Efficacy of rTMS in treating functional impairment in post-stroke patients: a systematic review and meta-analysis.

BACKGROUND: Most stroke patients suffer from an imbalance in blood supply, which causes severe brain damage leading to functional deficits in motor, sensory, swallowing, cognitive, emotional, and speech functions. Repetitive transcranial magnetic stimulation (rTMS) is thought to restore functions impaired during the stroke process and improve the quality of life of stroke patients. However, the efficacy of rTMS in treating post-stroke function impairment varies significantly. Therefore, we conducted a meta-analysis of the number of patients with effective rTMS in treating post-stroke dysfunction. METHODS: The PubMed, Embase, and Cochrane Library databases were searched. Screening and full-text review were performed by three investigators. Single-group rate meta-analysis was performed on the extracted data using a random variable model. Then subgroup analyses were performed at the levels of stroke acuity (acute, chronic, or subacute); post-stroke symptoms (including upper and lower limb motor function, dysphagia, depression, aphasia); rTMS stimulation site (affected side, unaffected side); and whether or not it was a combination therapy. RESULTS: We obtained 8955 search records, and finally 33 studies (2682 patients) were included in the meta-analysis. The overall analysis found that effective strength (ES) of rTMS was 0.53. In addition, we found that the ES of rTMS from acute/subacute/chronic post-stroke was 0.69, 0.45, and 0.52. We also found that the ES of rTMS using high-frequency stimulation was 0.56, while the ES of rTMS using low-frequency stimulation was 0.53. From post-stroke symptoms, we found that the ES of rTMS in sensory aspects, upper limb functional aspects, swallowing function, and aphasia was 0.50, 0.52, 0.51, and 0.54. And from the site of rTMS stimulation, we found that the ES of rTMS applied to the affected side was 0.51, while the ES applied to the unaffected side was 0.54. What's more, we found that the ES of rTMS applied alone was 0.53, while the ES of rTMS applied in conjunction with other therapeutic modalities was 0.53. CONCLUSIONS: By comparing the results of the data, we recommend rTMS as a treatment option for rehabilitation of functional impairment in patients after stroke. We also recommend that rehabilitation physicians or clinicians use combination therapy as one of the options for patients.

Hierarchical Negative Sampling Based Graph Contrastive Learning Approach for Drug-Disease Association Prediction.

Predicting potential drug-disease associations (RDAs) plays a pivotal role in elucidating therapeutic strategies for diseases and facilitating drug repositioning, making it of paramount importance. However, existing methods are constrained and rely heavily on limited domain-specific knowledge, impeding their ability to effectively predict candidate associations between drugs and diseases. Moreover, the simplistic definition of unknown information pertaining to drug-disease relationships as negative samples presents inherent limitations. To overcome these challenges, we introduce a novel hierarchical negative sampling-based graph contrastive model, termed HSGCLRDA, which aims to forecast latent associations between drugs and diseases. In this study, HSGCLRDA integrates the association information as well as similarity between drugs, diseases and proteins. Meanwhile, the model constructs a drug-disease-protein heterogeneous network. Subsequently, employing a hierarchical structural sampling technique, we establish reliable negative drug-disease samples utilizing PageRank algorithms. Utilizing meta-path aggregation within the heterogeneous network, we derive low-dimensional representations for drugs and diseases, thereby constructing global and local feature graphs that capture their interactions comprehensively. To obtain representation information, we adopt a self-supervised graph contrastive approach that leverages graph convolutional networks (GCNs) and second-order GCNs to extract feature graph information. Furthermore, we integrate a contrastive cost function derived from the cross-entropy cost function, facilitating holistic model optimization. Experimental results obtained from benchmark datasets not only showcase the superior performance of HSGCLRDA compared to various baseline methods in predicting RDAs but also emphasize its practical utility in identifying novel potential diseases associated with existing drugs through meticulous case studies.

Bridging the Gap between Field Experiments and Machine Learning: The EC H2020 B-GOOD Project as a Case Study towards Automated Predictive Health Monitoring of Honey Bee Colonies.

