13856 research outputs found
Sort by
Causal mediation analysis: A summary-data mendelian randomization approach
No full text[[abstract]]data Mendelian randomization (MR), a widely used approach in causal inference, has recently attracted attention for improving causal mediation analysis. Two existing methods corresponding to the difference method and product method of linear mediation analysis have been developed to perform MR-based mediation analysis using the inverse-variance weighted method (MR-IVW). Despite these developments, there is still a need for more rigorous, efficient, and precise MR-based mediation methodologies. In this study, we develop summary-data MR-based frameworks for causal mediation analysis. We improve the accuracy, statistical efficiency and robustness of the existing MR-based mediation analysis by implementing novel variance estimators for the mediation effects, deriving rigorous procedures for statistical inference, and accounting for widespread pleiotropic effects. Specifically, we propose Diff-IVW and Prod-IVW to improve upon the existing methods and provide the pleiotropy-robust methods (Diff-Egger, Diff-Median, Prod-Egger, and Prod-Median), adapted from MR-Egger and MR-Median, to enhance the robustness of the MR-based mediation analysis. We conduct comprehensive simulation studies to compare the existing and proposed methods. The results show that the proposed methods, Diff-IVW and Prod-IVW, improve statistical efficiency and type I error control over the existing approaches. Although all IVW-based methods suffer from directional pleiotropy biases, the median-based methods (Diff-Median and Prod-Median) can mitigate such biases. The differences among the methods can lead to discrepant statistical conclusions as demonstrated in real data applications. Based on our simulation results, we recommend the three proposed methods in practice: Diff-IVW, Prod-IVW, and Prod-Median, which are complementary under various scenarios
E3 ligase TRIM8 suppresses lung cancer metastasis by targeting MYOF degradation through K48-linked polyubiquitination
No full text[[abstract]]Ubiquitination is a posttranslational modification that regulates tumour progression-associated proteins through the ubiquitin-proteasome system, making E3 ligases potential antitumour targets. Here, we report that TRIM8, a member of the TRIM family and an E3 ligase, can act as a tumour suppressor in non-small cell lung cancer (NSCLC). Both gain- and loss-of-function experiments revealed that TRIM8 inhibits the proliferation, colony formation, migration and invasion of NSCLC cells. Experiments with a xenograft model showed that TRIM8 expression suppresses tumour metastasis in vivo. Moreover, low expression of TRIM8 was associated with poor overall survival in both the Taiwanese and GEO lung cancer cohorts. TRIM8 overexpression in lung cancer cells reduced MYOF expression, and restoring MYOF rescued cell migration in TRIM8-overexpressing cells. TRIM8 targeted MYOF for K48-linked ubiquitination, facilitating proteasome-mediated degradation and subsequently suppressing the extracellular secretion of MMPs. Our results provide new insights into the contribution of TRIM8 to lung cancer progression, suggesting that TRIM8 is a new biomarker and a novel therapeutic target for lung cancer
Predicting NSAID-associated cardiac mortality: A deep learning approach to 12-lead ECG analysis
No full text[[abstract]]Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used for pain relief, but their consumption is associated with a risk of myocardial infarction, heart failure, and increased mortality. In this study, we introduce a deep learning model that uses electrocardiograms (ECGs) data acquired before NSAID administration to identify patients at higher risk of cardiovascular mortality triggered by the use of NSAIDs. This model aims to predict potential cardiovascular deaths within five years of NSAID usage, facilitating proactive measures to reduce cardiovascular risks. A key feature of our approach is the spatial-and-channel attention block (SCA), which expands the multiscale receptive field of the feature map and enhances the identification of critical ECG signals across spatial and channel dimensions. Our results demonstrate that our model outperforms several existing benchmarks, achieving an area under the receiver operating characteristic curve (AUROC) of 0.825. Additionally, we investigate the advantages of applying transfer learning to further improve the model's performance. By utilizing a model trained on sinus rhythm data as the source model for transfer learning, our method exhibits an increased AUROC of 0.846. A thorough analysis of our approach is presented, suggesting that the proposed model could be instrumental in revising pain medication prescription protocols
