Singular Value Decomposition-Based Penalized Multinomial Regression for Classifying Imbalanced Medulloblastoma Subgroups Using Methylation Data.

Medulloblastoma (MB) is a molecularly heterogeneous brain malignancy with large differences in clinical presentation. According to genomic studies, there are at least four distinct molecular subgroups of MB: sonic hedgehog (SHH), wingless/INT (WNT), Group 3, and Group 4. The treatment and outcomes depend on appropriate classification. It is difficult for the classification algorithms to identify these subgroups from an imbalanced MB genomic data set, where the distribution of samples among the MB subgroups may not be equal. To overcome this problem, we used singular value decomposition (SVD) and group lasso techniques to find DNA methylation probe features that maximize the separation between the different imbalanced MB subgroups. We used multinomial regression as a classification method to classify the four different molecular subgroups of MB using the reduced DNA methylation data. Coordinate descent is used to solve our loss function associated with the group lasso, which promotes sparsity. By using SVD, we were able to reduce the 321,174 probe features to just 200 features. Less than 40 features were successfully selected after applying the group lasso, which we then used as predictors for our classification models. Our proposed method achieved an average overall accuracy of 99% based on fivefold cross-validation technique. Our approach produces improved classification performance compared with the state-of-the-art methods for classifying MB molecular subgroups.

C-reactive protein is associated with low-density lipoprotein cholesterol and obesity in type 2 diabetic Sudanese.

BACKGROUND: Type 2 diabetes is emerging in Sudan and is associated with obesity. Deregulated lipid metabolism and inflammatory states are suggested risk factors for cardiovascular disease, which is a leading cause of diabetic death. This study aimed to investigate C-reactive protein (CRP) levels and the lipid profile in type 2 diabetic adult Sudanese compared with nondiabetics, and to test their associations with other characteristics. METHODS: A cross-sectional study including 70 diabetics and 40 nondiabetics was conducted. Anthropometric measurements were assessed, and demographic and medical data were obtained using a structured questionnaire. Blood specimens were collected and biochemical parameters were analyzed applying standard methods. RESULTS: CRP and triglycerides were significantly higher in the diabetic group (P<0.001 and P=0.01, respectively). Differences in total cholesterol, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were not statistically significant between the diabetic and nondiabetic groups. In the diabetic group, correlation analysis revealed that the CRP level had a significant positive correlation with LDL-C (r=0.255, P=0.034) and body mass index (r=0.29, P=0.016). Body mass index showed a significant positive correlation with triglycerides (r=0.386, P=0.001). Within the lipid parameters, a number of significant correlations were observed. Elevated levels of CRP, LDL-C, and triglycerides were markedly more prevalent in the diabetic group of patients. Diabetics showed significantly higher CRP levels compared with nondiabetics (odds ratio 5.56, P=0.001). CONCLUSION: The high prevalence of obesity among diabetics, together with elevated levels of triglycerides and CRP, suggest coexistence of dyslipidemia and inflammation in diabetes. Our findings emphasize that diabetics were 5.6 times more likely to have high CRP levels than nondiabetics; as CRP is a predictor of cardiovascular disease risk, it can be recognized that diabetics are at more risk of cardiovascular disease than nondiabetics. Considering evaluation of CRP together with the lipid profile in prediction of cardiovascular disease risk in Sudanese diabetics should be further tested in large-scale studies.