Honey bee colonies have great societal and economic importance. The main challenge that beekeepers face is keeping bee colonies healthy under ever-changing environmental conditions. In the past two decades, beekeepers that manage colonies of Western honey bees (Apis mellifera) have become increasingly concerned by the presence of parasites and pathogens affecting the bees, the reduction in pollen and nectar availability, and the colonies' exposure to pesticides, among others. Hence, beekeepers need to know the health condition of their colonies and how to keep them alive and thriving, which creates a need for a new holistic data collection method to harmonize the flow of information from various sources that can be linked at the colony level for different health determinants, such as bee colony, environmental, socioeconomic, and genetic statuses. For this purpose, we have developed and implemented the B-GOOD (Giving Beekeeping Guidance by computational-assisted Decision Making) project as a case study to categorize the colony's health condition and find a Health Status Index (HSI). Using a 3-tier setup guided by work plans and standardized protocols, we have collected data from inside the colonies (amount of brood, disease load, honey harvest, etc.) and from their environment (floral resource availability). Most of the project's data was automatically collected by the BEEP Base Sensor System. This continuous stream of data served as the basis to determine and validate an algorithm to calculate the HSI using machine learning. In this article, we share our insights on this holistic methodology and also highlight the importance of using a standardized data language to increase the compatibility between different current and future studies. We argue that the combined management of big data will be an essential building block in the development of targeted guidance for beekeepers and for the future of sustainable beekeeping.

A study on table tennis landing point detection algorithm based on spatial domain information.

To address the limitations of computer vision-assisted table tennis ball detection, which heavily relies on vision acquisition equipment and exhibits slow processing speed, we propose a real-time calculation method for determining the landing point of table tennis balls. This novel approach is based on spatial domain information and reduces the dependency on vision acquisition equipment. This method incorporates several steps: employing dynamic color thresholding to determine the centroid coordinates of all objects in the video frames, utilizing target area thresholding and spatial Euclidean distance to eliminate interference balls and noise, optimizing the total number of video frames through keyframe extraction to reduce the number of operations for object recognition and landing point detection, and employing the four-frame difference slope method and polygonal area determination to detect the landing point and area of the target object, thereby obtaining precise coordinates and their corresponding areas. Experimental results on the above method on the Jetson Nano development board show that the dynamic color thresholding method achieves a detection speed of 45.3 fps. The keyframe extraction method correctly identifies the landing point frames with an accuracy rate exceeding 93.3%. In terms of drop point detection, the proposed method achieves 78.5% overall accuracy in detecting table tennis ball drop points while ensuring real-time detection. These experiments validate that the proposed method has the ability to detect table tennis ball drop points in real time and accurately in low frame rate vision acquisition devices and real environments.

Predicting circRNA-drug resistance associations based on a multimodal graph representation learning framework.

Circular RNA (circRNA) is a class of noncoding RNA that is highly conserved and exhibit exceptional stability. Due to its function as a microRNA sponge, circRNA has gained significant attention as an essential biomarker and potential drug target in the pathogenesis of several cancers. Although many circRNAs have been identified to play a role in cancer resistance, traditional methods are time-consuming and expensive. In this context, computational methods offer a promising way to facilitate the discovery process. However, most existing prediction models focus on the association between circRNAs and drug resistance, without considering the corresponding disease-related information in the circRNA-drug resistance association. Incorporating disease-related information into the prediction of circRNA-drug resistance associations could potentially improve the efficiency and speed of discovering and developing circRNA-targeting drugs. We propose a computational framework, named GraphCDD, for predicting the association between circRNA and drug resistance. Our model utilizes data from three sources, namely circRNA, disease, and drug, to construct three similarity networks that represent the features of circRNA, disease, and drug, respectively. We utilize a multimodal graph neural network to acquire efficient representations of circRNAs, diseases, and drugs by integrating various types of information, and establish a predictive model. The experimental results have validated the effectiveness of our model and provided a promising method in predicting potential associations between circRNA and drug resistance. The source code and dataset of GraphCDD can be found at https://github.com/Ziqiang-Liu/GraphCDD.

