Installation

ETFBA has been tested with Python versions 3.8, 3.9, 3.10 and 3.11. It can be installed using pip from PyPI:

python -m pip install --upgrade pip
pip install etfba

Alternatively, you can install it from source (assuming git is installed):

git clone https://github.com/Chaowu88/etfba.git /path/to/etfba
pip install /path/to/etfba

Note: It is recommended to install ETFBA within a virtual environment.

Solver installation

ETFBA uses the modeling language Pyomo to formulate linear programming (LP) and mixed integer linear programming (MILP) problems. You can install the freely available solver GLPK via conda:

conda install -c conda-forge glpk

For larger models, such as genome scale models, it is highly recommended to use the commercial optimizer Gurobi and install the Python support:

conda install -c gurobi gurobi

Example Usage

You can build an ETFBA model from scratch or convert it from a COBRA model (refer to here for more details). Below is an example of estimating the flux distribution constrained by enzyme protein allocation and thermodynamics:

from etfba import Model

model = Model.load('/path/to/model.bin')

res = model.optimize(
    'etfba',
    objective=objective,      # typically the growth rate
    flux_bound=flux_bound,    # bounds for metabolic fluxes
    conc_bound=conc_bound,    # bounds for metabolite concentrations
    preset_flux=preset_flux,  # preset values for specific metabolic fluxes
    preset_conc=preset_conc,  # preset values for specific metabolite concentrations
    ex_thermo_cons=ex_rxns,   # reactions excluded from thermodynamic constraint
    inc_enz_cons=eff_rxns,    # reactions included in enzyme protein constraint
    enz_prot_lb=enz_ub,       # upper bound on enzyme protein allocation
    parsimonious=True         # to obtain parsimonious flux distributions
).solve(solver='gurobi')

opt_growth_rate = res.opt_objective
opt_metabolic_fluxes = res.opt_fluxes

To estimate the variability of fluxes:

res = model.evaluate_variability(
    'etfva',
    objective=objective,
    obj_value=obj_value,   # optimal objective value obtained by "optimize"
    gamma=gamma,           # fraction of the optimum objective to achieve
    flux_bound=flux_bound,
    conc_bound=conc_bound,
    preset_flux=preset_flux,
    preset_conc=preset_conc,
    ex_thermo_cons=ex_rxns,
    inc_enz_cons=eff_rxns,
    enz_prot_lb=enz_ub
).solve(solver='gurobi', n_jobs=100)

metabolic_flux_ranges = res.flux_ranges

For more detailed information, please refer to the complete documentation.