1, install the COBRA2.0 toolbaox from palsson lab, and initializing the COBRA tools using command line:
initCobraToolbox
2, install the matlab toolbox SpectraLib_A for clustering analysis;
3, install the matlab toolbox Fast_SNP for solving sparse basis vectors;
Necessary:
You should first prepare your input files, example is as follows:
1, the genome-scale metabolic model of given species, e.g. bacillus_iYO844
2, the cofactor metabolites, e.g. bacillus_iUO844_cofactors
3, the elementary enviroment-cell exchange reactions, e.g. bacillus_iUO844_exchanges
4, the secreted reactions, e.g. bacillus_iUO844_secretion
5, the nutrient uptake reactions, e.g. bacillus_iUO844_nutrient
Optional:
1, experimental 13C measured reaction fluxes for model validation, e.g. bacillus_iUO844_13C_flux
2, experimental extracellular uptake/secrete rate, e.g. the carbon uptake, O2 rate, CO2 rate, actate or ethonal secretion et.al. which has a same file format with the 13C flux file.
3, the gene knockout list, each row should only store one knockout gene.
1, bulid the topologically-decoupled metabolic model
1, directly running the Decrem_Demo.m for bacillus iYO844 model test or modifing the default input files as user self-defined data.
Decrem_Demo
2, bulid the topologically-decoupled metabolic model and predict metabolic fluxes or growth rate using command line:
[Decrem_solution,Decrem_model,CBM_model] = Decrem(CBM_Model,cofactors,input_nutrient,secretion,general_IO,cluster_num,extraflux(:,[1,i]),intraflux(:,[1,i]),knockout_genes);
for given CBM_model model,e.g. bacillus iYO844 model or replace all the data in bacillus filefold with user self-defined data
three reconstructed decoupled metabolic models for Escherichia coli, Saccharomyces cerevisiae and Bacillus subtilis are found in the filefold of three reconstructed models
** Li, G., Liu, L., Du, W. et al. Local flux coordination and global gene expression regulation in metabolic modeling. Nat Commun 14, 5700 (2023). https://doi.org/10.1038/s41467-023-41392-6**