Construction of diagnostic and prognostic models based on gene signatures of nasopharyngeal carcinoma by machine learning methods.

Background: Diagnostic models based on gene signatures of nasopharyngeal carcinoma (NPC) were constructed by random forest (RF) and artificial neural network (ANN) algorithms. Least absolute shrinkage and selection operator (Lasso)-Cox regression was used to select and build prognostic models based on gene signatures. This study contributes to the early diagnosis and treatment, prognosis, and molecular mechanisms associated with NPC. Methods: Two gene expression datasets were downloaded from the Gene Expression Omnibus (GEO) database, and differentially expressed genes (DEGs) associated with NPC were identified by gene expression differential analysis. Subsequently, significant DEGs were identified by a RF algorithm. ANN were used to construct a diagnostic model for NPC. The performance of the diagnostic model was evaluated by area under the curve (AUC) values using a validation set. Lasso-Cox regression examined gene signatures associated with prognosis. Overall survival (OS) and disease-free survival (DFS) prediction models were constructed and validated from The Cancer Genome Atlas (TCGA) database and the International Cancer Genome Consortium (ICGC) database. Results: A total of 582 DEGs associated with NPC were identified, and 14 significant genes were identified by the RF algorithm. A diagnostic model for NPC was successfully constructed using ANN, and the validity of the model was confirmed on the training set AUC =0.947 [95% confidence interval (CI): 0.911-0.969] and the validation set AUC =0.864 (95% CI: 0.828-0.901). The 24-gene signatures associated with prognosis were identified by Lasso-Cox regression, and prediction models for OS and DFS of NPC were constructed on the training set. Finally, the ability of the model was validated on the validation set. Conclusions: Several potential gene signatures associated with NPC were identified, and a high-performance predictive model for early diagnosis of NPC and a prognostic prediction model with robust performance were successfully developed. The results of this study provide valuable references for early diagnosis, screening, treatment and molecular mechanism research of NPC in the future.

Sphingosine-1-phosphate derived from PRP-Exos promotes angiogenesis in diabetic wound healing via the S1PR1/AKT/FN1 signalling pathway.

Background: Sphingosine-1-phosphate (S1P), a key regulator of vascular homeostasis and angiogenesis, is enriched in exosomes derived from platelet-rich plasma (PRP-Exos). However, the potential role of PRP-Exos-S1P in diabetic wound healing remains unclear. In this study, we investigated the underlying mechanism of PRP-Exos-S1P in diabetic angiogenesis and wound repair. Methods: Exosomes were isolated from PRP by ultracentrifugation and analysed by transmission electron microscopy, nanoparticle tracking analysis and western blotting. The concentration of S1P derived from PRP-Exos was measured by enzyme-linked immunosorbent assay. The expression level of S1P receptor1-3 (S1PR1-3) in diabetic skin was analysed by Q-PCR. Bioinformatics analysis and proteomic sequencing were conducted to explore the possible signalling pathway mediated by PRP-Exos-S1P. A diabetic mouse model was used to evaluate the effect of PRP-Exos on wound healing. Immunofluorescence for cluster of differentiation 31 (CD31) was used to assess angiogenesis in a diabetic wound model. Results: In vitro, PRP-Exos significantly promoted cell proliferation, migration and tube formation. Furthermore, PRP-Exos accelerated the process of diabetic angiogenesis and wound closure in vivo. S1P derived from PRP-Exos was present at a high level, and S1PR1 expression was significantly elevated compared with S1PR2 and S1PR3 in the skin of diabetic patients and animals. However, cell migration and tube formation were not promoted by PRP-Exos-S1P in human umbilical vein endothelial cells treated with shS1PR1. In the diabetic mouse model, inhibition of S1PR1 expression at wounding sites decreased the formation of new blood vessels and delayed the process of wound closure. Bioinformatics analysis and proteomics indicated that fibronectin 1 (FN1) was closely related to S1PR1 due to its colocalization in the endothelial cells of human skin. Further study supported that FN1 plays an important role in the PRP-Exos-S1P-mediated S1PR1/protein kinase B signalling pathway. Conclusions: PRP-Exos-S1P promotes angiogenesis in diabetic wound healing via the S1PR1/protein kinase B/FN1 signalling pathway. Our findings provide a preliminary theoretical foundation for the treatment of diabetic foot ulcers using PRP-Exos in the future.

