RESUMEN
Type 2 Diabetes (T2D) is a chronic disease characterized by abnormally high blood glucose levels due to insulin resistance and reduced pancreatic insulin production. The challenge of this work is to identify T2D-associated features that can distinguish T2D sub-types for prognosis and treatment purposes. We thus employed machine learning (ML) techniques to categorize T2D patients using data from the Pima Indian Diabetes Dataset from the Kaggle ML repository. After data preprocessing, several feature selection techniques were used to extract feature subsets, and a range of classification techniques were used to analyze these. We then compared the derived classification results to identify the best classifiers by considering accuracy, kappa statistics, area under the receiver operating characteristic (AUROC), sensitivity, specificity, and logarithmic loss (logloss). To evaluate the performance of different classifiers, we investigated their outcomes using the summary statistics with a resampling distribution. Therefore, Generalized Boosted Regression modeling showed the highest accuracy (90.91%), followed by kappa statistics (78.77%) and specificity (85.19%). In addition, Sparse Distance Weighted Discrimination, Generalized Additive Model using LOESS and Boosted Generalized Additive Models also gave the maximum sensitivity (100%), highest AUROC (95.26%) and lowest logarithmic loss (30.98%) respectively. Notably, the Generalized Additive Model using LOESS was the top-ranked algorithm according to non-parametric Friedman testing. Of the features identified by these machine learning models, glucose levels, body mass index, diabetes pedigree function, and age were consistently identified as the best and most frequently accurate outcome predictors. These results indicate the utility of ML methods in constructing improved prediction models for T2D and successfully identified outcome predictors for this Pima Indian population. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s13755-021-00168-2.
RESUMEN
With the increasing number of immunoinflammatory complexities, cancer patients have a higher risk of serious disease outcomes and mortality with SARS-CoV-2 infection which is still not clear. In this study, we aimed to identify infectome, diseasome and comorbidities between COVID-19 and cancer via comprehensive bioinformatics analysis to identify the synergistic severity of the cancer patient for SARS-CoV-2 infection. We utilized transcriptomic datasets of SARS-CoV-2 and different cancers from Gene Expression Omnibus and Array Express Database to develop a bioinformatics pipeline and software tools to analyze a large set of transcriptomic data and identify the pathobiological relationships between the disease conditions. Our bioinformatics approach revealed commonly dysregulated genes (MARCO, VCAN, ACTB, LGALS1, HMOX1, TIMP1, OAS2, GAPDH, MSH3, FN1, NPC2, JUND, CHI3L1, GPNMB, SYTL2, CASP1, S100A8, MYO10, IGFBP3, APCDD1, COL6A3, FABP5, PRDX3, CLEC1B, DDIT4, CXCL10 and CXCL8), common gene ontology (GO), molecular pathways between SARS-CoV-2 infections and cancers. This work also shows the synergistic complexities of SARS-CoV-2 infections for cancer patients through the gene set enrichment and semantic similarity. These results highlighted the immune systems, cell activation and cytokine production GO pathways that were observed in SARS-CoV-2 infections as well as breast, lungs, colon, kidney and thyroid cancers. This work also revealed ribosome biogenesis, wnt signaling pathway, ribosome, chemokine and cytokine pathways that are commonly deregulated in cancers and COVID-19. Thus, our bioinformatics approach and tools revealed interconnections in terms of significant genes, GO, pathways between SARS-CoV-2 infections and malignant tumors.
