feat: convertCharArray function

- all functions in this commit are tested by a unit_test
SysBioChalmers · May 1, 2022 · ff462fd · JonathanRob · May 26, 2022 · edkerk
1 parent 1ad1109
commit ff462fd
Show file tree

Hide file tree

Showing 45 changed files with 785 additions and 216 deletions.
diff --git a/.gitattributes b/.gitattributes
@@ -3,7 +3,7 @@
 # Use MATLAB style diff
 *.m text diff=matlab
 # Mark as binary to prevent EOL changes. Also no need to track 
-mafft-linux*/*  binary
+mafft.bat  binary
 # Files/folders that should not be included when downloading repo as ZIP
 .gitattributes	export-ignore
 .gitignore	export-ignore

diff --git a/core/addExchangeRxns.m b/core/addExchangeRxns.m
@@ -23,6 +23,8 @@
 
 if nargin<3
     mets=model.mets;
+elseif ~islogical(mets) && ~isnumeric(mets)
+    mets=convertCharArray(mets);
 end
 J=getIndexes(model,mets,'mets',false);
 mets=model.mets(J);

diff --git a/core/addGenesRaven.m b/core/addGenesRaven.m
@@ -32,11 +32,8 @@
 if ~isfield(genesToAdd,'genes')
     EM='genes is a required field in genesToAdd';
     dispEM(EM);
-end
-
-if ~iscellstr(genesToAdd.genes)
-    EM='genesToAdd.genes must be a cell array of strings';
-    dispEM(EM);
+else
+    genesToAdd.genes=convertCharArray(genesToAdd.genes);
 end
 
 %Number of genes
@@ -68,14 +65,11 @@
 
 %Some more checks and if they pass then add each field to the structure
 if isfield(genesToAdd,'geneShortNames')
+    genesToAdd.geneShortNames=convertCharArray(genesToAdd.geneShortNames);
     if numel(genesToAdd.geneShortNames)~=nGenes
         EM='genesToAdd.geneShortNames must have the same number of elements as genesToAdd.genes';
         dispEM(EM);
     end
-    if ~iscellstr(genesToAdd.geneShortNames)
-        EM='genesToAdd.geneShortNames must be a cell array of strings';
-        dispEM(EM);
-    end
     %Add empty field if it doesn't exist
     if ~isfield(newModel,'geneShortNames')
         newModel.geneShortNames=largeFiller;

diff --git a/core/addMets.m b/core/addMets.m
@@ -71,35 +71,19 @@
 %Check some stuff regarding the required fields
 if ~isfield(metsToAdd,'mets')
     metsToAdd.mets=generateNewIds(newModel,'mets',prefix,numel(metsToAdd.metNames));
+else
+    metsToAdd.mets=convertCharArray(metsToAdd.mets);
 end
 if ~isfield(metsToAdd,'metNames')
     metsToAdd.metNames=metsToAdd.mets;
+else
+    metsToAdd.metNames=convertCharArray(metsToAdd.metNames);
 end
 if ~isfield(metsToAdd,'compartments')
     EM='compartments is a required field in metsToAdd';
     dispEM(EM);
-end
-if ischar(metsToAdd.mets)
-    metsToAdd.mets={metsToAdd.mets};
-elseif ~iscellstr(metsToAdd.mets)
-    EM='metsToAdd.mets must be a cell array of strings';
-    dispEM(EM);
-end
-if ischar(metsToAdd.metNames)
-    metsToAdd.metNames={metsToAdd.metNames};
-elseif ~iscellstr(metsToAdd.metNames)
-    EM='metsToAdd.metNames must be a cell array of strings';
-    dispEM(EM);
-end
-if ~iscellstr(metsToAdd.compartments)
-    if ischar(metsToAdd.compartments)
-        temp=cell(numel(metsToAdd.mets),1);
-        temp(:)={metsToAdd.compartments};
-        metsToAdd.compartments=temp;
-    else
-        EM='metsToAdd.compartments must be a cell array of strings';
-        dispEM(EM);
-    end
+else
+    metsToAdd.compartments=convertCharArray(metsToAdd.compartments);
 end
 
