Revealing common lncRNAs and gene signatures: computational analysis of public RNA-seq datasets in different in vitro senescence models

Recently, non-coding RNAs (ncRNAs) have emerged as crucial regulators of gene expression. To uncover common lncRNA signatures, we retrieved and analyzed seven publicly available RNA-seq datasets. These datasets included different in vitro models of senescence (replicative senescence, drug-induced senescence, and H₂O₂-induced senescence) in four human cell types: fibroblasts, mesenchymal stem cells, endothelial cells, and smooth muscle cells. The computational analysis consisted of several phases. The first phase involved querying and selecting RNA-seq datasets from open repositories such as the Gene Expression Omnibus (GEO). In the second phase, we performed RNA-seq data analysis, including data quality assessment of pre-processed read counts, filtering and normalization annotation, and calculation of differentially expressed genes (DEGs) and differentially expressed long non-coding RNAs (DELs). To enhance the biological relevance of the modulated genes derived from the different senescence datasets, we performed enrichment analysis using the Gprofiler R package, focusing on statistically modulated KEGGs terms. All analyses were performed in the R environment (version 2023.12.1.402), using open-source packages tailored to the specific features of each dataset. Finally, to identify common signatures, we compared the lists of up- and downregulated long non-coding genes and genes, with a particular focus on the replicative senescence model, which was the most represented senescence model across the datasets