Predictive model and risk analysis for coronary heart disease in people living with HIV using machine learning.

OBJECTIVE: This study aimed to construct a coronary heart disease (CHD) risk-prediction model in people living with human immunodeficiency virus (PLHIV) with the help of machine learning (ML) per electronic medical records (EMRs). METHODS: Sixty-one medical characteristics (including demography information, laboratory measurements, and complicating disease) readily available from EMRs were retained for clinical analysis. These characteristics further aided the development of prediction models by using seven ML algorithms [light gradient-boosting machine (LightGBM), support vector machine (SVM), eXtreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), decision tree, multilayer perceptron (MLP), and logistic regression]. The performance of this model was assessed using the area under the receiver operating characteristic curve (AUC). Shapley additive explanation (SHAP) was further applied to interpret the findings of the best-performing model. RESULTS: The LightGBM model exhibited the highest AUC (0.849; 95% CI, 0.814-0.883). Additionally, the SHAP plot per the LightGBM depicted that age, heart failure, hypertension, glucose, serum creatinine, indirect bilirubin, serum uric acid, and amylase can help identify PLHIV who were at a high or low risk of developing CHD. CONCLUSION: This study developed a CHD risk prediction model for PLHIV utilizing ML techniques and EMR data. The LightGBM model exhibited improved comprehensive performance and thus had higher reliability in assessing the risk predictors of CHD. Hence, it can potentially facilitate the development of clinical management techniques for PLHIV care in the era of EMRs.

Identification of ferroptosis related genes and pathways in prostate cancer cells under erastin exposure.

BACKGROUND: Few studies are focusing on the mechanism of erastin acts on prostate cancer (PCa) cells, and essential ferroptosis-related genes (FRGs) that can be PCa therapeutic targets are rarely known. METHODS: In this study, in vitro assays were performed and RNA-sequencing was used to measure the expression of differentially expressed genes (DEGs) in erastin-induced PCa cells. A series of bioinformatic analyses were applied to analyze the pathways and DEGs. RESULTS: Erastin inhibited the expression of SLC7A11 and cell survivability in LNCaP and PC3 cells. After treatment with erastin, the concentrations of malondialdehyde (MDA) and Fe2+ significantly increased, whereas the glutathione (GSH) and the oxidized glutathione (GSSG) significantly decreased in both cells. A total of 295 overlapping DEGs were identified under erastin exposure and significantly enriched in several pathways, including DNA replication and cell cycle. The percentage of LNCaP and PC3 cells in G1 phase was markedly increased in response to erastin treatment. For four hub FRGs, TMEFF2 was higher in PCa tissue and the expression levels of NRXN3, CLU, and UNC5B were lower in PCa tissue. The expression levels of SLC7A11 and cell survivability were inhibited after the knockdown of TMEFF2 in androgen-dependent cell lines (LNCaP and VCaP) but not in androgen-independent cell lines (PC3 and C4-2). The concentration of Fe2+ only significantly increased in TMEFF2 downregulated LNCaP and VCaP cells. CONCLUSION: TMEFF2 might be likely to develop into a potential ferroptosis target in PCa and this study extends our understanding of the molecular mechanism involved in erastin-affected PCa cells.

HERVK-mediated regulation of neighboring genes: implications for breast cancer prognosis.

Human endogenous retroviruses (HERVs) are the remnants of ancient retroviral infections integrated into the human genome. Although most HERVs are silenced or rendered inactive by various regulatory mechanisms, they retain the potential to influence the nearby genes. We analyzed the regulatory map of 91 HERV-Ks on neighboring genes in human breast cancer and investigated the impact of HERV-Ks on the tumor microenvironment (TME) and prognosis of breast cancer. Nine RNA-seq datasets were obtained from GEO and NCBI SRA. Differentially expressed genes and HERV-Ks were analyzed using DESeq2. Validation of high-risk prognostic candidate genes using TCGA data. These included Overall survival (multivariate Cox regression model), immune infiltration analysis (TIMER), tumor mutation burden (maftools), and drug sensitivity analysis (GSCA). A total of 88 candidate genes related to breast cancer prognosis were screened, of which CD48, SLAMF7, SLAMF1, IGLL1, IGHA1, and LRRC8A were key genes. Functionally, these six key genes were significantly enriched in some immune function-related pathways, which may be associated with poor prognosis for breast cancer (p = 0.00016), and the expression levels of these genes were significantly correlated with the sensitivity of breast cancer treatment-related drugs. Mechanistically, they may influence breast cancer development by modulating the infiltration of various immune cells into the TME. We further experimentally validated these genes to confirm the results obtained from bioinformatics analysis. This study represents the first report on the regulatory potential of HERV-K in the neighboring breast cancer genome. We identified three key HERV-Ks and five neighboring genes that hold promise as novel targets for future interventions and treatments for breast cancer.

Investigating Causal Relations between Genetic-Related Intermediate Endophenotype and Risk of Chronic Prostatitis: Mendelian Randomization Study.

Objective: Prostatitis is a common disease of the male genitourinary system, which seriously disturbs the physical and mental health of male patients. It is related to many factors such as living habits, age, and race, but the etiology has not been fully elucidated. This study investigated whether there is a causal relationship between clinical biochemical indicators (i.e., intermediate phenotype) and prostatitis through Mendelian randomization. The subjects of the study were prostatitis patients and related SNPs in the Guangxi Fangchenggang health examination cohort. Methods: According to the requirements of Mendelian randomization (MR), the single nucleotide polymorphisms (SNPs) related to prostatitis patients and 29 common SNPs related to clinical biochemical indicators were analyzed by linkage disequilibrium, and the calculated SNPs were selected. Finally, the related SNPs were analyzed by Mendelian randomization method. Results: 15 biochemical indicators such as complement C4, FOL, CRP, HCY, and estradiol have shared chronic prostatitis SNP sites, and five qualified SNPs were finally screened for complement C4. Finally, complement C4 was obtained by Mendelian randomization method (P = 0.039), which was statistically significant. The other 28 clinical endophenotypes were all negative. Conclusion: The results show that there was a causal relationship between complement C4 and prostatitis, and the more consistent SNP is rs2075799.