getINITModel_legacy
Generates a model using the INIT algorithm, based on proteomics and/or
transcriptomics and/or metabolomics and/or metabolic tasks. This is the original
implementation of tINIT, which is replaced by ftINIT.
Input:
refModel a model structure. The model should be in the
closed form (no exchange reactions open). Import
using import(filename,false). If the model is not
loaded using importModel, it might be that there
is no "unconstrained" field. In that case,
manually add the field like:
model.unconstrained=false(numel(model.mets),1);
tissue tissue to score for. Should exist in either
hpaData.tissues or arrayData.tissues
celltype cell type to score for. Should exist in either
hpaData.celltypes or arrayData.celltypes for this
tissue (optional, default is to use the best values
among all the cell types for the tissue. Use [] if
you want to supply more arguments)
hpaData HPA data structure from parseHPA (optional if arrayData is
supplied, default [])
arrayData gene expression data structure (optional if hpaData is
supplied, default [])
genes cell array with the unique gene names
tissues cell array with the tissue names. The list may not be
unique, as there can be multiple cell types per tissue
celltypes cell array with the cell type names for each tissue
levels GENESxTISSUES array with the expression level for
each gene in each tissue/celltype. NaN should be
used when no measurement was performed
threshold a single value or a vector of gene expression
thresholds, above which genes are considered to be
"expressed". (optional, by default, the mean expression
levels of each gene across all tissues in arrayData
will be used as the threshold values)
singleCells binary value selecting whether to use the
single-cell algorithm to identify expressed genes.
If used, specify cell subpopulations in CELLTYPES
(optional, default [])
plotResults true if single cell probability distributions
should be plotted (optional, default = False)
metabolomicsData cell array with metabolite names that the model
should produce (optional, default [])
taskFile a task list in Excel format. See parseTaskList for
details (optional, default [])
useScoresForTasks true if the calculated reaction scored should be used as
weights in the fitting to tasks (optional, default true)
printReport true if a report should be printed to the screen
(optional, default true)
taskStructure task structure as from parseTaskList. Can be used
as an alternative way to define tasks when Excel
sheets are not suitable. Overrides taskFile (optional,
default [])
params parameter structure as used by getMILPParams. This is
for the INIT algorithm. For the the MILP problems
solved to fit tasks, see paramsFT (optional, default [])
paramsFT parameter structure as used by getMILPParams. This is
for the fitTasks step. For the INIT algorithm, see
params (optional, default [])
Output:
model the resulting model structure
metProduction array that indicates which of the
metabolites in metabolomicsData that could be
produced. Note that this is before the
gap-filling process to enable defined tasks. To
see which metabolites that can be produced in
the final model, use canProduce.
-2: metabolite name not found in model
-1: metabolite found, but it could not be produced
1: metabolite could be produced
essentialRxnsForTasks cell array of the reactions which were
essential to perform the tasks
addedRxnsForTasks cell array of the reactions which were added in
order to perform the tasks
deletedDeadEndRxns cell array of reactions deleted because they
could not carry flux (INIT requires a
functional input model)
deletedRxnsInINIT cell array of the reactions which were deleted by
the INIT algorithm
taskReport structure with the results for each task
id cell array with the id of the task
description cell array with the description of the task
ok boolean array with true if the task was successful
essential cell array with cell arrays of essential
reactions for the task
gapfill cell array of cell arrays of reactions included
in the gap-filling for the task
This is the main function for automatic reconstruction of models based
on the (t)INIT algorithm (PLoS Comput Biol. 2012;8(5):e1002518,
Mol Syst Biol. 2014;10:721). Not all settings are possible using this
function, and you may want to call the functions scoreModel, runINIT
and fitTasks individually instead.
NOTE: Exchange metabolites should normally not be removed from the model
when using this approach, since checkTasks/fitTasks rely on putting specific
constraints for each task. The INIT algorithm will remove exchange metabolites
if any are present. Use importModel(file,false) to import a model with
exchange metabolites remaining.
