search

cortex-lab / allenCCF

Tools to work with Allen Inst CCF data in matlab
122 stars 78 forks source link

Preparation of histology slices #37

Open JannaAarse opened 4 years ago

JannaAarse commented 4 years ago

Hi, first of all I would like to thank you for this great tool. I just finished the biorxiv paper and was start using the SHARP-Track code on my slices. So far my pictures are not perfect (because they were done fast , immediately after the slicing, I have to redo them and rerun the code but I had some questions before) and without any staining, I just used the fluorescence of the electrode track and another fluorescence channel to overlay so that I can make some distinctions about the position I am at. In your paper I did read that you stained your slices with DAPI, I was wondering if any staining method is necessary ? Would a nissl staining also work (the staining is mentioned in the last part of your paper and that it could help to improve the identification of slices so I thought it might be "enough" to stain with Nissl ? ) Or are there any other pre-steps that need to be done to the slices before I make the pictures?

Best, Janna

JannaAarse commented 4 years ago

Hey,

I was able to get my hands on some DAPI and now the pre-processing works and I can set transformation points, however when I set multiple points and than click "h" to overlay the atlas with the slice the slice is cut and it is not aligned correctly. I attached a screenshot of it. I saw in a different post that some aligning issues also happened in a sagital view but that this was already solved by changing reference_sice=size(tv_plot). Do you have an idea how I can fix the aligning in the coronal plane ? I tried different pictures but it happens with all of them.

Untitled

JannaAarse commented 4 years ago

Just in case someone has the same problem, it was a scaling problem. With help I found out that my value for my microns/pixel was off, rerunning the code with a better value helped. No offset anymore, everything works great now.

philshams commented 4 years ago

Nice, glad that you could fix that issue.

As for staining, for future reference, any staining that allows you to see the landmarks that are present in the reference atlas is sufficient. Nissl staining should work fine.