Error message: iMAT-CobraTools

Hi Everyone! I'm a MatLab begginner and I'd like to run the iMAT plugin to reach a comparative proteomic analysis.

My input is an Excel file with data about 2 growth conditions: Wildtype and Fe-growth.

initCobraToolbox(false)
changeCobraSolver('gurobi');

% Load the expression data from an Excel file
expressionDataFile = 'C:\Users\usuario\cobratoolbox\160824_BP-Fe.xlsx';  % Replace with the correct path
[num, txt, raw] = xlsread(expressionDataFile);

% Load the model .mat file
modelFile = 'C:\Users\usuario\cobratoolbox\iBP1870.mat'; 
load(modelFile);

% Select the columns I want to analyze from the Excel file
geneNames = txt(2:end, 3);
expressionLevels = num(:, 17);

% Check if there are any cells with NaN values
nanIndices = isnan(expressionLevels);
if any(nanIndices)
    warning('There are %d NaN values in the expression levels.', sum(nanIndices));
end

% Remove NaN values from the expression data
expressionLevels = expressionLevels(~any(isnan(expressionLevels), 2), :);

% Assuming `nanIndices` is the index of the elements that were NaN in expressionLevels
% and that you have already used it to remove the NaN values:
% Create indices that are NOT NaN
nanIndices = ~isnan(expressionLevels);

% Filter geneNames using nanIndices, i.e., removing gene names that returned NaN values
filteredGeneNames = geneNames(nanIndices);

% Filter genes and expression levels that are in model.genes
isMember = ismember(filteredGeneNames, model.genes);
if any(~isMember)
    warning('Some genes in expressionDataadjusted are not in model.genes');
end

filteredGenes = filteredGeneNames(isMember);

% Run this:
expressionRxns = mapExpressionToReactions(model, filteredGenes, filteredExpressionLevels);

% Set the expression level thresholds
threshold_ub = 50.0; % Upper threshold for expression levels
threshold_lb = 5.0;  % Lower threshold for expression levels

% This is where I run iMAT:
tissueModel = iMAT(model, expressionRxns, threshold_lb, threshold_ub);

tissueModel = iMAT(model, expressionRxns, threshold_lb, threshold_ub);

Warning: There are 216 NaN values in the expression levels.

Warning: Some genes in expressionDataadjusted are not in model.genes

RHindex:

53

55

81

156

184

197

........

RLindex:

1

3

4

5

6

7

9

10

11

18

19

20

21

22

1652

1653

1655

1656

1658

1660

1662

1664

1665

1666

1667

1668

1669

1672

Verificando lb y ub para NaN o Inf...

Error using iMAT (line 61)

Vector ub contiene NaN o Inf valores

However, I can't run this code because I keep getting error messages relationated to RHindex.

There is not Nan o Inf values. It was checked

Someone knows how it can solve?

Thanks a lot

DEb

[Walter Roberson]

initCobraToolbox appears to be from https://www.mathworks.com/matlabcentral/fileexchange/69132-the-constraint-based-reconstruction-and-analysis-toolbox

[Walter Roberson]

We would need 'C:\Users\usuario\cobratoolbox\160824_BP-Fe.xlsx' to test with.

Answer 1

[Deborah]

ok..thank you..I attach the file

Answer 2

[Walter Roberson]

expressionDataFile = '160824_BP-Fe.xlsx';
T = readtable(expressionDataFile, 'VariableNamingRule', 'preserve');

Warning: Table variable names were truncated to the length namelengthmax. The original names are saved in the VariableDescriptions property.

summary(T)

T: 1309x25 table

Variables:

    Uniprot Accession: cell array of character vectors (Uniprot Accession)
    Uniprot ID: cell array of character vectors (Uniprot ID)
    locus: cell array of character vectors (locus)
    gene: cell array of character vectors (gene)
    description: cell array of character vectors (description)
    number total peptides [WT:WT Fe-:Hfq_Hfq Fe-, without OFF prote: double (number total peptides [WT:WT Fe-:Hfq_Hfq Fe-, without OFF protei...)
    number unique peptides [WT:WT Fe-:Hfq_Hfq Fe-, without OFF prot: double (number unique peptides [WT:WT Fe-:Hfq_Hfq Fe-, without OFF prote...)
    number of peptides for quantification [WT:WT Fe-:Hfq_Hfq Fe-, w: double (number of peptides for quantification [WT:WT Fe-:Hfq_Hfq Fe-, wi...)
    C1_WT_control_BR1: double (C1_WT_control_BR1)
    C2_WT_control_BR2: double (C2_WT_control_BR2)
    C3_WT_control_BR3: double (C3_WT_control_BR3)
    C4_WT_control_BR4: double (C4_WT_control_BR4)
    E1_WT_Fe-lim_BR1: double (E1_WT_Fe-lim_BR1)
    E2_WT_Fe-lim_BR2: double (E2_WT_Fe-lim_BR2)
    E3_WT_Fe-lim_BR3: double (E3_WT_Fe-lim_BR3)
    E4_WT_Fe-lim_BR4: double (E4_WT_Fe-lim_BR4)
    WT control average: double (WT control average)
    WT Fe- average: double (WT Fe- average)
    WT control standard deviation: double (WT control standard deviation)
    WT Fe- standard deviation: double (WT Fe- standard deviation)
    WT control CV [%]: double (WT control CV [%])
    WT Fe- CV [%]: double (WT Fe- CV [%])
    WT Fe-/WT control: double (WT Fe-/WT control)
    WT Fe-/WT control p-value: double (WT Fe-/WT control p-value)
    WT Fe-/WT control q-value: double (WT Fe-/WT control q-value)

