BACKGROUND:Studies have shown that rheumatoid arthritis and osteoporosis are positively correlated,but the causal relationship and related mechanisms have not yet been confirmed.With the cross-fertilization of computer science and life sciences,Mendelian randomization and bioinformatics analyses based on genome-wide association study(GWAS)and transcriptome sequencing data can assess the causal relationship between two diseases,explore the related mechanisms,and mine the therapeutic targets,which will be beneficial to the precision treatment of rheumatoid arthritis combined with osteoporosis.OBJECTIVE:To explore the causal relationship between rheumatoid arthritis and osteoporosis using two-sample Mendelian randomization and to mine potential co-morbid targets and potential targeted drugs through summary-data-based Mendelian randomization and bioinformatics analyses,aiming to provide theoretical basis for mechanism exploration and precision treatment in the field of rheumatoid arthritis combined with osteoporosis.METHODS:(1)Firstly,GWAS data of rheumatoid arthritis,osteoporosis,and cis-expression quantitative trait locus(cis-eQTL)in Asian and European populations were downloaded from the GWAS Catalog,IEU Open GWAS,FinnGen,and eQTLGen databases,and were used for two-sample Mendelian randomization analysis and summary-data-based Mendelian randomization analysis.(2)Transcriptome sequencing data of rheumatoid arthritis(GSE93272 and GSE15573)were downloaded from the GEO database for bioinformatics analysis.(3)Subsequently,forward and inverse Mendelian randomization analyses between rheumatoid arthritis and osteoporosis were performed,and inverse variance weighted was used as the main metric for the analyses,and the results were corroborated with MR Egger,simple mode,weighted median and weighted mode.(4)Then,the genes closely related to rheumatoid arthritis and osteoporosis were identified based on the summary-data-based Mendelian randomization analysis,and the co-disease targets of rheumatoid arthritis and osteoporosis were mined based on cross-analysis.Meanwhile,the biological functions of the co-morbid targets were verified based on bioinformatics analysis and cellular experiments.(5)In addition,a rheumatoid arthritis risk prediction nomogram was constructed based on DYRK2,and its prediction performance was verified by receiver operating characteristic curve,correction curve and decision curve.Finally,the target potential drugs were mined based on Enrichr database and molecular docking was performed.RESULTS AND CONCLUSION:(1)Forward Mendelian randomization analysis of rheumatoid arthritis and osteoporosis showed statistically significant results except for GCST90044540 and GCST90086118,and all other results indicated a significant causal relationship and positive correlation between rheumatoid arthritis and osteoporosis.(2)Inverse Mendelian randomization analysis suggested that no significant causal relationship was seen between osteoporosis and rheumatoid arthritis.(3)Summary-data-based Mendelian randomization analysis identified a total of 412 and 344 genes positively associated with rheumatoid arthritis and osteoporosis,and 421 and 347 genes negatively associated.Based on the cross-analysis,26 co-morbid genes were subsequently obtained.Among them,DYRK2 was a potential therapeutic target,and subsequent bioinformatics analysis and cellular experiments confirmed its important role in the progression of rheumatoid arthritis and osteoporosis.(4)Furthermore,the constructed nomogram has excellent predictive performance.Finally,four potential DYRK2-targeting drugs(undecanoic acid,metyrapone,JNJ-38877605,and ACA)were discovered and molecular docking also demonstrated reliable targeting ability.(5)In conclusion,based on GWAS data from Asian and European populations,we successfully demonstrated that rheumatoid arthritis and osteoporosis are causally related at the genetic level,DYRK2 is a potential therapeutic target,and four small molecules are potential target drugs.