Usage: [model, metProduction, essentialRxnsForTasks, addedRxnsForTasks,...
deletedDeadEndRxns, deletedRxnsInINIT, taskReport]=...
getINITModel(refModel, tissue, celltype, hpaData, arrayData,...
metabolomicsData, taskFile, useScoresForTasks, printReport,...
taskStructure, params, paramsFT)0001 function [model, metProduction, essentialRxnsForTasks, addedRxnsForTasks, deletedDeadEndRxns, deletedRxnsInINIT, taskReport]=getINITModel(refModel, tissue, celltype, hpaData, arrayData, metabolomicsData, taskFile, useScoresForTasks, printReport, taskStructure, params, paramsFT) 0002 % getINITModel_legacy 0003 % Generates a model using the INIT algorithm, based on proteomics and/or 0004 % transcriptomics and/or metabolomics and/or metabolic tasks. This is the original 0005 % implementation of tINIT, which is replaced by ftINIT. 0006 % 0007 % Input: 0008 % refModel a model structure. The model should be in the 0009 % closed form (no exchange reactions open). Import 0010 % using import(filename,false). If the model is not 0011 % loaded using importModel, it might be that there 0012 % is no "unconstrained" field. In that case, 0013 % manually add the field like: 0014 % model.unconstrained=false(numel(model.mets),1); 0015 % tissue tissue to score for. Should exist in either 0016 % hpaData.tissues or arrayData.tissues 0017 % celltype cell type to score for. Should exist in either 0018 % hpaData.celltypes or arrayData.celltypes for this 0019 % tissue (optional, default is to use the best values 0020 % among all the cell types for the tissue. Use [] if 0021 % you want to supply more arguments) 0022 % hpaData HPA data structure from parseHPA (optional if arrayData is 0023 % supplied, default []) 0024 % arrayData gene expression data structure (optional if hpaData is 0025 % supplied, default []) 0026 % genes cell array with the unique gene names 0027 % tissues cell array with the tissue names. The list may not be 0028 % unique, as there can be multiple cell types per tissue 0029 % celltypes cell array with the cell type names for each tissue 0030 % levels GENESxTISSUES array with the expression level for 0031 % each gene in each tissue/celltype. NaN should be 0032 % used when no measurement was performed 0033 % threshold a single value or a vector of gene expression 0034 % thresholds, above which genes are considered to be 0035 % "expressed". (optional, by default, the mean expression 0036 % levels of each gene across all tissues in arrayData 0037 % will be used as the threshold values) 0038 % singleCells binary value selecting whether to use the 0039 % single-cell algorithm to identify expressed genes. 0040 % If used, specify cell subpopulations in CELLTYPES 0041 % (optional, default []) 0042 % plotResults true if single cell probability distributions 0043 % should be plotted (optional, default = False) 0044 % metabolomicsData cell array with metabolite names that the model 0045 % should produce (optional, default []) 0046 % taskFile a task list in Excel format. See parseTaskList for 0047 % details (optional, default []) 0048 % useScoresForTasks true if the calculated reaction scored should be used as 0049 % weights in the fitting to tasks (optional, default true) 0050 % printReport true if a report should be printed to the screen 0051 % (optional, default true) 0052 % taskStructure task structure as from parseTaskList. Can be used 0053 % as an alternative way to define tasks when Excel 0054 % sheets are not suitable. Overrides taskFile (optional, 0055 % default []) 0056 % params parameter