Computational Analysis of Multilevel High-throughput Data from Cancer Tissue

Abstract

The emergence of high-throughput measurement technologies has greatly expanded the possibilities for detecting and quantifying biomolecules involved in various subcellular and biomolecular processes on a larger scale than previously possible. These measurements, along with their accurate and robust analysis, play a crucial role in deepening our understanding and explaining a wide range of biological phenomena, including the development and progression of cancers. Consequently, this knowledge can be harnessed to develop effective interventions, particularly in the management and treatment of cancer.

Within this dissertation, we have pursued two primary aims. Firstly, we aimed to develop novel computational and statistical tools and methods for the effective and efficient analysis of high-throughput data within the context of cancer. Secondly, using the tools and methods we developed, we sought to investigate single and multilevel high-throughput data to identify key alterations that drive the development and progression of prostate cancer.

To accomplish the first aim, we devised a computational tool capable of detecting somatic copy number alterations. Additionally, we developed other computational and statistical approaches to mitigate the inherent biases present in data obtained from high- throughput sequencing technologies. As for the second aim, using the methods and tools we developed, we analyzed single and multilevel high-throughput data from a cohort of prostate cancer patients at various stages of their disease, identified multiple alterations, and presented our observations in detail.

In summary, our study demonstrates the potential of the analysis of single and multilevel high-throughput data. Through this approach, we were able to replicate previous findings and uncover alterations that impact biological processes at different levels during the development and progression of prostate cancer.