createTissueSpecificModel problem (vol.2)

[Zhu Wenhan]

Hey guys, 

I was fiddling around with the create tissue specific model function a bit and was able to get pass the "Matrix dimensions must agree." error. Now I am getting a new error : 

>> [ModelGlc,RxnsGlc] = createTissueSpecificModel (model,ExpressionData,[],[], [],'GIMME')

Error using cell/setdiff>cellsetdiffR2012a (line 291)

Input A of class double and input B of class cell must be cell arrays of character vectors,

unless one is a character vector.

Error in cell/setdiff (line 83)

    [varargout{1:nlhs}] = cellsetdiffR2012a(varargin{:});

Error in createTissueSpecificModel>mapProbes (line 472)

unknown = setdiff(unknown,rxnExpressed);

Error in createTissueSpecificModel (line 145)

[ExpressedRxns,UnExpressedRxns,unknown] =

mapProbes(parsedGPR,corrRxn,locus,genePresence,match_strings);

I have made sure both the locus and data are double instead of cell. And the error still occurs if I change the 'model.genes' into double. I have attached the relevant files.

Any input is appreciated! 

Best,

Wenhan

[Aarash Bordbar]

Hi Zhu,

Do you have the matlab file of the variables so this can be replicated?

Best

Aarash

--

---
You received this message because you are subscribed to the Google Groups "COBRA Toolbox" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbox+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Aarash Bordbar, Ph.D.

[Zhu Wenhan]

Hi Aarash, 

Thanks for helping! Please find the files attached. The matlan file for the variables are in "ExpressionData.mat".  Please let me know if you need anything else:) 

Best, 

Wenhan 

To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbo...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Aarash Bordbar, Ph.D.

[Aarash Bordbar]

Hi Zhu,

What is happening is that the algorithm cannot match the GPRs in the model with the data you are entering. Thus there are no reactions that are expressed or unexpressed. When the algorithm calls the Matlab function setdiff it is comparing two empty arrays and thus crashes.

Looking at the variables, the ExpressionData.Locus identifiers should match exactly to what's in model.genes, which should match exactly to model.grRules. I would recommend setting them all up as strings and making sure they are exactly the same. (There are some differences I saw in Ids such as Id 59 vs 0059).

Hope this helps,

Aarash

To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbox+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Aarash Bordbar, Ph.D.

[Zhu Wenhan]

Hi Aarash, 

Thank you so much for the advice! 

I think what you are saying is absolutely correct, and that is what is causing the command to crash. I tried to make ExpressionData.Locus and model.genes identical (ExpressionData.Locus = model.genes) but that still did not work...I think actually when createtissuespecific model was implemented, it only took pure number as input, and that is why I previously changed Locus to double...

What do you think?

[Aarash Bordbar]

From my previous inspection, it seemed the grRule strings were not all doubles.

Best

Aarash

To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbox+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Aarash Bordbar, Ph.D.