Comprehensive Bioinformatics Analysis for the Identification of Hub Genes and Critical Signaling Pathways Differentiating Latent and Active Tuberculosis

Objectives Latent tuberculosis infection (LTBI) population is the principle source of active tuberculosis (ATB) patients. The identification of reliable diagnostic biomarkers is critical to the prevention and control of progression from LTBI to ATB. The aim of this study is to screen biomarkers that can distinguish LTBI from ATB patients by a comprehensive bioinformatic analysis strategy. Methods The transcriptomic datasets were obtained from GEO database. Hub genes and critical signal pathways for differentiating latent and active TB were identified by a comprehensive bioinformatic analysis strategy comprising weighted gene co-expression network analysis (WGCNA), differentially expressed gene (DEG), protein-protein interaction (PPI), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and hub genes were verified by RT-qPCR in this study. Results The transcriptome profiles of GSE193777, GSE157657, GSE168519, GSE107991, and GSE107992 were extracted from the GEO database, in which a total of 18,397 protein-coding genes from 206 samples were included in the bioinformatics analysis. Combined with weighted gene co-expression network, differentially expressed gene, functional enrichment and proteinprotein interaction analyses, six hub genes were identified. RT-qPCR confirmed significantly lower expression of HLA-DOA (p=0.002), ECH1 (p=7.19e-8), PARN (p=9.22e-9), and TRAPPC4 (p=6.81e-7) in LTBI versus ATB. Conclusions Our findings may provide crucial clues to potential biomarkers that can distinguish patients with LTBI from those with ATB, aiding the understanding of the mechanism of LTBI progression to ATB.