Presentation Information
[4GteX-15]Computational Analysis of Metabolic Stability and Dynamics
○Yusuke HIMEOKA Himeoka1 (1. The University of Tokyo (Japan))
Keywords:
Kinetic modeling,Metabolic network,Microbial metabolism
In order to understand microbial metabolism, an in silico approach to modelling metabolic reaction networks is useful. Flux balance analysis (FBA) is one of the most impactful computational methods for elucidating what a given metabolic network can produce, based on the assumption of steady metabolic state and that cells regulate gene expression levels to maximize growth rate.
While FBA predictions are often useful, it is known that FBA can fail to capture metabolic responses due to its ignorance of metabolic dynamics. There are two types of steady state: stable and unstable. If a steady state is stable, the system is tolerant of small perturbations in metabolite concentrations. Conversely, if the steady state is unstable, the system cannot maintain the metabolic state against even infinitesimally small perturbations. Due to metabolic state instability, metabolic engineering strategies can fail, as confirmed both computationally and experimentally.
In order to construct the mathematical basis for next-generation metabolic engineering, the viewpoint of system stability is indispensable. To this end, we have carried out a series of studies on metabolic stability using kinetic models of cellular metabolism and dynamical systems theory.
We present the stability characteristics of E.coli central metabolism to small perturbations. Since kinetic models of cellular metabolism are expressed as non-linear ordinary differential equations, the computational expense of steady-state calculations is one of the main challenges of this type of study. To address this issue, we have developed an ultrafast steady-state sampler for metabolic kinetic models. Using this sampler, we can generate one trillion stable and unstable steady states by changing the parameter values within a day. Using this extensive set of steady states, we studied the stability statistics of the steady states. For example, we found that stability is closely related to the irreversibility of reaction fluxes. We found that some steady flux distributions are easy/hard to stabilize; there are plenty of stable steady states that realize a certain steady flux distribution, while the number of stable steady states that realize other steady flux distributions is quite limited. It also turned out that the frequency of stable steady states for each flux distribution could be inferred from the irreversibility of reaction fluxes, i.e. the forward-to-backward ratio of steady fluxes.
In my talk, I will briefly explain the method and demonstrate why the sampler allows such quick sampling. I will also talk about the stabilizability of flux distributions and the possible applications of the method in metabolic engineering.
While FBA predictions are often useful, it is known that FBA can fail to capture metabolic responses due to its ignorance of metabolic dynamics. There are two types of steady state: stable and unstable. If a steady state is stable, the system is tolerant of small perturbations in metabolite concentrations. Conversely, if the steady state is unstable, the system cannot maintain the metabolic state against even infinitesimally small perturbations. Due to metabolic state instability, metabolic engineering strategies can fail, as confirmed both computationally and experimentally.
In order to construct the mathematical basis for next-generation metabolic engineering, the viewpoint of system stability is indispensable. To this end, we have carried out a series of studies on metabolic stability using kinetic models of cellular metabolism and dynamical systems theory.
We present the stability characteristics of E.coli central metabolism to small perturbations. Since kinetic models of cellular metabolism are expressed as non-linear ordinary differential equations, the computational expense of steady-state calculations is one of the main challenges of this type of study. To address this issue, we have developed an ultrafast steady-state sampler for metabolic kinetic models. Using this sampler, we can generate one trillion stable and unstable steady states by changing the parameter values within a day. Using this extensive set of steady states, we studied the stability statistics of the steady states. For example, we found that stability is closely related to the irreversibility of reaction fluxes. We found that some steady flux distributions are easy/hard to stabilize; there are plenty of stable steady states that realize a certain steady flux distribution, while the number of stable steady states that realize other steady flux distributions is quite limited. It also turned out that the frequency of stable steady states for each flux distribution could be inferred from the irreversibility of reaction fluxes, i.e. the forward-to-backward ratio of steady fluxes.
In my talk, I will briefly explain the method and demonstrate why the sampler allows such quick sampling. I will also talk about the stabilizability of flux distributions and the possible applications of the method in metabolic engineering.
Comment
To browse or post comments, you must log in.Log in