Monkeypox Virus Crosstalk with HIV: An Integrated Skin Transcriptome and Machine Learning Study.

The emergence of the monkeypox virus (MPXV) outbreak presents a formidable challenge to human health. Emerging evidence suggests that individuals with HIV have been disproportionately affected by MPXV, with adverse clinical outcomes and higher mortality rates. However, the shared molecular mechanisms underlying MPXV and HIV remain elusive. We identified differentially expressed genes (DEGs) from two public data sets, GSE219036 and GSE184320, and extracted common DEGs between MPXV and HIV. We further performed gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), protein-protein interactions (PPI), candidate drug assessment, and immune correlation of hub genes analysis. We validated the key biomarkers using multiple machine learning (ML) methods including random forest (RF), t-distributed stochastic neighbor embedding (tSNE), and uniform manifold approximation and projection (UMAP). A total of 59 common DEGs were identified between MPXV and HIV. Our functional analysis highlighted multiple pathways, including the ERK cascade, NF-κB signaling, and various immune responses, playing a collaborative role in the progression of both diseases. The PPI and gene co-expression networks were constructed, and five key genes with significant immune correlations were identified and validated by multiple ML models, including SPRED1, SPHK1, ATF3, AKT3, and AKT1S1. Our study emphasizes the common pathogenesis of HIV and MPXV and highlights the pivotal genes and shared pathways, providing new opportunities for evidence-based management strategies in HIV patients co-infected with MPXV.