Integrative transcriptomic analysis and expression validation identify R-loop-related genes CDYL and SF3B1 as candidate biomarkers for osteoporosis

Osteoporosis (OP) is characterized by low bone mass and microarchitectural deterioration. Although R-loops are implicated in transcription, DNA repair, and epigenetic regulation, their roles in OP remain unclear. By integrating differential expression analysis of public transcriptomic datasets with a curated list of R-loop-related genes (RLRGs), we identified 203 overlapping candidates. Subsequent machine-learning prioritization and receiver operating characteristic curve (ROC) analysis identified CDYL and SF3B1 as top candidate biomarkers. Both genes exhibited increased expression in the analyzed datasets and in a small clinical OP cohort relative to controls. In silico subcellular localization analysis predicted predominant nuclear enrichment. Correlation-based enrichment analyses linked their expression to proteasome- and spliceosome-related pathways. Their expression was negatively correlated with macrophage- and mast-cell-related single-sample gene set enrichment analysis (ssGSEA) scores. Network analysis identified putative upstream transcription factors (TFs) and microRNAs (miRNAs). Exploratory molecular docking predicted favorable binding affinities for selected vitamin D metabolites. Quantitative reverse transcription PCR(RT-qPCR) using clinical blood samples further corroborated the upregulation of both genes, providing preliminary support for CDYL and SF3B1 as candidate blood-based biomarkers for OP. Although these genes were selected from a curated R-loop-related gene set, the findings do not establish altered R-loop abundance in OP or an R-loop-dependent mechanism involving either gene.