Installation

Creating your virtual environment

When working with Python projects, it is highly recommended to use a virtual environment. Virtual environments help to: - Isolate dependencies - Avoid version conflicts

An environment (example: env_test_cFBA) can easilly be created: .. code-block:: python

conda create –name env_test_cFBA

Once this is created, you can activate said environment: ..

conda activate env_test_cFBA

If you want to work with Jupyter Notebooks you need to add this environment: ..

conda install jupyter
conda install ipykernel
python -m ipykernel install --user --name env_test_cFBA --display-name "Python (env_test_cFBA)"

Now you have your own custom enviornment where yo can install packages, change versions, etc. without affecting your main environment.

Installation of py-cfba

Our package is available at PyPI, so it can be easilly installed with the following command: ..

pip install py-cfba

Once this is installed correctly, you can import the functions of this package simply as: ..

import py_cfba as cFBA

Usage

Requirements

  • Stoichiometric Matrix that represents the biochemical reactions of the metabolic network.

  • Imbalanced and Balanced Metabolites to be simulated.

  • When Enzyme Capacities will be tested, you need to know the Enzyme-Reaction Mapping with their corresponding kcat values.

Step-by-Step instructions to get started

  1. Import the py_cFBA module into your Python environment:

    import py_cfba as cFBA

  2. (Optional) Generate the basic model structure in excel using the cFBA_backbone_from_S_matrix function. It only requires an S matrix and the function will guide you to input required data.

  3. Include additional information to the model structure in excel. Parameters to fill out: w_matrix weights, upper and lower bounds, enzyme capacities.

  4. Generate an SBML file with the excel_to_sbml function.

  5. Define (if needed) quotas for the simulation.

  6. Test the model works with a µ = 1 (no growth), by using the function create_lp_problem, using an alpha = 1.

  7. Apply the optimization algorithm by using the find_alpha function.

  8. Retrieve the simulation fluxes and amounts with get_fluxes_amounts function.

These general steps are followed in the two examples avaialable in this documentation.

cFBA functions

The following are the general functions within the method and how you can apply them:

  • cFBA_backbone_from_S_matrix(S_matrix): Function to help you generate an Excel backbone for a cFBA model based on the provided Stoichiometric matrix. Alternatively, you can generate your own model structure following the template.

  • excel_to_sbml(excel_file, output_file): Takes the excel file in the format from the previous function and generates an SBML file for the cFBA method.

  • generate_LP_cFBA(sbml_file, quotas): Generates the basic structure of the cFBA problem.

  • create_lp_problem( alpha, cons , Mk, imbalanced_mets ): Tests the linear program given a growth of the system (alpha = µ)

  • find_alpha(cons, Mk, imbalanced_mets): Optimization algorithm searching for the largest growth of the system (alpha = µ)

  • get_fluxes_amounts(sbml_file, prob): Retrieves the fluxes and metabolites concentrations in the optimal simulation