(507e) Estimation of Genome-Scale Metabolic Fluxes From Tn-Seq Single Mutant Growth Measurements

Metabolic flux is the key to understand both wild-type and mutant metabolic phenotypes. Minimization of metabolic adjustment (MOMA) has been extensively used to estimate mutant flux maps by choosing the nearest feasible flux solution to a wild-type flux map. However, a genome-wide wild-type flux map is never available, and an FBA solution space is usually used instead of a unique solution. MOMA predictions are therefore a solution space as well, where each prediction is a function of the location in the FBA solution space from which the MOMA projection is made.  

To overcome this limitation, we developed a novel algorithm to calculate the wild-type flux map by minimizing the difference between predicted mutant growth rates (using MOMA) and measured mutant growth rates from Tn-seq data. Using the estimated wild-type flux solution, we compared mutant growth predictions for a control set to predictions made by sampling the FBA solution space. The Tn-seq data was acquired for the Shewanella oneidensis strain. More generally, this work provides a new approach to calculate the unique wild-type flux map which provides better predictions for mutant phenotypes.