Molecular Subtypes and Machine Learning-Based Predictive Models for Intracranial Aneurysm Rupture.

BACKGROUND: The determination of biological mechanisms and biomarkers related to intracranial aneurysm (IA) rupture is of utmost significance for the development of effective preventive and therapeutic strategies in the clinical field. METHODS: GSE122897 and GSE13353 datasets were downloaded from Gene Expression Omnibus. Data extracted from GSE122897 were used for analyzing differential gene expression, and consensus clustering was performed to identify stable molecular subtypes. Clinical characteristics were compared between subgroups, and fast gene set enrichment analysis and weighted gene coexpression network analysis were performed. Hub genes were identified via least absolute shrinkage and selection operator analysis. Predictive models were constructed based on hub genes using the Light Gradient Boosting Machine, eXtreme Gradient Boosting, and logistic regression algorithm. Immune cell infiltration in IA samples was analyzed using Microenvironment Cell Population counter, CIBERSORT, and xCell algorithm. The correlation between hub genes and immune cells was analyzed. The predictive model and immune cell infiltration were validated using data from the GSE13353 dataset. RESULTS: A total of 43 IA samples were classified into 2 subgroups based on gene expression profiles. Subgroup I had a higher risk of rupture, while 70% of subgroup II remained unruptured. In subgroup I, specific genes were associated with inflammation and immunity, and weighted gene coexpression network analysis revealed that the black module genes were linked to IA rupture. We identified 4 hub genes (spermine synthase, macrophage receptor with collagenous structure, zymogen granule protein 16B, and LIM and calponin-homology domains 1), which constructed predictive models with good diagnostic performance in differentiating between ruptured and unruptured IA samples. Monocytic lineage was found to be a significant factor in IA rupture, and the 4 hub genes were linked to monocytic lineage (P < 0.05). CONCLUSIONS: We reveal a new molecular subtype that can reflect the actual pathological state of IA rupture, and our predictive models constructed by machine learning algorithms can efficiently predict IA rupture.

Hemolysis associated with Russula subnigricans ingestion in a patient with glucose-6-phosphate dehydrogenase deficiency.

INTRODUCTION: Russula subnigricans is now one of the leading lethal mushroom species in China, with a mortality rate of more than 50%. The typical clinical manifestation of Russula subnigricans poisoning is rhabdomyolysis, and we are unaware of previous reports of Russula subnigricans-associated hemolysis. CASE SERIES: Herein we report a cluster of five patients with confirmed Russula subnigricans poisoning. Four of the patients who ingested sun-dried Russula subnigricans never developed rhabdomyolysis. However, in one patient, acute hemolysis developed on the second day following ingestion and was associated with a fall in hemoglobin concentration and a rise in unconjugated bilirubin concentration. Further investigation revealed that the patient had glucose-6-phosphate dehydrogenase deficiency. CONCLUSION: This case cluster suggests that the toxin of Russula subnigricans could cause hemolysis in a susceptible patient and warrants further study.

Association of Dietary intake of vitamin E with chronic obstructive pulmonary disease events in US adults: A cross-sectional study of NHANES 2013-2018.

Introduction: Several studies have demonstrated that vitamin E intake is negatively associated with the development of several diseases, but the relationship between vitamin E intake and COPD in different groups of people is not clear. The aim was to investigate the relationship between vitamin E intake and COPD in different groups of people. Methods: This study used data from NHANES (National Health and Nutrition Examination Survey) from 2013-2018. A final total of 4,706 participants were included, univariate versus multivariate logistic regression and restricted cubic spline models adjusted for confounders were used to explore the relationship between vitamin E intake and COPD, and subgroup analyses were conducted to assess whether there are differences in the relationship between vitamin E intake and COPD in different populations or conditions. Results: After adjusting for potential confounders, higher vitamin E intake showed a significant negative association with COPD [Model 1(unadjusted covariates, OR = 0.48;95% CI:0.33-0.70; p < 0.001), Model 2(adjusted for age, sex, and race, OR = 0.48;95% CI:0.31-0.73; p < 0.01), and Model 3(adjusted for all covariates, OR = 0.57;95% CI:0.36-0.91; p = 0.02)]. And a restricted cubic spline curve showed a significant negative correlation between vitamin E intake and COPD (p for nonlinear = 0.2036). In the subgroup analysis, we found a negative association between vitamin E intake and COPD in all subgroups as well. Conclusion: After analyzing data based on the NHANES database from 2013-2018, the results showed that vitamin E intake among U.S. adults was well below the recommended levels and that higher vitamin E intake was negatively associated with COPD incidence.

Intersession Instability in fNIRS-Based Emotion Recognition.

Emotion recognition based on neural signals is a promising technique for the detection of patients' emotions for enhancing healthcare. However, emotion-related neural signals, such as from functional near infrared spectroscopy (fNIRS), can be affected by various psychophysiological and environmental factors. There is a paucity of literature regarding data instability and classification instability in fNIRS-based emotion recognition systems, phenomenon which may lead to user dissatisfaction and abandonment. We collected data in an fNIRS-based 2-class emotion recognition test-retest experiment (3 week interval) with visual stimuli emotion induction to examine data instability and its impact on classification accuracy. We found a 22.2% average deterioration of emotion classification accuracy between the two sessions, suggesting that classification instability is a serious problem. We found that the changes in the distributions of the selected neural signal features, as evaluated by Kullback-Leibler (KL) divergence, were a likely cause of the accuracy decline. We analyzed the data instability and our results showed that instability of spatial activation patterns and instability of the hemodynamic response in the most activated region are correlated with accuracy decline. Finally, we propose a method for mitigating classification instability in fNIRS-based emotion recognition based on feature selection for stable features, the first such method to our knowledge. This new feature selection criterion considers not only the separability of features (evaluated by Fisher Score) but also their stability over time (evaluated by KL divergence between feature distributions at different time points). Testing showed that this method led to an approximately 5% improvement in cross-session generalization accuracy.

A novel fractal approach for predicting G-protein-coupled receptors and their subfamilies with support vector machines.

G-protein-coupled receptors (GPCRs) are seven membrane-spanning proteins and regulate many important physiological processes, such as vision, neurotransmission, immune response and so on. GPCRs-related pathways are the targets of a large number of marketed drugs. Therefore, the design of a reliable computational model for predicting GPCRs from amino acid sequence has long been a significant biomedical problem. Chaos game representation (CGR) reveals the fractal patterns hidden in protein sequences, and then fractal dimension (FD) is an important feature of these highly irregular geometries with concise mathematical expression. Here, in order to extract important features from GPCR protein sequences, CGR algorithm, fractal dimension and amino acid composition (AAC) are employed to formulate the numerical features of protein samples. Four groups of features are considered, and each group is evaluated by support vector machine (SVM) and 10-fold cross-validation test. To test the performance of the present method, a new non-redundant dataset was built based on latest GPCRDB database. Comparing the results of numerical experiments, the group of combined features with AAC and FD gets the best result, the accuracy is 99.22% and Matthew's correlation coefficient (MCC) is 0.9845 for identifying GPCRs from non-GPCRs. Moreover, if it is classified as a GPCR, it will be further put into the second level, which will classify a GPCR into one of the five main subfamilies. At this level, the group of combined features with AAC and FD also gets best accuracy 85.73%. Finally, the proposed predictor is also compared with existing methods and shows better performances.