BACKGROUND:Studies have shown that rheumatoid arthritis and osteoporosis are positively correlated,but the causal relationship and related mechanisms have not yet been confirmed.With the cross-fertilization of computer science and life sciences,Mendelian randomization and bioinformatics analyses based on genome-wide association study(GWAS)and transcriptome sequencing data can assess the causal relationship between two diseases,explore the related mechanisms,and mine the therapeutic targets,which will be beneficial to the precision treatment of rheumatoid arthritis combined with osteoporosis.OBJECTIVE:To explore the causal relationship between rheumatoid arthritis and osteoporosis using two-sample Mendelian randomization and to mine potential co-morbid targets and potential targeted drugs through summary-data-based Mendelian randomization and bioinformatics analyses,aiming to provide theoretical basis for mechanism exploration and precision treatment in the field of rheumatoid arthritis combined with osteoporosis.METHODS:(1)Firstly,GWAS data of rheumatoid arthritis,osteoporosis,and cis-expression quantitative trait locus(cis-eQTL)in Asian and European populations were downloaded from the GWAS Catalog,IEU Open GWAS,FinnGen,and eQTLGen databases,and were used for two-sample Mendelian randomization analysis and summary-data-based Mendelian randomization analysis.(2)Transcriptome sequencing data of rheumatoid arthritis(GSE93272 and GSE15573)were downloaded from the GEO database for bioinformatics analysis.(3)Subsequently,forward and inverse Mendelian randomization analyses between rheumatoid arthritis and osteoporosis were performed,and inverse variance weighted was used as the main metric for the analyses,and the results were corroborated with MR Egger,simple mode,weighted median and weighted mode.(4)Then,the genes closely related to rheumatoid arthritis and osteoporosis were identified based on the summary-data-based Mendelian randomization analysis,and the co-disease targets of rheumatoid arthritis and osteoporosis were mined based on cross-analysis.Meanwhile,the biological functions of the co-morbid targets were verified based on bioinformatics analysis and cellular experiments.(5)In addition,a rheumatoid arthritis risk prediction nomogram was constructed based on DYRK2,and its prediction performance was verified by receiver operating characteristic curve,correction curve and decision curve.Finally,the target potential drugs were mined based on Enrichr database and molecular docking was performed.RESULTS AND CONCLUSION:(1)Forward Mendelian randomization analysis of rheumatoid arthritis and osteoporosis showed statistically significant results except for GCST90044540 and GCST90086118,and all other results indicated a significant causal relationship and positive correlation between rheumatoid arthritis and osteoporosis.(2)Inverse Mendelian randomization analysis suggested that no significant causal relationship was seen between osteoporosis and rheumatoid arthritis.(3)Summary-data-based Mendelian randomization analysis identified a total of 412 and 344 genes positively associated with rheumatoid arthritis and osteoporosis,and 421 and 347 genes negatively associated.Based on the cross-analysis,26 co-morbid genes were subsequently obtained.Among them,DYRK2 was a potential therapeutic target,and subsequent bioinformatics analysis and cellular experiments confirmed its important role in the progression of rheumatoid arthritis and osteoporosis.(4)Furthermore,the constructed nomogram has excellent predictive performance.Finally,four potential DYRK2-targeting drugs(undecanoic acid,metyrapone,JNJ-38877605,and ACA)were discovered and molecular docking also demonstrated reliable targeting ability.(5)In conclusion,based on GWAS data from Asian and European populations,we successfully demonstrated that rheumatoid arthritis and osteoporosis are causally related at the genetic level,DYRK2 is a potential therapeutic target,and four small molecules are potential target drugs.

A highly efficient oxidase-mimetic silver phosphate/nickel hydroxystannate(Ag3PO4/NiSn(OH)6)composite was synthesized via a precipitation method using nickel hydroxystannate(NiSn(OH)6)as the support.The abundant hydroxyl groups(—OH)on NiSn(OH)6 not only provided nucleation sites for Ag3PO4 nanoparticles but also improved their dispersion and overall material stability.Based on oxidase-like activity of Ag3PO4/NiSn(OH)6 and inhibitory effect of sodium dodecylbenzenesulfonate(SDBS)on this catalytic activity,a novel colorimetric sensing method for SDBS detection was developed.Under optimized experimental conditions,the method exhibited a linear range of 3.69-42.7 μmol/L,with a detection limit of 0.135 μmol/L(S/N=3).The regression equation was ΔA652=0.01125C(μmol/L)+0.1498,with a correlation coefficient(R2)of 0.992.Practical application in dishwashing liquid analysis achieved satisfactory recoveries of 96.9%-106.4%,demonstrating the method's reliability for real sample detection.