Association of intraoperative hypotension and severe postoperative complications during non-cardiac surgery in adult patients: A systematic review and meta-analysis.

Background: Intraoperative hypotension (IOH) is a common side effect of non-cardiac surgery that might induce poor postoperative outcomes. The relationship between the IOH and severe postoperative complications is still unclear. Thus, we summarized the existing literature to evaluate whether IOH contributes to developing severe postoperative complications during non-cardiac surgery. Methods: We conducted a comprehensive search of PubMed, Embase, Cochrane Library, Web of Science, and the CBM from inception to 15 September 2022. The primary outcomes were 30-day mortality, acute kidney injury (AKI), major adverse cardiac events (myocardial injury or myocardial infarction), postoperative cognitive dysfunction (POCD), and postoperative delirium (POD). Secondary outcomes included surgical-site infection (SSI), stroke, and 1-year mortality. Results: 72 studies (3 randomized; 69 non-randomized) were included in this study. Low-quality evidence showed IOH resulted in an increased risk of 30-day mortality (OR, 1.85; 95% CI, 1.30-2.64; P < .001), AKI (OR, 2.69; 95% CI, 2.15-3.37; P < .001), and stroke (OR, 1.33; 95% CI, 1.21-1.46; P < .001) after non-cardiac surgery than non-IOH. Very low-quality evidence showed IOH was associated with a higher risk of myocardial injury (OR, 2.00; 95% CI, 1.17-3.43; P = .01), myocardial infarction (OR, 2.11; 95% CI, 1.41-3.16; P < .001), and POD (OR, 2.27; 95% CI, 1.53-3.38; P < .001). Very low-quality evidence showed IOH have a similar incidence of POCD (OR, 2.82; 95% CI, 0.83-9.50; P = .10) and 1-year-mortality (OR, 1.66; 95% CI, 0.65-4.20; P = .29) compared with non-IOH in non-cardiac surgery. Conclusion: Our results suggest IOH was associated with an increased risk of severe postoperative complications after non-cardiac surgery than non-IOH. IOH is a potentially avoidable hazard that should be closely monitored during non-cardiac surgery.

Predicting Potential Drug-Disease Associations Based on Hypergraph Learning with Subgraph Matching.

The search for potential drug-disease associations (DDA) can speed up drug development cycles, reduce costly wasted resources, and accelerate disease treatment by repurposing existing drugs that can control further disease progression. As technologies such as deep learning continue to mature, many researchers tend to use emerging technologies to predict potential DDA. The performance of DDA prediction is still challenging and there is some space for improvement due to issues such as the small number of existing associations and possible noise in the data. To better predict DDA, we propose a computational approach based on hypergraph learning with subgraph matching (HGDDA). In particular, HGDDA first extracts feature subgraph information in the validated drug-disease association network and proposes a negative sampling strategy based on similarity network to reduce the data imbalance. Second, the hypergraph Unet module is used by extracting Finally, the potential DDA is predicted by designing a hypergraph combination module to convolution and pooling the two constructed hypergraphs separately, and calculating the difference information between the subgraphs using cosine similarity for node matching. The performance of HGDDA is verified under two standard datasets by 10-fold cross-validation (10-CV), and the results outperform existing drug-disease prediction methods. In addition, to validate the overall utility of the model, the top 10 drugs for the specific disease are predicted through the case study and validated using the CTD database.

ISLMI:Predicting lncRNA-miRNA Interactions Based on Information Injection and Second-Order Graph Convolution Network.

