Modelproperties

checkModelProperties(model, printLevel)[source]

calculates various generic properties of a model, not specific to any biochemistry

USAGE:

model = checkModelProperties (model, printLevel)

INPUT:

model –

COBRA model structure with at least the fields:

  • .S - m x n stoichiometric matrix

  • .mets - m x 1 cell array of metabolite identifiers

  • .rxns - n x 1 cell array of reaction identifiers

  • .lb - n x 1 lower bounds on reaction flux

  • .ub - n x 1 upper bounds on reaction flux

Read and reused if already present, otherwise computed and added to the output model (see OUTPUT below):

  • .SIntRxnBool - n x 1 boolean of reactions heuristically thought to be internal (non-exchange) reactions

  • .SIntMetBool - m x 1 boolean of metabolites heuristically thought to be involved only in internal reactions

  • .fluxConsistentMetBool - m x 1 boolean vector indicating flux consistent mets

  • .fluxConsistentRxnBool - n x 1 boolean vector indicating flux consistent rxns

Only read (and, if so, overwritten) within the coefficient scaling branch, which is disabled by default:

  • .b - m x 1 right hand side vector of S*v = b

  • .c - n x 1 linear objective coefficient vector

OPTIONAL INPUT:

printLevel – verbose level, silent if 0 (default 1)

OUTPUT:

model – the input model, augmented with the following fields:

  • .rankFRvanilla - rank of [F R], when using all rows and cols (F/R are the negative/positive parts of model.S)

  • .rankFRVvanilla - rank of [F;R], when using all rows and cols

  • .SIntRxnBool - n x 1 boolean of reactions heuristically thought to be internal (non-exchange) reactions

  • .SIntMetBool - m x 1 boolean of metabolites heuristically thought to be involved only in internal reactions

  • .SIntRxnBool_findSExRxnInd - n x 1 copy of .SIntRxnBool as first returned by findSExRxnInd, kept for comparison against any later stoichiometric-consistency-based refinement of .SIntRxnBool

  • .leakMetBool - m x 1 boolean of metabolites in a positive leakage mode

  • .leakRxnBool - n x 1 boolean of reactions exclusively involved in a positive leakage mode

  • .siphonMetBool - m x 1 boolean of metabolites in a negative leakage mode

  • .siphonRxnBool - n x 1 boolean of reactions exclusively involved in a negative leakage mode

  • .SConsistentMetBool - m x 1 boolean vector indicating stoichiometrically consistent mets

  • .SConsistentRxnBool - n x 1 boolean vector indicating stoichiometrically consistent rxns

  • .SInConsistentMetBool - m x 1 boolean vector indicating stoichiometrically inconsistent mets

  • .SInConsistentRxnBool - n x 1 boolean vector indicating stoichiometrically inconsistent rxns

  • .fluxConsistentMetBool - m x 1 boolean vector indicating flux consistent mets

  • .fluxConsistentRxnBool - n x 1 boolean vector indicating flux consistent rxns

  • .thermoFluxConsistentMetBool - m x 1 boolean vector indicating thermodynamically flux consistent mets

  • .thermoFluxConsistentRxnBool - n x 1 boolean vector indicating thermodynamically flux consistent rxns

  • .thermoFwdFluxConsistentRxnBool - n x 1 boolean vector indicating forward thermodynamically flux consistent rxns

  • .thermoRevFluxConsistentRxnBool - n x 1 boolean vector indicating reverse thermodynamically flux consistent rxns

  • .FRrows - m x 1 boolean of rows of [F R] that are nonzero, unique upto positive scaling and part of the maximal conservation vector, selecting the subset of mets used for rank analysis

  • .FRVcols - n x 1 boolean of cols of [F;R] that are nonzero, unique upto positive scaling and part of the maximal conservation vector, selecting the subset of rxns used for rank analysis

  • .FRnonZeroRowBool1 - m x 1 boolean of rows of [F,R] that are nonzero, on the first (checkTrivial) pass

  • .FRnonZeroColBool1 - n x 1 boolean of cols of [F;R] that are nonzero, on the first (checkTrivial) pass

  • .FRuniqueRowBool - m x 1 boolean of rows of [F,R] that are unique upto scalar multiplication

  • .FRuniqueColBool - n x 1 boolean of cols of [F;R] that are unique upto scalar multiplication

  • .FRnonZeroRowBool - m x 1 boolean of rows of [F,R] that are nonzero, for the final subset of mets/rxns

  • .FRnonZeroColBool - n x 1 boolean of cols of [F;R] that are nonzero, for the final subset of mets/rxns

  • .largestConnectedRowsFRBool - m x 1 boolean, currently set to true for all mets (connectedness test bypassed)

  • .largestConnectedColsFRVBool - n x 1 boolean, currently set to true for all rxns (connectedness test bypassed)

  • .Fr - the forward half of model.S, restricted to .FRrows and .FRVcols

  • .Rr - the reverse half of model.S, restricted to .FRrows and .FRVcols

  • .Fc - same restricted forward half of model.S as .Fr, used for the column-rank computation