 %Number of metabolites
@@ -195,14 +179,11 @@
 end
 
 if isfield(metsToAdd,'inchis')
+    metsToAdd.inchis=convertCharArray(metsToAdd.inchis);
     if numel(metsToAdd.inchis)~=nMets
         EM='metsToAdd.inchis must have the same number of elements as metsToAdd.mets';
         dispEM(EM);
     end
-    if ~iscellstr(metsToAdd.inchis)
-        EM='metsToAdd.inchis must be a cell array of strings';
-        dispEM(EM);
-    end
     %Add empty field if it doesn't exist
     if ~isfield(newModel,'inchis')
         newModel.inchis=largeFiller;
@@ -216,14 +197,11 @@
 end
 
 if isfield(metsToAdd,'metFormulas')
+    metsToAdd.metFormulas=convertCharArray(metsToAdd.metFormulas);
     if numel(metsToAdd.metFormulas)~=nMets
         EM='metsToAdd.metFormulas must have the same number of elements as metsToAdd.mets';
         dispEM(EM);
     end
-    if ~iscellstr(metsToAdd.metFormulas)
-        EM='metsToAdd.metFormulas must be a cell array of strings';
-        dispEM(EM);
-    end
     %Add empty field if it doesn't exist
     if ~isfield(newModel,'metFormulas')
         newModel.metFormulas=largeFiller;
@@ -257,14 +235,11 @@
 end
 
 if isfield(metsToAdd,'metNotes')
+    metsToAdd.metNotes=convertCharArray(metsToAdd.metNotes);
     if numel(metsToAdd.metNotes)~=nMets
         EM='metsToAdd.metNotes must have the same number of elements as metsToAdd.mets';
         dispEM(EM);
     end
-    if ~iscellstr(metsToAdd.metNotes)
-        EM='metsToAdd.metNotes must be a cell array of strings';
-        dispEM(EM);
-    end
     %Add empty field if it doesn't exist
     if ~isfield(newModel,'metNotes')
         newModel.metNotes=largeFiller;

diff --git a/core/addRxns.m b/core/addRxns.m
@@ -116,7 +116,7 @@
 end
 
 if allowNewGenes & isfield(rxnsToAdd,'grRules')
-    genesToAdd.genes = strjoin(rxnsToAdd.grRules);
+    genesToAdd.genes = strjoin(convertCharArray(rxnsToAdd.grRules));
     genesToAdd.genes = regexp(genesToAdd.genes,' |)|(|and|or','split'); % Remove all grRule punctuation
     genesToAdd.genes = genesToAdd.genes(~cellfun(@isempty,genesToAdd.genes));  % Remove spaces and empty genes
     genesToAdd.genes = setdiff(unique(genesToAdd.genes),model.genes); % Only keep new genes
@@ -161,7 +161,8 @@
 if ~isfield(rxnsToAdd,'rxns')
     EM='rxns is a required field in rxnsToAdd';
     dispEM(EM);
-elseif iscell(rxnsToAdd.rxns)
+else
+    rxnsToAdd.rxns=convertCharArray(rxnsToAdd.rxns);
     %To fit with some later printing
     rxnsToAdd.rxns=rxnsToAdd.rxns(:);
 end
@@ -175,30 +176,16 @@
     dispEM(EM);
 end
 
-if ~iscellstr(rxnsToAdd.rxns) && ~ischar(rxnsToAdd.rxns)
-    %It could also be a string, but it's not encouraged
-    EM='rxnsToAdd.rxns must be a cell array of strings';
-    dispEM(EM);
-else
-    rxnsToAdd.rxns=cellstr(rxnsToAdd.rxns);
-end
-
 %Normal case: equations provided
 if isfield(rxnsToAdd,'equations')
-    if ~iscellstr(rxnsToAdd.equations) && ~ischar(rxnsToAdd.equations)
-        %It could also be a string, but it's not encouraged
-        EM='rxnsToAdd.equations must be a cell array of strings';
-        dispEM(EM);
-    else
-        rxnsToAdd.equations=cellstr(rxnsToAdd.equations);
-    end
-
+    rxnsToAdd.equations=convertCharArray(rxnsToAdd.equations);
+
     %Alternative case: mets+stoichiometry provided
 else
     %In the case of 1 rxn added (array of strings + vector), transform to
     %cells of length=1:
     if iscellstr(rxnsToAdd.mets)
-        rxnsToAdd.mets = {rxnsToAdd.mets};
+        rxnsToAdd.mets={rxnsToAdd.mets};
     end
     if isnumeric(rxnsToAdd.stoichCoeffs)
         rxnsToAdd.stoichCoeffs = {rxnsToAdd.stoichCoeffs};
@@ -258,6 +245,7 @@
 newModel.rxns=[newModel.rxns;rxnsToAdd.rxns(:)];
 
