An approach combining bioinformatics and machine learning to identify eight autophagy-related biomarkers and construct molecular mechanisms underlying COVID-19 and major depressive disorders.

Abstract

A lack of objective biomarkers is preventing the screening and diagnosis of COVID-19 combined with major depression disorder (COVID-19-MDD). The purpose of this study was to identify diagnostic biomarkers and gene regulatory mechanisms associated with autophagy; a crucial process significantly involved in the pathogenesis of COVID-19-MDD. In this study, differentially expressed genes (DEGs) were screened using GSE98793 from the GEO2R analysis (GEO) database, and intersected with the COVID-19-related gene (CRGs) and autophagy-related genes (ARGs) to obtain common genes involved in. Then, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of these common genes were performed. Subsequently, the transcription factor (TF)-gene regulatory network and comorbidity network were constructed. In addition, 10 drug candidates were screened using the DSigDB database. To identify diagnostic markers, we used LASSO regression. In total, 13 common genes were screened, which were primarily enriched in lysosomes, endoplasmic reticulum membranes, and other endomembrane systems also associated with autophagy. Additionally, these genes were involved in neurological cell signaling and have a functional role in pathways related to vascular endothelial growth factor, tyrosine kinase, autophagy, inflammation, immunity, and carcinogenesis. Tumors and psychiatric disorders were the most highly linked diseases to COVID-19. Finally, ten drug candidates and eight diagnostic markers (STX17, NRG1, RRAGD, XPO1, HERC1, HSP90AB1, EPHB2, and S1PR3) were screened. This is the first study to screen eight diagnostic markers and construct a gene regulatory network for COVID-19-MDD from the perspective of autophagy. The findings of our study provide novel insights into the diagnosis and treatment of COVID-19-MDD.