Identification of biomarkers associated with diagnosis of acute lung injury based on bioinformatics and machine learning

Abstract

Background: Acute lung injury (ALI) is an acute inflammatory disease characterized by excess production of inflammatory factors in lung tissue and has a high mortality. This research was designed for the identification of novel diagnostic biomarkers for ALI and analyzing the possible association between critical genes and infiltrated immune cells.

Methods: The study used 2 datasets (GSE2411 and GSE18341) to identify differentially expressed genes (DEGs) between 2 groups. Then we performed Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses to identify the functions of these DEGs. The study also used SVM-recursive feature elimination analysis and least absolute shrinkage and selection operator regression model to screen possible markers. The study further analyzed immune cell infiltration via CIBERSORT. Gene Set Enrichment Analysis was used to explore the molecular mechanism of the critical genes.

Results: DEGs were identified between 2 groups. In total, 690 DEGs were obtained: 527 genes were upregulated and 163 genes were downregulated. We identified PDZK1IP1, CCKAR, and CXCL2 as critical genes. And we then found that these critical genes correlated with Mast Cells, Neutrophil Cells, M1 Macrophage, dendritic cell Actived, Eosinophil Cells, B Cells Naive, Mast Cells, and dendritic cell Immature. Furthermore, we investigated the specific signaling pathways involved in key genes and derived some potential molecular mechanisms by which key genes affect disease progression by use of Gene Set Enrichment Analysis. Moreover, we predict transcription factors. Also, we obtained critical gene-related microRNAs through the targetscan database, and visualized the microRNA network of the genes.

Conclusion: Our findings might provide some novel clue for the exploration of novel markers for ALI diagnosis. The critical genes and their associations with immune infiltration may offer new insight into understanding ALI developments.