Network-Based Computational Approach to Identify Delineating Common Cell Pathways Influencing Type 2 Diabetes and Diseases of Bone and Joints

Abstract

Developing type 2 diabetes (T2D) can increase patient risk of developing other common diseases and exacerbate their severity, including diseases that affect bone and joints. Such comorbidity interactions are hard to study in detail by traditional endocrinological methods. Thus, we developed tissue transcript analytical approaches to identify common pathways through which these diseases can interact. We examined RNAseq and microarray transcript datasets from studies of T2D and chronic bone and joint diseases, namely rheumatoid arthritis (RA), osteoarthritis (OA), juvenile idiopathic arthritis (JIA) and low peak bone density, a key osteoporosis (OP) determinant. These datasets contained data from affected individuals and matched controls. Differentially expressed genes (DEGs) for each condition were compared with T2D DEG. Overlapping DEGs (i.e., those common to T2D and a bone or joint condition) were subjected to gene enrichment by pathway analyses and by gene ontology methods, and the results were evaluated by using SNP-disease linkage (dbGaP) and gene-disease association (OMIM) databases that indicate gene involvement in pathologies. By examining gene targets of transcription factors (TFs) and microRNA (miRNAs), we also constructed DEG-TF and DEG-miRNA interactions networks for analysis. We identified strong candidate genes in common pathways, notably including SYK, UCP3, ROR1, PPARG, BUB1, AKT2, ADCY2 and CCR5. The DEG-TF network and DEG-miRNA interactions network analyses revealed a number of TFs (GATA2, FOXC1, USF2, YY1, E2F1, JUN, RELA, CREB1, TFAP2A, NFB1) and miRNAs (mir-335-5p, mir-16-5p, mir-26b-5p, mir-124-3p, mir-218-5p, mir-98-5p, mir-29b-3p, mir-3135b, mir-29c-3p, mir-1-1) that can regulate the identified DEGs at the transcriptional and post-transcriptional levels. Thus this data-driven approach has enabled identification and validation of regulatory factors and cell pathways by which T2D may influence bone and joint conditions, which may suggest new ways to interfere with the pathogenic processes involved.INDEX TERMS T2 diabetes, network-based approach, joint diseases, bone diseases.

A Network-Based Methodology to Identify Subnetwork Markers for Diagnosis and Prognosis of Colorectal Cancer

“…Traditional statistical approaches are not suitable for detecting gene interactions, especially when interactions appear between more than two genes, or when the data are high-dimensional, meaning the data have many attributes or independent variables ( McKinney et al, 2006 ; Lai et al, 2019 ). Machine learning approaches have been widely used to identify disease biomarkers ( Lim et al, 2019 ; Moni et al, 2019 ; Tabl et al, 2019 ; Sanchez and Mackenzie, 2020 ). Recently, Sanchez et al identified methylation biomarkers for leukemia by investigating PPI for differentially methylated genes (DMGs) and differentially expressed genes (DEGs) using machine learning approach ( Sanchez and Mackenzie, 2020 ).…”

Machine learning models for prediction of co-occurrence of diabetes and cardiovascular diseases: a retrospective cohort study

“…A number of studies have developed models for predicting several diseases and comorbidities [9,[43][44][45][46][47][48]. Chun and colleagues [49] have introduced a comorbidity prediction method using filtering technique to predict likely comorbid conditions for individuals and a trajectory prediction graph model to reveal progression paths of the conditions.…”

Disclosing Potential Key Genes, Therapeutic Targets and Agents for Non-Small Cell Lung Cancer: Evidence from Integrative Bioinformatics Analysis

“…Identification of disease-causing crucial biomarkers may shed light on a deeper understanding of the molecular mechanism of disease [ 58 , 59 , 60 , 61 , 62 , 63 ]. The present study was conducted to analyze the NSCLC gene expression data to determine the DEGs, extensive molecular pathways, significant hub proteins, and associated regulatory biomolecules in order to pick up the potential therapeutic targets for NSCLC through a multi-omics data integration framework.…”

“…The higher expression pattern of YY1 transcription factor triggered the patients having larger tumor size, differentiation, higher TNM stage, and lymph node metastasis [ 83 ]. The reported TFs are also involved in other cancer diseases [ 58 , 59 , 60 , 61 , 62 , 63 ]. In various types of cancer tissues, the miR-26b-5p acts as a tumor suppressor [ 84 ].…”