(643g) Enabling More Predictive Modeling of Photoautotrophic Growth Using a Multi-Scale Multi-Paradigm Approach

Photosynthetic microorganisms have the potential to be a renewable source of fuels and chemicals; unfortunately, their potential have not yet been realized due to the complexity of these organisms and a general lack of tools when compared to premiere model organisms such as E. coli and yeast. The use of metabolic models for the design of large-scale production strains of heterotrophic bacteria and yeast has enabled a more streamlined approach and drastically cut down on research and development time. There are numerous algorithms to automate the creation of metabolic models from genome sequences, but automated annotations are skewed towards model organisms and do not perform well for photosynthetic strains. Therefore, we have developed an automated platform specifically for photosynthetic organisms that leverages manually curated models and high-quality annotations from sequenced organisms. This will expedite the creation and use of metabolic models for the growing number of photosynthetic strains used for metabolic engineering efforts. The most economical approach to grow these organisms are in large outdoor ponds, but this introduces a difficult challenge for metabolic modeling due to the dynamic nature of diurnal growth. We are also developing the next generation of metabolic model that can more accurately predict growth and phenotype of diurnal growth, track cell movement and nutrient availability. We are using a multi-scale multi-paradigm modeling approach, which also enables us to investigate the governing biological objectives and rules of behavior. I will discuss our model of Trichodesmium erythraeum, a filamentous diazotrophic cyanobacterium, validation of the model with in situ data and what we have discovered about the rules of behavior that define this organism.