Simulating pharmacologic inhibition through FBA?

[agirgis3]

Hi there, thanks for developing this tool! I wanted to ask about a potential use for FBA. If I wanted to model, say inhibition of a specific metabolic enzyme, would it be conceptually reasonable to manually edit the RNA count information for that gene (ie set expresison level to zero or something), and feed this modified counts data into FBA?

From my understanding, this will modify reaction penalties and the corresponding flux of related reactions accordingly and in a way simulate pharmacologic intervention, correct? Would this approach be reasonable? If this violates some fundamental assumptions of the algorithm, please let me know.

Thanks!