Mesoscopic structure graphs for interpreting uncertainty in non-linear embeddings.

Probabilistic-based non-linear dimensionality reduction (PB-NL-DR) methods, such as t-SNE and UMAP, are effective in unfolding complex high-dimensional manifolds, allowing users to explore and understand the structural patterns of data. However, due to the trade-off between global and local structure preservation and the randomness during computation, these methods may introduce false neighborhood relationships, known as distortion errors and misleading visualizations. To address this issue, we first conduct a detailed survey to illustrate the design space of prior layout enrichment visualizations for interpreting DR results, and then propose a node-link visualization technique, ManiGraph. This technique rethinks the neighborhood fidelity between the high- and low-dimensional spaces by constructing dynamic mesoscopic structure graphs and measuring region-adapted trustworthiness. ManiGraph also addresses the overplotting issue in scatterplot visualization for large-scale datasets and supports examining in unsupervised scenarios. We demonstrate the effectiveness of ManiGraph in different analytical cases, including generic machine learning using 3D toy data illustrations and fashion-MNIST, a computational biology study using a single-cell RNA sequencing dataset, and a deep learning-enabled colorectal cancer study with histopathology-MNIST.

A pathology foundation model for cancer diagnosis and prognosis prediction.

Histopathology image evaluation is indispensable for cancer diagnoses and subtype classification. Standard artificial intelligence methods for histopathology image analyses have focused on optimizing specialized models for each diagnostic task1,2. Although such methods have achieved some success, they often have limited generalizability to images generated by different digitization protocols or samples collected from different populations3. Here, to address this challenge, we devised the Clinical Histopathology Imaging Evaluation Foundation (CHIEF) model, a general-purpose weakly supervised machine learning framework to extract pathology imaging features for systematic cancer evaluation. CHIEF leverages two complementary pretraining methods to extract diverse pathology representations: unsupervised pretraining for tile-level feature identification and weakly supervised pretraining for whole-slide pattern recognition. We developed CHIEF using 60,530 whole-slide images spanning 19 anatomical sites. Through pretraining on 44 terabytes of high-resolution pathology imaging datasets, CHIEF extracted microscopic representations useful for cancer cell detection, tumour origin identification, molecular profile characterization and prognostic prediction. We successfully validated CHIEF using 19,491 whole-slide images from 32 independent slide sets collected from 24 hospitals and cohorts internationally. Overall, CHIEF outperformed the state-of-the-art deep learning methods by up to 36.1%, showing its ability to address domain shifts observed in samples from diverse populations and processed by different slide preparation methods. CHIEF provides a generalizable foundation for efficient digital pathology evaluation for patients with cancer.

Characterization of Conserved Evolution in H7N9 Avian Influenza Virus Prior Mass Vaccination.

Vaccination is crucial for the prevention and mitigation of avian influenza infections in China. The inactivated H7N9 vaccine, when administered to poultry, significantly lowers the risk of infection among both poultry and humans, while also markedly decreasing the prevalence of H7N9 detections. Highly pathogenic (HP) H7N9 viruses occasionally appear, whereas their low pathogenicity (LP) counterparts have been scarcely detected since 2018. However, these contributing factors remain poorly understood. We conducted an exploratory investigation of the mechanics via the application of comprehensive bioinformatic approaches. We delineated the Yangtze River Delta (YRD) H7N9 lineage into 5 clades (YRD-A to E). Our findings highlight the emergence and peak occurrence of the LP H7N9-containing YRD-E clade during the 5th epidemic wave in China's primary poultry farming areas. A more effective control of LP H7N9 through vaccination was observed compared to that of its HP H7N9 counterpart. YRD-E exhibited a tardy evolutionary trajectory, denoted by the conservation of its genetic and antigenic variation. Our analysis of YRD-E revealed only minimal amino acid substitutions along its phylogenetic tree and a few selective sweep mutations since 2016. In terms of epidemic fitness, the YRD-E was measured to be lower than that of the HP variants. Collectively, these findings underscore the conserved evolutionary patterns distinguishing the YRD-E. Given the conservation presented in its evolutionary patterns, the YRD-E LP H7N9 is hypothesized to be associated with a reduction following the mass vaccination in a relatively short period owing to its lower probability of antigenic variation that might affect vaccine efficiency.

Recovery of the Infraorbital Nerve Following Open Reduction and Fixation Surgery of Zygomaticomaxillary Complex Fractures-A Prospective Cohort Study Based on Quantitative Sensory Testing.