Objective To observe the depressive behavior and neuronal damage induced by different doses of corticosterone(CORT)in mice,and to explore the optimal dose for a corticosterone-induced depression model in mice.Methods Forty male C57BL/6J mice were divided randomly into four groups:control group and low,medium,and high CORT groups(20,40,and 60 mg/kg,respectively),treated with the corresponding drug dose by subcutaneous injection for 4 weeks.Behavioral c hanges in mice after corticosterone administration for 3 and 4 weeks were detected by sugar water preference,forced swimming,tail suspension,and open field tests.Morphological changes in neurons in the hippocampal CA1 area and forebrain cortex area were observed by hematoxylin-eosin(HE)and Nissl staining.Serum levels of 5-hydroxytryptamine(5-HT)were detected by enzyme-linked immunosorbent assay.Depression-related behavioral changes induced by different doses of corticosterone were compared.Results The bodyweights of mice in all three CORT groups(20,40,and 60 mg/kg)decreased(P＜0.05)and the preference for sucrose solution decreased(P＜0.01)compared with the findings in the control group.The immobility time in the forced swimming test was prolonged in the CORT 20 and 40 mg/kg groups(P＜0.01)and the immobility time of mice in the tail suspension test was prolonged in the CORT 40 mg/kg group(P＜0.05).The total distance,the length of time spent in the peripheral area was prolonged and the time entering the central area in the open-field experiment were decreased in the CORT 40 and 60 mg/kg groups(P＜0.05),and average speed were decreased in the CORT 40 mg/kg group(P＜0.05).In addition,CORT injection result ed in abnormal neuronal cell morphology,cell deformation,and nuclear condensation in the hippocampal CA1 and forebrain cortex areas,to different degrees.Serum 5-hydroxytryptamine levels were reduced in the CORT 40 and 60 mg/kg groups(P＜0.05).Conclusions CORT 20,40,and 60 mg/kg can induce depression-like behavioral changes and neuronal damage in mice to varying degrees,with the most notable effect at 40 mg/kg.Under experimental conditions,we consider that 40 mg/kg is the best dose for replicating corticosterone-induced depression in model mice.

Addressing the enduring challenge of evaluating traditional Chinese medicines (TCMs), the integrated evidence chain-based effectiveness evaluation of TCMs (Eff-iEC) has emerged. This paper explored its capacity through a demonstration study that evaluated the effectiveness evidence of six commonly used anti-hepatic fibrosis Chinese patent medicines (CPMs), including Biejiajian Pill (BP), Dahuang Zhechong Pill (DZP), Biejia Ruangan Compound (BRC), Fuzheng Huayu Capsule (FHC), Anluo Huaxian Pill (AHP), and Heluo Shugan Capsule (HSC), using both Eff-iEC and the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system. The recognition of these CPMs within the TCM academic community was also assessed through their inclusion in relevant medical documents. Results showed that the evidence of BRC and FHC received higher assessments in both Eff-iEC and GRADE system, while the assessments for others varied. Analysis of community recognition revealed that Eff-iEC more accurately reflects the clinical value of these CPMs, exhibiting superior evaluative capabilities. By breaking through the conventional pattern of TCMs effectiveness evaluation, Eff-iEC offers a novel epistemology that better aligns with the clinical realities and reasoning of TCMs, providing a coherent methodology for clinical decision-making, new drug evaluations, and health policy formulation.

