Hi Sepideh,

You are saying that you dont have any experimental source for the organism, but I suggest you try databases like MMCD, KEGG etc. If you dont get the data then go for the specie with the maximum homology.

I think your approach is stoichiometry based, but if you can change your approach you can go for Constraint Based analysis.

Regarding the model validation you can try first to check if "known" pathways are funcitonal in your model (e.g. conversion of the carbon source to CO2 or fermentative formation of major (by)products such as ethanol or lactate proceeds as expected. Then you should check whether you can form biomass at appreciable yields (e.g. maximum of 40-50% carbon yield for a typical respiratory regime). In case cell growth does not work, test individual biomass precursors first.

There is no general rule as to how many rates you should measure to validate the model as this depends both on the network structure and on the cultivation conditions you try to model. Dedicated software tools like Insilico Discovery or CellNetAnalyzer can support you in such tasks as they also visualize the flux solutions graphically

Hi,

To answer your question, if you have 300 metabolic reactions available, and if you use all of them, then you would have 300 fluxes, one for each reaction in your model. As Dirk pointed out, there isn't a fast rule. It depends on what your modelling goals are.

As one example, if you are using a constraints-based approach, like the optimization technique Flux Balance Analysis, for example, your stoichiometric matrix of constraints will have that number of columns. The number or rows depends on the number of metabolites that take part in these reactions.

Also, even if you do not have any experimental source, try searching specific databases for your organism or closely related ones, like Amlan suggested.