This study aimed to assess the sensory function of the infraorbital nerve in patients with fractures of the zygomatic complex who underwent open reduction and internal fixation at different time points using quantitative sensory testing, which was established by the German Neuropathic Pain Research Network, comprising a 7-item mechanical and thermal sensory test. A total of 21 patients (age range 17-46 y, 14 males) with unilateral zygomatic complex fractures were included. Quantitative sensory testing was conducted before the operation and at 1 week, 3 months, and 6 months operatively, with cold detection threshold, warmth detection threshold, cold pain threshold, heat pain threshold, mechanical detection threshold, mechanical pain threshold, pressure pain threshold, and vibration detection threshold being measured in bilateral infraorbital regions. Notable changes in sensitivity were observed in all values except for the mechanical pain threshold. In the majority of patients with zygomaticomaxillary complex fractures, infraorbital hypoesthesia was significantly improved within 3 months postoperatively, and almost complete recovery could be achieved by 6 months postoperatively.

Predicting physical functioning status in older adults: insights from wrist accelerometer sensors and derived digital biomarkers of physical activity.

OBJECTIVE: Conventional physical activity (PA) metrics derived from wearable sensors may not capture the cumulative, transitions from sedentary to active, and multidimensional patterns of PA, limiting the ability to predict physical function impairment (PFI) in older adults. This study aims to identify unique temporal patterns and develop novel digital biomarkers from wrist accelerometer data for predicting PFI and its subtypes using explainable artificial intelligence techniques. MATERIALS AND METHODS: Wrist accelerometer streaming data from 747 participants in the National Health and Aging Trends Study (NHATS) were used to calculate 231 PA features through time-series analysis techniques-Tsfresh. Predictive models for PFI and its subtypes (walking, balance, and extremity strength) were developed using 6 machine learning (ML) algorithms with hyperparameter optimization. The SHapley Additive exPlanations method was employed to interpret the ML models and rank the importance of input features. RESULTS: Temporal analysis revealed peak PA differences between PFI and healthy controls from 9:00 to 11:00 am. The best-performing model (Gradient boosting Tree) achieved an area under the curve score of 85.93%, accuracy of 81.52%, sensitivity of 77.03%, and specificity of 87.50% when combining wrist accelerometer streaming data (WAPAS) features with demographic data. DISCUSSION: The novel digital biomarkers, including change quantiles, Fourier transform (FFT) coefficients, and Aggregated (AGG) Linear Trend, outperformed traditional PA metrics in predicting PFI. These findings highlight the importance of capturing the multidimensional nature of PA patterns for PFI. CONCLUSION: This study investigates the potential of wrist accelerometer digital biomarkers in predicting PFI and its subtypes in older adults. Integrated PFI monitoring systems with digital biomarkers would improve the current state of remote PFI surveillance.

Deep learning-based diagnosis and survival prediction of patients with renal cell carcinoma from primary whole slide images.

There is an urgent clinical demand to explore novel diagnostic and prognostic biomarkers for renal cell carcinoma (RCC). We proposed deep learning-based artificial intelligence strategies. The study included 1752 whole slide images from multiple centres. Based on the pixel-level of RCC segmentation, the diagnosis diagnostic model achieved an area under the receiver operating characteristic curve (AUC) of 0.977 (95% CI 0.969-0.984) in the external validation cohort. In addition, our diagnostic model exhibited excellent performance in the differential diagnosis of RCC from renal oncocytoma, which achieved an AUC of 0.951 (0.922-0.972). The graderisk for the recognition of high-grade tumour achieved AUCs of 0.840 (0.805-0.871) in the Cancer Genome Atlas (TCGA) cohort, 0.857 (0.813-0.894) in the Shanghai General Hospital (General) cohort, and 0.894 (0.842-0.933) in the Clinical Proteomic Tumor Analysis Consortium (CPTAC) cohort, for the recognition of high-grade tumour. The OSrisk for predicting 5-year survival status achieved an AUC of 0.784 (0.746-0.819) in the TCGA cohort, which was further verified in the independent general cohort and the CPTAC cohort, with AUCs of 0.774 (0.723-0.820) and 0.702 (0.632-0.765), respectively. Moreover, the competing-risk nomogram (CRN) showed its potential to be a prognostic indicator, with a hazard ratio (HR) of 5.664 (3.893-8.239, p<0.0001), outperforming other traditional clinical prognostic indicators. Kaplan-Meier survival analysis further illustrated that our CRN could significantly distinguish patients with high survival risk. Deep learning-based artificial intelligence could be a useful tool for clinicians to diagnose and predict the prognosis of RCC patients, thus improving the process of individualised treatment.

