastal ecosystems.

Nasal mucosal fibroblasts produce IL-4 to induce Th2 response.

Th2 polarization is essential for the pathogenesis of allergic rhinitis (AR). Th2 polarization's mechanism requires further understanding. IL-4 is the primary cytokine involved in Th2 response. Fibroblasts play a role in immune regulation. This study aims to elucidate the role of nasal mucosal fibroblast-derived IL-4 in the induction of Th2 responses. Nasal mucosal tissues were obtained from surgically removed samples from patients with nasal polyps, whether with or without AR. Fibroblasts were isolated from the tissues by flow cytometry cell sorting, and analyzed by RNA sequencing (RNAseq). The data from RNAseq showed that nasal fibroblasts expressed genes of GATA3, CD80, CD83, CD86, STAT6, IL2, IL4, IL5, IL6, IL13 and costimulatory factor. The data were verified by RT-qPCR. The level of gene activity was positively correlated with those of AR-related cytokines present in nasal secretions. Nasal fibroblasts release IL-4 upon activation. Nasal fibroblasts had the ability to transform naive CD4+ T cells into Th2 cells, which can be eliminated by inhibiting IL-4 receptor or CD28 in CD4+ T cells. To sum up, nasal mucosal fibroblasts produce IL-4, which can induce Th2 cell development. The data implicate that nasal fibroblasts are involved in the pathogenesis of nasal allergy.

Resistive Memristors Using Robust Electropolymerized Porous Organic Polymer Films as Switchable Materials.

Porous organic polymers (POPs) with inherent porosity, tunable pore environment, and semiconductive property are ideally suitable for application in various advanced semiconductor-related devices. However, owing to the lack of processability, POPs are usually prepared in powder forms, which limits their application in advanced devices. Herein, we demonstrate an example of information storage application of POPs with film form prepared by an electrochemical method. The growth process of the electropolymerized films in accordance with the Volmer-Weber model was proposed by observation of atomic force microscopy. Given the mechanism of the electron transfer system, we verified and mainly emphasized the importance of porosity and interfacial properties of porous polymer films for memristor. As expected, the as-fabricated memristors exhibit good performance on low turn-on voltage (0.65 ± 0.10 V), reliable data storage, and high on/off current ratio (104). This work offers inspiration for applying POPs in the form of electropolymerized films in various advanced semiconductor-related devices.

Carbon Dots Derived from Curcumae Radix and Their Heartprotective Effect.

Background: Acute myocardial infarction (AMI) is a common cardiovascular disease in clinic. Currently, there is no specific treatment for AMI. Carbon dots (CDs) have been reported to show excellent biological activities, which hold promise for the development of novel nanomedicines for the treatment of cardiovascular diseases. Methods: In this study, we firstly prepared CDs from the natural herb Curcumae Radix Carbonisata (CRC-CDs) by a green, simple calcination method. The aim of this study is to investigate the cardioprotective effect and mechanism of CRC-CDs on isoproterenol (ISO) -induced myocardial infarction (MI) in rats. Results: The results showed that pretreatment with CRC-CDs significantly reduced serum levels of cardiac enzymes (CK-MB, LDH, AST) and lipids (TC, TG, LDL) and reduced st-segment elevation and myocardial infarct size on the ECG in AMI rats. Importantly, cardiac ejection fraction (EF) and shortening fraction (FS) were markedly elevated, as was ATPase activity. In addition, CRC-CDs could significantly increase the levels of superoxide dismutase (SOD), reduced glutathione (GSH), catalase (CAT), and reduce the levels of malondialdehyde (MDA) and reactive oxygen species (ROS) in myocardial tissue, thereby exerting cardioprotective effect by enhancing the antioxidant capacity of myocardial tissue. Moreover, the TUNEL staining image showed that positive apoptotic cells were markedly declined after CRC-CDs treatment, which indicate that CRC-CDs could inhibit cardiomyocyte apoptosis. Importantly, The protective effect of CRC-CDs on H2O2 -pretreated H9c2 cells was also verified in vitro. Conclusion: Taken together, CRC-CDs has the potential for clinical application as an anti-myocardial ischemia drug candidate, which not only provides evidence for further broadening the biological application of cardiovascular diseases, but also offers potential hope for the application of nanomedicine to treat intractable diseases.

Biocompatible Electronic Skins for Cardiovascular Health Monitoring.

