This study aimed to characterize the role of non-coding RNAs—particularly circRNAs—in HIV pathogenesis by constructing immune-related regulatory networks that integrate bulk expression and single-cell data. The investigators sought to identify differentially expressed genes and microRNAs, infer circRNA-mediated competing endogenous RNA (ceRNA) interactions, determine protein interaction hubs, and nominate candidate biomarkers or therapeutic targets in HIV viremic patients.
Publicly available HIV-related mRNA and microRNA expression datasets were systematically retrieved from the Gene Expression Omnibus (GEO) repository. The authors state that analyses were performed in compliance with ethical standards and institutional guidelines. The abstract does not report specific GEO accession numbers or detailed sample counts.
Bulk expression datasets were processed using the limma package in R to detect differentially expressed genes (DEGs) and differentially expressed microRNAs (DEmiRNAs). Validated miRNA-target interactions from miRBase and ENCORI were used to construct circRNA–DEmiRNA and DEmiRNA–DEG regulatory networks. Protein–protein interaction (PPI) networks were generated using the STRING database; hub genes were defined as those lying within the highest 5% of interaction degrees in the PPI network.
Single-cell RNA sequencing (scRNA-seq) datasets from HIV-infected individuals were analyzed using the Seurat package to identify DEGs within T cell subsets. The reported single-cell findings include substantial gene expression shifts in T cell populations: in CD4+ T cells there were 194 upregulated and 813 downregulated genes, and in CD8+ T cells 244 upregulated and 567 downregulated genes. Cross-validation between microarray-derived DEGs and scRNA-seq results was performed to strengthen confidence in hub gene selection.
Integrated analysis across bulk and single-cell datasets produced five hub genes consistently dysregulated across T cell subsets. Two genes were identified as upregulated: STAT1 and DDX39B. Three were identified as downregulated: CXCL8, CXCL12, and PTGS2. These hub genes were selected based on their high degree in the PPI network and consistent dysregulation in single-cell analyses.
Several microRNAs were differentially expressed in HIV-infected samples versus controls; highlighted miRNAs include hsa-miR-21-5p, hsa-miR-146a-5p, and hsa-let-7b-5p. DEmiRNA–DEG network analysis showed reciprocal regulatory axes linking up- and down-regulated DEGs with down- and up-regulated miRNAs, respectively. A competing endogenous RNA (ceRNA) analysis identified circRNAs predicted to interact with these DEmiRNA–DEG pairs; representative reported axes include:
These axes propose potential circRNA-mediated modulation of gene expression relevant to HIV-associated immune responses.
Pathway enrichment analysis linked the identified hub genes and associated networks to regulatory pathways relevant to infectious disease mechanisms. The abstract indicates that these pathways are pertinent to HIV biology and immune regulation, but it does not list specific pathway names or enrichment statistics in the abstract text.
From the integrative workflow, the authors propose several candidates for further investigation as biomarkers or therapeutic targets in HIV viremic patients. The principal candidates named in the abstract are the five hub genes (STAT1, DDX39B, CXCL8, CXCL12, PTGS2) and the circRNA–miRNA–mRNA axes exemplified above. The analysis also highlights differentially expressed miRNAs such as hsa-miR-21-5p, hsa-miR-146a-5p, and hsa-let-7b-5p as potentially relevant regulators.
The abstract confirms use of public GEO datasets and validation using scRNA-seq, but it does not report detailed sample sizes, cohort characteristics, or GEO accession identifiers in the abstract text. Detailed methods, full datasets, statistical thresholds, and experimental validation (for example, in vitro or clinical validation of candidate circRNAs) are not described in the abstract; readers must consult the full article for those specifics.
This integrative computational study suggests that circRNA-mediated regulatory mechanisms intersecting with miRNA and mRNA networks may contribute to gene expression changes in HIV viremic patients. Five hub genes and several circRNA–miRNA–mRNA axes are nominated as candidate biomarkers or therapeutic targets for further investigation. The authors report no competing interests. Experimental validation and reporting of dataset identifiers and cohort metadata would be necessary next steps to move these candidates toward clinical or translational applications.