 if isfield(rxnsToAdd,'rxnNames')
+    rxnsToAdd.rxnNames=convertCharArray(rxnsToAdd.rxnNames);
     if numel(rxnsToAdd.rxnNames)~=nRxns
         EM='rxnsToAdd.rxnNames must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -341,6 +329,7 @@
 end
 
 if isfield(rxnsToAdd,'eccodes')
+    rxnsToAdd.eccodes=convertCharArray(rxnsToAdd.eccodes);
     if numel(rxnsToAdd.eccodes)~=nRxns
         EM='rxnsToAdd.eccodes must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -415,6 +404,7 @@
     end
 end
 if isfield(rxnsToAdd,'grRules')
+    rxnsToAdd.grRules=convertCharArray(rxnsToAdd.grRules);
     if numel(rxnsToAdd.grRules)~=nRxns
         EM='rxnsToAdd.grRules must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -432,6 +422,7 @@
 end
 
 if isfield(rxnsToAdd,'rxnFrom')
+    rxnsToAdd.rxnFrom=convertCharArray(rxnsToAdd.rxnFrom);
     if numel(rxnsToAdd.rxnFrom)~=nRxns
         EM='rxnsToAdd.rxnFrom must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -449,6 +440,7 @@
 end
 
 if isfield(rxnsToAdd,'rxnNotes')
+    rxnsToAdd.rxnNotes=convertCharArray(rxnsToAdd.rxnNotes);
     if numel(rxnsToAdd.rxnNotes)~=nRxns
         EM='rxnsToAdd.rxnNotes must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -466,6 +458,7 @@
 end
 
 if isfield(rxnsToAdd,'rxnReferences')
+    rxnsToAdd.rxnReferences=convertCharArray(rxnsToAdd.rxnReferences);
     if numel(rxnsToAdd.rxnReferences)~=nRxns
         EM='rxnsToAdd.rxnReferences must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);
@@ -483,6 +476,7 @@
 end
 
 if isfield(rxnsToAdd,'pwys')
+    rxnsToAdd.pwys=convertCharArray(rxnsToAdd.pwys);
     if numel(rxnsToAdd.pwys)~=nRxns
         EM='rxnsToAdd.pwys must have the same number of elements as rxnsToAdd.rxns';
         dispEM(EM);

diff --git a/core/addRxnsGenesMets.m b/core/addRxnsGenesMets.m
@@ -5,8 +5,8 @@
 %
 %   model           draft model where reactions should be copied to
 %   sourceModel     model where reactions and metabolites are sourced from
-%   rxns            cell array with reaction IDs (from source model)
-%                   string allowed if only one reaction is added
+%   rxns            cell array with reaction IDs (from source model). Can also
+%                   be string if only one reaction is added
 %   addGene         three options:
 %                   false   no genes are annotated to the new reactions
 %                   true    grRules ared copied from the sourceModel and
@@ -16,10 +16,12 @@
 %                           array, and any new genes are added when
 %                           required
 %                   (opt, default false)
-%   rxnNote         string explaining why reactions were copied to model,
-%                   is included as newModel.rxnNotes (opt, default
+%   rxnNote         cell array with strings explaining why reactions were copied
+%                   to the model, to be included as newModel.rxnNotes. Can also
+%                   be string if same rxnNotes should be added for each new
+%                   reaction, or only one reaction is to be added (opt, default
 %                   'Added via addRxnsAndMets()')
-%   confidence      double specifying confidence score for all reactions.
+%   confidence      integer specifying confidence score for all reactions.
 %                   4:  biochemical data: direct evidence from enzymes
 %                       assays
 %                   3:  genetic data: knockout/-in or overexpression
@@ -45,19 +47,21 @@
 if nargin<6
     confidence=0;
 end
+rxns=convertCharArray(rxns);
 if nargin<5
-    rxnNote='Added via addRxnsGenesMets()';
+    rxnNote={'Added via addRxnsGenesMets()'};
+else
+    rxnNote=convertCharArray(rxnNote);
+    if numel(rxnNote)==1 && numel(rxns)>1
+        rxnNoteArray=cell(numel(rxns),1);
+        rxnNoteArray{:}=rxnNote;
+        rxnNote=rxnNoteArray;
+    end
 end
 if nargin<4
     addGene=false;
 end
 