Cardiovascular diseases represent a significant threat to the overall well-being of the global population. Continuous monitoring of vital signs related to cardiovascular health is essential for improving daily health management. Currently, there has been remarkable proliferation of technology focused on collecting data related to cardiovascular diseases through daily electronic skin monitoring. However, concerns have arisen regarding potential skin irritation and inflammation due to the necessity for prolonged wear of wearable devices. To ensure comfortable and uninterrupted cardiovascular health monitoring, the concept of biocompatible electronic skin has gained substantial attention. In this review, biocompatible electronic skins for cardiovascular health monitoring are comprehensively summarized and discussed. The recent achievements of biocompatible electronic skin in cardiovascular health monitoring are introduced. Their working principles, fabrication processes, and performances in sensing technologies, materials, and integration systems are highlighted, and comparisons are made with other electronic skins used for cardiovascular monitoring. In addition, the significance of integrating sensing systems and the updating wireless communication for the development of the smart medical field is explored. Finally, the opportunities and challenges for wearable electronic skin are also examined.

Impacts of active mobility on individual health mediated by physical activities.

Active mobility, encompassing walking and cycling for transportation, is a potential solution to health issues arising from inadequate physical activity in modern society. However, the extent of active mobility's impact on individual physical activity levels, and its association with health as mediated by physical activities, is not fully quantified. This study aims to clarify the direct relationship between active mobility usage and individual health, as well as the indirect relationship mediated by physical activity, with a focus on varying levels of physical activity intensity. Utilizing data from the 2017 U.S. National Household Travel Survey (NHTS), we employed Poisson regression to predict active mobility usage based on socio-demographic and household socio-economic characteristics. A Structural Equation Model (SEM) was then used to investigate the direct and indirect effects of active mobility on individual health, mediated by physical activity. We further segmented individuals according to their intensity of physical activity to examine how such effect differs between different levels of physical activity. The study demonstrates that active mobility usage positively correlates with both the amount and intensity of physical activity. The effect of active mobility on individual health includes a direct positive effect (29% for intensity, 67.7% for amount) and an indirect effect mediated by physical activity (71% for intensity, 32.3% for amount). Notably, the mediation effect of active mobility on health is more substantial in the context of vigorous physical activities compared to light or moderate activities. Our findings reveal a significant positive influence of active mobility on individual health, encompassing both direct and indirect effects mediated by physical activities. These results quantitatively underscore the health benefits of active mobility and suggest the importance of promoting active mobility as a strategy to improve public health.

Phytochemicals against Osteoarthritis by Inhibiting Apoptosis.

Osteoarthritis (OA) is a chronic joint disease that causes pathological changes in articular cartilage, synovial membrane, or subchondral bone. Conventional treatments for OA include surgical and non-surgical methods. Surgical treatment is suitable for patients in the terminal stage of OA. It is often the last choice because of the associated risks and high cost. Medication of OA mainly includes non-steroidal anti-inflammatory drugs, analgesics, hyaluronic acid, and cortico-steroid anti-inflammatory drugs. However, these drugs often have severe side effects and cannot meet the needs of patients. Therefore, safe and clinically appropriate long-term treatments for OA are urgently needed. Apoptosis is programmed cell death, which is a kind of physiologic cell suicide determined by heredity and conserved by evolution. Inhibition of apoptosis-related pathways has been found to prevent and treat a variety of diseases. Excessive apoptosis can destroy cartilage homeostasis and aggravate the pathological process of OA. Therefore, inhibition of apoptosis-related factors or signaling pathways has become an effective means to treat OA. Phytochemicals are active ingredients from plants, and it has been found that phytochemicals can play an important role in the prevention and treatment of OA by inhibiting apoptosis. We summarize preclinical and clinical studies of phytochemicals for the treatment of OA by inhibiting apoptosis. The results show that phytochemicals can treat OA by targeting apoptosis-related pathways. On the basis of improving some phytochemicals with low bioavailability, poor water solubility, and high toxicity by nanotechnology-based drug delivery systems, and at the same time undergoing strict clinical and pharmacological tests, phytochemicals can be used as a potential therapeutic drug for OA and may be applied in clinical settings.

Ditch control and land reclamation promote vegetation recovery in Loess Plateau.

