Extensive epigenetic reprogramming involves in memory CD8+ T-cell differentiation. The elaborate epigenetic rewiring underlying the heterogeneous functional states of CD8+ T cells remains hidden. Here, we profile single-cell chromatin accessibility and map enhancer-promoter interactomes to characterize the differentiation trajectory of memory CD8+ T cells. We reveal that under distinct epigenetic regulations, the early activated CD8+ T cells divergently originated for short-lived effector and memory precursor effector cells. We also uncover a defined epigenetic rewiring leading to the conversion from effector memory to central memory cells during memory formation. Additionally, we illustrate chromatin regulatory mechanisms underlying long-lasting versus transient transcription regulation during memory differentiation. Finally, we confirm the essential roles of Sox4 and Nrf2 in developing memory precursor effector and effector memory cells, respectively, and validate cell state-specific enhancers in regulating Il7r using CRISPR-Cas9. Our data pave the way for understanding the mechanism underlying epigenetic memory formation in CD8+ T-cell differentiation.
CD8-Positive T-Lymphocytes/metabolism* ; Cell Differentiation ; Chromatin/immunology* ; Animals ; Mice ; Immunologic Memory ; Epigenesis, Genetic ; SOXC Transcription Factors/immunology* ; NF-E2-Related Factor 2/immunology* ; Mice, Inbred C57BL ; Gene Regulatory Networks ; Enhancer Elements, Genetic

Objective To construct a machine learning prediction model for diabetic kidney disease(DKD)in type 2 diabetes mellitus(T2DM)patients in the plain-sand and loess hilly areas of Gansu Province,and analyze the interpretability of the model.Methods A multi-stage stratified random sampling method was used to collect the data of T2DM patients in the two areas.After key feature screening,eight ML prediction models were constructed for the risk of DKD in the two areas.The receiver operating characteristic(ROC)curve,accuracy and F1 index were used to evaluate the model,and Shapley additive explanation(SHAP)algorithm was used for model interpretation.Results A total of 1599 patients with T2DM were enrolled in this study.After feature screening,ten variables were selected for model construction in the plain-sand areas.Among the eight models,the gradient boosting decision tree(GBDT)model had the highest prediction efficiency.The area under the curve(AUC)of the test dataset was 0.972,the accuracy was 0.949,and the F1 index was 0.884.In the loess hilly region,12 variables were included in the model,and the best model was the random forest(RF).The AUC of the test set was 0.966,the accuracy was 0.951,and the F1 index was 0.861.SHAP analysis showed that in addition to serum creatinine,age,LDL-C,HbA1c,DM duration,serum uric acid and urinary microalbumin were also closely related to the high risk of DKD.Conclusions The GBDT and RF models have good predictive efficiency for the occurrence of DKD in the two areas,which can be used for the screening of DKD high-risk populations and the in-depth exploration of potential risk factors in the two areas.

Objective:To systematically review the prediction models for extubation failure in critically ill patients undergoing mechanical ventilation, providing a reference for healthcare professionals in selecting appropriate models to identify high-risk populations.Methods:Literature on the construction of prediction models for extubation failure risk in critically ill patients undergoing mechanical ventilation was retrieved from China National Knowledge Infrastructure, Wanfang Database, VIP, SinoMed, PubMed, Web of Science, Embase, and Cochrane Library. The search was limited from database inception to February 2024. Two researchers independently screened the literature and extracted data, using bias risk assessment tools to evaluate the bias risk and applicability of the prediction models.Results:A total of nine studies were included, with the most common predictive factors being mechanical ventilation duration, Glasgow Coma Scale score, cough reflex strength, age, and 24-hour input/output volume. The area under the receiver operating characteristic curve for the models ranged from 0.689 to 0.926, indicating good predictive performance. However, the risk of bias was high, mainly due to small sample sizes, the selection of predictive factors based on univariate analysis, and lack of proper internal validation.Conclusions:Existing prediction models show good predictive performance, but they carry high bias risk. Future studies should improve research design, adhere to model development and reporting guidelines, and develop well-performing, user-friendly prediction models to more accurately identify high-risk populations for extubation failure.

