Metabolic engineering and synthetic biology in general are powerful areas of research, but they can only realize a quantum leap in breakthroughs when computational models involved in the Design-Build-Test-Learn engineering paradigm are standardized and implemented in the field. As part of our iGEM project, the models developed are motivated by different objectives and approaches in order to inform and be shaped by our endeavors in the lab. To do so, two different computational models were used to determine the best approaches to improve PHA yield and productivity in recombinant E. coli. One of the ways to model a system in silico is through the dynamic model, which many systems biologists should be familiar with. By defining a pathway (such as the heterologous PHA biosynthetic pathway) or metabolic network as a series of ordinary differential equations (ODE), dynamic behaviors in the system over time may be predicted. A considerable limitation arises from the need for highly detailed information to parameterize the kinetics of the pathway; therefore, this type of model may be difficult to implement. On the other hand, an organism as well-studied as E. coli enjoys a vast wealth of literature surrounding its genome and metabolic network. As a result, stoichiometric models such as genome-scale models (GEM) are available to be used for in silico simulations without necessitating the rigorous characterization associated with dynamic or kinetic models. Using this, we can investigate problems such as the best conditions to grow a recombinant SBM E. coli and how to improve the titer of PHA.
To read more about the dynamic model developed by the Stanford-Brown 2015 team, please click here!
Using the dynamic model, we can investigate problems related to the behavior of the system and the consequences in production. The questions that we seek to investigate using this model are related most closely to the amount of feedstock (e.g. acetate and propionate) and their relative proportions that would achieve a desired end composition of PHBV co-polymer . In addition, we also used the model to theoretically survey the effects that can be achieved towards this end through the engineering of enzymes (that is, changing the binding affinity and catalytic activity).
To learn more about the COBRA Toolbox developed by the Palsson Group, please click here!
For those interested in the code behind the implementation, please click here to access our GitHub repository!
