shyamsrinivasan / mfa_design Goto Github PK

Might have to notice ATP synthase KO in aerobic condition. For implementation on E. coli core model, the model switch to acetate and formate fermentation. Having pseudo-anaerobic flux distribution is dangerous when determine kinetic parameters for aerobic condition (https://www.nature.com/articles/ncomms13806).

There are three choices for these:

1. (currently implemented) Take difference of absolute flux values for two mutants
   sqrt[ sum[ abs(v1_j) - abs(v2_j) ]^2 ]
2. Difference of flux values for two mutants
   sqrt[ sum[ v1_j - v2_j ]^2 ]
3. Split to two reactions and consider them separately, thus there are two separate term in the sum:
   abs(v1_j_fwd) - abs(v2_j_fwd)
   abs(v1_j_rev) - abs(v2_j_rev)

Finding mutants' flux vectors using MOMA in relation to experimentally measured flux might be more appropriate. This method will possibly be developed in a next update of the script.

Since the goal is to find a set(s) of mutants with largest dispersion (using norm-2 distance) to generate MFA data used in reconstruction of (near) genome-scale kinetic model, the reaction flux (v_j) used in calculate the norm-2 distance might have to be restricted to the set of reactions that will be in the kinetic model.