Studies have shown that IncRNA-miRNA interactions can affect cellular expression at the level of gene molecules through a variety of regulatory mechanisms and have important effects on the biological activities of living organisms. Several biomolecular network-based approaches have been proposed to accelerate the identification of lncRNA-miRNA interactions. However, most of the methods cannot fully utilize the structural and topological information of the lncRNA-miRNA interaction network. In this article, we proposed a new method, ISLMI, a prediction model based on information injection and second order graph convolution network(SOGCN). The model calculated the sequence similarity and Gaussian interaction profile kernel similarity between lncRNA and miRNA, fused them to enhance the intrinsic interaction between the nodes, using SOGCN to learn second-order representations of similarity matrix information. At the same time, multiple feature representations obtain using different graph embedding methods were also injected into the second-order graph representation. Finally, matrix complementation was used to increase the model accuracy. The model combined the advantages of different methods and achieved reliable performance in 5-fold cross-validation, significantly improved the performance of predicting lncRNA-miRNA interactions. In addition, our model successfully confirmed the superiority of ISLMI by comparing it with several other model algorithm.

GraphCDA: a hybrid graph representation learning framework based on GCN and GAT for predicting disease-associated circRNAs.

MOTIVATION: CircularRNA (circRNA) is a class of noncoding RNA with high conservation and stability, which is considered as an important disease biomarker and drug target. Accumulating pieces of evidence have indicated that circRNA plays a crucial role in the pathogenesis and progression of many complex diseases. As the biological experiments are time-consuming and labor-intensive, developing an accurate computational prediction method has become indispensable to identify disease-related circRNAs. RESULTS: We presented a hybrid graph representation learning framework, named GraphCDA, for predicting the potential circRNA-disease associations. Firstly, the circRNA-circRNA similarity network and disease-disease similarity network were constructed to characterize the relationships of circRNAs and diseases, respectively. Secondly, a hybrid graph embedding model combining Graph Convolutional Networks and Graph Attention Networks was introduced to learn the feature representations of circRNAs and diseases simultaneously. Finally, the learned representations were concatenated and employed to build the prediction model for identifying the circRNA-disease associations. A series of experimental results demonstrated that GraphCDA outperformed other state-of-the-art methods on several public databases. Moreover, GraphCDA could achieve good performance when only using a small number of known circRNA-disease associations as the training set. Besides, case studies conducted on several human diseases further confirmed the prediction capability of GraphCDA for predicting potential disease-related circRNAs. In conclusion, extensive experimental results indicated that GraphCDA could serve as a reliable tool for exploring the regulatory role of circRNAs in complex diseases.

Development of a novel combined nomogram model integrating Rad-score, age and ECOG to predict the survival of patients with hepatocellular carcinoma treated by transcatheter arterial chemoembolization.

Background: Liver cancer is affecting more and more people's health. Transcatheter arterial chemoembolization (TACE) has become a routine treatment option, but the prognosis of patients is not optimistic. Effectively prediction of prognosis can provide clinicians with an objective basis for patient prognosis and timely adjustment of treatment strategies, thus improving the quality of patient survival. However, the current prediction methods have some limitations. Therefore, this study aims to develop a radiomics nomogram for predicting survival after TACE in patients with advanced hepatocellular carcinoma (HCC). Methods: Seventy advanced HCC patients treated with TACE were enrolled from January 2013 to July 2019. Clinical information included age, sex, and Eastern Cooperative Oncology Group (ECOG) score. Overall survival (OS) was confirmed by postoperative follow-up. Radiomics features were extracted using 3D Slicer (version 4.11.20210226) software, then obtain radiomics signature and calculate radiomics score (Rad-score) for each patient. Univariate and multivariate Cox regression were used to analyze the baseline clinical data of patients and establish clinical models. The obtained radiomics signature was incorporated into the clinical model to establish the radiomics nomogram. The predictive performance and calibration ability of the model were assessed by the area under the receiver operating characteristic (ROC) curve (AUC), C-index, and calibration curve. Results: Three significant features were selected from 851 radiomics features by the least absolute shrinkage and selection operator (LASSO) Cox regression model to construct the radiomics signature, and were significantly correlated with overall survival (P<0.001). Rad-score, age, and ECOG score were combined to construct a radiomics nomogram. The AUC, sensitivity, and specificity of the radiomics nomogram were 0.801 (95% CI: 0.693-0.909), 0.822 (95% CI: 0.674-0.915), and 0.720 (95% CI: 0.674-0.915), respectively. The C-index of the radiomics nomogram was 0.700 (95% CI: 0.547-0.853). Calibration curves showed better agreement between the predicted and actual probabilities in the radiomics nomogram among the 3 features. Conclusions: The Rad-score was a strong risk predictor of survival after TACE for HCC patients. The radiomics nomogram might be improved the predictive efficacy of survival after TACE and it may also provide assistance to physicians in making treatment decisions.