NUPR1 contributes to endocrine therapy resistance by modulating BIRC5 expression and inducing luminal B-ERBB2<SUP>+</SUP> subtype-like characteristics in estrogen breast cancer cells
No full text[[abstract]]Acquired resistance to endocrine therapy is a major clinical challenge in the treatment of luminal A [estrogen receptor (ER)(+)and/or progesterone receptor (PR)+, human epidermal growth factor receptor 2 (ERBB2/HER2)(-), and low Ki-67] breast cancer. Recently, molecular subtype conversion has been suggested as one of the possible causes of the development of drug-resistant breast cancer. However, the molecular mechanism underlying the molecular subtype conversion and the induction of endocrine therapy resistance in luminal A breast cancer is still incompletely understood. Here, we found that the ER+ MCF7-derived endocrine therapy-resistant MCF7-TamC3 breast cancer cells exhibit increased expression of an intrinsically disordered chromatin protein, NUPR1, compared to the parental luminal-A subtype like MCF7 breast cancer cells. Intriguingly, MCF7-TamC3 cells also exhibit characteristics that resemble the luminal B-ERBB2(+) breast tumor subtype, like the increased expression of ERBB2 and the increased sensitivity to monoclonal ERBB2-targeting antibody Trastuzumab in vitro. Kaplan-Meier analysis of expression cohorts of breast tumors showed that high NUPR1 mRNA expression levels correlate with poor overall and relapse-free survival in both endocrine therapy-treated ER+ and ERBB2-enriched breast cancer patients. Results of the bioinformatics analysis showed that the NUPR1 mRNA expression level is also correlated with the clinical grading of the Tamoxifen-treated ER+ primary breast cancer. The qPCR and the western blot analysis results revealed that NUPR1 positively regulates the expression of the epigenetic regulator HDAC5, the anti-apoptotic molecule BIRC5, and the mitogenic receptor ERBB2 in MCF7-TamC3 and the ERBB2-enriched subtype like SK-BR-3 breast cancer cells. Downregulation of NUPR1 increased the sensitivity to estrogen deprivation in MCF7-TamC3 cells and decreased the viability of SK-BR-3 cells in vitro. These findings indicate that dysregulation of NUPR1 promotes the development of estrogen independence in ER+ breast cancer cells in part through expression regulation of HDAC5, ERBB2, and BIRC5. Targeting NUPR1 or its downstream regulating molecules may offer a potential strategy for overcoming resistance to endocrine therapy in patients with ER+ breast cancer
Is history of cardiovascular disease associated with increased caries experience among Taiwanese adults?
No full text[[abstract]]INTRODUCTION: This study examined the association between cardiovascular disease (CVD) history and their dental caries experience status. METHODS: Conducted from January 2021 to June 2023, this cross-sectional cohort study involved 7,138 participants who underwent oral examinations. Data on demographic background, oral health-related behaviors, and smoking status were collected using a structured questionnaire. Dental caries was diagnosed at the cavitation level according to the World Health Organization criteria and calculated into caries experience indices including decayed, missing, and filled teeth (DMFT), decayed teeth, missing teeth and filled teeth. Information on CVD history was obtained from the Taiwan National Health Insurance Research Database, including acute myocardial infarction, ischemic stroke, and coronary artery disease. Multivariate linear regression models were used to assess the association between CVD history and its dental caries experience status. RESULTS: Of the participants, 158 (2.2%) had a prior diagnosis of CVD. Participants with CVD history had a significantly higher mean DMFT index (21.21 ± 8.37) than did those without CVD history (13.4 ± 7.82; p < 0.0001). After adjusting for confounding factors, participants with CVD history had a mean DMFT index that was 2.11 higher (95% CI = 0.99, 3.24, p < 0.01) and 2.21 more missing teeth (95% CI = 1.42, 3.00, p < 0.0001) than did those without CVD history. Subgroup analyses indicated that participants aged ≥65 years were predominantly affected. CONCLUSION: Older participants with CVD history were associated with an increased number of missing teeth. The present study design could not conclude a positive association between CVD history and its DMFT status, partly due to the lack of data on the reason for missing teeth
