Modelgeneration

• Massbalance
• Modelproperties
• Fluxconsistency
  • Greedy
  • Dca
• Stoichconsistency
• Modelverification

checkCobraModelUnique(model, renameFlag)

Checks uniqueness of reaction and metabolite names

USAGE:

model = checkCobraModelUnique (model, renameFlag)

INPUT:

model – COBRA model structure with fields:

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

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

OPTIONAL INPUT:

renameFlag – Renames non-unique reaction names and metabolites (Default = false)

OUTPUT:

• model – COBRA model structure

• isUnique – true if the model has unique reaction and metabolite names

checkDuplicateRxn(model, method, removeFlag, printLevel, boundsFlag)

Checks model for duplicate reactions and removes them. By default, it detects the columns of S that are identical upto scalar multiplication

USAGE:

[modelOut, removedRxnInd, keptRxnInd] = checkDuplicateRxn (model, method, removeFlag, printLevel, boundsFlag)

INPUTS:

model – Cobra model structure with fields:

• .S - m x n stoichiometric matrix

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

• .lb - n x 1 lower bounds on fluxes (used if boundsFlag)

• .ub - n x 1 upper bounds on fluxes (used if boundsFlag)

OPTIONAL INPUTS:

• method – S –> checks rxn S matrix (default), rxnAbbr –> checks rxn abbreviations, FR –> checks F + R matrix, where \(S:=-F + R\), which ignores reaction direction

• removeFlag – {(1), 0} if 1, duplicate reactions are removed from modelOut; if 0, they are only detected and reported (default 1)

• printLevel – verbose level (default 0)

• boundsFlag – {(0), 1} if 1, include .lb and .ub when comparing reactions for duplication (default 0)

OUTPUTS:

• modelOut – COBRA model structure without (with) duplicate reactions

• removedRxnInd – Reaction numbers in model that were (should be) removed

• keptRxnInd – Reaction numbers in model that were (should be) kept

checkModelPreFBA(model, param)

Checks if a model is (stoichiometrically and flux) consistent, which are necessary conditions prior to FBA

USAGE:

isConsistent = checkModelPreFBA (model, param)

INPUTS:

model – COBRA model structure, or a char with the file name of a model, with fields:

• .modelID - model identifier (char)

• .S - m x n stoichiometric matrix

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

• .SIntRxnBool - n x 1 boolean of internal (non-exchange) reactions (returned by findStoichConsistentSubset)

• .SConsistentRxnBool - n x 1 boolean of stoichiometrically consistent reactions (returned by findStoichConsistentSubset)

• .SConsistentMetBool - m x 1 boolean of stoichiometrically consistent metabolites (returned by findStoichConsistentSubset)

• .EXRxnBool - n x 1 boolean of exchange reactions (computed internally from reaction identifiers prefixed EX_)

• .DMRxnBool - n x 1 boolean of demand reactions (computed internally from reaction identifiers prefixed DM_, excluding named ATP maintenance reactions)

• .SinkRxnBool - n x 1 boolean of sink reactions (computed internally from reaction identifiers prefixed sink_)

OPTIONAL INPUTS:

param – parameters structure with fields:

• .printLevel - verbose level (default 0)

• .epsilon - set internally before testing flux consistency with findFluxConsistentSubset

• .modeFlag - set internally before testing flux consistency with findFluxConsistentSubset

• .method - set internally before testing flux consistency with findFluxConsistentSubset

OUTPUT:

isConsistent – {(1), 0} 1 if the model is stoichiometrically and flux consistent, 0 otherwise

checkObjective(model)

Prints out the Stoichiometric Coefficients for each Metabolite, with the name of the objective

USAGE:

objectiveAbbr = checkObjective (model)

INPUT:

model – COBRA model structure with fields:

• .c - n x 1 objective coefficient vector

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

• .S - m x n stoichiometric matrix

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

OUTPUT:

objectiveAbbr – Objective reaction abbreviation

detectDeadEnds(model, removeExternalMets)

Returns a list of indices of metabolites which either participate in only one reaction or can only be produced or consumed (i.e. all reactions involving the metabolite either only produce or only consume it, respecting the reaction lower and upper bounds).

USAGE:

mets = detectDeadEnds (model, removeExternalMets)

INPUT:

model – COBRA model structure with fields:

• .S - m x n stoichiometric matrix

• .lb - n x 1 lower bounds on fluxes

• .ub - n x 1 upper bounds on fluxes

• .SInConsistentRxnBool - n x 1 boolean of stoichiometrically inconsistent reactions (only if removeExternalMets; returned by findStoichConsistentSubset)

• .SExMetBool - m x 1 boolean of exchanged/external metabolites (only if removeExternalMets; returned by findStoichConsistentSubset)

OPTIONAL INPUT:

removeExternalMets – Dont return metabolites that participate in reactions of the following type: “A <=>/-> ” or ” <=>/-> A” or exclusively present in inconsistent reactions as defined in Gevorgyan et al, Bioinformatics, 2008

OUTPUT:

mets – List of indicies of metabolites which can ether only be produced or consumed.

fastLeakTest(model, testRxns, demandTest)

Tests if any metabolites in a model are leaking. A metabolite is leaking if the exchange reaction can carry secretion flux in the closed model (no uptake flux through any exchange reactions is permitted).