-%If the supplied object is a character array, then convert it to a cell
-%array
-if ischar(rxns)
-    rxns={rxns};
-end
-
 % Obtain indexes of reactions in source model
 [notNewRxn,oldRxn]=ismember(rxns,model.rxns);
 rxns=rxns(~notNewRxn);
@@ -150,8 +154,7 @@
 rxnToAdd.rxns=sourceModel.rxns(rxnIdx);
 rxnToAdd.lb=sourceModel.lb(rxnIdx);
 rxnToAdd.ub=sourceModel.ub(rxnIdx);
-rxnToAdd.rxnNotes=cell(1,numel(rxnToAdd.rxns));
-rxnToAdd.rxnNotes(:)={rxnNote};
+rxnToAdd.rxnNotes(:)=rxnNote(~notNewRxn);
 rxnToAdd.rxnConfidenceScores=NaN(1,numel(rxnToAdd.rxns));
 if ~isnumeric(confidence)
     EM='confidence score must be numeric';

diff --git a/core/addTransport.m b/core/addTransport.m
@@ -32,9 +32,7 @@
     EM='fromComps must have exactly one match in model.comps';
     dispEM(EM);
 end
-if ischar(toComps)
-    toComps={toComps};
-end
+toComps=convertCharArray(toComps);
 [I, toIDs]=ismember(toComps,model.comps);
 if ~all(I)
     EM='All compartments in toComps must have a match in model.comps';
@@ -43,12 +41,11 @@
 if nargin<4
     %Find all metabolites in fromComp
     metNames=model.metNames(model.metComps==fromID);
-end
-
-%If an empty set was given
-if isempty(metNames)
+elseif isempty(metNames)
     %Find all metabolites in fromComp
     metNames=model.metNames(ismember(model.metComps,model.comps(fromID)));
+else
+    metNames=convertCharArray(metNames);
 end
 
 if nargin<5
@@ -64,9 +61,6 @@
 end
 
 %Check that the names are unique
-if ischar(metNames)
-    metNames={metNames};
-end
 if numel(unique(metNames))~=numel(metNames)
     dispEM('Not all metabolite names are unique');
 end

diff --git a/core/buildEquation.m b/core/buildEquation.m
@@ -2,7 +2,7 @@
 % buildEquation
 %   Construct single equation string for a given reaction
 %
-%   mets            string array with metabolites involved in the reaction.
+%   mets            cell array with metabolites involved in the reaction.
 %   stoichCoeffs    vector with corresponding stoichiometric coeffs.
 %   isrev           logical indicating if the reaction is or not
 %                   reversible.
@@ -11,12 +11,7 @@
 %
 %    Usage: equationString=buildEquation(mets,stoichCoeffs,isrev)
 
-if ~iscellstr(mets) && ~ischar(mets)
-    EM = 'mets must be a cell array of strings';
-    dispEM(EM);
-else
-    mets = cellstr(mets);
-end
+mets=convertCharArray(mets);
 if ~isnumeric(stoichCoeffs)
     EM = 'stoichCoeffs must be a numeric vector';
     dispEM(EM);

diff --git a/core/canConsume.m b/core/canConsume.m
@@ -15,6 +15,8 @@
 
 if nargin<2
     mets=model.mets;
+elseif ~islogical(mets) && ~isnumeric(mets)
+    mets=convertCharArray(mets);
 end
 
 [model, rxns]=addExchangeRxns(model,'in',mets);

diff --git a/core/canProduce.m b/core/canProduce.m
@@ -16,6 +16,8 @@
 
 if nargin<2
     mets=model.mets;
+elseif ~islogical(mets) && ~isnumeric(mets)
+    mets=convertCharArray(mets);
 end
 
 [model, rxns]=addExchangeRxns(model,'out',mets);