The combination of differential expression and differential network connectivity analyses identifies RNA splicing and processing as common pathways altered with age across human tissues

Caio M.P.F. Batalha,André Fujita,Nadja C. de Souza-Pinto

DOI: https://doi.org/10.1101/2023.05.26.542445

2024-03-07

Abstract:Transcriptomic changes occur with age, but the extent of their similarities across tissues is not clear. Previous studies have identified no similarity in age-modulated genes in different tissues. In this study, we sought to identify transcriptional changes with age across tissues using differential network analysis, with the premise that differential expression analysis alone is not capable of detecting all the changes in the transcriptional landscape that occur in tissues with age. Our results show major transcriptional alterations not detected by differential expression analysis that can be detected by differential connectivity analysis. Combining these two analyses, we detected genes changing across tissues, enriched in “RNA splicing” and “RNA processing”, and highly connected in protein-protein interaction networks. Co-expression module analyses demonstrated that other genes with tissue-specific variations with age are enriched in pathways that combat accumulation of aberrant RNAs and proteins, which are caused by defective splicing. Additionally, tissues displayed a major reorganization of their genes’ connectivities with age, with most demonstrating convergent connectivity patterns. Our analyses identify genes and processes which transcriptional changes are conserved across tissues, demonstrating a central role for RNA splicing and processing genes and highlighting the importance of differential network analysis for understanding the ageing transcriptome.

Bioinformatics