Objective:To predict mechanical complications (MC) following spinal deformity surgery for adult spine deformity (ASD) using machine learning models, identify key risk factors, and develop a visualizable tool for individualized risk assessment.Methods:Clinical and radiological data from 525 patients with ASD who underwent surgery in our hospital between January 2017 and December 2021 were collected. Patients were randomly assigned to a training set (70%) and a test set (30%) for model development. The cohort included 88 males and 437 females, with a mean age of 42.2±18.1 years. Variables included demographic data, comorbidities, local and systemic radiological parameters, paraspinal muscle fat infiltration (FI), and vertebral bone quality (VBQ) scores. Multiple machine learning algorithms: Random Forest (RF), Gaussian Naive Bayes (GNB), Light GBM, Support Vector Machine (SVM), XGBoost (XGB), and Logistic Regression (LR) were trained and evaluated. Model performance was compared using the receiver operating characteristic curve (ROC) and precision-recall curve (PRC). SHAP (Shapley Additive Explanations) was used to rank risk factors, while LIME (Local Interpretable Model-Agnostic Explanations) was applied to visualize MC risk in individual cases.Results:Of the 525 patients, 135 (25.7%) developed postoperative MC. Among these, 80 (59.3%) experienced proximal junction kyphosis or failure (PJK/PJF), 7 (5.2%) had distal junction kyphosis or failure (DJK/DJF), 28 (20.7%) sustained rod fractures, and 29 (21.5%) showed significant loss of correction. In the validation cohort, the RF model achieved the highest area under the curve (AUC=0.80), followed by GNB (0.77), XGB (0.76), LR (0.74), LightGBM (0.73), and SVM (0.66). The RF model also demonstrated the best PRC value (0.58), highest sensitivity (0.65), and lowest Brier score (0.20). GNB, Light GBM, and LR models achieved the highest accuracy (0.78 each), while LightGBM exhibited the highest specificity (0.93). SHAP analysis identified higher preoperative VBQ scores, larger T 1 pelvic angle (TPA), and higher paraspinal muscle FI as the main risk factors for MC. Based on the RF model, a LIME-based tool was successfully constructed for individualized MC risk estimation. Conclusion:The RF model demonstrated the best overall predictive performance for MC. A machine learning-based prediction model has the potential to provide valuable guidance for surgical decision-making in ASD patients.

Objective To propose a lead attention network-based ST-T change recognition algorithm to detect ECG ST-T changes accurately.Methods Firstly,heartbeat signals were extracted through R-wave localization,and a 12-lead heartbeat matrix was generated by correlation-based screening and merging to realize data augmentation.Secondly,a lead attention module was constructed by combining depthwise convolution(DWConv)with the channel attention squeeze-and-excitation block(SE-block)structure to perceive the differences in ST-T status among electrocardiogram leads.Thirdly,the mapping output by two independent attention modules was fused and splicing with the original signal residual was carried out,so that attention information extraction and original information transfer were enhanced effectively.Finally,SE-ResNet was used as the backbone network to extract signal features to complete the classification and identification of ST-T changes.To validate the recognition performance of the proposed algorithm for ST-T changes in ECG,the 12-lead ECG data of 97 472 patients containing different ECG rhythms were collected for ablation and comparison experiments at the First Affiliated Hospital of Army Medical University.Results The proposed algorithm achieved an AUC of 0.965 with a sensitivity of 90.51%,specificity of 90.23%,positive predictive value of 89.24%and overall accuracy of 90.36%on an independent test set.Comparative analysis demonstrated superior performance to four benchmark architectures,including VGG16,ResNet18,MobileNetV3-Small and ShuffleNet,in terms of both classification accuracy and computational efficiency.Conclusion The algorithm designed can accurately detect ST-T changes and can be used for wearable ECG automatic analysis to assist in the early warning of cardiovascular diseases in both acute and chronic patients and highland residents.[Chinese Medical Equipment Journal,2025,46(7):1-11]

The innate immune system can be boosted in response to subsequent triggers by pre-exposure to microbes or microbial products, known as “trained immunity”. Compared to classical immune memory, innate trained immunity has several different features. Firstly, the molecules involved in trained immunity differ from those involved in classical immune memory. Innate trained immunity mainly involves innate immune cells (e.g., myeloid immune cells, natural killer cells, innate lymphoid cells) and their effector molecules (e.g., pattern recognition receptor (PRR), various cytokines), as well as some kinds of non-immune cells (e.g., microglial cells). Secondly, the increased responsiveness to secondary stimuli during innate trained immunity is not specific to a particular pathogen, but influences epigenetic reprogramming in the cell through signaling pathways, leading to the sustained changes