Thesis project at Chalmers University of Technology, LIFE, Polster Lab.
Contact: Gaston Sandstig@chalmers.se
Alzheimer’s Disease (AD) is the most common form of dementia and more than 10 million people are diagnosed with it annually. Because there is currently no method of curing or preventing AD, improving our understanding of the disease will be integral to developing such methods. However, the pathophysiological mechanisms of AD are determined by a complex combination of factors which will differ between patients and vary significantly with time. Therefore there is a need for holistic approaches which can be personalised to each patient and integrate the many factors into more actionable data.
The metabolism of organisms and cells can be mathematically represented by genome-scale metabolic models (GEMs), which are built on collected omics-data from that organism. By modelling the entire system, we are able to see phenotypes resulting from complex gene interactions which are not obvious from reductionist methods. GEMs can also be constructed from an individual (patient’s) genome, and is therefore suitable for both personalized and precision medicine. By tailoring GEMs to an individual patient, we are able to facilitate a comprehensive understanding of relevant pathophysiological mechanisms for that specific individual.
In the thesis I aim to build single-sample GEMs based on the genomes deceased people with and without AD. The models will subsequently be used to investigate associations between it and the clinical phenotype of the individual. I will be looking at differences between affected and non-impaired individuals as well as the heterogeneity within the Alzeheimer’s samples and examine if there is further clustering within it. Finally I will investigate if we can predict future outcomes of patients by their prior condition.
Attached to the repository is version info of the packages in R and Matlab. The project also relies on the SysBio-projects RAVEN and Human-GEM.
todo