Quantitative Multi-Omics Investigation of Cellular Senescence Regulatory Networks Associated with Age-Related Tissue Degeneration

Abstract

Cellular senescence represents a basic phenomenon of ageing resulting from complex molecular interactions, which leads to ageing-related tissue impairment. In this thesis, a computational framework is proposed that combines multi-omics data, such as transcriptomic, proteomic, epigenomic and metabolomic data in a quantitative manner to reconstruct senescence regulatory networks. To identify key regulatory mechanisms, differential molecular analysis, network reconstruction, topological assessment, functional enrichment and machine learning based biomarker prioritization was performed. The integrated framework showed coordinated dysregulation in several omics layers, highlighted highly connected hub regulators and revealed highly significant pathways including extracellular matrix remodeling, inflammatory signaling, mitochondrial dysfunction, oxidative stress and cell-cycle arrest. The biomarker classification was strong with an accuracy of 90.4%, F1 score of 0.89 and ROC-AUC of 0.93 in predictive evaluation. Model stability and model reproducibility were confirmed by cross validation. In summary, the proposed framework offers detailed systems-level understanding of aging mechanisms of cells, and can facilitate the discovery of biomarkers, prioritizing the therapeutic targets, and future precision medicine approaches for the age-related degenerative diseases.