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

Comprehensive machine learning models for predicting therapeutic targets in type 2 diabetes utilizing molecular and biochemical features in rats

Introduction

With the increasing prevalence of type 2 diabetes mellitus (T2DM), there is an urgent need to discover effective therapeutic targets for this complex condition. Coding and non-coding RNAs, with traditional biochemical parameters, have shown promise as viable targets for therapy. Machine learning (ML) techniques have emerged as powerful tools for predicting drug responses.

Method

In this study, we developed an ML-based model to identify the most influential features for drug response in the treatment of type 2 diabetes using three medicinal plant-based drugs (Rosavin, Caffeic acid, and Isorhamnetin), and a probiotics drug (Z-biotic), at different doses. A hundred rats were randomly assigned to ten groups, including a normal group, a streptozotocin-induced diabetic group, and eight treated groups. Serum samples were collected for biochemical analysis, while liver tissues (L) and adipose tissues (A) underwent histopathological examination and molecular biomarker extraction using quantitative PCR. Utilizing five machine learning algorithms, we integrated 32 molecular features and 12 biochemical features to select the most predictive targets for each model and the combined model.

Results and discussion

Our results indicated that high doses of the selected drugs effectively mitigated liver inflammation, reduced insulin resistance, and improved lipid profiles and renal function biomarkers. The machine learning model identified 13 molecular features, 10 biochemical features, and 20 combined features with an accuracy of 80% and AUC (0.894, 0.93, and 0.896), respectively. This study presents an ML model that accurately identifies effective therapeutic targets implicated in the molecular pathways associated with T2DM pathogenesis.

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

Coiled-coil domain-containing 58 (CCDC58) is a member of the CCDC protein family. Similar to other members, CCDC58 exhibits potential tumorigenic roles in a variety of malignancies. However, there is no systematic and comprehensive pan-cancer analysis to investigate the diagnosis, prognosis, immune infiltration, and other related functions of CCDC58. We used several online websites and databases, such as TCGA, GTEx, UALCAN, HPA, CancerSEA, BioGRID, GEPIA 2.0, TIMER 2.0, and TISIDB, to extract CCDC58 expression data and clinical data of patients in pan-cancer. Then, the relationship between CCDC58 expression and diagnosis, prognosis, genetic alterations, DNA methylation, genomic heterogeneity, and immune infiltration level were determined. In addition, the biological function of CCDC58 in liver hepatocellular carcinoma (LIHC) was investigated. Pan-cancer analysis results showed that CCDC58 was differentially expressed in most tumors and showed excellent performance in diagnosis and prediction of prognosis. The expression of CCDC58 was highly correlated with genetic alterations, DNA methylation, and genomic heterogeneity in some tumors. In addition, the correlation analysis of CCDC58 with the level of immune infiltration and immune checkpoint marker genes indicated that CCDC58 might affect the composition of the tumor immune microenvironment. Enrichment analysis showed that CCDC58-related genes were mainly linked to mitosis, chromosome, and cell cycle. Finally, biological function experiments demonstrated that CCDC58 plays an important role in tumor cell proliferation and migration. CCDC58 was first identified as a pan-cancer biomarker. It may be used as a potential therapeutic target to improve the prognosis of patients in the future.

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

Omics technologies accumulated an enormous amount of data that advanced knowledge about the molecular pathogenesis of type 1 diabetes mellitus and identified a number of fundamental problems focused on the transition to personalized diabetology in the future. Among them, the most significant are the following: (1) clinical and genetic heterogeneity of type 1 diabetes mellitus; (2) the prognostic significance of DNA markers beyond the HLA genes; (3) assessment of the contribution of a large number of DNA markers to the polygenic risk of disease progress; (4) the existence of ethnic population differences in the distribution of frequencies of risk alleles and genotypes; (5) the infancy of epigenetic research into type 1 diabetes mellitus. Disclosure of these issues is one of the priorities of fundamental diabetology and practical healthcare. The purpose of this review is the systemization of the results of modern molecular genetic, transcriptomic, and epigenetic investigations of type 1 diabetes mellitus in general, as well as its individual forms. The paper summarizes data on the role of risk HLA haplotypes and a number of other candidate genes and loci, identified through genome-wide association studies, in the development of this disease and in alterations in T cell signaling. In addition, this review assesses the contribution of differential DNA methylation and the role of microRNAs in the formation of the molecular pathogenesis of type 1 diabetes mellitus, as well as discusses the most currently central trends in the context of early diagnosis of type 1 diabetes mellitus.

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

BACKGROUND

Diabetes (hyperglycemia) is defined as a multifactorial metabolic disorder in which insulin resistance and defects in pancreatic β-cell dysfunction are two major pathophysiologic abnormalities that underpin towards gestational diabetes mellitus (GDM). TCF7L2, KCNQ1, and KCNJ11 genes are connected to the mechanism of β-cell dysfunction. The purpose of this study was to investigate the genes associated with β-cell dysfunction and their genetic roles in the rs7903146, rs2237892, and rs5219 variants in Saudi women diagnosed with type 2 diabetes mellitus and GDM.

MATERIALS AND METHODS

In this case-control study, 100 women with GDM and 100 healthy volunteers (non-GDM) were recruited. Genotyping was performed using polymerase chain reaction (PCR), followed by restriction fragment length analysis. Validation was performed using Sanger sequencing. Statistical analyses were performed using multiple software packages.

RESULTS

Clinical studies showed a β-cell dysfunction positive association in women with GDM when compared to non-GDM women (p < 0.05). Both rs7903146 (CT vs. CC: OR-2.12 [95%CI: 1.13-3.96]; p = 0.01 & T vs. C: (OR-2.03 [95%CI: 1.32-3.11]; p = 0.001) and rs5219 SNPs (AG vs. AA: OR-3.37 [95%CI: 1.63-6.95]; p = 0.0006 & G vs. A: OR-3.03 [95%CI: 1.66-5.52]; p = 0.0001) showed a positive association with genotype and allele frequencies in women with GDM. ANOVA analysis confirmed that weight (p = 0.02), BMI (p = 0.01), and PPBG (p = 0.003) were associated with rs7903146 and BMI (p = 0.03) was associated with rs2237892 SNPs.

CONCLUSIONS

This study confirms that the SNPs rs7903146 (TCF7L2) and rs5219 (KCNJ11) are strongly associated with GDM in the Saudi population. Future studies should address the limitations of this study.