Predict microbial metabolic traits from genomes

Zeqian Li, Ahmed Selim, Seppe Kuehn

Last update: Mar 24th, 2024

Files

• snakemake: snakemake pipeline for bioinformatic pipeline and fba simulation. See snakemake/README.
• organize_data.ipynb: organize the three datasets
• figures.ipynb: Jupyter notebook to generate figures
  • figures_*.ipynb: Jupyter notebook to generate specific panels
• figures_SI.ipynb: code to generate SI figures
• workflows/gapfill: snakemake pipelines to study the effects of gap-filled carbons.

Data

Data files are deposited to the Open Science Framework (OSF)

Setup

1. If you use conda environments or pyenv, create and switch to a new environment.
2. Install packages

pip install -r requirements.txt

3. Download data

mkdir data && cd data
osf -p jwkr7 clone
mv -v data/jwkr7/osfstorage/* .
rm -rf jwkr7

figure_data.zip contains pre-trained data saved in pickle.

Citation

Li, Zeqian, Ahmed Selim, and Seppe Kuehn. "Statistical prediction of microbial metabolic traits from genomes." PLOS Computational Biology 19, no. 12 (2023): e1011705.

Contact

Zeqian Li (zeqianli.chicago@gmail.com)