(62s) Predicting Metabolic Engineering Knockout Strategies for Chemical Production: Accounting for Competing Pathways

Metabolic engineering aims to use microbes as factories that can produce and degrade organic molecules for industrial and biomedical purposes. In recent years metabolic engineering was successfully employed to produce various fuels and chemicals, and significant efforts are made for over-producing additional chemicals. Various computational methods are used to design genetic manipulations that can achieve metabolic engineering goals. These methods aim to anticipate the effect of genetic manipulations on cellular metabolism via a metabolic network, searching for specific manipulations that would lead to maximized production rate of chemicals of interest. However, a thorough examination of these methods reveals that they do not account for the presence of alternative pathways in a metabolic network, leading to over-optimistic predictions regarding achievable chemical production rates. In this paper we describe a novel computational method called RobustKnock that predicts genetic manipulations that maximize the guaranteed production rate of chemicals of interest, accounting for the presence of alternative pathways in the network. We show that RobustKnock predicts genetic manipulations that lead to guaranteed production rates of various chemicals in E.coli (e.g. ethanol that can be used as a bio-fuel), while strategies predicted by previous methods lead to zero guaranteed production rates.