HiCervix: An Extensive Hierarchical Dataset and Benchmark for Cervical Cytology Classification.

Cervical cytology is a critical screening strategy for early detection of pre-cancerous and cancerous cervical lesions. The challenge lies in accurately classifying various cervical cytology cell types. Existing automated cervical cytology methods are primarily trained on databases covering a narrow range of coarse-grained cell types, which fail to provide a comprehensive and detailed performance analysis that accurately represents real-world cytopathology conditions. To overcome these limitations, we introduce HiCervix, the most extensive, multi-center cervical cytology dataset currently available to the public. HiCervix includes 40,229 cervical cells from 4,496 whole slide images, categorized into 29 annotated classes. These classes are organized within a three-level hierarchical tree to capture fine-grained subtype information. To exploit the semantic correlation inherent in this hierarchical tree, we propose HierSwin, a hierarchical vision transformer-based classification network. HierSwin serves as a benchmark for detailed feature learning in both coarse-level and fine-level cervical cancer classification tasks. In our comprehensive experiments, HierSwin demonstrated remarkable performance, achieving 92.08% accuracy for coarse-level classification and 82.93% accuracy averaged across all three levels. When compared to board-certified cytopathologists, HierSwin achieved high classification performance (0.8293 versus 0.7359 averaged accuracy), highlighting its potential for clinical applications. This newly released HiCervix dataset, along with our benchmark HierSwin method, is poised to make a substantial impact on the advancement of deep learning algorithms for rapid cervical cancer screening and greatly improve cancer prevention and patient outcomes in real-world clinical settings.

Biochemical characterization and metabolic reprogramming of amino acids in Soybean roots under drought stress.

Amino acids play important roles in stress resistance, plant growth, development, and quality, with roots serving as the primary organs for drought response. We conducted biochemical and multi-omics analyses to investigate the metabolic processes of root amino acids in drought-resistant (HN44) and drought-sensitive (HN65) soybean (Glycine max) varieties. Our analysis revealed an increase in total amino acid content in both varieties, with phenylalanine, proline, and methionine accumulating in both. Additionally, several amino acids exhibited significant decreases in HN65 but slight increases in HN44. Multi-omics association analysis identified 13 amino acid-related pathways. We thoroughly examined the changes in genes and metabolites involved in various amino acid metabolism/synthesis and determined core genes and metabolites through correlation networks. The phenylalanine, tyrosine, and tryptophan metabolic pathways and proline, glutamic acid and sulfur-containing amino acid pathways were particularly important for drought resistance. Some candidate genes, such as ProDH and P4HA family genes, and metabolites, such as O-acetyl-L-serine, directly affected up- and downstream metabolism to induce drought resistance. This study provided a basis for soybean drought resistance breeding.

The application potential of mepiquat chloride in soybean: improvement of yield characteristics and drought resistance.

BACKGROUND: Drought can result in yield losses, the application of plant growth regulators is an effective measure to improve drought resistance and yield. The objective of the study was to explore the application potential of mepiquat chloride (MC) in regulating soybean yield and drought resistance. METHODS: In this study, a three-year field experiment was designed and combined with drought experiments to measure the yield of popularized varieties during 2021-2022 and drought-resistant and drought-sensitive varieties were selected, and planted in the field in 2023. RESULTS: MC increased the yield of HN84 and HN87 for two consecutive years from 2021 to 2022 and improved their physiological characteristics under field conditions. Under M200 treatment, the yield of HN84 increased by 6.93% and 9.46%, and HN87 increased by 11.11% and 15.72%. Different concentrations of MC have different effects on soybeans. The maximum increase of SOD, POD and proline in HN84 under M400 treatment reached 71.92%, 63.26% and 71.54%, respectively; the maximum increase of SOD, POD and proline in HN87 under M200 treatment reached 21.96%, 93.49% and 40.45%, respectively. In 2023, the foliar application of MC improved the physiological characteristics of HN44 and HN65 under drought-stress conditions. On the eighth day of drought treatment, compared to the drought treatment, the leaf and root dry weight of HN44 under M100 treatment increased by 17.91% and 32.76%, respectively; the dry weight of leaves and roots of HN65 increased by 20.74% and 29.29% under M200 treatment, respectively. MC also reduced malondialdehyde (MDA) content, decreased antioxidant enzyme activity and proline content. In addition, different concentrations of MC increased the chlorophyll fluorescence parameters (Fs, Fv/Fm, YII, and SPAD). In the field, the plant height of the two varieties decreased significantly, the yield increased, the number of two-grain and three-grain pods increased, and the stem length at the bottom and middle decreased with MC induction. CONCLUSIONS: The application of 100-200 mg/L MC effectively improved drought resistance and increased yield. This study provided support for the rational application of MC in soybean production.

