Decoding diabetes biomarkers and related molecular mechanisms using machine learning, text mining, and gene expression analysis

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 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 ten 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 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. ML 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.