A Network-Based Approach to Identify Molecular Signatures and Comorbidities of Thyroid Cancer

Abstract

The molecular mechanisms driving thyroid cancer (TC) development and progression are poorly understood, but identifying molecular pathways and hub genes abnormally active in TC tissues may highlight some of the key pathogenic processes involved. We therefore analysed TC tissue transcriptome profiles to identify such pathways which were then functionally characterized. Thus, we studied microarray gene expression dataset to identify the differentially expressed genes (DEGs) in TC compared to normal thyroid tissues. By using topological and neighbourhood based benchmark methods, we built comorbidity relationship networks to clarify how TC molecular pathways related to those of other diseases considering our identified significant DEGs of TC and assessed their functions using Online Mendelian Inheritance in Man (OMIM) databases, protein-protein interaction (PPI) networks. TC tissues showed dysregulation in expression of 598 genes, of which 133 had increased and 465 had decreased expression. From the analysis of the comorbidity network (gene-disease associations network; GDN), we identified 24 genes previously shown to be strongly associated with cancer, 24 genes associated with neurological disorders and 14 genes associated with cardiovascular disease. PPI network were built around the identified 79 unique genes in GDN of TC to identify the shared proteins group of different diseases. From the analysis of PPI network, we identified 10 significant hub genes namely EGFR, KIT, IRS1, KDR, BUB1B, CDH1, BUB1, TEK, TPM2 and NR4A2 based on the degree and betweenness centrality. Moreover, transcription factors (TFs) that may influence the observed TC gene expression were identified. Thus, our study identified DEGs, molecular pathways, hub genes, TFs and miRNAs of TC, as well as comorbidities with other diseases. The TC-associated genes thus identified comprised candidates for further studies to identify new TC biomarkers and pathological processes that underlie TC.