Intervertebral Disc Biology through Experiments and Knowledge: Network Modeling, Proteomics, and Machine Learning

Tseranidou, Sofia

Intervertebral Disc Biology through Experiments and Knowledge: Network Modeling, Proteomics, and Machine Learning

Document Type

Doctoral thesis
Published Version

Author

Tseranidou, Sofia

Director

Chorlton, Christine Lyn
Piñero González, Janet
Noailly, Jérôme

Tutor

Noailly, Jérôme

Publication Date

2025-10-03

Pages

206 p.

License

Citation

Tseranidou, S. Intervertebral Disc Biology through Experiments and Knowledge: Network Modeling, Proteomics, and Machine Learning. Universitat Pompeu Fabra; 2025. handle: http://hdl.handle.net/10803/695645

This citation was generated automatically.

Doctoral program

Universitat Pompeu Fabra. Doctorat en Tecnologies de la Informació i les Comunicacions

Abstract

Low back pain is a leading cause of global disability and is strongly linked to intervertebral disc degeneration (IDD). IDD arises from changes in the disc microenvironment that impair structure and function. It is marked by extracellular matrix breakdown, altered cell phenotype, loss of active cells, increased senescence, and increased production of inflammatory mediators, which drive catabolism. Although numerous studies have investigated the molecular basis of IDD, the variability in disc cell behavior and the complexity of the intracellular signaling pathways that govern inflammatory and catabolic processes hinder a unified understanding of the underlying mechanisms of IDD. Systems modeling, including network-based models, provide scalable means to integrate these diverse data and clarify the interplay among cytokines, growth factors, and other soluble mediators. This thesis develops a literature-based regulatory network model (RNM) focused on protein–protein interactions among key soluble mediators regulating disc cell behavior. The developed network captures interactions among soluble proteins, accurately reflecting control values from proteomics analyses. To further elucidate the roles of specific ligands, and understand how they modulate other soluble mediators, the network was expanded to incorporate critical signaling pathways identified in initial cell network modeling, providing a more holistic view of signaling cascades. This expanded model has been validated using experimental data from human nucleus pulposus cells. Additionally, to address incomplete knowledge regarding interactions among these mediators, a machine learning-based link prediction approach was applied to uncover missing or poorly characterized connections. Identifying these gaps not only underscores where additional experimental evidence is needed to understand the role of factors and signaling pathways but also helps pinpoint novel therapeutic targets for IDD. Altogether, this framework advances our understanding of the complex signaling landscape underlying IDD and offers a roadmap for future research to address the remaining gaps in disc cell biology.

Keywords

Disc intervertebral, Intervertebral disc, Disco intervertebral, Regulatory network model, Link prediction model, Signaling pathway model, Proteomics analysis, Nucelus pulposus, Systems Biology