superficial layers (L2/3) vs. deeper layer (L5/6) miniScope signals

Skip to first unread message

Yaoying Ma

Apr 22, 2022, 5:25:08 PMApr 22
to Miniscope

I am wondering if anyone has experience in collecting the data from superficial layers (L2/3) and deeper layer (L5/6) miniscope signals. Any tricks to implant the lens for imaging data collection in the cortical areas? We did not scoop out any brain tissue. Let me know if you did it in a different way. 

If there is any dura left in between the lens and brain tissue, how about the quality of ca2+ imaging? 

Also, I am wondering if there are any well-identified layer-markers for CRE-FLOX system to be used to specifically express GCAMP in specific layers (e.g., PT vs. IT).


xiaoyu peng

Apr 24, 2022, 12:53:03 AMApr 24
to Miniscope
Hi, Yaoying,

Check our paper  for motor cortex L2/3 imaging (V3 mostly,  but V4 scope is even easier). 

For V4 scope,  you just need a glass window,  no lens implantation for L2/3.  Don't need to open dura.   Just make sure to put on very thin layer of dental cement so the scope can get to the window closely so you can access more neurons.   You need high quality cranionomy window surgery.  

For V3 scope,  depends on the lens combination and working distance.  I used to use a combination that I implant the objective lens against dura (press down  ~200 um) .   That is what I used to collect most data.   Later I found out that the lens combination in this product (  can give you ~ 500 um working distance so you can also use it to image through glass window,  the objective lens stays with the scope and outside of the window.  

In my hand,  keep the dura intact is better,  the window will be clearer.  

I don't have experience with L5/6 neurons. 



Daniel Aharoni

Jun 17, 2022, 12:21:28 AMJun 17
to Miniscope
Hi! I second everything Xiaoyu said and will add that very very likely, layer 5/6 is just going to be too deep to be able to resolve single neurons. Usually you will be limited to 250um to 300um of tissue thickness before scattering becomes too big of an issue.
Reply all
Reply to author
0 new messages