Export to Excel


After some problems with the initiation, Cell Profiler is now running nicely on my Windows Vista. However, Export to Excel is not working. Trying to export with either ExportToExcel or ExportData, I get an error message.

In case of ExportData:

Error using ==> cat

CAT arguments dimensions are not consistent.

CPwritemeasurements in …\CPSubfunctions\CPwritemeasurements.m(82)
ExportData in …\DataTools\ExportData.m(112)

I don’t know whether this could be due to the fact that I am using Excel 2003 on Windows Vista.

Hope you can help me.


Which example pipeline are you trying to use?

Hi, it is not an example pipeline.


Please send me an image set and your pipeline to mvokes@broad.mit.edu as a zip file.

I have done that

Sorry, forgot to put ExportToExcel at the end.

Hi DS,

Yesterday I found a bug in the Relate Module for another user that caused a similar error. Basically, if there is an object that is being “related” that has a zero count in an image, then Export* has trouble writing the data. I haven’t seen your output file so I can’t be certain, but it sounds like your problem.

You could try either:
(1) Removing the Relate Module until it get fixed for the next CP release
(2) Tweak other settings to make sure that at least one object being Related is found in each image

Let us know if this helps, and regardless I will fix this for the next release.


Yes, I think you are right. I also get error messages because of division by zero.

I can try to avoid the relate module. However, I need to know which of the identified objects belongs to a transfected cell and for this I will need a link between the cell and this object.

best wishes


The problem will be resolved in the next release of CellProfiler. Sorry for the delay.

I am new to CP, but I’m guessing that there has been a release since the last post.

I am still getting this error and it does seem to directly relate to the zero count. When I take away image data that will cause zero count, the export to excel module works fine.

I’ve had a look at using data tool: clear data, but it doesn’t seem to work - I can’t select the defaultOUT.mat file (or is it meant to be some other file?)
Consequently, I have to run the pipeline again, and with so much data, it takes my machine over 3 hours

Any ideas?


Hi Timothy,

Yes, there has been a new release since November (and another is coming in the next few weeks). Check out the Downloads page. The 5122 version release ought to fix the zero object count exporting problem.

ClearData opens a dialog box in which you first choose the directory which contains an output file. This behavior is not well documented, and is especially worrisome on Macs, since Matlab does not allow us to change the title of a Mac dialog box to indicate that the user should look for a folder (and not a file). In any case, we will try to make this clearer. Let us know if this helps you. And yes, this file defaults to DefaultOUT.mat, but could be anything you specified.

To speed up your processing time, be sure to turn off displaying images. We also distribute a computing cluster version of CellProfiler which can dramatically lower your processing time, but you need access to a computing cluster.


Hi David,

Looking forward to the new version! :slight_smile:

It seems I still get the export to excel error window when there is a zero count. I’m not sure if this helps, but I use the relate module and I think someone mention this module is related.

I have yet to try the cleardata tool, in the case I did not remove image data that will cause zero count.

With the cluster version of CP, is this an entirely different download, or is it a case of using the batch processing modules?
And with the “cluster” terminology, do you mean simply a few networked computers?

Edit: Sorry - found CPCluster on the download section - didn’t realise it was there!


Hi, I get this error quite frequently:

There was a problem running the analysis module ExportToExcel which is number 13. Error using ==> CPwritemeasurements at 203
Error in handles.Measurements structure. The field AreaShapeFeatures does not have an associated measurement field.

CPwritemeasurements in /Applications/MACI_1.0.5122/CompiledCP/CellProfiler_mcr/CPsubfunctions/CPwritemeasurements.m (203)
ExportToExcel in /Applications/MACI_1.0.5122/CompiledCP/CellProfiler_mcr/Modules/ExportToExcel.m (124)
AnalyzeImagesButton_Callback in /Applications/MACI_1.0.5122/CompiledCP/CellProfiler_mcr/CellProfiler/CellProfiler.m (9409)
gui_mainfcn in /Applications/MACI_1.0.5122/CompiledCP/CellProfiler_mcr/CellProfiler/CellProfiler.m (11138)
CellProfiler in /Applications/MACI_1.0.5122/CompiledCP/CellProfiler_mcr/CellProfiler/CellProfiler.m (55)

I use a MAC with OS X 1.4 (if that helps to track the problem). Do you have a suggestion, how I could work around that?


Ups…moved my Relate module…now it works. Still I get the info that some measurements have not been made and are therefore not saved. This is not an error message, though.

Sorry for the post… Dani

Yes, the Relate Module should be placed after all Measurement modules, since it calculates means of all measurements across children in a each parent object. If there are measurement modules below Relate in the pipeline, then these measurements will not have parent/children means. This will not necessarily cause an error, but we have inserted a warning in case your output looks odd.


See also ExportData chokes on fields with zero identified objects

I had the same problem and still looking to find the solution .any help ??