getExpressionStructure

 getExpressionStructure
   Loads a representation of an experiment from an Excel file (see
   comments further down)

   fileName            an Excel representation on an experiment

   experiment          an experiment structure
       data            matrix with expression values
       orfs            the corresponding ORFs
       experiments     the titles of the experiments
       boundNames      reaction names for the bounds
       upperBoundaries matrix with the upper bound values
       fitNames        reaction names for the measured fluxes
       fitTo           matrix with the measured fluxes

   A very common data set when working with genome-scale metabolic models
   is that you have measured fermentation data, gene expression data,
   and some different 'bounds' (for example different carbon sources
   or genes that are knocked out) in a number of conditions. This function
   reads an Excel representation of such an experiment.
   The Excel file must contain three sheets, 'EXPRESSION', 'BOUNDS',
   'FITTING'. Below are some examples to show how they should be
   formatted:

   -EXPRESSION
       ORF    dsm_paa    wisc_paa
       Pc00e00030    79.80942723    78.14755338
   Shows the expression of the gene Pc00e00030 under two different
   conditions (in this case a DSM strain and a Wisconsin strain of P.
   chrysogenum with PSS in the media)

   -BOUNDS
       Fixed Upper    dsm_paa    wisc_paa
       paaIN    0.1    0.2
   The upper bound for the reaction paaIN should be 0.1 for the first
   condition and 0.2 for the second

   -FITTING
       Fit to    dsm_paa    wisc_paa
       co2OUT    2.85    3.05
       glcIN   1.2     0.9
   The measured fluxes for CO2 production and glucose uptake for the two
   conditions. The model(s) can later be fitted to match these values as
   good as possible.

 Usage: experiment=getExpressionStructure(fileName)

0001 function experiment=getExpressionStructure(fileName)
0002 % getExpressionStructure
0003 %   Loads a representation of an experiment from an Excel file (see
0004 %   comments further down)
0005 %
0006 %   fileName            an Excel representation on an experiment
0007 %
0008 %   experiment          an experiment structure
0009 %       data            matrix with expression values
0010 %       orfs            the corresponding ORFs
0011 %       experiments     the titles of the experiments
0012 %       boundNames      reaction names for the bounds
0013 %       upperBoundaries matrix with the upper bound values
0014 %       fitNames        reaction names for the measured fluxes
0015 %       fitTo           matrix with the measured fluxes
0016 %
0017 %   A very common data set when working with genome-scale metabolic models
0018 %   is that you have measured fermentation data, gene expression data,
0019 %   and some different 'bounds' (for example different carbon sources
0020 %   or genes that are knocked out) in a number of conditions. This function
0021 %   reads an Excel representation of such an experiment.
0022 %   The Excel file must contain three sheets, 'EXPRESSION', 'BOUNDS',
0023 %   'FITTING'. Below are some examples to show how they should be
0024 %   formatted:
0025 %
0026 %   -EXPRESSION
0027 %       ORF    dsm_paa    wisc_paa
0028 %       Pc00e00030    79.80942723    78.14755338
0029 %   Shows the expression of the gene Pc00e00030 under two different
0030 %   conditions (in this case a DSM strain and a Wisconsin strain of P.
0031 %   chrysogenum with PSS in the media)
0032 %
0033 %   -BOUNDS
0034 %       Fixed Upper    dsm_paa    wisc_paa
0035 %       paaIN    0.1    0.2
0036 %   The upper bound for the reaction paaIN should be 0.1 for the first
0037 %   condition and 0.2 for the second
0038 %
0039 %   -FITTING
0040 %       Fit to    dsm_paa    wisc_paa
0041 %       co2OUT    2.85    3.05
0042 %       glcIN   1.2     0.9
0043 %   The measured fluxes for CO2 production and glucose uptake for the two
0044 %   conditions. The model(s) can later be fitted to match these values as
0045 %   good as possible.
0046 %
0047 % Usage: experiment=getExpressionStructure(fileName)
0048 
0049 [type, sheets]=xlsfinfo(fileName);
0050 
0051 %Check if the file is a Microsoft Excel Spreadsheet
0052 if ~strcmp(type,'Microsoft Excel Spreadsheet')
0053     EM='The file is not a Microsoft Excel Spreadsheet';
0054     dispEM(EM);
0055 end
0056 
0057 %Check that all sheets are present and saves the index of each
0058 exprSheet=find(strcmp('EXPRESSION', sheets));
0059 boundSheet=find(strcmp('BOUNDS', sheets));
0060 fitSheet=find(strcmp('FITTING', sheets));
0061 
0062 if length(exprSheet)~=1 || length(boundSheet)~=1 || length(fitSheet)~=1
0063     EM='Not all required spreadsheets are present in the file';
0064     dispEM(EM);
0065 end
0066 
0067 %Load the expression data
0068 [discard,dataSheet]=xlsread(fileName,exprSheet);
0069 experiment.data=discard;
0070 experiment.orfs=dataSheet(2:size(dataSheet,1),1);
0071 experiment.experiments=dataSheet(1,2:size(dataSheet,2));
0072 
0073 %Loads the maximal boundaries
0074 [discard,dataSheet]=xlsread(fileName,boundSheet);
0075 experiment.boundNames=dataSheet(2:size(dataSheet,1),1);
0076 experiment.upperBoundaries=discard;
0077 
0078 %Loads the experimental data to fit to
0079 [discard,dataSheet]=xlsread(fileName,fitSheet);
0080 experiment.fitNames=dataSheet(2:size(dataSheet,1),1);
0081 experiment.fitTo=discard;
0082 
0083 %Check to see that the dimensions are correct
0084 if length(experiment.orfs)~=size(experiment.data,1) || (length(experiment.experiments)~=size(experiment.data,2) && ~isempty(experiment.data))
0085     EM='The expression data does not seem to be formated in the expected manner';
0086     dispEM(EM);
0087 end
0088 end