loadBRENDAdata

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

0001 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
0002 function [KCATcell, SAcell] = loadBRENDAdata(modelAdapter)
0003 
0004 if nargin < 1 || isempty(modelAdapter)
0005     modelAdapter = ModelAdapterManager.getDefault();
0006     if isempty(modelAdapter)
0007         error('Either send in a modelAdapter or set the default model adapter in the ModelAdapterManager.')
0008     end
0009 end
0010 
0011 basePath      = modelAdapter.getBrendaDBFolder();
0012 KCAT_file      = fullfile(basePath,'max_KCAT.txt');
0013 SA_file        = fullfile(basePath,'max_SA.txt');
0014 MW_file        = fullfile(basePath,'max_MW.txt');
0015 
0016 %Extract BRENDA DATA from files information
0017 KCATcell       = openDataFile(KCAT_file,1);
0018 scalingFactor = 1/60;    %[umol/min/mg] -> [mmol/s/g]    Old: 60 [umol/min/mg] -> [mmol/h/g]
0019 SA            = openDataFile(SA_file,scalingFactor);
0020 scalingFactor = 1/1000;  %[g/mol] -> [g/mmol]
0021 MW            = openDataFile(MW_file,scalingFactor);
0022 
0023 for i=1:4
0024     SAcell{i} = [];
0025 end
0026 previousEC = []; EC_indexes = [];
0027 
0028 %build an index on MW{1} to speed things up a bit
0029 %first just extract the genus (i.e. the first part of the name)
0030 MWECNum = upper(unique(MW{1}));
0031 MWECNumIndices = cell(length(MWECNum),1);
0032 MWECNumHashMap = containers.Map(MWECNum,1:length(MWECNum));
0033 for i = 1:length(MW{1})
0034     matchInd = cell2mat(values(MWECNumHashMap, MW{1}(i)));
0035     MWECNumIndices{matchInd} = [MWECNumIndices{matchInd};i];
0036 end
0037 
0038 
0039 for i=1:length(SA{1})
0040     %Gets the indexes of the EC repetitions in the MW cell for every
0041     %new (different) EC
0042     if ~strcmpi(SA{1}(i), previousEC)
0043         key = upper(SA{1}(i));
0044         if isKey(MWECNumHashMap,key) %annoyingly, this seems to be needed
0045             matchInd = cell2mat(values(MWECNumHashMap,key));
0046             EC_indexes = MWECNumIndices{matchInd};
0047         else
0048             EC_indexes = [];
0049         end
0050     end
0051     mwEC{1} = MW{3}(EC_indexes); mwEC{2} = MW{4}(EC_indexes);
0052     % just looks for the first match because just the maximal value for
0053     % each EC# / Orgaism is reported on the file
0054     org_index = find(strcmpi(SA{3}(i),mwEC{1}),1);
0055     if ~isempty(org_index)
0056         SAcell{1} = [SAcell{1};SA{1}(i)];
0057         SAcell{2} = [SAcell{2};SA{3}(i)];
0058         SAcell{3} = [SAcell{3}; SA{4}(i)*mwEC{2}(org_index)]; %[1/hr]
0059         SAcell{4} = [SAcell{4}; mwEC{2}(org_index)];
0060     end
0061     previousEC = SA{1}(i);
0062 end
0063 
0064 %remove EC in front of all the EC numbers
0065 if ~isempty(KCATcell{1})
0066     KCATcell{1} = extractAfter(KCATcell{1},2);
0067 end
0068 if ~isempty(SAcell{1})
0069     SAcell{1} = extractAfter(SAcell{1},2);
0070 end
0071 
0072 function data_cell = openDataFile(fileName,scalingFactor)
0073 fID          = fopen(fileName);
0074 data_cell    = textscan(fID,'%q %q %q %f %q','delimiter','\t');
0075 fclose(fID);
0076 data_cell{4} = data_cell{4}*scalingFactor;
0077 %Split string for each organism in the BRENDA data
0078 %{name, taxonomy, KEGG code}
0079 data_cell{3}  = regexprep(data_cell{3},'\/\/.*','');
0080 end
0081 end