Statistics for applicable variables:

                                                               NumMissing         Min              Median             Max               Mean              Std      

    UniprotAccession                                                0                                                                                              
    UniprotID                                                       0                                                                                              
    locus                                                           0                                                                                              
    gene                                                            0                                                                                              
    description                                                     0                                                                                              
    numberTotalPeptides_WT_WTFe__Hfq_HfqFe__WithoutOFFProte       121              0.0854            0.2550                19            0.9510            1.4072  
    numberUniquePeptides_WT_WTFe__Hfq_HfqFe__WithoutOFFProt       119              0.0854            0.2549                19            0.9452            1.4000  
    numberOfPeptidesForQuantification_WT_WTFe__Hfq_HfqFe__W         0              0.0854                 3                 3            2.0535            1.2091  
    C1_WT_control_BR1                                             311          1.1642e+06         217696632        3.6296e+10        7.1089e+08        2.0728e+09  
    C2_WT_control_BR2                                             320          2.2051e+05         217695616        4.6086e+10        6.8294e+08        2.2208e+09  
    C3_WT_control_BR3                                             310          3.2960e+06         217695616        4.0850e+10        6.4484e+08        2.0203e+09  
    C4_WT_control_BR4                                             332          1.5889e+05         217695616        4.3209e+10        8.0607e+08        2.4268e+09  
    E1_WT_Fe_lim_BR1                                              220          7.8416e+04         217696032        3.8837e+10        7.6263e+08        2.1097e+09  
    E2_WT_Fe_lim_BR2                                              216          9.6895e+04         217695616        2.7517e+10        6.9164e+08        1.7477e+09  
    E3_WT_Fe_lim_BR3                                              221          1.0184e+05         217695904        4.1295e+10        7.5235e+08        2.2386e+09  
    E4_WT_Fe_lim_BR4                                              217          7.0382e+05         217696088        3.2369e+10        7.4632e+08        2.0445e+09  
    WTControlAverage                                              310          4.2700e+06         218899698        4.1610e+10        7.0495e+08        2.1508e+09  
    WTFe_Average                                                  214             2170451         222730012        3.5005e+10        7.3520e+08        2.0007e+09  
    WTControlStandardDeviation                                    311          3.4850e+05        3.9544e+07        6.0446e+09        1.4589e+08        3.9954e+08  
    WTFe_StandardDeviation                                        216          8.5928e+05        3.7720e+07        6.8503e+09        1.4324e+08        4.3356e+08  
    WTControlCV___                                                311              2.6730           20.8692          141.2817           27.2149           20.2196  
    WTFe_CV___                                                    216              1.1092           19.7966          151.4677           26.0117           20.3106  
    WTFe__WTControl                                               404              0.0354            1.2057          317.7651            1.9056           10.6628  
    WTFe__WTControlP_value                                        410          3.4392e-08            0.0539            0.9974            0.1940            0.2651  
    WTFe__WTControlQ_value                                        410          3.0919e-05            0.1077            0.9974            0.2408            0.2812  

The "num" output of xlsread() would skip leading character vector arrays, so would skip the first four columns.

expressionLevels = num(:, 17);

That would refer to column 17 of the num output, which would be column 21 of T.

summary(T(:,21))

1309x1 table

Variables:

    WT control CV [%]: double (WT control CV [%])

Statistics for applicable variables:

                      NumMissing       Min          Median           Max            Mean            Std    

    WTControlCV___       311          2.6730        20.8692        141.2817        27.2149        20.2196  

That column is missing 311 inputs. Let's find them

idxmissing = ismissing(T{:,21});
obj = detectImportOptions(expressionDataFile, 'VariableNamingRule', 'preserve');
obj = setvartype(obj, 21, 'char');
T2 = readtable(expressionDataFile, obj);

Warning: Table variable names were truncated to the length namelengthmax. The original names are saved in the VariableDescriptions property.

summary(T2(:,21))

1309x1 table

Variables:

    WT control CV [%]: cell array of character vectors (WT control CV [%])

Statistics for applicable variables:

                      NumMissing

    WTControlCV___        0     

T2(idxmissing, 21)

ans = 311x1 table
    WT control CV [%]
    _________________

         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      
         {'na'}      

So the entries might not be inf or nan, but they are 'na'

[Deborah]

Thank you!

I already checked the Excel table. You were right, the numeric column is the 12th one. I ran it again, and it gave the same error. I reviewed each of the outputs, and a doubt arose regarding the limits for the model and the expression data configuration. For the model, it's 99999, and for the expression data, it's more than 1*10^8. Could this be the cause of the error message about NaN and Inf?

Verificando lb y ub para NaN o Inf...

Error using iMAT (line 61)

Vector ub contiene NaN o Inf valores

I looked for information, and it seems possible to normalize the expression data. Or is there another method that could be used to adjust my numerical data?

[Walter Roberson]

No, the base problem is that the file contains 'na' entries that are being converted to nan or inf by xlsread().