To create and verify a machine learning model that integrates social determinants of health (SDoH) for assessing post-stroke depression (PSD) and examining the association between SDoH and disease outcomes. Data were acquired from the National Health and Nutrition Examination Survey. Logistic regression was employed to analyse the association between SDoH and PSD, whereas Cox regression was utilized to assess the correlation between SDoH and all-cause mortality in PSD. The Boruta algorithm was employed for feature selection, and four machine learning models were constructed (CatBoost, Logistic, Multilayer Perceptron, and Random Forest) to evaluate the predictive effectiveness, calibration, and clinical applicability of these ML models. SHAP values were computed to ascertain the predictive significance of each feature in the model that exhibited the highest predictive performance. Logistic regression analysis revealed a significant positive correlation between SDoH and PSD prevalence(p for trend < 0.0001). Compared to the other three models, CatBoost (AUC = 0.966) demonstrated the best overall predictive performance. Moreover, the decision curve analysis (DCA) and calibration curve findings demonstrated that the CatBoost model possessed considerable clinical utility and consistent predictive efficacy. The ten-fold cross-validation method further confirmed the model’s robustness and generalization ability. A linear relationship exists between SDoH and PSD, with CatBoost demonstrating the best performance in predicting PSD. SHAP values emphasize the importance of SDoH.

Background: Substantial evidence suggests that immunoproteasome is implicated in the various neurological diseases such as stroke, multiple sclerosis and neurodegenerative diseases. However, whether the immunoproteasome itself deficiency causes brain disease is still unclear. Therefore, the aim of this study was to explore the contribution of the immunoproteasome subunit low molecular weight protein 2 (LMP2) in neurobehavioral functions. Methods: Male LMP2 gene completed knockout (LMP2-KO) and littermate wild type (WT) Sprague–Dawley (SD) rats aged 12-month-old were used for neurobehavioral testing and detection of proteins expression by western blotting and immunofluorescence. A battery of neurobehavioral test tools including Morris water maze (MWM), open field maze, elevated plus maze were used to evaluate the neurobehavioral changes in rats. Evans blue (EB) assay, Luxol fast blue (LFB) and Dihydroethidium (DHE) staining were applied to explore the blood–brain barrier (BBB) integrity, brain myelin damage and brain intracellular reactive oxygen species (ROS) levels, respectively. Results: We firstly found that LMP2 gene deletion did not cause significantly difference in rats’ daily feeding activity, growth and development as well as blood routine, but it led to metabolic abnormalities including higher levels of low-density lipoprotein cholesterol, uric acid and blood glucose in the LMP2-KO rats. Compared with the WT rats, LMP2-KO rats displayed obviously cognitive impairment and decreased exploratory activities, increased anxiety-like behavior and without strong effects on gross locomotor abilities. Furthermore, multiple myelin loss, increased BBB leakage, downregulation of tight junction proteins ZO-1, claudin-5 and occluding, and enhanced amyloid-β protein deposition were observed in brain regions of LMP2-KO rats. In addition, LMP2 deficiency significantly enhanced oxidative stress with elevated levels of ROS, caused the reactivation of astrocytes and microglials and markedly upregulated protein expression levels of interleukin (IL)-1 receptor-associated kinase 1 (IRAK1), IL-6 and tumor necrosis factor-α (TNF-α) compared to the WT rats, respectively. Conclusion: These findings highlight LMP2 gene global deletion causes significant neurobehavioral dysfunctions. All these factors including metabolic abnormalities, multiple myelin loss, elevated levels of ROS, increased BBB leakage and enhanced amyloid-β protein deposition maybe work together and eventually led to chronic oxidative stress and neuroinflammation response in the brain regions of LMP2-KO rats, which contributed to the initial and progress of cognitive impairment.