This study aimed to assess the impact of land consolidation projects and climate change on changes in vegetation in the Loess Plateau during 2012-2021. The study also explored the impacts of human activities and climate change on the ecological quality of the Loess Plateau during this period. The spatial and temporal normalized difference combined meteorological monitoring data, project data, and normalized difference vegetation index (NDVI) data that was used to create the vegetation index dataset spanning from 2012-2021. The study discussed and assessed the effectiveness of the project, revealing the following results: 1) A significant increase was observed in the vegetation index of the Loess Plateau region from 2012 to 2021, with an upward trend of 0.0024 per year (P < 0.05). 2) Contributions to changes in vegetation attributed to climatic factors and the anthropogenic factors of the ditch construction project were 82.74 and 17.62%, respectively, with climatic factors dominating and the degree of response of the ditch construction project increasing annually. 3) In the Loess Plateau, climatic variables dominated changes in vegetation. However, land consolidation projects in vegetation factors played a key role in changes in vegetation, and the degree of influence was gradually increasing.

Spatial transcriptomics reveals heterogeneity of histological subtypes between lepidic and acinar lung adenocarcinoma.

BACKGROUND: Patients who possess various histological subtypes of early-stage lung adenocarcinoma (LUAD) have considerably diverse prognoses. The simultaneous existence of several histological subtypes reduces the clinical accuracy of the diagnosis and prognosis of early-stage LUAD due to intratumour intricacy. METHODS: We included 11 postoperative LUAD patients pathologically confirmed to be stage IA. Single-cell RNA sequencing (scRNA-seq) was carried out on matched tumour and normal tissue. Three formalin-fixed and paraffin-embedded cases were randomly selected for 10× Genomics Visium analysis, one of which was analysed by digital spatial profiler (DSP). RESULTS: Using DSP and 10× Genomics Visium analysis, signature gene profiles for lepidic and acinar histological subtypes were acquired. The percentage of histological subtypes predicted for the patients from samples of 11 LUAD fresh tissues by scRNA-seq showed a degree of concordance with the clinicopathologic findings assessed by visual examination. DSP proteomics and 10× Genomics Visium transcriptomics analyses revealed that a negative correlation (Spearman correlation analysis: r = -.886; p = .033) between the expression levels of CD8 and the expression trend of programmed cell death 1(PD-L1) on tumour endothelial cells. The percentage of CD8+ T cells in the acinar region was lower than in the lepidic region. CONCLUSIONS: These findings illustrate that assessing patient histological subtypes at the single-cell level is feasible. Additionally, tumour endothelial cells that express PD-L1 in stage IA LUAD suppress immune-responsive CD8+ T cells.

Dual-environment-sensitive probe to detect protein aggregation in stressed laryngeal carcinoma cells and tissues.

The interplay between protein folding and biological activity is crucial, with the integrity of the proteome being paramount to ensuring effective biological function execution. In this study, we report a dual-environment-sensitive probe A1, capable of selectively binding to protein aggregates and dynamically monitoring their formation and degradation. Through in vitro, cellular, and tissue assays, A1 demonstrated specificity in distinguishing aggregated from folded protein states, selectively partitioning into aggregated proteins. Thermal shift assays revealed A1 could monitor the process of protein aggregation upon binding to misfolded proteins and preceding to insoluble aggregate formation. In cellular models, A1 detected stress-induced proteome aggregation in TU212 cells (laryngeal carcinoma cells), revealing a less polar microenvironment within the aggregated proteome. Similarly, tissue samples showed more severe proteome aggregation in cancerous tissues compared to paracancerous tissues. Overall, A1 represents a versatile tool for probing protein aggregation with significant implications for both fundamental research and clinical diagnostics.

Circadian rhythms of melatonin and its relationship with anhedonia in patients with mood disorders: a cross-sectional study.

BACKGROUND: Mood disorders are strongly associated with melatonin disturbances. However, it is unclear whether there is a difference in melatonin concentrations and melatonin circadian rhythm profiles between depression and bipolar disorder. In addition, the relationship between anhedonia, a common symptom of affective disorders, and its melatonin circadian rhythm remains under-investigated. METHODS: Thirty-four patients with depression disorder, 20 patients diagnosed with bipolar disorder and 21 healthy controls participated in this study. The Revised Physical Anhedonia Scale (RPAS) was performed to assess anhedonia. Saliva samples were collected from all subjects at fixed time points (a total of 14 points) in two consecutive days for measuring the melatonin concentrations to fit circadian rhythms of subjects. Melatonin circadian rhythms were compared between the three groups using ANOVA. Partial correlation analysis and linear regression analysis were used to explore the correlation between melatonin rhythm variables and anhedonia. RESULTS: We found that the peak phase of melatonin in the depression group was significantly advanced compared to the control group (P < 0.001) and the bipolar disorder group (P = 0.004). The peak phase of melatonin and RPAS showed a negative correlation (P = 0.003) in depression patients, which was also demonstrated in the multiple linear regression model (B=-2.47, P = 0.006). CONCLUSIONS: These results suggest that circadian rhythms of melatonin are differentiated in depression and bipolar disorder and correlate with anhedonia in depression. Future research needs to explore the neurobiological mechanisms linking anhedonia and melatonin circadian rhythms in depressed patients.

