Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

OBJECTIVE: To preliminarily investigate the effects and mechanism of action of Yishen Yangsui Formula (, YSYSF)on the recovery of neurological function in rats with degenerative cervical myelopathy. METHODS: Fifty adult SD female rats were randomly divided into control group, sham group, model group, YSYSF group and positive drug group by using randomized numerical table method. In the model group, YSYSF group and positive drug group, polyvinyl alcohol acrylamide interpenetrating network hydrogel(water-absorbent swelling material) was used to construct a rat spinal cord chronic compression model. The sham group was implanted with the water-absorbent swelling material and then removed without causing spinal cord compression. The control group, the sham group and the model group were given equal amounts of saline by gavage, the group of YSYSF was given Chinese herbal medicine soup by gavage 9.1 g·kg^-1 once a day, and the positive drug group was given tetrahexylsalicylglucoside sodium monosialate ganglioside by intraperitoneal injection 4.2 mg·kg^-1 once a day. The motor function of the rats was assessed by the BBB method after 1, 3, 7, and 14 d of drug administration. The spinal cord tissues were taken from rats executed 14 d after drug administration, and the morphological changes of the spinal cord compression site were observed by HE staining, and the expression levels of Caspase-1, GSDMD, NLRP3, PYCARD, IL-1β, and IL-18 were detected in the area of spinal cord injury by Western blot method. RESULTS: The BBB scores of the control group and the sham group were normal at all time points after modeling, which were higher than the BBB scores of the model group, the YSYSF, and the positive drug group (P<0.05). From the 3rd day after gavage, at all time points, the BBB scores of rats in the YSYSF group and the positive drug group were higher than those of rats in the model group (P<0.05). The staining pattern of HE spinal cord tissue was normal in the control group and the sham group, and the HE spinal cord in the model group was severely damaged with a large number of neuron deaths, whereas the damage to the spinal cord and neuron cells was reduced in the YSYSF group and the positive drug group. The expression levels of caspase-1, GSDMD, NLRP3, PYCARD, IL-1β and IL-18 in the spinal cord of the model group were significantly higher than those of the sham group (P<0.0001), and the expression levels of caspase-1, GSDMD, NLRP3, PYCARD, IL-1β, and IL-18 in the YSYSF group and the drug group were significantly lower than those in the model group (P<0.05). CONCLUSION YSYSF can improve the motor function of rats with degenerative cervical spinal cord disease, alleviate the pathological changes, and promote the recovery of spinal cord neurological function. The specific mechanism may be related to the inhibition of the activation of inflammatory vesicles NLRP3 and PYCARD, the reduction of the release of inflammatory factors IL-1β and IL-18, the reduction of the expression of caspase-1 and GSDMD, the reduction of cellular death, and the inhibition of inflammatory response.
Animals ; Female ; Drugs, Chinese Herbal/administration & dosage* ; Rats ; Rats, Sprague-Dawley ; Pyroptosis/drug effects* ; Spinal Cord/pathology* ; NLR Family, Pyrin Domain-Containing 3 Protein ; Spinal Cord Diseases/drug therapy* ; Interleukin-1beta/metabolism*

This study aims to establish BHK-21 stable cell lines expressing APN from four species(human,pig,dog,and cat),the APN fragments were amplified from pEGFP-C1-APN plasmids of the four species stored in the laboratory to generate the recombinant plasmids pcDNA4.0-APN.Af-ter the recombinant plasmids were transfected into BHK-21 cells,the stable BHK-21 cell lines ex-pressing the APNs were selected by two rounds of limited dilution.The constructed BHK-21 cell lines were identified by indirect immunofluorescence assay(IFA),and their susceptibility to PD-CoV and TGEV was tested for these four cell lines.Virus infection experiments revealed that PD-CoV infected cells expressing human,pig,and dog APNs,while it did not infect cells expressing cat APN.Simultaneously,TGEV infected cells expressing pig,dog,and cat APNs,but did not infect cells expressing human APN.The results suggest that the risk of cross-species infection for different coronaviruses and the established cell line can be used effectively to evaluate the virus in-fection.The findings also revealed that PDCoV has the potential risk of cross-species infection of human and dog,and TGEV has the potential risk of cross-species infection of dog and cat.These results provide a basis for the prevention and control strategy of coronaviruses.