structure as used by getMILPParams. This is 0057 % for the INIT algorithm. For the the MILP problems 0058 % solved to fit tasks, see paramsFT (optional, default []) 0059 % paramsFT parameter structure as used by getMILPParams. This is 0060 % for the fitTasks step. For the INIT algorithm, see 0061 % params (optional, default []) 0062 % 0063 % 0064 % Output: 0065 % model the resulting model structure 0066 % metProduction array that indicates which of the 0067 % metabolites in metabolomicsData that could be 0068 % produced. Note that this is before the 0069 % gap-filling process to enable defined tasks. To 0070 % see which metabolites that can be produced in 0071 % the final model, use canProduce. 0072 % -2: metabolite name not found in model 0073 % -1: metabolite found, but it could not be produced 0074 % 1: metabolite could be produced 0075 % essentialRxnsForTasks cell array of the reactions which were 0076 % essential to perform the tasks 0077 % addedRxnsForTasks cell array of the reactions which were added in 0078 % order to perform the tasks 0079 % deletedDeadEndRxns cell array of reactions deleted because they 0080 % could not carry flux (INIT requires a 0081 % functional input model) 0082 % deletedRxnsInINIT cell array of the reactions which were deleted by 0083 % the INIT algorithm 0084 % taskReport structure with the results for each task 0085 % id cell array with the id of the task 0086 % description cell array with the description of the task 0087 % ok boolean array with true if the task was successful 0088 % essential cell array with cell arrays of essential 0089 % reactions for the task 0090 % gapfill cell array of cell arrays of reactions included 0091 % in the gap-filling for the task 0092 % 0093 % This is the main function for automatic reconstruction of models based 0094 % on the (t)INIT algorithm (PLoS Comput Biol. 2012;8(5):e1002518, 0095 % Mol Syst Biol. 2014;10:721). Not all settings are possible using this 0096 % function, and you may want to call the functions scoreModel, runINIT 0097 % and fitTasks individually instead. 0098 % 0099 % NOTE: Exchange metabolites should normally not be removed from the model 0100 % when using this approach, since checkTasks/fitTasks rely on putting specific 0101 % constraints for each task. The INIT algorithm will remove exchange metabolites 0102 % if any are present. Use importModel(file,false) to import a model with 0103 % exchange metabolites remaining. 0104 % 0105 % Usage: [model, metProduction, essentialRxnsForTasks, addedRxnsForTasks,... 0106 % deletedDeadEndRxns, deletedRxnsInINIT, taskReport]=... 0107 % getINITModel(refModel, tissue, celltype, hpaData, arrayData,... 0108 % metabolomicsData, taskFile, useScoresForTasks, printReport,... 0109 % taskStructure, params, paramsFT) 0110 0111 if nargin<3 0112 celltype=[]; 0113 else 0114 celltype=char(celltype); 0115 end 0116 if nargin<4 0117 hpaData=[]; 0118 end 0119 if nargin<5 0120 arrayData=[]; 0121 end 0122 if nargin<6 0123 metabolomicsData=[]; 0124 end 0125 if nargin<7 0126 taskFile=[]; 0127 else 0128 taskFile=char(taskFile); 0129 end 0130 if nargin<8 || isempty(useScoresForTasks) 0131 useScoresForTasks=true; 0132 end 0133 if nargin<9 || isempty(printReport) 0134 printReport=true; 0135 end 0136 if nargin<10 0137 taskStructure=[]; 0138 end 0139 if nargin<11 0140 params=[]; 0141 end 0142 if nargin<12 0143 paramsFT=[]; 0144 end 0145 0146 %Check that the model is in the closed form 0147 if ~isfield(refModel,'unconstrained') 0148 EM='Exchange metabolites should normally