The PanGEM Toolbox is designed for the reconstruction of strain-level genome-scale models (GEMs) of metabolism, using a pangenome as a template. A series of steps are taken which guide the user from database creation, to draft model reconstruction, to annotation, to testing, statistical summary and visualization. All available genomes from a taxonomic bin are compiled and used to generate a pangenome metabolic network, from which functional, mass and charge balanced and stoichiometrically consistent strain models can be automatically generated.

What’s needed to run PanGEM:

  • Genomes representing a taxonomic bin (typically a genus or species)
  • Time! One of the steps requires significant manual curation and is better guided by other software (like COBRA or RAVEN)