Current views of drought research: experimental methods, adaptation mechanisms and regulatory strategies.

Drought stress is one of the most important abiotic stresses which causes many yield losses every year. This paper presents a comprehensive review of recent advances in international drought research. First, the main types of drought stress and the commonly used drought stress methods in the current experiment were introduced, and the advantages and disadvantages of each method were evaluated. Second, the response of plants to drought stress was reviewed from the aspects of morphology, physiology, biochemistry and molecular progression. Then, the potential methods to improve drought resistance and recent emerging technologies were introduced. Finally, the current research dilemma and future development direction were summarized. In summary, this review provides insights into drought stress research from different perspectives and provides a theoretical reference for scholars engaged in and about to engage in drought research.

Federated attention consistent learning models for prostate cancer diagnosis and Gleason grading.

Artificial intelligence (AI) holds significant promise in transforming medical imaging, enhancing diagnostics, and refining treatment strategies. However, the reliance on extensive multicenter datasets for training AI models poses challenges due to privacy concerns. Federated learning provides a solution by facilitating collaborative model training across multiple centers without sharing raw data. This study introduces a federated attention-consistent learning (FACL) framework to address challenges associated with large-scale pathological images and data heterogeneity. FACL enhances model generalization by maximizing attention consistency between local clients and the server model. To ensure privacy and validate robustness, we incorporated differential privacy by introducing noise during parameter transfer. We assessed the effectiveness of FACL in cancer diagnosis and Gleason grading tasks using 19,461 whole-slide images of prostate cancer from multiple centers. In the diagnosis task, FACL achieved an area under the curve (AUC) of 0.9718, outperforming seven centers with an average AUC of 0.9499 when categories are relatively balanced. For the Gleason grading task, FACL attained a Kappa score of 0.8463, surpassing the average Kappa score of 0.7379 from six centers. In conclusion, FACL offers a robust, accurate, and cost-effective AI training model for prostate cancer pathology while maintaining effective data safeguards.

SIRT2 promotes base excision repair by transcriptionally activating OGG1 in an ATM/ATR-dependent manner.

Sirtuin 2 (SIRT2) regulates the maintenance of genome integrity by targeting pathways of DNA damage response and homologous recombination repair. However, whether and how SIRT2 promotes base excision repair (BER) remain to be determined. Here, we found that independent of its catalytic activity SIRT2 interacted with the critical glycosylase OGG1 to promote OGG1 recruitment to its own promoter upon oxidative stress, thereby enhancing OGG1 promoter activity and increasing BER efficiency. Further studies revealed that SIRT2 was phosphorylated on S46 and S53 by ATM/ATR upon oxidative stress, and SIRT2 phosphorylation enhanced the SIRT2-OGG1 interaction and mediated the stimulatory effect of SIRT2 on OGG1 promoter activity. We also characterized 37 cancer-derived SIRT2 mutants and found that 5 exhibited the loss of the stimulatory effects on OGG1 transcription. Together, our data reveal that SIRT2 acts as a tumor suppressor by promoting OGG1 transcription and increasing BER efficiency in an ATM/ATR-dependent manner.

Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge.

Recognition of mitotic figures in histologic tumor specimens is highly relevant to patient outcome assessment. This task is challenging for algorithms and human experts alike, with deterioration of algorithmic performance under shifts in image representations. Considerable covariate shifts occur when assessment is performed on different tumor types, images are acquired using different digitization devices, or specimens are produced in different laboratories. This observation motivated the inception of the 2022 challenge on MItosis Domain Generalization (MIDOG 2022). The challenge provided annotated histologic tumor images from six different domains and evaluated the algorithmic approaches for mitotic figure detection provided by nine challenge participants on ten independent domains. Ground truth for mitotic figure detection was established in two ways: a three-expert majority vote and an independent, immunohistochemistry-assisted set of labels. This work represents an overview of the challenge tasks, the algorithmic strategies employed by the participants, and potential factors contributing to their success. With an F1 score of 0.764 for the top-performing team, we summarize that domain generalization across various tumor domains is possible with today's deep learning-based recognition pipelines. However, we also found that domain characteristics not present in the training set (feline as new species, spindle cell shape as new morphology and a new scanner) led to small but significant decreases in performance. When assessed against the immunohistochemistry-assisted reference standard, all methods resulted in reduced recall scores, with only minor changes in the order of participants in the ranking.

