Robust, realistic, and reliable measurements of metabolism
New measuring techniques have always been on of the most important driving forces for medical development. For instance, through the development of magnetic resonance imaging we can today see into out bodies in ways not previously possible. We can see where cancerous tumours have developed or where dangerous amounts of fat tissue have accumulated, and thus both detect and treat such diseases more effectively. The methods developed in this project could allow us to look even further into the body: into the highly detailed metabolism of the human cells.
What metabolic reactions that exists are already known, but there are no good methods for measuring how fast the exchange rate of these reactions are, these rates are also called the metabolic fluxes. The most promising techniques for measuring the metabolic fluxes relies on measuring which of a cell’s metabolites contain labelled carbon atoms. By combining such measurements with the knowledge of how the different reactions in the cell rearrange these carbon atoms between metabolites, one can calculate the rate of each reaction in the cell.
However, in practice this analysis is quite complicated, and mistakes are easily made. It is common to only look at a couple of metabolites but by doing this one misses most of the information contained within the system, which in turn lead to the wrong conclusion. In order to handle the complexity of human metabolic systems in a correct way mathematically models are required. The problem is that the methods that are currently used for model analysis are not developed for such complex systems found in human cells. Hence, model analysis could with current method could take months and one could still be unsure regarding the reliability of the results.
In this project we will develop new methods to improve the analysis of these model systems, by adopting principles from other modelling fields. These improvements will concern three important aspects of the analysis i) by accounting for that human cells do not necessarily fulfil the ideal conditions that simples systems can be expected to do, we will be able to easily develop models that a more robust; ii) by developing a new system for evaluating the models ability to predict new data, we will get more realistic models; iii) by differentiating between different types of model predictions, we will be able to increase the reliability of our results. All these aspects will be combined into a more intuitive analysis, where one will be able to easier reach new conclusions that can be experimentally tested.
If the project is successful, we will have a first method for reliably being able to measure the detailed metabolism in human cells, something that will create a lot of new opportunities and that will help treat some of societies most widespread diseases. We will evaluate our method in collaboration with both the Swedish heal care system and the pharmaceutical industry in order to study important disease such as cancer and metabolic syndrome.