Background: Immunity and neuroinflammation play crucial roles in the pathogenesis of Parkinson’s disease (PD). Nonetheless, prior investigations into the correlation between immune inflammation and PD have produced varying results. Identifying specific immune cell phenotypes that are truly associated with PD is challenging, and the causal relationship between immune cells and PD remains elusive.Methods: This study conducted a comprehensive two-sample Mendelian randomization (MR) analysis, employing five distinct analytical approaches, to clarify the causal connection between immune cell characteristics and the risk of PD. Utilizing GWAS data, we investigated the causal relationship between 731 immune cell traits and PD. These immune cell phenotypes encompass absolute cell (AC) counts, median fluorescence intensity (MFI), and relative cell (RC) counts for B cells, cDCs, mature stage T cells, monocytes, myeloid cells, TBNK (T cells, B cells, and natural killer cells), and Tregs, as well as the logistic parameter (MP) for cDCs and TBNK.Results: The inverse variance weighted (IVW) analysis indicated that Myeloid DCs (p = 0.004), HVEM expression on CD45RA− CD4+ T cells (p = 0.007), CD62L− CD86+ Myeloid DCs (p = 0.015), and HLA DR expression on monocytes (p = 0.019) were associated with a reduced risk of PD. CD14+ CD16+ monocytes (p = 0.005), HLA DR+ NK cells within CD3− lymphocytes (p = 0.023), and CD28 expression on activated & secreting Tregs (p = 0.032) were associated with an increased risk of PD.Conclusion: This study establishes a causal link between immune cell phenotype and the pathogenesis of PD, identifying several specific immune cell characteristics associated with PD. This could inspire researchers to delve into the pathogenesis of PD at the cellular subtype level, and aid in the identification of potential pharmacological protein targets for PD.

Background: The epidemiological association between frailty and insomnia is well established, yet the presence of a common genetic etiology is still uncertain. Further exploration is needed to ascertain the causal relationship between frailty and insomnia.Methods: Utilizing data obtained from genome-wide association studies (GWAS) summaries, we utilized the linkage disequilibrium score regression (LDSC) to determine the genetic correlation existing between frailty and insomnia. The determination of causality was achieved through the application of two-sample Mendelian randomization. We investigated the enrichment of single nucleotide polymorphism (SNP) at various tissue types utilizing stratified LD score regression (S-LDSC) and multimarker analysis of genome annotation (MAGMA). Common risk SNPs were identified using Multi-Trait Analysis of GWAS (MTAG) and Cross-Phenotype Association (CPASSOC). We further investigated the expression profiles of risk genes in tissues using Summary-data-based Mendelian randomization(SMR) based on pooled data, to explore potential functional genes.Results: Our findings indicated a significant genetic correlation between frailty and insomnia, highlighting SNPs sharing risk (rs34290943, rs10865954), with a pronounced correlation in the localized genomic region 3p21.31. Partitioned genetic analysis revealed 24 functional elements significantly associated with both frailty and insomnia. Furthermore, mendelian randomization revealed a causal connection between frailty and insomnia. The genetic correlation between frailty and insomnia showed enrichment in 11 brain regions (S-LDSC) and 9 brain regions (MAGMA), where four functional genes (RMB6, MST1R, RF123, and FAM212A) were identified.Conclusion: This study suggests the existence of a genetic correlation and common risk genes between frailty and insomnia, contributing to a deeper comprehension of their pathogenesis and assists in identifying potential therapeutic targets.

Objective The optimal management of a small intracranial aneurysm (sIA) remains a challenge due to the lack of a size-specific risk predictive model for aneurysm rupture. We aimed to develop and validate a nomogram-based risk predictive model for sIA. Methods A total of 382 patients harboring 215 ruptured and 167 unruptured small intracranial aneurysms (uSIAs) (≤ 7 mm) were recruited and divided into training and validation cohorts. Risk factors for the construction of a nomogram were selected from clinical and aneurysmal features by least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression. The nomogram for risk of rupture was evaluated in both the training and validation cohorts for discrimination, calibration, and clinical usefulness. Results Hyperlipidemia (odds ratio (OR)=2.74, 95% confidence interval (CI)=1.322~5.956, P=0.008), the presence of a daughter dome (OR=3.068, 95%CI=1.311~7.598, P=0.012), larger size-to-neck ratio (SN) (OR=1.807, 95%CI=1.131~3.063, P=0.021) and size ratio (SR) (OR=2.221, 95%CI=1.262~4.025, P=0.007) were selected as independent risk factors for sIA rupture and used for construction of nomogram. Internal validation by bootstrap sampling showed the Concordance index (C index) of 0.756 for the nomogram. The calibration by the Hosmer-Lemeshow test showed a P value of 0.847, indicating the model was well-fitted. Additionally, decision curve analysis (DCA) demonstrated that the predictive model has good clinical usefulness, providing net benefits across a range of threshold probabilities, thus supporting its application in clinical decision-making. Conclusion The risk prediction model can reliably predict the risk of sIA rupture, which may provide an important reference for optimizing the therapeutic strategy.