Paternal age, de novo mutation, and age at onset among co-affected schizophrenia sib-pairs: Whole-genome sequencing in multiplex families
No full text[[abstract]]Whether delaying fatherhood leads to more mutations, thereby resulting in adverse psychiatric outcomes in offspring, remains under debate. No study has directly examined the role of de novo mutations (DNMs) between paternal age and offspring psychiatric outcomes. This study aimed to explore the association between paternal age, the number of DNMs, and age at onset of schizophrenia by sequencing the whole genome of multiplex schizophrenia families. Whole-genome sequencing (30x) was performed in 5 Taiwanese families, each comprising 3 co-affected siblings and healthy parents. Causal mediation analyses were used to explore the mediating role of DNMs in the paternal age effect. Paternal age predicted increased DNMs (+1.50 DNMs/year, 95% CI: 0.81, 2.19, p < 0.0001) over maternal age (+0.09 DNMs/year, 95% CI: -1.01, 1.19, p = 0.87). The effect of paternal age on the number of DNMs varied across families. Each additional DNM resulted in a 0.16-year earlier onset age of schizophrenia (95% CI: 0.04, 0.27, p = 0.009). The estimated direct effect of paternal age on the onset of schizophrenia was -0.82 (95% CI: -0.90, -0.73), while the indirect effect through DNMs was -0.32 (95% CI: -0.47, -0.17). The proportion mediated via DNMs was 28.04% (95% CI: 18.19%, 37.89%). The mediation analyses showed that 30% of the observed association of paternal age with onset age of schizophrenia might be mediated through paternal age-related DNMs. Our study, the first to directly quantify the mediating effect of DNMs, provides support for a causal role of paternal age-related mutations in the increased psychiatric risk in offspring
[[alternative]]Suppression of NF-κB and downstream XBP1 by DcR3 contributes to a decrease in antibody secretion
No full text[[abstract]]Decoy receptor 3 (DcR3), a soluble receptor in the tumor necrosis factor receptor superfamily, regulates the functions of monocytes, macrophages, dendritic cells, and T cells. Previous studies have demonstrated that DcR3 suppresses B cell proliferation in vitro and ameliorates autoimmune diseases in animal models; however, whether and how DcR3 regulates antibody production is unclear. Using a DcR3 transgenic mouse model, we found that DcR3 impaired the T cell–dependent antigen-stimulated antibody response. The number of Ag-specific antibody-secreting cells was transiently reduced, but the concentration of specific antibodies continued to decrease in the DcR3 transgenic mice, implying a direct suppression of antibody production by DcR3. In vitro assays showed that the DcR3-Fc fusion protein attenuated T cell–dependent induced antibody production and reduced the expression of secretory Igh and Xbp1. We found that nuclear factor κB (NF-κB) activity was essential for the expression of Xbp1 in activated B cells. DcR3-Fc attenuated anti-CD40-induced NF-κB activity and Xbp1 promoter activity. Furthermore, DcR3-Fc decreased the expression of Xbp1 in Blimp1þ antibody-secreting cells. Restoration of spliced XBP1 (X-box binding protein 1) in DcR3-treated B cells increased the secretory Ighg1 transcript levels, suggesting that reducing XBP1 is one of the mechanisms by which DcR3 regulates antibody production both in vitro and in vivo. Collectively, these results indicate that in addition to blocking proliferation, DcR3 impairs NF-κB activation, subsequently decreasing the expression of Xbp1, eventually leading to a reduction in antibody secretion
A comparative study of GPT3.5 fine tuning and rule-based approaches for de-identification and normalization of sensitive health information in electronic medical record notes
No full text[[abstract]]Electronic Medical Records (EMR) systems enhance medical care efficiency through the consolidation of data analysis, bolstering patient safety, and diminishing storage expenses. Nevertheless, EMR text notes have the potential to reveal sensitive personal information, necessitating the implementation of robust security measures to protect sensitive information. Furthermore, the representation of temporal data in EMR text notes differs among institutions, which has a direct effect on the precision and dependability of time data analysis. In light of this, the AI CUP 2023-Privacy Protection and Standardization of Electronic Medical Record competition released a dataset annotated with sensitive health information (SHI) and normalized temporal information. We participated in the challenge and conducted a comparative