Merge pull request #361 from SysBioChalmers/develop

yeast 9.0.0

.gitignore

@@ -49,6 +49,7 @@ Thumbs.db
 *.mex*
 *.mlappinstall
 *.mltbx
+*.mat
 helpsearch*/
 
 # Python-related things #
@@ -59,6 +60,7 @@ helpsearch*/
 
 # Non-complying tables and files #
 ##################################
+*.xls*
 *.tab
 *.doc*
 *.ppt*

README.md

@@ -33,21 +33,7 @@ This repository contains the current consensus genome-scale metabolic model of _
 
 | Taxonomy | Latest update | Version | Reactions | Metabolites | Genes |
 |:-------|:--------------|:------|:------|:----------|:-----|
-| _Saccharomyces cerevisiae_ | 17-Oct-2023 | 8.7.1 | 4131 | 2806 | 1163 |
-
-### Gene essentiality prediction
-
-- Accuracy: 0.881
-- True non-essential genes: 928
-- True essential genes: 63
-- False non-essential genes: 96
-- False essential genes: 38
-
-### Growth prediction
-
-- Correlation coefficient R<sup>2</sup>: 0.842
-
-![Growth curve](growth.png)
+| _Saccharomyces cerevisiae_ | 02-Dec-2023 | develop | 4130 | 2805 | 1162 |
 
 # Installation & usage
 

code/increaseVersion.m

@@ -47,7 +47,7 @@ function increaseVersion(bumpType)
 
 %Load model:
 disp('Loading model file')
-model = importModel('../model/yeast-GEM.xml');
+model = readYAMLmodel('../model/yeast-GEM.yml');
 
 %Run tests
 cd modelTests
@@ -109,6 +109,7 @@ function increaseVersion(bumpType)
 %Include tag and save model:
 disp('Write model files')
 model.id = ['yeastGEM_v' newVersion];
+model.version = newVersion;
 saveYeastModel(model,true,true,true)   %only save if model can grow
 
 %Check for any unexpected file changes
@@ -127,9 +128,9 @@ function increaseVersion(bumpType)
                     change = true;
                 end
             case 'model/yeast-GEM.yml'
-                %.yml file: 2 lines should be added & 2 lines should be deleted
-                %(1 with version information, 1 with current date)
-                if eval([diff_i{1} ' > 2']) || eval([diff_i{2} ' > 2'])
+                %.yml file: 3 lines should be added & 3 lines should be deleted
+                %(2 with version information, 1 with current date)
+                if eval([diff_i{1} ' > 3']) || eval([diff_i{2} ' > 3'])
                     disp(['NOTE: File ' diff_i{3} ' is changing more than expected'])
                     change = true;
                 end                

code/modelCuration/v9_0_0.m

@@ -0,0 +1,67 @@
+% This scripts applies curations to be applied on yeast-GEM release 8.7.1, to
+% get to yeast-GEM release 9.0.0.
+% Indicate which Issue/PR are addressed. If multiple curations are performed
+% before a new release is made, just add the required code to this script. If
+% more extensive coding is required, you can write a separate (generic) function
+% that can be kept in the /code/modelCuration folder. Otherwise, try to use
+% existing functions whenever possible. In particular /code/curateMetsRxnsGenes
+% can do many types of curation.
+
+%% Load yeast-GEM 8.7.1 (requires local yeast-GEM git repository)
+cd ..
+codeDir=pwd();
+model = getEarlierModelVersion('8.7.1');
+model.id='yeastGEM_develop';
+model.version='';
+%dataDir=fullfile(pwd(),'..','data','modelCuration','v9.0.0'); % No dataDir required for these curations
+cd modelCuration
+
+%% Correct inbalanced reactions, based on metFormulas
+% While dolichol can have any number of isoprenoid units, in yeast-GEM it is
+% defined as 4 units. This means that there is no need to keep R-subgroups as
+% part of dolichol-derived metabolites to indicate that unspecified length.
+% Define which metabolites are dolichol-derived and remove the R from their
+% metabolite formula.
+dolMets = getIndexes(model,{'s_3765','s_3767','s_3888','s_3911'},'mets');
+model.metFormulas(dolMets) = regexprep(model.metFormulas(dolMets),'R','');
+
+% r_4722 (polyphosphate polymerase) is unbalanced, 2 ADP is missing as product
+model = changeRxns(model,'r_4722','2 ATP[c] + H2O[c] => H+[c] + polyphosphate[v] + 2 ADP[c]',3);
+
+% r_4240 is unbalanced, but also has a generic reactant (protein asparagine)
+% and describes a process (protein modification) that is out-of-scope of a
+% metabolic model. Remove the reactions to resolve all 3 issues at once.
+model = removeReactions(model,'r_4240',true,true,true);
+
+% Some glycan metFormulas were incorrect absent. They were manually curated
+% by drawing out the structures.
+glycMets = getIndexes(model,{'s_3932','s_4003','s_4002'},'mets');
+model.metFormulas(glycMets) = {'C50H82N4O37R','C38H70N2O36P2R2','C32H60N2O31P2R2'};
+
+% r_0774 and r_0775 are unbalanced due to a missing H2O.
+model = changeRxns(model,{'r_0774','r_0775'},...
+    {'ATP[c] + H+[c] + nicotinate[c] + PRPP[c] => ADP[c] + diphosphate[c] + nicotinic acid D-ribonucleotide[c] + phosphate[c]',...
+     'ATP[m] + H+[m] + nicotinate[m] + PRPP[m] => ADP[m] + diphosphate[m] + nicotinic acid D-ribonucleotide[m] + phosphate[m]'},3);
+
+% r_4196 (NADH:ferricytochrome-b5 oxidoreductase) is unbalanced, NAD is missing as product
+model = changeRxns(model,'r_4196','NADH[erm] + 2 Ferricytochrome b5[erm] <=> H+[erm] + 2 Ferrocytochrome b5[erm] + NAD[erm]',3);
+
+%% DO NOT CHANGE OR REMOVE THE CODE BELOW THIS LINE.
+% Show some metrics:
+cd(fullfile(codeDir,'modelTests'))
+disp('Run gene essentiality analysis')
+[new.accuracy,new.tp,new.tn,new.fn,new.fp] = essentialGenes(model);
+fprintf('Genes in model: %d\n',numel(model.genes));
+fprintf('Gene essentiality accuracy: %.4f\n', new.accuracy);
+fprintf('True non-essential genes: %d\n', numel(new.tp));
+fprintf('True essential genes: %d\n', numel(new.tn));
+fprintf('False non-essential genes: %d\n', numel(new.fp));
+fprintf('False essential genes: %d\n', numel(new.fn));
+fprintf('\nRun growth analysis\n')
+R2=growth(model);
+fprintf('R2 of growth prediction: %.4f\n', R2);
+
+% Save model:
+cd ..
+saveYeastModel(model)
+cd modelCuration

data/databases/model_metDeltaG.csv

@@ -2372,7 +2372,6 @@ s_3882,10000000
 s_3883,-356.59
 s_3884,10000000
 s_3885,10000000
-s_3886,10000000
 s_3887,10000000
 s_3888,83.01
 s_3889,10000000

data/databases/model_rxnDeltaG.csv

@@ -3602,7 +3602,6 @@ r_4236,16.42
 r_4237,10000000
 r_4238,10000000
 r_4239,10000000
-r_4240,10000000
 r_4241,10000000
 r_4242,10000000
 r_4243,10000000