Pfc 2: MATLAB file, missing clu files and relation

The documentation says that the MATLAB file EE.188_example.mat, in the EE0708 folder, contains the final data actually used for the main conclusions in the Nature 2008 paper it cites. But two versions of MATLAB say it cannot be loaded, file corrupt. (To be sure we downloaded and decompressed many times.) Two versions of octave say it cannot be loaded because zlib cannot decompress numerical value. To be honest, since it is only 3 megabytes, is there a way to get it as simple email attachment? That would be a lifesaver.

Failing that, we could try to get the information by processing the clu and res files in the same EE188 folder which we get from decompressing the TAR. Those load fine in MATLAB, octave and even excel. But I note that clu seems to be missing for other trials we glanced at, like EE040 through EE048. More seriously, can we say we are using the same data that fujisawa and buzsaki used if we just merge all the spikes good which clu>1?

Since I don't see a copy of that Anton m file mentioned in the documentation, I can't tell. I have a suspicion that we should be using the fet and spk file information as well, but not sure how: fet loads up nicely in excel at least. 

Am puzzled why there are 13 or 14 clu files in each complete session but only 8 shanks. Also, are the neurons all cerebral cortex, not based on the 32 channels of original data from hippocampus? Or are the extra shanks hippocampus ????

Thanks for your patience. The MATLAB file is all we need urgently, but it would be really nice go know the rest.

OK, I will send it as email attachment. Please try the file.

About the second question: There are several versions for clu files and we might've uploaded older files for some folders. I'll check them ASPS.

About the third question: clu.1-.8 were recorded from PFC with using a 8-shank probe, and clu.9-.12 were simultaneously recorded from CA1 with a 4-shank probe. Please neglect clu.13 and .14, these were from external signals.

Thank you ever so much, Dr. Fujisawa! The new example MATLAB file works fine. It is also good to know which is cerebral cortex and which is hippocampus, simultaneous. It also seems, for the other files, the res.x and clu.x are all I would need. At arxiv.org, paper 1309.2848 by Kadir et AL provides important documentation on today's best spike sorting, substantially improved, and suggests I do not need to use fet.x or spk.x or mask.x files unless someday we try to develop a new way to do the spike sorting.

Also, I am sorry I did not think of looking for clu.x files in the merged tar when they were missing from tar of some individual trials. Of course your paper in Nature 2008 is such an important starting point for the most fundamental questions.

Sorry, the original 'EE188_example.mat' includes only PFC units, and does not include CA1.

I attach another file which also includes CA1 units. (I will send it to you by email.)

The shanks #1-8 were from PFC, and #9-12 were from CA1, which are indicated in the second column of 'CellsInfo' matrix.