Discovering Cancer's Evolutionary Patterns: A Multi-Phylogenetic Approach to Tumor Trajectories

Cancer results from the accumulation of somatic alterations conferring a selective advantage to cells. The accumulation of such alterations is an evolutionary process, with sub-populations of tumor cells having distinct genomic alterations that arise as a tumor grows.\\ Alterations can happen in a single cell's life cycle at any time. These alterations can be caused by external factors such as exposure to radiation or chemical products, or by internal factors such as mistakes during the duplication of a cell. As a result, the cell's \textit{DNA} will be altered. Accumulating more alterations leads to different \textit{sub-populations}. A \textit{sub-population} in a tumor refers to a group of cells that share distinct characteristics that distinguish them from other cells in the same tumor. These characteristics can enhance their adaptability, promote survival and proliferation, and ultimately contribute to the progression of cancer. Alterations occur in a stochastic manner, making it unpredictable what kind of alteration will appear in the future and when. Despite this unpredictability, some features are shared by the progression of a certain \textit{sub-population}, such as the order in which alterations arise. \\ Leveraging the insights derived from this relation we can provide valuable clues about potential future alterations. This approach lays the foundation for the development of targeted interventions that could revolutionize cancer treatment, improving patient outcomes.\newline In recent years, several methods have been developed to identify evolutionary patterns in cancer, such as \textit{RECAP}(\cite{RECAP}) ,an integer programming approach to identify evolutionary patterns in cancer, \textit{CONETT}(\cite{CONETT}) an integer programming approach to identify a consensus phylogenetic tree from the trees describing the evolution of a number of tumors, or \textit{MASTRO}(\cite{MASTRO}) that exploits the relationships between alterations to identify \textit{maximal frequent trajectories}.\\ The aim of this thesis is to leverage multiple phylogenetic trees for each patient, in order to identify significant cancer evolutionary trajectories. By analyzing multiple evolutionary paths, we want to improve predictions of future evolutionary steps in cancer progression.