study to assess the efficacy of two approaches, the fine-tuning of a large language model based on the chat generative pre-trained transformer (ChatGPT) and the rule-based approach, in enhancing the privacy of clinical texts. For the ChatGPT-based approach, we fine-tuned the gpt-3.5-turbo-1106 model to efficiently identify labels. In contrast, in the rule-based approach, we compiled several rules along with dictionaries collected from the internet and released datasets for recognizing and normalizing SHIs. The ChatGPT-based approach achieved macro-F-scores of 0.752 and 0.799 for SHI recognition and temporal information normalization, while the rule-based approach highlighted 0.866 and 0.869 for the two respective subtasks, exhibiting lower latency and power consumption. Both strategies exhibited their respective benefits in tackling privacy concerns. The implementations of both rule-based and ChatGPT-based methods can be accessed through the following GitHub repositories: https://github.com/zhao-rui-NB/worker_intelligence_ner (rule-based method) and https://github.com/Chou-po-chen/openai_ner (ChatGPT-based method)
Mitochondrial fission in insulin-like growth factor binding protein 3-mediated radiosensitivity of oral cancer cells
No full text[[abstract]]We have previously demonstrated that ectopic expression of insulin-like growth factor-binding protein 3 (IGFBP3) enhanced the radiosensitivity of oral squamous cell carcinoma (OSCC) cells. We observed changes in mitochondrial morphology in IGFBP3-expressing OSCC cells after ionizing radiation (IR) treatment; however, the roles of IGFBP3 in mitochondrial dynamics still remain unclear. The analyses of cell survival, reactive oxygen species production, and loss of mitochondrial membrane potential were used to verify the IGFBP3-mediated radiosensitivity. Using mitochondria by 3D cell explorer and quantification of mitochondria dynamics by immunofluorescence staining, the fission and fusion mitochondria in irradiated IGFBP3-expressing cells and IGFBP3knockdown cells were increased, respectively. By western blot, we found phosphorylated dynamin-related protein 1 (DRP1)positively associated with IGFBP3 levels in irradiated cells. By transiently expressing DRP1 in IGFBP3-expressing cells, we found that IGFBP3-mediated radiosensitivity and mitochondrial fission were flattened by DRP1 inhibition, suggesting the role of DRP1 inIGFBP3-mediated mitochondrial fission
The age effect upon fnirs measured brain function between schizophrenia patients and normal controls
No full text[[abstract]]Background: Schizophrenia has been considered as a brain disorder which deteriorates cognitive and social function in affected people. Whether the aging effect will further decline the brain function of schizophrenic patients or not is rarely discussed before. Aims & Objectives: We here applied fNIRS (functional infrared spectrometry) to measure brain function of schizophrenia to tell the difference from normal with age effect adjustment. Method: We recruited schizophrenia people and normal controls with different age from 2015 to 2018 with IRB approval in one medical center in central Taiwan. After consent form completed, their brain function performance was measured by using Hitachi ETG 4000 with both centroid and integral data over frontal and bilateral temporal regions channels once. The standard VFT (verbal fluency test) Taiwan version was tested to record the change of oxygenated and deoxygenated hemoglobin reflection curve during the task. Results: The sample size is 87 for schizophrenic patients and 88 for normal controls. The average integral value of frontal function is lower in people with schizophrenia compared to normal controls, and between group difference is decreasing while people getting older (Fig 1). The average centroid value of frontal function is higher in people with schizophrenia over normal one at any age stage, and the average group gap is increasing by years of aging (Fig 2). Discussion & Conclusion: The brain function by measuring integral value showed the total working performance during VFT, which showed a greater slope of decline in normal people compared to schizophrenia. However, the reaction lag time after task start during VFT by observing centroid value is prolonged among old people with schizophrenia compared to normal control ones, it might reveal a worse response on cognitive performance for older schizophrenic patients at the same age level. In general, the brain function measured by fNIRS showed a worse outcome in people with schizophrenia compared to normal controls with compatible age level