Which model is more efficient in carbon emission prediction research? A comparative study of deep learning models, machine learning models, and econometric models.

Accurately predicting future carbon emissions is of great significance for the government to scientifically promote carbon emission reduction policies. Among the current technologies for forecasting carbon emissions, the most prominent ones are econometric models and deep learning, but few works have systematically compared and analyzed the forecasting performance of the methods. Therefore, the paper makes a comparison for deep learning model, machine learning model, and the econometric model to demonstrate whether deep learning is an efficient method for carbon emission prediction research. In model mechanism, neural network for deep learning refers to an information processing model established by simulating biological neural system, and the model can be further extended through bionic characteristics. So the paper further optimizes the model from the perspective of bionics and proposes an innovative deep learning model based on the memory behavior mechanism of group creatures. Comparison results show that the prediction accuracy of the heuristic neural network is higher than that of the econometric model. Through in-depth analysis, the heuristic neural network is more suitable for predicting future carbon emissions, while the econometric model is more suitable for clarifying the impact of influencing factors on carbon emissions.

Molecular mechanism of drought resistance in soybean roots revealed using physiological and multi-omics analyses.

Soybeans are one of the most cultivated crops worldwide and drought can seriously affect their growth and development. Many studies have elucidated the mechanisms through which soybean leaves respond to drought; however, little is known about these mechanisms in roots. We used two soybean varieties with different drought tolerances to study the morphological, physiological, and molecular response mechanisms of the root system to drought stress in seedlings. We found that drought stress led to a significant decrease in the root traits and an increase in antioxidant enzyme activity in the two varieties. Drought-resistant varieties accumulate large amounts of flavonoids and phenolic acids at the metabolic level, which causes variations in drought resistance. Additionally, differences in gene expression and drought-resistance pathways between the two varieties were clarified using transcriptome analysis. Through a multi-omics joint analysis, phenylpropanoid and isoflavonoid biosynthesis were identified as the core drought resistance pathways in soybean roots. Candidate genes and marker metabolites affecting drought resistance were identified. The phenylpropanoid pathway confers drought tolerance to roots by maintaining a high level of POD activity and mediates the biosynthesis of various secondary drought-resistant metabolites to resist drought stress. This study provides useful data for investigating plant root drought responses and offers theoretical support for plant breeding for drought resistance.

Spatial pattern characteristics and influencing factors of mountainous rural settlements in metropolitan fringe area: A case study of Pingnan County, Fujian Province.

Amid global industrialization and urbanization, mountainous rural settlements, especially those in metropolitan fringe area, are experiencing significant spatial changes in location and scale. This study takes Pingnan County, Fujian Province, China, as an example. Utilizing land use data and employing methods including standard deviation ellipse, average nearest neighbor index, kernel density estimation, spatial hotspot detection, binary logistic regression model, and Geodetector, this study aims to scientifically identify the spatial pattern characteristics and influencing factors of its settlements. The results show that: (1) The spatial distribution of settlements in Pingnan County is biased toward the southern part of the county; the center of settlement's spatial distribution is located south of the junction of Gufeng Town and Pingcheng Town; the spatial distribution trend of settlements is north-east-southwest. Settlements are generally aggregated, and the aggregation degree of Gufeng Town is obviously lower than that of other towns. (2) The density distribution of settlements presents a "core-periphery" structure and a "north-south linear" structure in space; the spatial pattern characteristics show high-density, large patches in Gufeng Town, high-density, small patches in Changqiao Town, Tangkou Town and Gantang Town, and medium-density or low-density, small patches in other towns. (3) Settlement location is mainly affected by the elevation, distance to cultivated land, and distance to main roads, while settlement scale is mainly affected by slope, relief degree of land surface, and distance to urban centers. The interaction between these factors exhibits enhancement effects, with natural terrain and location conditions exerting the most prominent influence. These findings underscore the strong constraints posed by natural topography on mountainous rural settlements in metropolitan fringe areas, coupled with a more pronounced influence from socio-economic factors. The study's results hold significant implications for optimizing the layout of such settlements, guiding land spatial planning, and promoting rural revitalization.

