/tb_pangenome_workflow

Reproducible workflow for analyzing the pan-genome of 1595 M. Tuberculosis strains

Primary LanguageJupyter NotebookMIT LicenseMIT

Workflow for statistical and genome-scale modeling of 1595 M. Tuberculosis strains

The study is divided into four sections.

  1. Pan-genome analysis

    1. Pan-genome partitioning
    2. Heaps Law applied to pan-genome

  2. Statistical determination of antimicrobial genes (Gene-AMR associations)

    1. Pairwise assocations with Mutual information
    2. Machine Learning with a Support Vector Machine

  3. iEK1020

    1. iEK1020.json (new reconstruction of M. tuberculosis H37Rv, iEK1020, in json format)
    2. iEK1020_supplementary.xlsx
      1. Sheet 1- Reactions in iEK1020 and details of reactions.
      2. Sheet 2- Metabolites in iEK1020 and details of metabolites.
      3. Sheet 3- Gene Essentiality simulation information
      4. Sheet 4- Various in silico media conditions
    3. SUPP_reproducible_modeling.ipynb: an ipython notebook that runs the simulations described in "iEK1020 simulates flux states indicative of physiological conditions"
    4. Escher Maps folder; contains four escher maps of M. tuberculosis metabolic subsystems

  4. Game of Allele Coordination (System-AMR assocations)

    1. SUPP_game_simulations.ipynb: an ipython notebook that sets up and simulates a game.
    2. dfdsa