Files
Abstract
Metabolic reprogramming, recognized as one of the hallmarks of cancer, refers to the ability of cancer cells to rapidly adapt to microenvironment stresses via genetic mutations, transcriptomic regulations, and epigenetic modulations. In this dissertation, I have developed a framework to investigate the dysregulated metabolic reaction chains using transcriptomics data collected from cancer tissues. In the first chapter, we systematically analyzed 50 reprogrammed metabolic pathways across 14 cancer types and proposed proposed a model that these alterations are induced to neutralize a intracellular alkaline stress caused by chronic inflammation and iron overload. The close scrutiny of a specific pathway, sialic acid synthesis and utilization, confirmed our previous hypothesis of pH-stress driven metabolic reprogramming. Further analysis also identified a link between metabolic changes and tumor proliferation and metastasis. Our model provided strong evidence that sialic acid accumulation on the surface of cancer cells promotes cell-cell repulsion and ultimately drives migration. Finally, we developed a open-source tool named metabolike that can transform pre-specified systems biology models into a knowledge graph. The graph representation within Metabolike enables efficient identification and visualization of novel reaction routes, which makes it a great resource for studying complex biochemical reaction networks under diseased conditions such as cancer.