M.Sc Thesis | |

M.Sc Student | Srour Orr |
---|---|

Subject | Fluxomers: A New Approace for 13C Metabolic Flux Analysis |

Department | Department of Electrical and Computer Engineering |

Supervisor | PROF. Yonina Eldar |

Full Thesis text |

The ability to quantify intracellular metabolic fluxes is critical for detecting pathway bottlenecks and elucidating control architectures in biochemical networks. In this regard, stoichiometric balance equations provide a set of linear constraints that describe all feasible flux states. Unfortunately, this set of equations is typically underdetermined, yielding an infinite number of possible values for the metabolic fluxes. In order to fully quantify intracellular metabolic fluxes, various experimental procedures have been developed, exposing hidden features of the metabolic network and thus adding additional equations to the stoichiometric set.

One of these methods is ^{13}C
metabolic flux analysis (MFA), in which the isotopic labeling patterns (i.e.,
isotopomers) of downstream metabolites are measured in response to the
introduction of a 13C-labeled substrate. Mathematically, these experiments are
traditionally analyzed using separate variables representing the fluxes and
labeling states of the metabolic system. Analysis of tracer experiments using
this set of variables results in a difficult non-convex optimization problem,
suffering from both implementation complexity and various convergence issues.
Furthermore, the error model is relaxed in order to ease the optimization
process, which detracts from the statistical quality of the results and
necessitates estimation of additional measurement normalization factors.

This work addresses the mathematical
and computational problems of ^{13}C MFA experiments using a new set of
variables referred to as fluxomers. Combining both fluxes and isotopomers,
fluxomer variables result in a simply-posed problem with a corrected error
model that enables the analysis of incomplete measurement data without the need
to estimate additional normalization factors. A simple yet powerful fluxomer
iterative algorithm (FIA) is constructed to solve the MFA optimization problem.
The algorithm is shown to outperform the commonly used 13CFLUX and the modern
OpenFLUX algorithms in both convergence time and variances of the results under
different scenarios.

Implementation of the FIA algorithm has been developed and published for common usage as an open-source software project on our web-site. This software shows substantial improvements in convergence time and results quality, along with the ability to analyze experiments that were impossible to analyze previously.