USAGE:

[LeakMets, modelClosed, FluxExV] = fastLeakTest (model, testRxns, demandTest)

INPUTS:

• model – Model structure with fields:

  • .S - m x n stoichiometric matrix

  • .lb - n x 1 lower bounds on fluxes

  • .ub - n x 1 upper bounds on fluxes

• testRxns – List of exchange reactions to be testetd for leaks

• demandTest – Optional: if ‘true’ is entered, demand reactions for all metabolites in the model are created

OUTPUTS:

• LeakMets – List of exchange reactions for leaking metabolites

• modelClosed – Model strucutre that has been tested for leaks

• FluxExV – Flux vector for computed exchange reactions in the closed model

printObjective(model)

Prints out the Stoichiometric Coefficients for each Metabolite, with the name of the objective

USAGE:

objectiveAbbr = printObjective (model)

INPUT:

model – COBRA model structure with fields:

• .c - n x 1 objective coefficient vector

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

• .S - m x n stoichiometric matrix

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

OUTPUT:

objectiveAbbr – Objective reaction abbreviation

removeCompartments(model, comp, newComp)

Removes a list of compartments from metabolite names, decompartmentalizing the model. Removes duplicate and empty entries.

USAGE:

[modelDecomp] = removeCompartments (model, comp, newComp)

INPUTS:

• model – Model structure with fields:

  • .genes - g x 1 cell array of gene identifiers

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

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

  • .S - m x n stoichiometric matrix

  • .lb - n x 1 lower bounds on fluxes

  • .ub - n x 1 upper bounds on fluxes

  • .c - n x 1 objective coefficient vector

  • .rev - n x 1 boolean of reversible reactions (computed from .lb if absent)

  • .subSystems - n x 1 cell array of subsystem assignments

  • .grRules - n x 1 cell array of gene-reaction rules

• comp – List of compartments to remove or replace (e.g., {‘c’, ‘n’})

OPTIONAL INPUTS:

newComp – New compartment (default: ‘[c]’)

OUTPUT:

modelDecomp – Decompartmentalized model structure

removeDeadEnds(model)

Removes all dead end metabolites and reactions from the model

USAGE:

[model, removedMets, removedRxns] = removeDeadEnds (model)

INPUT:

model – COBRA model structure

OUTPUTS:

• model – COBRA model structure w/o dead end metabolites and reactions

• removedMets – List of removed metabolites

• removedRxns – List of removed reactions

setConstraintsIrrevModel(constrOpt, model, modelIrrev, rev2irrev)

Sets constraints for a subset of rxns while converting reversible to irreversible reaction names and handling the constraint directions correctly

USAGE:

constrOptIrrev = setConstraintsIrrevModel (constrOpt, model, modelIrrev, rev2irrev)

INPUTS:

• constrOpt – Constraint options with fields:

  • .rxnList - Reaction selection cell array (for reversible representation)

  • .values - Constraint values

  • .sense - Constraint senses ordered as rxnNameList

• model – Model in reversible format with field:

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

• modelIrrev – Model in irreversible format with field:

  • .rxns - cell array of irreversible reaction identifiers

• rev2irrev – Reversible to irreversible reaction index conversion obtained from convertToIrreversible

OUTPUTS:

constrOptIrrev – Constraint options in irrev model with fields:

• .rxnList - Reaction selection cell array

• .rxnInd - Selection index for constraints in irreversible model (e.g. [2 4 5 9 10])

• .values - Correctly ordered constraint values

• .sense - Correctly ordered constraint senses

test4HumanFctExt(model, test, optionSinks)

test for the ~288 human functions

USAGE:

[TestSolution, TestSolutionName, TestedRxns, PercTestedRxns] = test4HumanFctExt (model, test, optionSinks)

INPUT:

• model – model structure (Recon1, with desired in silico condition) with fields:

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

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

  • .c - n x 1 objective coefficient vector (reset for each test)

  • .SinkRxnBool - n x 1 boolean of sink reactions (used if optionSinks)

• test – possible statements: Recon1, IECori, IEC, all (default) (choose IECori if you intend to test the IEC model OR a model that contains lumen (‘u’) as compartment otw choose IEC); all check for Recon1 and IEC

• optionSinks – if true = set sink reactions to 0 (default, leave unchanged). Note that all lb’s of exchanges and demands will be set to 0

OUTPUT:

• TestSolution – array containing the optimal value for the different tests

• TestSolutionName – array containing the names for the different tests

• TestedRxns – cell array of reaction identifiers that carried flux (above tolerance) in any of the tests

• PercTestedRxns – percentage of model.rxns contained in TestedRxns

testATPYieldFromCsources(model, modelName, extraCellCompIn, extraCellCompOut, minCard)

computes the ATP yield from various carbon sources in Recon2 or Recon3.

USAGE:

[Table_csources, TestedRxns, PercTestedRxns] = testATPYieldFromCsources (model, modelName)

INPUT:

model – model structure with field:

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

OPTIONAL INPUTS:

• modelName – name of the model structure, by default Recon3

• extraCellCompIn – abbreviation for extracellular compartment (in-going), default [e]

• extraCellCompOut – abbreviation for extracellular compartment (out-going), default [e]

• minCard – {(0), 1} default false for option to minimize the cardinality of the flux vector

OUTPUT:

• Table_csources – table listing ATP yield computed for the carbon sources

• TestedRxns – list of reactions that are contributing to ATP production from carbon sources

• PercTestedRxns – Fraction that tested reactions make up compared with all reactions in model