Identification of Type 2 Diabetes Based on a Ten-Gene Biomarker Prediction Model Constructed Using a Support Vector Machine Algorithm

Abstract

Background. Type 2 diabetes is a major health concern worldwide. The present study is aimed at discovering effective biomarkers for an efficient diagnosis of type 2 diabetes. Methods. Differentially expressed genes (DEGs) between type 2 diabetes patients and normal controls were identified by analyses of integrated microarray data obtained from the Gene Expression Omnibus database using the Limma package. Functional analysis of genes was performed using the R software package clusterProfiler. Analyses of protein-protein interaction (PPI) performed using Cytoscape with the CytoHubba plugin were used to determine the most sensitive diagnostic gene biomarkers for type 2 diabetes in our study. The support vector machine (SVM) classification model was used to validate the gene biomarkers used for the diagnosis of type 2 diabetes. Results. GSE164416 dataset analysis revealed 499 genes that were differentially expressed between type 2 diabetes patients and normal controls, and these DEGs were found to be enriched in the regulation of the immune effector pathway, type 1 diabetes mellitus, and fatty acid degradation. PPI analysis data showed that five MCODE clusters could be considered as clinically significant modules and that 10 genes (IL1B, ITGB2, ITGAX, COL1A1, CSF1, CXCL12, SPP1, FN1, C3, and MMP2) were identified as “real” hub genes in the PPI network using algorithms such as Degree, MNC, and Closeness. The sensitivity and specificity of the SVM model for identifying patients with type 2 diabetes were 100%, with an area under the curve of 1 in the training as well as the validation dataset. Conclusion. Our results indicate that the SVM-based model developed by us can facilitate accurate diagnosis of type 2 diabetes.

Methylation haplotypes of the insulin gene promoter in children and adolescents with type 1 diabetes: Can a dimensionality reduction approach predict the disease?

“…Studies on the widespread use of machine-learning methods in diabetes have already been published (10,12,14). The large data available after the implementation of internationally accepted recommendations in diabetes, allow the use of artificial intelligence machine-learning methods to extract new knowledge and develop predictive tools (10).…”

“…Towards this direction, there are reports that explore the use of machine learning methods in T1D (10) and identify epigenetic differentiations with prognostic value (11). Specifically, studies are using machine learning algorithms, fed with a gene signature deriving from gene expression (12) or daily life data (13) to diagnose diabetes, while others use deep learning algorithms to classify diabetic and healthy cohorts (14). Regarding the prognostic value of methylation haplotypes, research has focused on the early detection of carcinogenesis (15) and only very recently as an autoimmunity biomarker (16).…”

Intracellular complement: Evidence, definitions, controversies, and solutions

“…C3 expression was also upregulated in pancreatic islets from human donors with type 2 diabetes (T2D) compared to healthy donors, 24 and has been recently identified as one of a set of just 10 differentially expressed genes in pancreatic islets that accurately predict T2D status. 26 The independently confirmed interaction between C3 and ATG16L1, and a defective autophagic phenotype in C3‐knockout cells, suggests that endogenous C3 must be found within the cytosol in this cell type, to interact with ATG16L1. This raises questions as to how C3, a typically secreted protein, should enter the cytosol, what conformation the protein will have in the cytosolic environment, and what other complement factors, if any, it interacts with here.…”

“…C3‐knockout autophagy‐defective beta‐cells also underwent increased apoptosis when faced with diabetogenic cell stresses such as gluco‐ and lipo‐toxicity, consistent with a known protective role of autophagy in beta‐cells, 25 and demonstrating a role of C3 in cytoprotective homeostasis. C3 expression was also upregulated in pancreatic islets from human donors with type 2 diabetes (T2D) compared to healthy donors, 24 and has been recently identified as one of a set of just 10 differentially expressed genes in pancreatic islets that accurately predict T2D status 26 . The independently confirmed interaction between C3 and ATG16L1, and a defective autophagic phenotype in C3‐knockout cells, suggests that endogenous C3 must be found within the cytosol in this cell type, to interact with ATG16L1.…”

Classification of gene expression dataset for type 1 diabetes using machine learning methods

“…Table 3 shows the evaluation of ML models were used in the related works. GSE55098 for T1D LASSO-SVM AUC=0.918 [7] Single-cell RNA-sequencing for T2D Bayesian network, SVM, RF, LR and NN ACC=0.907 [8] lncRNA expression for T2D KNN, SVM, LR and ANN AUC=0.95 [9] GSE38642 and GSE13760 for T2D LR and SVM ACC=90.23% [10] GSE164416 for T2D SVM Sensitivity=100%…”