Using Bayesian and evolutionary statistical learning to integrate temperature dependence in enzyme-constrained GEMs

Description of folders

• code/ contains all scripts and detailed descrition can be found in code/README.md.
• data/ contains all input data needed, including experimental and estimated thermal parameters.
• models/ contains a list of yeast genome scale models with different settings used in this study.

Dependences

austin 3.3.0 cobra 0.21.0 numpy 1.21.0 pandas 1.3.3 scikit-learn 1.0.0 scipy 1.7.1 reframed 1.2.1 dill 0.3.4 jupyter 1.0.0 matplotlib 3.5.2 Gurobi 9.1.2 pebble 5.0.0 The repository was tested with Python 3.8.12 and Ubuntu 20.04.4. The easiest way to install the dependencies is through the Conda package manager. Using Conda, the environment can be set up by running conda env create --file .condaconfig.yml.

Hardware

Since the Bayesian calculation method and evolutionary algorithm are computational expensive, all scripts except the visualization Jupyter notebook may be too heavy to run of a desktop computer. Some of those scripts have been designed for parallel computation through the use of SLURM. The visualization notebook takes several seconds or minutes on a normal PC.

Reproduce the figures

(1) Clone this repository.
(2) Install all required packages. This step takes at most several minutes. (3) Download the pre-computed results from Figshare (doi.org/10.6084/m9.figshare.21436668). Download the results.tar.gz file to the current directory and uncompress with tar -xzvf results.tar.gz Then the figures in the manuscript can be reproduced by using Jupyter notebook, Note that the result repository does not contain the heavy intermediate files which are not directly required for creating the figures.

One can also recompute those results by following the introductions in code/README.md, but again this might be infeasible on a usual desktop computer.