not be removed from the model when using getINITModel. Use importModel(file,false) to import a model with exchange metabolites remaining (see the documentation for details)'; 0149 dispEM(EM); 0150 end 0151 0152 %Create the task structure if not supplied 0153 if any(taskFile) && isempty(taskStructure) 0154 taskStructure=parseTaskList(taskFile); 0155 end 0156 0157 0158 % sc-tINIT to identify confidence levels of gene expression 0159 if ~isempty(arrayData) && isfield(arrayData,'singleCells') 0160 if arrayData.singleCells == 1 0161 % Check to ensure cell type is defined 0162 if ~isfield(arrayData,'celltypes') 0163 dispEM('arrayData must contain cell type information if sc-tINIT is to be used','false'); 0164 end 0165 if ~ismember(upper(celltype),upper(arrayData.celltypes)) 0166 dispEM('The cell type name does not match'); 0167 end 0168 0169 % Analyze only cell type of interest 0170 J= strcmpi(arrayData.celltypes,celltype); 0171 0172 % Analyze only genes included in the reference model 0173 I=ismember(arrayData.genes,refModel.genes); 0174 0175 % Convert expression data to population fractions 0176 binary_levels = arrayData.levels(I,J)~=0; % Classify each gene as detected (1) or not (0) in each cell 0177 cell_count_levels = sum(binary_levels,2); % Number of cells expressing each transcript 0178 cell_frac_levels = cell_count_levels/size(binary_levels,2); % Number of cells expressing each transcript 0179 0180 % Bin cell_frac_counts manually 0181 x = 0:.01:1; 0182 for(i = 1:length(x)) 0183 cell_frac_count(i) = sum(round(cell_frac_levels,2)==x(i)); 0184 end 0185 0186 % Fit four beta distributions 0187 cell_frac_count(cell_frac_count==0) = NaN; % Remove zeros from optimization 0188 cell_frac_count(1) = NaN; % Remove non-expressed genes from optimization 0189 x_lim = 1; % Somewhat arbitrary parameter to fit left tail of distr. 0190 myfun = @(par) nansum((cell_frac_count(1:find(x>=x_lim,1)) - ... 0191 abs(par(1))*betapdf(x(1:find(x>=x_lim,1)),abs(par(2)),abs(par(3))) - ... 0192 abs(par(4))*betapdf(x(1:find(x>=x_lim,1)),abs(par(5)),abs(par(6))) - ... 0193 abs(par(7))*betapdf(x(1:find(x>=x_lim,1)),abs(par(8)),abs(par(9))) - ... 0194 abs(par(10))*betapdf(x(1:find(x>=x_lim,1)),abs(par(11)),abs(par(12)))).^2); 0195 0196 par0 = [4,2,100,7,2,30,7,5,20,5,15,20]; 0197 opts = optimset('Display','off'); 0198 [par,f_val] = fminsearch(myfun,par0,opts); 0199 par = abs(par); 0200 0201 % Plot results 0202 if (isfield(arrayData,'plotResults')) 0203 if arrayData.plotResults == true 0204 figure(); hold on; plot(x,cell_frac_count,'ko','MarkerSize',5); 0205 plot(x,abs(par(1))*betapdf(x,abs(par(2)),abs(par(3))),'b-','LineWidth',1) 0206 plot(x,abs(par(4))*betapdf(x,abs(par(5)),abs(par(6))),'b-','LineWidth',1) 0207 plot(x,abs(par(7))*betapdf(x,abs(par(8)),abs(par(9))),'b-','LineWidth',1) 0208 plot(x,abs(par(10))*betapdf(x,abs(par(11)),abs(par(12))),'b-','LineWidth',1) 0209 plot(x,abs(par(1))*betapdf(x,abs(par(2)),abs(par(3))) + ... 0210 abs(par(4))*betapdf(x,abs(par(5)),abs(par(6))) + ... 0211 abs(par(7))*betapdf(x,abs(par(8)),abs(par(9))) + ... 0212 abs(par(10))*betapdf(x,abs(par(11)),abs(par(12))),'-','Color',[.5 .5 .5],'LineWidth',2) 0213 xlabel('Expression Probability');ylabel('# of genes');set(gca,'FontSize',14,'LineWidth',1.25); 0214 title('Expression prediction','FontSize',18,'FontWeight','bold') 0215 end 0216 end 0217 0218 % Score genes based on population expression (p = .05) 0219 exprs_cutoff_1 = find(cdf('beta',x,par(2),par(3)) >.95,1)-1; % Find index of no confidence genes 0220 exprs_cutoff_2 = find(cdf('beta',x,par(5),par(6)) >.95,1)-1; % Find index of low confidence genes 0221 exprs_cutoff_3 = find(cdf('beta',x,par(8),par(9)) >.95,1)-1; % Find index of low confidence genes 0222 exprs_cutoffs = sort([exprs_cutoff_1,exprs_cutoff_2,exprs_cutoff_3]); 0223 gene_scores = cell_frac_levels*0; 0224 gene_scores(cell_frac_levels <= x(exprs_cutoffs(1))) = 4; % Not detected 0225 gene_scores(logical((cell_frac_levels >= x(exprs_cutoffs(1))).