Aim To mine the competing ceRNA net-works associated with programmed cell death in the pathophysiological mechanisms of atherosclerosis(AS)based on bioinformatics,in order to identify new targets for the diagnosis and treatment of AS.Methods Firstly,the GSE97210 and GSE28858 datasets were screened from the GEO database.Differentially ex-pressed lncRNA,mRNA and miRNA were identified,following which a IncRNA-miRNA-mRNA regulatory network was constructed in Cytoscape 3.7.2 software based on ceRNA theory.Second,GO and KEGG en-richment analysis of mRNA in the ceRNA network was performed.Finally,the mRNAS within the ceRNA net-work were compared with genes related to autophagy,pyroptosis and ferroptosis to establish a ceRNA network related to programmed cell death.Results A total of 1208 DElncRNAS,4723 DEmRNAS and 139 DEmiR-NAS were identified.A ceRNA network was estab-lished,comprising 64 lncRNAS,8 miRNAS and 167 mRNAS.The mRNAS within the CeRNA network were mainly enriched in biological processes such as positive regulation of transcription and migration,protein bind-ing,and signaling pathways including PI3K-Akt signa-ling pathway,and mTOR signaling pathway.Finally,this study established 7 lncRNA-mediated ceRNA regu-latory pathways associated with pyroptosis and 23 ln-cRNA-mediated regulatory pathways for ferroptosis and autophagy.Conclusion This study has successfully constructed a ceRNA network related to programmed cell death,which helps us understand the mechanism by which programmed cell death leads to AS.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Osteoporosis(OP) is a senile bone disease characterized by an imbalance between bone remodeling and bone formation. Targeting pathogenesis of kidney deficiency, spleen deficiency, and blood stasis, Bushen Jianpi Huoxue Decoction has a significant effect on the treatment of OP by tonifying kidney, invigorating spleen, and activating blood circulation. MicroRNA(miRNA) and the anti-apoptotic protein B-cell lymphoma-2-like protein 1(BCL2L1) are closely related to bone cell metabolism. Therefore, in this study, the binding of miR-140-5p to BCL2L1 was detected by dual luciferase assay and polymerase chain reaction(PCR). After silencing or overexpressing miR-140-5p, the apoptosis, autophagy, and osteogenic function of human fetal osteoblast cell line 1.19(HFOB1.19) were observed by flow cytometry and Western blot. Bushen Jianpi Huoxue Decoction-containing serum was prepared by intragastric administration of Bushen Jianpi Huoxue Decoction in rats. Different concentrations of Bushen Jianpi Huoxue Decoction-containing serum were used to treat HFOB1.19 with or without miR-140-5p mimic. The expression of osteogenic proteins in each group was observed, and the role of miR-140-5p/BCL2L1 in apoptosis and autophagy of HFOB1.19 was studied, along with the effect of Bushen Jianpi Huoxue Decoction on these processes. As indicated by the dual luciferase assay, miR-140-5p bound to BCL2L1. Flow cytometry and Western blot showed that miR-140-5p promoted apoptosis and inhibited autophagy in HFOB1.19. After intervention with high, medium, and low doses of Bushen Jianpi Huoxue Decoction-medicated serum, compared with the miR-140-5p NC group, the expression of osteocalcin(OCN), osteopontin(OPN), Runt-related transcription factor 2(RUNX2), and transforming growth factor beta 1(TGF-β1) decreased in the miR-140-5p mimic group, while the expression of bone morphogenetic protein 2(BMP2) showed no significant difference under high-dose intervention. Therefore, miR-140-5p/BCL2L1 can promote apoptosis and inhibit autophagy in HFOB1.19. Bushen Jianpi Huoxue Decoction can affect the osteogenic effect of miR-140-5p through BMP2.
MicroRNAs/metabolism* ; Autophagy/drug effects* ; Apoptosis/drug effects* ; Humans ; Drugs, Chinese Herbal/administration & dosage* ; Animals ; Cell Line ; bcl-X Protein/metabolism* ; Osteoblasts/metabolism* ; Rats ; Osteoporosis/physiopathology* ; Male ; Rats, Sprague-Dawley ; Osteogenesis/drug effects*

Objective To construct an aptamer-functionalized carboxylated graphene oxide(CGO)fluo-rescent sensor to achieve highly sensitive and specific detection of ketamine(KET)and its metabolite norketamine(NK)using an aptamer capable of simultaneously recognizing KET and NK.Methods A specific aptamer for simultaneous recognition of KET and NK was screened using graphene oxide-sys-tematic evolution of ligand by exponential enrichment(GO-SELEX)and molecular docking tech-niques.The aptamer,labeled with Cy5 fluorescence,was chemically conjugated to CGO to construct an aptamer-functionalized CGO fluorescent sensor.By optimizing detection conditions,including the mass concentration of CGO,aptamer concentration,reaction temperature,and incubation time,quantita-tive analysis of the target analytes was achieved using the ratio of fluorescence intensity changes be-fore and after target addition.The stability of the sensor in biological matrices was evaluated by moni-toring fluorescence intensity changes over incubation time in blank blood and urine,in comparison with the traditional physical adsorption-based CGO fluorescent sensor.Spiked recovery experiments in blank blood and urine were conducted to compare performance with that of HPLC-MS/MS.Results A specific aptamer A5 was selected and chemically conjugated with CGO to construct the aptamer-functionalized CGO fluorescent sensor.Under optimized conditions,the proposed fluorescent sensor ex-hibited a linear detection range of 1.0-5.0 ng/mL for KET,with a limit of detection(LOD)of 0.86 ng/mL;while for NK,the linear detection range was 1.0-5.0 ng/mL,with an LOD of 0.70 ng/mL.Com-pared with the CGO fluorescent sensor constructed via physical adsorption,this sensor demonstrated greater stability in blood and urine.The spiked recovery rates of KET and NK in blank blood and urine ranged from 81.50%to 110.03%,exhibiting detection performance comparable to that of HPLC-MS/MS.Conclusion The aptamer screening method offers a novel approach for selecting aptamers tar-geting drugs and their metabolites.The constructed aptamer-functionalized CGO fluorescent sensor pro-vides an efficient and reliable strategy for the high-performance detection of KET and NK.

Lung cancer has the highest incidence and mortality among malignant tumors in China.Persistent subsolid nodules(SSNs)are closely associated with early-stage lung adenocarcinoma.Artificial intelligence(AI),as an emerging technology,is capable of performing in-depth analysis of large-scale imaging data through autonomous learning and possesses the ability to predict outcomes from new data,demonstrating great potential and application prospects in the assessment of SSNs.AI can not only effectively assist radiologists in diagnosis and treatment,but also improve work efficiency while reducing misdiagnosis and missed diagnosis rates.This review summarizes the recent applications and research progress of AI in the assessment of SSNs,to provide new insights for the diagnosis and treatment of SSNs.