Parsimoniousfba

pFBA(model, varargin)[source]

Parsimoneous enzyme usage Flux Balance Analysis - method that optimizes the user’s objective function and then minimizes the flux through the model and subsequently classifies each gene by how it contributes to the optimal solution. See Lewis, et al. Mol Syst Bio doi:10.1038/msb.2010.47

USAGE:

[GeneClasses, RxnClasses, modelIrrevFM, MinimizedFlux] = pFBA (model, varargin)

INPUTS:

  • model – COBRA model structure with fields:

    • .rxns - n x 1 reaction identifiers

    • .genes - g x 1 gene identifiers

    • .c - n x 1 objective coefficients

    • .lb - n x 1 lower flux bounds

    • .rxnGeneMat - n x g reaction-gene incidence matrix

  • varargin – parameter name / value pairs:

    • ‘geneoption’ - 0 = minimize the sum of all fluxes in the network, 1 = only minimize the sum of the flux through gene-associated fluxes (default), 2 = only minimize the sum of the flux through non-gene-associated fluxes

    • ‘tol’ - tolerance below which a flux is treated as zero (default 1e-6)

    • ‘map’ - map structure from readCbMap.m (no map written if empty)

    • ‘mapoutname’ - File Name for map

    • ‘skipclass’ - 1 = Don’t classify genes and reactions. Only return model with the minimium flux set as upper bound. 0 = classify genes and reactions (default).

OUTPUTS:

  • GeneClasses – Structure with fields for each gene class

  • RxnClasses – Structure with fields for each reaction class

  • modelIrrevFM – Irreversible model used for minimizing flux with the minimum flux set as a flux upper bound

  • MinimizedFlux – minimized flux

Note

Note on maps: Red (6) = Essential reactions, Orange (5) = pFBA optima reaction, Yellow (4) = ELE reactions, Green (3) = MLE reactions, blue (2) = zero flux reactions, purple (1) = blocked reactions, black (0) = not classified

Example

[GeneClasses, RxnClasses, modelIrrevFM] = pFBA(model, ‘geneoption’, 0, ‘tol’, 1e-7)