Diagnostic strategy of metagenomic next-generation sequencing for gram negative bacteria in respiratory infections.

OBJECTIVE: This study aims to identify the most effective diagnostic method for distinguishing pathogenic and non-pathogenic Gram-negative bacteria (GNB) in suspected pneumonia cases using metagenomic next-generation sequencing (mNGS) on bronchoalveolar lavage fluid (BALF) samples. METHODS: The effectiveness of mNGS was assessed on BALF samples collected from 583 patients, and the results were compared with those from microbiological culture and final clinical diagnosis. Three interpretational approaches were evaluated for diagnostic accuracy. RESULTS: mNGS outperformed culture significantly. Among the interpretational approaches, Clinical Interpretation (CI) demonstrated the best diagnostic performance with a sensitivity of 87.3%, specificity of 100%, positive predictive value of 100%, and negative predictive value of 98.3%. CI's specificity was significantly higher than Simple Interpretation (SI) at 37.9%. Additionally, CI excluded some microorganisms identified as putative pathogens by SI, including Haemophilus parainfluenzae, Haemophilus parahaemolyticus, and Klebsiella aerogenes. CONCLUSION: Proper interpretation of mNGS data is crucial for accurately diagnosing respiratory infections caused by GNB. CI is recommended for this purpose.

GSB: Group superposition binarization for vision transformer with limited training samples.

Vision Transformer (ViT) has performed remarkably in various computer vision tasks. Nonetheless, affected by the massive amount of parameters, ViT usually suffers from serious overfitting problems with a relatively limited number of training samples. In addition, ViT generally demands heavy computing resources, which limit its deployment on resource-constrained devices. As a type of model-compression method, model binarization is potentially a good choice to solve the above problems. Compared with the full-precision one, the model with the binarization method replaces complex tensor multiplication with simple bit-wise binary operations and represents full-precision model parameters and activations with only 1-bit ones, which potentially solves the problem of model size and computational complexity, respectively. In this paper, we investigate a binarized ViT model. Empirically, we observe that the existing binarization technology designed for Convolutional Neural Networks (CNN) cannot migrate well to a ViT's binarization task. We also find that the decline of the accuracy of the binary ViT model is mainly due to the information loss of the Attention module and the Value vector. Therefore, we propose a novel model binarization technique, called Group Superposition Binarization (GSB), to deal with these issues. Furthermore, in order to further improve the performance of the binarization model, we have investigated the gradient calculation procedure in the binarization process and derived more proper gradient calculation equations for GSB to reduce the influence of gradient mismatch. Then, the knowledge distillation technique is introduced to alleviate the performance degradation caused by model binarization. Analytically, model binarization can limit the parameter's search space during parameter updates while training a model. Therefore, the binarization process can actually play an implicit regularization role and help solve the problem of overfitting in the case of insufficient training data. Experiments on three datasets with limited numbers of training samples demonstrate that the proposed GSB model achieves state-of-the-art performance among the binary quantization schemes and exceeds its full-precision counterpart on some indicators. Code and models are available at: https://github.com/IMRL/GSB-Vision-Transformer.

Knowledge-guided machine learning can improve carbon cycle quantification in agroecosystems.

Accurate and cost-effective quantification of the carbon cycle for agroecosystems at decision-relevant scales is critical to mitigating climate change and ensuring sustainable food production. However, conventional process-based or data-driven modeling approaches alone have large prediction uncertainties due to the complex biogeochemical processes to model and the lack of observations to constrain many key state and flux variables. Here we propose a Knowledge-Guided Machine Learning (KGML) framework that addresses the above challenges by integrating knowledge embedded in a process-based model, high-resolution remote sensing observations, and machine learning (ML) techniques. Using the U.S. Corn Belt as a testbed, we demonstrate that KGML can outperform conventional process-based and black-box ML models in quantifying carbon cycle dynamics. Our high-resolution approach quantitatively reveals 86% more spatial detail of soil organic carbon changes than conventional coarse-resolution approaches. Moreover, we outline a protocol for improving KGML via various paths, which can be generalized to develop hybrid models to better predict complex earth system dynamics.