*(cell_frac_levels < x(exprs_cutoffs(2))))) = 3; % Low detection 0226 gene_scores(logical((cell_frac_levels >= x(exprs_cutoffs(2))).*(cell_frac_levels < x(exprs_cutoffs(3))))) = 2; % Medium detection 0227 gene_scores(cell_frac_levels > x(exprs_cutoffs(3))) = 1; % High detection 0228 0229 % Replace hpaData with singleCellData 0230 if printReport==true 0231 dispEM('Single cell data is not currently compatible with HPA data. \n Replacing hpaData with single cell-based scoring.',false); 0232 end 0233 hpaData.genes = arrayData.genes; 0234 hpaData.tissues = arrayData.tissues; 0235 hpaData.celltypes = arrayData.celltypes; 0236 hpaData.levels = [{'High'},{'Medium'},{'Low'},{'None'}]; 0237 hpaData.gene2Level = zeros(length(arrayData.genes),length(arrayData.celltypes)); 0238 for i = 1:length(find(J)) 0239 find_var = find(J,i); 0240 hpaData.gene2Level(I,find_var(end)) = gene_scores; 0241 end 0242 0243 % Remove arrayData from the analysis (Might be a bad idea) 0244 clear arrayData 0245 arrayData=[]; 0246 end 0247 end 0248 0249 0250 if printReport==true 0251 if any(celltype) 0252 fprintf(['***Generating model for: ' tissue ' - ' celltype '\n']); 0253 else 0254 fprintf(['***Generating model for: ' tissue '\n']); 0255 end 0256 if ~isempty(hpaData) 0257 fprintf('-Using HPA data\n'); 0258 end 0259 if ~isempty(arrayData) 0260 fprintf('-Using array data\n'); 0261 end 0262 if ~isempty(metabolomicsData) 0263 fprintf('-Using metabolomics data\n'); 0264 end 0265 if ~isempty(taskFile) || ~isempty(taskStructure) 0266 fprintf('-Using metabolic tasks\n'); 0267 end 0268 fprintf('\n'); 0269 0270 printScores(refModel,'Reference model statistics',hpaData,arrayData,tissue,celltype); 0271 end 0272 0273 %Remove dead-end reactions to speed up the optimization and to 0274 %differentiate between reactions removed by INIT and those that are 0275 %dead-end 0276 [~, deletedDeadEndRxns]=simplifyModel(refModel,true,false,true,true,true); 0277 cModel=removeReactions(refModel,deletedDeadEndRxns,false,true); 0278 0279 %Store the connected model like this to keep track of stuff 0280 if printReport==true 0281 printScores(cModel,'Pruned model statistics',hpaData,arrayData,tissue,celltype); 0282 end 0283 0284 %If tasks have been defined, then go through them and get essential 0285 %reactions 0286 if ~isempty(taskStructure) 0287 [taskReport, essentialRxnMat]=checkTasks(cModel,[],printReport,true,true,taskStructure); 0288 0289 essentialRxnsForTasks=cModel.rxns(any(essentialRxnMat,2)); 0290 0291 %Remove tasks that cannot be performed 0292 taskStructure(taskReport.ok==false)=[]; 0293 if printReport==true 0294 printScores(removeReactions(cModel,setdiff(cModel.rxns,essentialRxnsForTasks),true,true),'Reactions essential for tasks',hpaData,arrayData,tissue,celltype); 0295 end 0296 else 0297 essentialRxnsForTasks={}; 0298 end 0299 0300 %Score the connected model 0301 [rxnScores, geneScores]=scoreModel(cModel,hpaData,arrayData,tissue,celltype); 0302 0303 %Run the INIT algorithm. The exchange reactions that are used in the final 0304 %reactions will be open, which doesn't fit with the last step. Therefore 0305 %delete reactions from the original model instead of taking the output. The 0306 %default implementation does not constrain reversible reactions to only 0307 %carry flux in one direction. Runs without the constraints on reversibility 0308 %and with all output allowed. This is to reduce the complexity of the 0309 %problem. 