Artificial intelligence and multi-omic empowered-tracing macrophage diagnostic and druggable implications in coronavirus pneumonia.

BACKGROUND

As the innate immune barrier, macrophages play an important role in coronavirus infection pneumonia and post-infection regeneration.

OBJECTIVE

Deeper understanding of macrophages can provide additional insights into treatment of coronavirus pneumonia.

METHOD: Single-cell transcriptomic dataset of coronavirus pneumonia (GSE171524) was acquired from GEO database and then underwent hdWGCNA analysis for identified macrophage-associated shared module and differentially expressed genes (DEGs). Next, we also performed Lasso framework in coronavirus pneumonia patient bulk dataset (GSE205099) for identification of diagnostic model and hub gene, and then model performance in GSE183533. Besides, molecular and immune features of hub gene was estimated at bulk and single-cell level, especially evaluation of changes in molecular functions after knockout (KO) of targeted gene in macrophage via artificial intelligence (AI)-simulated KO. Next, therapeutic agent targeting hub gene was estimated by CTD database and validated by molecular docking.

RESULT: Macrophages play an important role in the progression of coronavirus pneumonia, and FMN1 can be considered as target for diagnostic and druggable target for coronavirus pneumonia. Quercetin should be considered as optimal natural compounds for the treatment of coronavirus pneumonia.

CONCLUSION

Our study first discovered the dynamic role of macrophage in coronavirus pneumonia and creatively pointed out that the FMN1 can be considered as central molecule involved in coronavirus pneumonia pathogenesis. Quercetin also can be considered as additional choice for coronavirus pneumonia clinical translation.