Decoding Diabetes Biomarkers and Related Molecular Mechanisms by Using Machine Learning, Text Mining, and Gene Expression Analysis

Abstract

The molecular basis of diabetes mellitus is yet to be fully elucidated. We aimed to identify the most frequently reported and differential expressed genes (DEGs) in diabetes by using bioinformatics approaches. Text mining was used to screen 40,225 article abstracts from diabetes literature. These studies highlighted 5939 diabetes-related genes spread across 22 human chromosomes, with 112 genes mentioned in more than 50 studies. Among these genes, HNF4A, PPARA, VEGFA, TCF7L2, HLA-DRB1, PPARG, NOS3, KCNJ11, PRKAA2, and HNF1A were mentioned in more than 200 articles. These genes are correlated with the regulation of glycogen and polysaccharide, adipogenesis, AGE/RAGE, and macrophage differentiation. Three datasets (44 patients and 57 controls) were subjected to gene expression analysis. The analysis revealed 135 significant DEGs, of which CEACAM6, ENPP4, HDAC5, HPCAL1, PARVG, STYXL1, VPS28, ZBTB33, ZFP37 and CCDC58 were the top 10 DEGs. These genes were enriched in aerobic respiration, T-cell antigen receptor pathway, tricarboxylic acid metabolic process, vitamin D receptor pathway, toll-like receptor signaling, and endoplasmic reticulum (ER) unfolded protein response. The results of text mining and gene expression analyses used as attribute values for machine learning (ML) analysis. The decision tree, extra-tree regressor and random forest algorithms were used in ML analysis to identify unique markers that could be used as diabetes diagnosis tools. These algorithms produced prediction models with accuracy ranges from 0.6364 to 0.88 and overall confidence interval (CI) of 95%. There were 39 biomarkers that could distinguish diabetic and non-diabetic patients, 12 of which were repeated multiple times. The majority of these genes are associated with stress response, signalling regulation, locomotion, cell motility, growth, and muscle adaptation. Machine learning algorithms highlighted the use of the HLA-DQB1 gene as a biomarker for diabetes early detection. Our data mining and gene expression analysis have provided useful information about potential biomarkers in diabetes.

Genetic and Epigenetic Aspects of Type 1 Diabetes Mellitus: Modern View on the Problem

“…The National Center for Biotechnology Information (NCBI) database contains more than 20 thousand papers devoted to the investigation of genetic factors that are involved in the pathogenesis of T1DM. In addition to the genes of the major histocompatibility complex, scientific interest is focused on investigating the significance of a number of other genes such as INS (insulin), PTPN22 (tyrosine phosphatase), IFIH1 (RIG-I-like receptor), SH2B3 (adapter protein), CD226 (immunoglobulin superfamily protein), TYK2 (tyrosine kinase 2), FUT2 (galactoside-2-alpha-L-fucosyltransferase 2), SIRPG (signal-regulatory protein gamma), CTLA4 (immunoglobulin superfamily protein), CTSH (cathepsin H), CTLA4 (cytotoxic T-lymphocyte glycoprotein), and UBASH3A (ubiquitin-associated protein A containing the SH3 domain) [9]. There is evidence of augmentation in the T1DM risk through the mechanisms of genomic imprinting with the participation of the INS gene [10] and the alternative splicing of islet cell autoantigen IA-2 mRNA [11], as well as through gene-gene and gene-environment interactions stipulated by epigenetic modifications [12,13] or retrovirus-mediated changes in different cells involved in pancreatic β-cell functioning [14,15].…”

Saudi Community-Based Screening Study on Genetic Variants in β-Cell Dysfunction and Its Role in Women with Gestational Diabetes Mellitus

“…Chronically abnormal blood sugar levels are a hallmark of this potentially lethal metabolic condition [ 1 ]. According to the International Diabetes Federation, approximately 700 million adults aged 18–99 will be diagnosed with diabetes worldwide by 2045 [ 2 ]. Type 2 diabetes mellitus (T2DM) is a common form of diabetes, accounting for 90–95% of cases, with the remainder classified as T1DM, gestational diabetes mellitus (GDM), and other types of diabetes [ 3 ].…”

CCDC58 is a potential biomarker for diagnosis, prognosis, immunity, and genomic heterogeneity in pan-cancer

“…CCDC58 had also been identified as a gene associated with tumor progression in endometrial cancer and ovarian cancer 14,15 . Among the differentially expressed genes between diabetic patients and normal controls, CCDC58 ranked in the top ten, suggesting that CCDC58 might be related to the occurrence and development of diabetes 16 .…”