0310 [~, deletedRxnsInINIT, metProduction]=runINIT(simplifyModel(cModel),rxnScores,metabolomicsData,essentialRxnsForTasks,0,true,false,params); 0311 initModel=removeReactions(cModel,deletedRxnsInINIT,true,true); 0312 if printReport==true 0313 printScores(initModel,'INIT model statistics',hpaData,arrayData,tissue,celltype); 0314 printScores(removeReactions(cModel,setdiff(cModel.rxns,deletedRxnsInINIT),true,true),'Reactions deleted by INIT',hpaData,arrayData,tissue,celltype); 0315 end 0316 0317 %The full model has exchange reactions in it. fitTasks calls on fillGaps, 0318 %which automatically removes exchange metabolites (because it assumes that 0319 %the reactions are constrained when appropriate). In this case the 0320 %uptakes/outputs are retrieved from the task sheet instead. To prevent 0321 %exchange reactions being used to fill gaps, they are delete from the 0322 %reference model here. 0323 initModel.id='INITModel'; 0324 0325 %If gaps in the model should be filled using a task list 0326 if ~isempty(taskStructure) 0327 %Remove exchange reactions and reactions already included in the INIT 0328 %model 0329 refModelNoExc=removeReactions(refModel,union(initModel.rxns,getExchangeRxns(refModel)),true,true); 0330 0331 %At this stage the model is fully connected and most of the genes with 0332 %good scores should have been included. The final gap-filling should 0333 %take the scores of the genes into account, so that "rather bad" 0334 %reactions are preferred to "very bad" reactions. However, reactions 0335 %with positive scores will be included even if they are not connected 0336 %in the current formulation. Therefore, such reactions will have to be 0337 %assigned a small negative score instead. 0338 if useScoresForTasks==true 0339 refRxnScores=scoreModel(refModelNoExc,hpaData,arrayData,tissue,celltype); 0340 [outModel, addedRxnMat]=fitTasks(initModel,refModelNoExc,[],true,min(refRxnScores,-0.1),taskStructure,paramsFT); 0341 else 0342 [outModel, addedRxnMat]=fitTasks(initModel,refModelNoExc,[],true,[],taskStructure,paramsFT); 0343 end 0344 if printReport==true 0345 printScores(outModel,'Functional model statistics',hpaData,arrayData,tissue,celltype); 0346 printScores(removeReactions(outModel,intersect(outModel.rxns,initModel.rxns),true,true),'Reactions added to perform the tasks',hpaData,arrayData,tissue,celltype); 0347 end 0348 0349 addedRxnsForTasks=refModelNoExc.rxns(any(addedRxnMat,2)); 0350 else 0351 outModel=initModel; 0352 addedRxnMat=[]; 0353 addedRxnsForTasks={}; 0354 end 0355 0356 %The model can now perform all the tasks defined in the task list. The 0357 %algorithm cannot deal with gene-complexes at the moment. It is therefore 0358 %ok to remove bad genes from a reaction (as long as at least one gene is 0359 %kept) 0360 model=outModel; 0361 0362 [~, I]=ismember(model.genes,cModel.genes); %All should be found 0363 %This is a little weird way to make sure that only one bad gene is included 0364 %if there are no good ones (since all -Inf==max(-Inf)) 0365 geneScores(isinf(geneScores))=-1000+rand(sum(isinf(geneScores)),1); 0366 0367 model.grRules(:)={''}; 0368 for i=1:numel(model.rxns) 0369 ids=find(model.rxnGeneMat(i,:)); 0370 if