  • .Rc - same restricted reverse half of model.S as .Rr, used for the column-rank computation

  • .rankFR - rank of [Fr Rr], i.e. [F R] restricted to .FRrows

  • .rankFRV - rank of [Fc;Rc], i.e. [F;R] restricted to .FRVcols

  • .FRp - row permutation returned by getRankLUSOL for [Fr Rr]

  • .FRq - column permutation returned by getRankLUSOL for [Fr Rr]

  • .FRVp - row permutation returned by getRankLUSOL for [Fc;Rc]

  • .FRVq - column permutation returned by getRankLUSOL for [Fc;Rc]

  • .FRirows - m x 1 boolean of rows of [Fr Rr] that are independent

  • .FRdrows - m x 1 boolean of rows of [Fr Rr] that are dependent

  • .FRVicols - n x 1 boolean of cols of [Fc;Rc] that are independent

  • .FRVdcols - n x 1 boolean of cols of [Fc;Rc] that are dependent

  • .FRrowRankDeficiency - number of dependent rows of [Fr Rr] (nnz(.FRrows) - .rankFR)

  • .FRcolRankDeficiency - number of dependent cols of [Fc;Rc] (nnz(.FRVcols) - .rankFRV)

  • .FRW - (only if .FRrowRankDeficiency > 0) sparse matrix expressing each dependent row of [F R] as a linear combination of the independent rows

  • .FRwrows - m x 1 boolean (only if .FRrowRankDeficiency > 0) of independent rows that other rows amongst .FRdrows depend on

  • .FRVW - (only if .FRcolRankDeficiency > 0) sparse matrix expressing each dependent col of [F;R] as a linear combination of the independent cols

  • .FRVwcols - n x 1 boolean (only if .FRcolRankDeficiency > 0) of independent cols that other cols depend on

  • .Frb - forward half of the bilinear decomposition of Fr - Rr (only set if the if 0 debug branch is enabled)

  • .Rrb - reverse half of the bilinear decomposition of Fr - Rr (only set if the if 0 debug branch is enabled)

  • .rankBilinearFrRr - rank of [Frb Rrb] (only set if the if 0 debug branch is enabled)

  • .rankS - rank of model.S

  • .maxSij - maximum magnitude stoichiometric coefficient in model.S

  • .minSij - minimum nonzero magnitude stoichiometric coefficient in model.S

makeModelPropertiesTable(modelResults, modelMetaData, resultsDirectory, resultsFileName, tableFilename)[source]

Makes a table of model property results

USAGE:

[modelResultsTable, modelResults] = makeModelPropertiesTable (modelResults, modelMetaData, resultsDirectory, resultsFileName, tableFilename)

INPUT:

modelResults – output of checkModelProperties

OPTIONAL INPUTS:

  • modelMetaData – Cell array, where each row is metadata for one model with five columns: species, modelID, fileName, PMID, doi. See function modelMetaData=modelCitations() for example. Table columns ordered by order of rows in modelMetaData.

  • resultsDirectory – directory where output of checkModelProperties has been saved

  • resultsFileName – filename where output of checkModelProperties has been saved

  • tableFilename – If provided, the table of results is written out to a csv file, with specified filename

OUTPUTS:

  • modelResultsTable – table displaying the results of checkModelProperties

  • modelResults – output of checkModelProperties

plotModelConsistency(modelResults, modelMetaData, schematicFlag, nRows, nCols, resultsDirectory, figureFileName, resultsFileName)[source]

Plots stoichiometric and flux consistency figures, given a modelResults structure or by loading the modelResults structure from a specified location

USAGE:

plotModelConsistency (modelResults, modelMetaData, schematicFlag, nRows, nCols, resultsDirectory, figureFileName, resultsFileName)

INPUT:

  • modelResults – output of checkModelProperties

  • modelMetaData – Cell array, where each row is metadata for one model with five columns: species, modelID, fileName, PMID, doi.

  • schematicFlag – top corner plot illustrating the different colours

  • nRows – number of rows in the subplot

  • nCols – number of rows in the subplot (nRows*nCols >= length(modelResults))

OPTIONAL INPUTS:

  • resultsDirectory – directory where output of checkModelProperties has been saved, same directory where the figure will be saved

  • figureFileName – filename of the figure (without the directory)

  • resultsFileName – filename where output of checkModelProperties has been saved

printBiomass(model, BiomassNumber)[source]

Prints the metabolites and their stoichiometric coefficients that participate in a given reaction (typically the biomass reaction) and returns them as well

USAGE:

[Component, Fraction] = printBiomass (model, BiomassNumber)

INPUTS:

  • model – COBRA model structure with fields:

    • .S - m x n stoichiometric matrix (used if .A is absent)

    • .A - stoichiometric/coupling matrix, used in place of .S when present

    • .mets - m x 1 cell array of metabolite identifiers

  • BiomassNumber – column index of the biomass reaction (or any other reaction) in model.S

OUTPUTS:

  • Component – cell array of metabolite identifiers with a nonzero coefficient in reaction BiomassNumber

  • Fraction – stoichiometric coefficients of Component in reaction BiomassNumber