Reactions flux analysis
Bruno Ribeiro Pinto
Hello everyone, how are you?
I would like to ask two questions. First: How can I compare the flux of a specific reaction, which is not my objective function, under two different conditions? For example, wild type and knock-out. Is it correct to use FBA and examine the individual flux of a reaction, or are other approaches such as flux sampling or FVA more suitable?
And second: For instance, under one condition my model secretes formate, but after performing a knock-out of a specific reaction, my model reduces formate secretion. What types of analyses can I perform to try to determine the mechanism that may explain the reason for this decrease in formate secretion?
Luis Vitores Valcárcel García
How are the conditions different? If there is a change in nutrients or different or knock-down of several reactions, and the objective function is not sure, you can use sampling techniques:
Connecting extracellular metabolomic measurements to intracellular flux states in yeast
https://link.springer.com/article/10.1186/1752-0509-3-37
with these alternatives you can exploe the changes in all reactions of the system, to look for the most different ones
Kind regards,
Luis V. Valcárcel
Matthias König
Hi Bruno,
> Is it correct to use FBA and examine the individual flux of a reaction, or are other approaches such as flux sampling or FVA more suitable?
A major challenge in constraint-based modeling is the non-uniqueness of solutions. The fluxes of most internal reactions are typically not unique, meaning it’s usually not meaningful to compare individual flux values directly—except for the objective function or certain exchange reactions.
In addition, reproducibility can become an issue: when multiple optimal solutions exist, solvers are free to return any of them. This means that running the same optimization twice can result in different flux distributions, even though all are equally valid. Most solvers implement internal heuristics to consistently pick one solution, which can give the false impression that individual flux values have an intrinsic meaning—when in fact they do not.
To obtain unique or interpretable flux values, you can introduce additional constraints or use iterative approaches, such as parsimonious FBA (pFBA), which minimizes the total flux as a secondary objective.
However, a more robust approach is to perform Flux Variability Analysis (FVA), which provides flux ranges for each reaction, revealing the feasible solution space. Alternatively, flux sampling allows you to explore and compare distributions of feasible fluxes, offering a better understanding of variability across solutions.
> What types of analyses can I perform to try to determine the mechanism that may explain the reason for this decrease in formate secretion?
You could perform gene or reaction knockout analyses to identify which genes or reactions most strongly affect your target—formate secretion in this case. This can help pinpoint mechanisms responsible for the observed decrease.
Best,
Matthias
Ronan M.T. Fleming
--
---
You received this message because you are subscribed to the Google Groups "COBRA Toolbox" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbo...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cobra-toolbox/939a5ad9-86f2-43bd-be17-ac5c5cc3c4b6n%40googlegroups.com.
Mr. Ronan MT Fleming B.V.M.S. Dip. Math. Ph.D.
----------------------------------------------------------------------------
Associate Professor,
University of Galway.
Peer-reviewed publications: https://goo.gl/FZPG23
Mobile: +353 852 109 806
Github: @rmtfleming
----------------------------------------------------------------------------