Use of genome scale modelling of Saccharomyces cerevisiae to aid second generation biofuels research

Genome scale modelling has a large potential to mine the wealth of published knowledge in order to identify genes and pathways which are potential targets for metabolic engineering. Engineering organisms, such as Escherichia coli, Clostridia and Saccharomyces cerevisiae, for biofuel production could lead to economically viable renewable fuels. S. cerevisiae is a commonly used organism in biotechnology, with a number of in silico models available. We have coupled large scale metabolic modelling with experimental techniques to investigate the production of second generation biofuels from yeast. Constraint based analysis techniques were applied to a S. cerevisiae model to identify pathways that may lead to an increased isobutanol production. By simulating in silico knockouts, we found that genes from amino acid pathways greatly affected isobutanol output. This was further investigated by testing the effect of excess amino acids on the alcohol production. The predicted levels of isobutanol were experimentally tested by growing a selection of S. cerevisiae strains in media containing a specific amino acid as a nitrogen source and analysing the alcohol output. We further refined our predictions by incorporating quantitative omics data in the constraint-based model. A number of key predictions were in line with the experimental results; in particular, valine was confirmed to produce the highest volume of isobutanol. This research demonstrates the ability of genome-scale metabolic modelling to direct biotechnological enhancement of biofuel production by microorganisms.