Effectiveness of aquatic exercise in lower limb osteoarthritis: a meta-analysis of randomized controlled trials.

The objective of this study was to evaluate the short-term and follow-up effectiveness of aquatic training on the health status of lower limb osteoarthritis. Randomized controlled trials (RCTs) on related topics were systematically searched in PubMed, Embase, Web of Science, the Cochrane Library, Physiotherapy Evidence Database (PEDro), the China National Knowledge Infrastructure and Wanfang databases from inception to January 2021. RevMan 5.3 was used for statistical analysis, and the standardized mean difference (SMD) was used to present pooled effect sizes. As a result, 19 RCTs (1592 patients) were included. Compared with unsupervised home exercise or usual care (land-based training excluded), aquatic training showed short-term pain relief (SMD, -0.54; 95% CI, -0.81 to -0.28), physical function improvement (SMD, -0.64; 95% CI, -1.00 to -0.28), stiffness reduction (SMD, -0.40; 95% CI, -0.79 to -0.01) and improved function in sport and recreation (SMD, -0.30; 95% CI, -0.59 to -0.02). Analyses restricted to patients with knee osteoarthritis only also confirmed the positive effects of aquatic training on most dimensions excluding physical function. At medium-term follow-ups, improvements in physical function and function in sport and recreation were observed. No significant difference was observed between arms in the above four outcomes at long-term follow-ups. All studies reported no major adverse event with relation to aquatic training, and the minor adverse events were not common. It is concluded that aquatic training likely has short-term benefits on pain, physical function, stiffness and sport ability in lower limb osteoarthritis patients, but these positive effects may not last long.

RLF-LPI: An ensemble learning framework using sequence information for predicting lncRNA-protein interaction based on AE-ResLSTM and fuzzy decision.

Long non-coding RNAs (lncRNAs) play a regulatory role in many biological cells, and the recognition of lncRNA-protein interactions is helpful to reveal the functional mechanism of lncRNAs. Identification of lncRNA-protein interaction by biological techniques is costly and time-consuming. Here, an ensemble learning framework, RLF-LPI is proposed, to predict lncRNA-protein interactions. The RLF-LPI of the residual LSTM autoencoder module with fusion attention mechanism can extract the potential representation of features and capture the dependencies between sequences and structures by k-mer method. Finally, the relationship between lncRNA and protein is learned through the method of fuzzy decision. The experimental results show that the ACC of RLF-LPI is 0.912 on ATH948 dataset and 0.921 on ZEA22133 dataset. Thus, it is demonstrated that our proposed method performed better in predicting lncRNA-protein interaction than other methods.

The Evaluation of Gait and Balance for Patients with Early Diabetic Peripheral Neuropathy: A Cross-Sectional Study.

