RidgeFBA_test.jl

# This script can be used to test the funcionality of RidgeFBA 
# Here, the metabolic model 'e_coli_core' is  downloaded and by assumption of biomass as objective function,
# flux dixtribution is computed and the value for objective function is tested

using JuMP, Ipopt
using SBML
using HTTP
using Test

# Downloading the metabolic model
ecoli_model=HTTP.get("http://bigg.ucsd.edu/static/models/e_coli_core.xml")
write("e_coli_core.xml",ecoli_model.body)
ecoli_metabolic_model=readSBML("e_coli_core.xml")


#The name of the reactions are extracted as a vector.
reaction_names_keySet=keys(ecoli_metabolic_model.reactions)
reaction_names=[]
for item in reaction_names_keySet
    push!(reaction_names,item)

end

#Obtaining the biomass reaction index
for n in 1:length(ecoli_metabolic_model.reactions)
    
    if occursin("BIOMASS" , reaction_names[n])
        global biomass_index=n
    end
end


#As biomass is set as objective function the corresponding element in c vector is set to 1
c_test=zeros(1,length(ecoli_metabolic_model.reactions))
c_test[biomass_index]=1



flux_vector=RidgeFBA(ecoli_metabolic_model,c_test,10^-5)                                                          

                                                           
@test isapprox(flux_vector[biomass_index] ,46.17244338638067; atol=0.001)