in genes transcriptional process, which ultimately affects cellular physiology without permanent genetic changes (e.g., mutations or recombination). Finally, innate trained immunity relies on an altered functional state of innate immune cells that could persist for weeks to months after initial stimulus removal. An appropriate inducer could induce trained immunity in innate lymphocytes, such as exogenous stimulants (including vaccines) and endogenous stimulants, which was firstly discovered in bone marrow derived immune cells. However, mature bone marrow derived immune cells are short-lived cells, that may not be able to transmit memory phenotypes to their offspring and provide long-term protection. Therefore, trained immunity is more likely to be relied on long-lived cells, such as epithelial stem cells, mesenchymal stromal cells and non-immune cells such as fibroblasts. Epigenetic reprogramming is one of the key molecular mechanisms that induces trained immunity, including DNA modifications, non-coding RNAs, histone modifications and chromatin remodeling. In addition to epigenetic reprogramming, different cellular metabolic pathways are involved in the regulation of innate trained immunity, including aerobic glycolysis, glutamine catabolism, cholesterol metabolism and fatty acid synthesis, through a series of intracellular cascade responses triggered by the recognition of PRR specific ligands. In the view of evolutionary, trained immunity is beneficial in enhancing protection against secondary infections with an induction in the evolutionary protective process against infections. Therefore, innate trained immunity plays an important role in therapy against diseases such as tumors and infections, which has signature therapeutic effects in these diseases. In organ transplantation, trained immunity has been associated with acute rejection, which prolongs the survival of allografts. However, trained immunity is not always protective but pathological in some cases, and dysregulated trained immunity contributes to the development of inflammatory and autoimmune diseases. Trained immunity provides a novel form of immune memory, but when inappropriately activated, may lead to an attack on tissues, causing autoinflammation. In autoimmune diseases such as rheumatoid arthritis and atherosclerosis, trained immunity may lead to enhance inflammation and tissue lesion in diseased regions. In Alzheimer’s disease and Parkinson’s disease, trained immunity may lead to over-activation of microglial cells, triggering neuroinflammation even nerve injury. This paper summarizes the basis and mechanisms of innate trained immunity, including the different cell types involved, the impacts on diseases and the effects as a therapeutic strategy to provide novel ideas for different diseases.

Objective:To predict mechanical complications (MC) following spinal deformity surgery for adult spine deformity (ASD) using machine learning models, identify key risk factors, and develop a visualizable tool for individualized risk assessment.Methods:Clinical and radiological data from 525 patients with ASD who underwent surgery in our hospital between January 2017 and December 2021 were collected. Patients were randomly assigned to a training set (70%) and a test set (30%) for model development. The cohort included 88 males and 437 females, with a mean age of 42.2±18.1 years. Variables included demographic data, comorbidities, local and systemic radiological parameters, paraspinal muscle fat infiltration (FI), and vertebral bone quality (VBQ) scores. Multiple machine learning algorithms: Random Forest (RF), Gaussian Naive Bayes (GNB), Light GBM, Support Vector Machine (SVM), XGBoost (XGB), and Logistic Regression (LR) were trained and evaluated. Model performance was compared using the receiver operating characteristic curve (ROC) and precision-recall curve (PRC). SHAP (Shapley Additive Explanations) was used to rank risk factors, while LIME (Local Interpretable Model-Agnostic Explanations) was applied to visualize MC risk in individual cases.Results:Of the 525 patients, 135 (25.7%) developed postoperative MC. Among these, 80 (59.3%) experienced proximal junction kyphosis or failure (PJK/PJF), 7 (5.2%) had distal junction kyphosis or failure (DJK/DJF), 28 (20.7%) sustained rod fractures, and 29 (21.5%) showed significant loss of correction. In the validation cohort, the RF model achieved the highest area under the curve (AUC=0.80), followed by GNB (0.77), XGB (0.76), LR (0.74), LightGBM (0.73), and SVM (0.66). The RF model also demonstrated the best PRC value (0.58), highest sensitivity (0.65), and lowest Brier score (0.20). GNB, Light GBM, and LR models achieved the highest accuracy (0.78 each), while LightGBM exhibited the highest specificity (0.93). SHAP analysis identified higher preoperative VBQ scores, larger T 1 pelvic angle (TPA), and higher paraspinal muscle FI as the main risk factors for MC. Based on the RF model, a LIME-based tool was successfully constructed for individualized MC risk estimation. Conclusion:The RF model demonstrated the best overall predictive performance for MC. A machine learning-based prediction model has the potential to provide valuable guidance for surgical decision-making in ASD patients.