Objective: Falls often occur in patients with diabetic neuropathy due to biomechanical alternation. The implication of diabetic peripheral neuropathy (DPN) on gait and balance remains poorly understood. Methods: A total of 11 dynamic gait, balance and electrophysiological parameters were evaluated in 176 participants. The biomechanical parameters were compared between groups. Results: Stride length and stride velocity were significantly lower in all subgroups of DPN compared with healthy subjects (p<0.05). Stance phase and double support phase were significantly higher, but swing phase were significantly lower across all subgroups of DPN than healthy subjects (p<0.05). Under eyes-open standing, the ML and AP range parameters of CoM sway, ankle sway and hip sway, CoM sway index, ankle swing index in both subclinical and confirmed DPN patients were all significantly higher compared to healthy subjects (p<0.05). Under eyes-closed standing, AP range parameters of CoM sway in subclinical DPN and confirmed DPN patients were significantly higher than healthy subjects (p<0.05). The hip sway areas in diabetics were significantly higher compared to healthy subjects (p<0.05). Conclusion: The abnormal biomechanical parameters existed in the early stages of patients with DPN. The static balance under eyes-open and eye-closed condition is maintained by ankle joint compensation strategy and hip joint protection strategy. An early evaluation and better risk management is essential for diabetic patients with a history of more than 5 years even without DPN clinical symptoms and signs. Clinical Trial Registration Number: No. ChiCTR1800019179, www.chictr.org.cn.

MD-MLI: Prediction of miRNA-lncRNA Interaction by Using Multiple Features and Hierarchical Deep Learning.

Long non-coding RNA(lncRNA) can interact with microRNA(miRNA) and play an important role in inhibiting or activating the expression of target genes and the occurrence and development of tumors. Accumulating studies focus on the prediction of miRNA-lncRNA interaction, and mostly are concerned with biological experiments and machine learning methods. These methods are found with long cycles, high costs, and requiring over much human intervention. In this paper, a data-driven hierarchical deep learning framework was proposed, which was composed of a capsule network, an independent recurrent neural network with attention mechanism and bi-directional long short-term memory network. This framework combines the advantages of different networks, uses multiple sequence-derived features of the original sequence and features of secondary structure to mine the dependency between features, and devotes to obtain better results. In the experiment, five-fold cross-validation was used to evaluate the performance of the model, and the zea mays data set was compared with the different model to obtain better classification effect. In addition, sorghum, brachypodium distachyon and bryophyte data sets were used to test the model, and the accuracy reached 0.9850, 0.9859 and 0.9777, respectively, which verified the model's good generalization ability.

Canagliflozin promotes osteoblastic MC3T3-E1 differentiation via AMPK/RUNX2 and improves bone microarchitecture in type 2 diabetic mice.

Individuals with type 2 diabetes mellitus (T2DM) have an increased risk of bone metabolic disorders and bone fracture due to disease progression and clinical treatment. The effect of sodium-glucose cotransporter 2 (SGLT2) inhibitors, now greatly prescribed for the treatment of T2DM, on bone metabolism is not clear. This study aimed to explore the possible influence of bone metabolic disorder and the underlying mechanism through a comparison of three different SGLT2 inhibitors (canagliflozin, dapagliflozin, and empagliflozin) in the treatment of type 2 diabetic mice. For the in vivo experiments, four groups (DM, DM+Cana, DM+Dapa, and DM+Empa) were established using micro-CT to detect the bone microarchitecture and bone-related parameters. The study results indicated that canagliflozin, but not dapagliflozin or empagliflozin, increased bone mineral density (p<0.05) and improved bone microarchitecture in type 2 diabetic mice. Furthermore, canagliflozin promoted osteoblast differentiation at a concentration of 5 µM under high glucose concentration (HG). Phosphorylated adenosine 5'-monophosphate (AMP)-activated protein kinase (AMPK) α (Thr172) has been confirmed to activate run-related transcription factor-2 (RUNX2) to perform this function. This effect can be partially reversed by the AMPK inhibitor dorsomorphin (compound C) and strengthened by the AMPK activator acadesine (AICAR) in vitro. The level trend of RUNX2 and p-AMPK in vivo were consistent with those in vitro. This study suggested that canagliflozin played a beneficial role in bone metabolism in type 2 diabetic mice compared with dapagliflozin and empagliflozin. It provides some theoretical support for the chosen drugs, especially for patients with osteoporosis or a high risk of fracture.