numel(ids)>1 0371 scores=geneScores(I(ids)); 0372 %Only keep the positive ones if possible 0373 model.rxnGeneMat(i,ids(~(scores>0 | scores==max(scores))))=0; 0374 end 0375 %Rewrite the grRules to be only OR 0376 if isfield(model,'grRules') 0377 J=find(model.rxnGeneMat(i,:)); 0378 for j=1:numel(J) 0379 model.grRules{i}=[model.grRules{i} '(' model.genes{J(j)} ')']; 0380 if j<numel(J) 0381 model.grRules{i}=[model.grRules{i} ' or ']; 0382 end 0383 end 0384 end 0385 end 0386 0387 %Find all genes that are not used and delete them 0388 I=sum(model.rxnGeneMat)==0; 0389 model.genes(I)=[]; 0390 model.rxnGeneMat(:,I)=[]; 0391 if isfield(model,'geneShortNames') 0392 model.geneShortNames(I)=[]; 0393 end 0394 if isfield(model,'proteins') 0395 model.proteins(I)=[]; 0396 end 0397 if isfield(model,'geneMiriams') 0398 model.geneMiriams(I)=[]; 0399 end 0400 if isfield(model,'geneFrom') 0401 model.geneFrom(I)=[]; 0402 end 0403 if isfield(model,'geneComps') 0404 model.geneComps(I)=[]; 0405 end 0406 0407 %At this stage the model will contain some exchange reactions but probably 0408 %not all (and maybe zero). This can be inconvenient, so all exchange 0409 %reactions from the reference model are added, except for those which 0410 %involve metabolites that are not in the model. 0411 0412 %First delete and included exchange reactions in order to prevent the order 0413 %from changing 0414 model=removeReactions(model,getExchangeRxns(model)); 0415 0416 %Create a model with only the exchange reactions in refModel 0417 excModel=removeReactions(refModel,setdiff(refModel.rxns,getExchangeRxns(refModel)),true,true); 0418 0419 %Find the metabolites there which are not exchange metabolites and which do 0420 %not exist in the output model 0421 I=~ismember(excModel.mets,model.mets) & excModel.unconstrained==0; 0422 0423 %Then find those reactions and delete them 0424 [~, J]=find(excModel.S(I,:)); 0425 excModel=removeReactions(excModel,J,true,true); 0426 0427 %Merge with the output model 0428 model=mergeModels({model;excModel},'metNames'); 0429 model.id='INITModel'; 0430 model.name=['Automatically generated model for ' tissue]; 0431 if any(celltype) 0432 model.name=[model.name ' - ' celltype]; 0433 end 0434 0435 if printReport==true 0436 printScores(model,'Final model statistics',hpaData,arrayData,tissue,celltype); 0437 end 0438 0439 %Add information about essential reactions and reactions included for 0440 %gap-filling and return a taskReport 0441 if ~isempty(taskStructure) 0442 I=find(taskReport.ok); %Ignore failed tasks 0443 for i=1:numel(I) 0444 taskReport.essential{I(i),1}=cModel.rxns(essentialRxnMat(:,I(i))); 0445 taskReport.gapfill{I(i),1}=refModelNoExc.rxns(addedRxnMat(:,i)); 0446 end 0447 else 0448 taskReport=[]; 0449 end 0450 0451 %Fix grRules and reconstruct rxnGeneMat 0452 [grRules,rxnGeneMat] = standardizeGrRules(model,true); 0453 model.grRules = grRules; 0454 model.rxnGeneMat = rxnGeneMat; 0455 end 0456 0457 %This is for printing a summary of a model 0458 function [rxnS, geneS]=printScores(model,name,hpaData,arrayData,tissue,celltype) 0459 [a, b]=scoreModel(model,hpaData,arrayData,tissue,celltype); 0460 rxnS=mean(a); 0461 geneS=mean(b(~isinf(b))); 0462 fprintf([name ':\n']); 0463 fprintf(['\t' num2str(numel(model.rxns)) ' reactions, ' num2str(numel(model.genes)) ' genes\n']); 0464 fprintf(['\tMean reaction score: ' num2str(rxnS) '\n']); 0465 fprintf(['\tMean gene score: ' num2str(geneS) '\n']); 0466 fprintf(['\tReactions with positive scores: ' num2str(100*sum(a>0)/numel(a)) '%%\n\n']); 0467 end