Objective To construct a machine learning prediction model for diabetic kidney disease(DKD)in type 2 diabetes mellitus(T2DM)patients in the plain-sand and loess hilly areas of Gansu Province,and analyze the interpretability of the model.Methods A multi-stage stratified random sampling method was used to collect the data of T2DM patients in the two areas.After key feature screening,eight ML prediction models were constructed for the risk of DKD in the two areas.The receiver operating characteristic(ROC)curve,accuracy and F1 index were used to evaluate the model,and Shapley additive explanation(SHAP)algorithm was used for model interpretation.Results A total of 1599 patients with T2DM were enrolled in this study.After feature screening,ten variables were selected for model construction in the plain-sand areas.Among the eight models,the gradient boosting decision tree(GBDT)model had the highest prediction efficiency.The area under the curve(AUC)of the test dataset was 0.972,the accuracy was 0.949,and the F1 index was 0.884.In the loess hilly region,12 variables were included in the model,and the best model was the random forest(RF).The AUC of the test set was 0.966,the accuracy was 0.951,and the F1 index was 0.861.SHAP analysis showed that in addition to serum creatinine,age,LDL-C,HbA1c,DM duration,serum uric acid and urinary microalbumin were also closely related to the high risk of DKD.Conclusions The GBDT and RF models have good predictive efficiency for the occurrence of DKD in the two areas,which can be used for the screening of DKD high-risk populations and the in-depth exploration of potential risk factors in the two areas.

Objective:To systematically review the prediction models for extubation failure in critically ill patients undergoing mechanical ventilation, providing a reference for healthcare professionals in selecting appropriate models to identify high-risk populations.Methods:Literature on the construction of prediction models for extubation failure risk in critically ill patients undergoing mechanical ventilation was retrieved from China National Knowledge Infrastructure, Wanfang Database, VIP, SinoMed, PubMed, Web of Science, Embase, and Cochrane Library. The search was limited from database inception to February 2024. Two researchers independently screened the literature and extracted data, using bias risk assessment tools to evaluate the bias risk and applicability of the prediction models.Results:A total of nine studies were included, with the most common predictive factors being mechanical ventilation duration, Glasgow Coma Scale score, cough reflex strength, age, and 24-hour input/output volume. The area under the receiver operating characteristic curve for the models ranged from 0.689 to 0.926, indicating good predictive performance. However, the risk of bias was high, mainly due to small sample sizes, the selection of predictive factors based on univariate analysis, and lack of proper internal validation.Conclusions:Existing prediction models show good predictive performance, but they carry high bias risk. Future studies should improve research design, adhere to model development and reporting guidelines, and develop well-performing, user-friendly prediction models to more accurately identify high-risk populations for extubation failure.

Objective To investigate the possible mechanism of action of Jinqi Qingshu granules(JQQSG)in the treatment of community-acquired pneumonia(CAP)by network pharmacology and molecular docking technology.Methods The TCMSP database and SwissTargetPrediction database were used to obtain and screen the active ingredients and targets of JQQSG,and GeneCards,OMIM,TTD,and DisGeNET databases were used to search for the predicted targets of CAP,and the two targets were mapped and then imported into STRING database to construct a PPI network to screen the key targets,and then the GO and KEGG pathway enrichment were analyzed by the DAVID database,and molecular docking was carried out by the AutoDock Tools software.Results 209 active ingredients and 1 041 targets of JQQSG were obtained after screening;312 targets were co-activated with CAP,and 64 core targets were obtained after PPI network screening.571 biological processes,68 cellular components,and 199 molecular functions were analyzed by GO enrichment,and 165 KEGG pathways were analyzed by KEGG pathway enrichment,mainly involved in protein action,apoptosis and MAPK signaling pathway.Molecular docking suggests that the core target and the core components all have good binding ability.Conclusion The mechanism of action of JQQSG in the treatment of CAP may be related to its regulation of Akt,MAPK signaling pathway,improvement of oxidative stress,and other pathways to exert anti-inflammatory and antioxidant effects,which could lay the foundation for further in-depth study of its specific mechanism of action.