CoNIC Challenge: Pushing the frontiers of nuclear detection, segmentation, classification and counting.

Nuclear detection, segmentation and morphometric profiling are essential in helping us further understand the relationship between histology and patient outcome. To drive innovation in this area, we setup a community-wide challenge using the largest available dataset of its kind to assess nuclear segmentation and cellular composition. Our challenge, named CoNIC, stimulated the development of reproducible algorithms for cellular recognition with real-time result inspection on public leaderboards. We conducted an extensive post-challenge analysis based on the top-performing models using 1,658 whole-slide images of colon tissue. With around 700 million detected nuclei per model, associated features were used for dysplasia grading and survival analysis, where we demonstrated that the challenge's improvement over the previous state-of-the-art led to significant boosts in downstream performance. Our findings also suggest that eosinophils and neutrophils play an important role in the tumour microevironment. We release challenge models and WSI-level results to foster the development of further methods for biomarker discovery.

Graph model for multiple scattering in lithium niobate on insulator integrated photonic networks.

We present a graph-based model for multiple scattering of light in integrated lithium niobate on insulator (LNOI) networks, which describes an open network of single-mode integrated waveguides with tunable scattering at the network nodes. We first validate the model at small scale with experimental LNOI resonator devices and show consistent agreement between simulated and measured spectral data. Then, the model is used to demonstrate a novel platform for on-chip multiple scattering in large-scale optical networks up to few hundred nodes, with tunable scattering behaviour and tailored disorder. Combining our simple graph-based model with material properties of LNOI, this platform creates new opportunities to control randomness in large optical networks.

PhaseFIT: live-organoid phase-fluorescent image transformation via generative AI.

Organoid models have provided a powerful platform for mechanistic investigations into fundamental biological processes involved in the development and function of organs. Despite the potential for image-based phenotypic quantification of organoids, their complex 3D structure, and the time-consuming and labor-intensive nature of immunofluorescent staining present significant challenges. In this work, we developed a virtual painting system, PhaseFIT (phase-fluorescent image transformation) utilizing customized and morphologically rich 2.5D intestinal organoids, which generate virtual fluorescent images for phenotypic quantification via accessible and low-cost organoid phase images. This system is driven by a novel segmentation-informed deep generative model that specializes in segmenting overlap and proximity between objects. The model enables an annotation-free digital transformation from phase-contrast to multi-channel fluorescent images. The virtual painting results of nuclei, secretory cell markers, and stem cells demonstrate that PhaseFIT outperforms the existing deep learning-based stain transformation models by generating fine-grained visual content. We further validated the efficiency and accuracy of PhaseFIT to quantify the impacts of three compounds on crypt formation, cell population, and cell stemness. PhaseFIT is the first deep learning-enabled virtual painting system focused on live organoids, enabling large-scale, informative, and efficient organoid phenotypic quantification. PhaseFIT would enable the use of organoids in high-throughput drug screening applications.

Mechanism of Mepiquat Chloride Regulating Soybean Response to Drought Stress Revealed by Proteomics.

Soybeans are the main sources of oil and protein for most of the global population. As the population grows, so does the demand for soybeans. However, drought is a major factor that limits soybean production. Regulating soybean response to drought stress using mepiquat chloride (MC) is a feasible method; however, its mechanism is still unclear. This study used PEG-6000 to simulate drought stress and quantitative proteomic techniques to reveal changes in Heinong44 (HN44) and Heinong65 (HN65) subjected to drought following the application of 100 mg/L of MC. The results showed that SOD in HN44 did not change significantly but decreased by 22.61% in HN65 after MC pretreatment, and MDA content decreased by 22.75% and 21.54% in HN44 and HN65, respectively. Furthermore, MC improved the GSH-ASA cycle and simultaneously promoted the Calvin cycle process to enable the plant to maintain a certain carbon assimilation rate under osmotic stress. In addition, MC upregulated some proteins during gluconeogenesis and starch metabolism and increased soluble sugar content by 8.41% in HN44. MC also reduced ribosomal protein abundance, affecting translation and amino acid metabolism. In summary, MC improved GSH-ASA cycle and Calvin cycle under stress to alleviate oxidative damage and maintain crop growth. Our study is the first to report the mechanism of MC regulation in soybean under osmotic stress, providing new insights for the rational application of MC in soybean.