PAModelpy - Protein Allocation Model reconstruction in Python

What is PAModelpy?

Models of metabolism are powerful tools to explore the metabolic potential of microorganism. Powerful tools have been created to support Python-based analysis of genome-scale models. These models, however, cannot capture all metabolic phenotypes and the simulation results have high flux variability. Adding protein to each reaction increases the simulation fidelity. The PAModelpy package is designed to integrate protein constraints and protein sectors as described by Alter et al. (2021) to metabolic models. It is the Python implementation of the PAM MATLAB framework to create GECKO like protein-constrained models.

The PAModelpy package builds upon the community-wide used COBRApy. We have extended this package with the following features:

  • protein-reaction associations

  • infrastructure to include isozymes and promiscuous enzymes

  • protein sectors

  • specialized objects to build protein allocation models

  • the possibility to perform a computational efficient sensitivity analysis

Installation

IMPORTANT: PAModelpy is only compatible with Python >= 3.9 and <=3.11

PAModelpy is a PiPy package which allows for easy installation with pip:

pip install PAModelpy

Note that the package has been tested with the Gurobi solver. In order for Gurobi to work properly, please install gurobipy with a version matching your license.

For example for the version used for the development of PAModelpy:

pip install gurobipy==9.5.2

What can you find where in this repository?

This repository contains not only the source code, but also examples and scripts which were used in van den Bogaard et al. (2024).

  • Data

    • eGFP_expression_Bienick2014: measured growth rate and eGFP expression by Bienick et al. (2014)

    • proteinAllocationModel_iML1515_EnzymaticData_py: information about the proteinsectors of the PAM for Escherichia coli (E.coli)

    • proteome_data_extract_schmidt2016: quantitative proteomics data from Schmidt et al. (2016) used to parametrize the E.coli core PAM

    • Ecoli_phenotypes/Ecoli_phenotypes_py_rev: experimental physiology measurements to validate the model simulations

  • Examples: example notebook on how to build, run and validate a PAM using the PAModelpy package

  • Figures: scripts used to create Figure 1-3 and supplementary figures

  • MATLAB: MATLAB code for doing simulations with the E. coli core PAM and the toy model (validating the sensitivity relationships)

  • Models: models used (iML1515 and core E. coli model)

  • Results: results of computational performance analysis

  • Scripts: scripts used for gathering results

    • computational performance analysis: compare_computational_efficiency_esc.py and numeric_error_estimation_schemes_esc.py

    • E.coli core PAM creation: analyze_proteome.ipynb and create_ecolicore_pam_incl_UE.ipynb

    • Sensitivity analysis: toy_ec_pam.py and Ecoli_core_sensitivity_analysis.ipynb

  • src/PAModelpy: source code for PAModelpy package

Code structure:

  • EnzymeSectors: The objects which are used to store the data of the different enzyme sectors which are added to the genome-scale model

  • PAModel: Proteome Allocation (PA) model class. This class builds on to the cobra.core.Model class from the COBRApy toolbox with functions to build enzyme sectors, to add enzyme kinetics parameters and in the future to perform a sensitivity analysis on the enzyme variables.

  • Enzyme: Different classes which relate enzymes to the model with enzyme constraints and variables.

  • CatalyticEvent: A class which serves as an interface between reactions and enzyme. This allows for easy lookup of Protein-Reaction assocations.

  • PAMValidator: Functions to validate the model predictions with physiology data and giving a graphical overview. The script uses data for E.coli (found in ./Data/Ecoli_physiology) by default.

For the technical users, the following UML diagram gives an overview of the model structure and the software architecture. Please be aware that the Config object is not shown in this UML diagram, as this object is merely used to transfer identifiers from one object to another in the software.

PAModelUML

UML diagram of the PAModelpy software. Not all attributes and functions are shown. Those who are shown are assumes to be most descriptive for the objects function.

Dependencies

IMPORTANT: PAModelpy is only compatible with Python >= 3.9 and <=3.11

Dependencies for the scripts in this repository, not included in the PAModelpy package:

  • PAModelpy

  • gurobipy

  • plotly

  • matplotlib

  • scipy

  • time

  • resource

  • PIL

All dependencies can be installed in one go by downloading this repository and running:

python setup.py install

from the src directory

The dependencies of the PAModelpy package can be found in src/pyproject.toml

License

Copyright institute of Applied Microbiology, RWTH Aachen University, Aachen, Germany (2024)

PAModelpy is free of charge open source software, which can be used and modified for your particular purpose under the MIT or Apache 2.0 of the users choice.

Please note that according to these licenses, the software is provided ‘as is’, WITHOUT WARRANTY OF ANY KIND, without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.