The aim of this thesis is to statistically evaluate potential relationships among heterogeneous clinical, proteomic, and experimental data collected from 56 ovarian cancer patients to identify factors associated with distinct treatment responses-classified as sensitive (SE), resistant (RT), and partially resistant (RP). Given that the dataset includes quantitative protein intensity data for established OC markers, a secondary objective is to explore interrelationships among these proteins to determine whether they form functional networks linked to treatment outcomes.

The current dataset, although representing a cohort of 56 patients, will serve as the foundation for a larger, comprehensive database encompassing all patients treated within our research framework. This extended database will enable more robust modeling and validation of observed relationships. Furthermore, a full-scale proteomic dataset of extracellular vesicles (EVs), comprising approximately 7,000 quantified proteins, is available for the secondary objective and can be used to explore regulatory and functional associations among EV proteins relevant to treatment response prediction.

Technically, the goal will be to annotate the provided data sets from pseudoanonymised patient data, design and implement suitable data processing pipelines and provide detailed statistical analysis allowing to assess the relationships among the hypothesised factors and distinct treatment responses.

In addition, the thesis will include a structured review of relevant literature focusing on OC,  available analytical approaches for proteomic data, as well as statistical methods applicable to heterogeneous datasets, particularly those derived from ascites in OC. All analyses will be implemented in the R programming language, and the thesis will document the individual analytical pipelines to ensure transparency and reproducibility of the proposed workflow.