Compensation slide acquisition

Looking for ways to analyze your dataset? Need help with a software package?
Computational image analysis of multiplexed Imaging Mass Cytometry acquisitions
Post Reply
User avatar
sbooth
Posts: 8
Joined: Wed Apr 25, 2018 9:40 pm

Compensation slide acquisition

Post by sbooth » Mon Nov 12, 2018 6:27 pm

Hi all,

We are running a 33x Ab panel on the Hyperion at the moment and I would like to try to generate a compensation slide. I think the protocol from Vito is very clear for preparing the slide, but I am wondering how you actually acquire such a slide on the Hyperion? i.e. do you do a single acquisition over the whole area, or do you try to acquire each Ab spot individually? I had thought to lay out the Ab spots on the slide like that attached, unless there is a better way.

There are definitely pros and cons to each; acquiring the whole area would take a long time (= money) and generate a very large mcd file with largely 0 value pixels between the Ab spots. On the other hand acquiring each Ab spot individually could also be very tricky to precisely ablate each very small ROI.

Any advice would be greatly appreciated! Also if you have any photos or diagrams to assist that would be great :) . As we are not performing the acquisition at our site I will have to describe the process to the Hyperion core operator.
Best,
Steve
ss.jpg
ss.jpg (20.96 KiB) Viewed 1111 times

ssivajothi
Posts: 30
Joined: Thu Jan 11, 2018 9:00 pm

Re: Compensation slide acquisition

Post by ssivajothi » Mon Nov 12, 2018 6:40 pm

Hi Steve,

Your diagram is on track. That is how I have patterned the comp slide. The way I do the acquisition is to generate a panorama of the entire spotted region and go in and draw 5px tall ROIs inside each spot. Use the output from this to generate your compensation matrix. Hope this helps.

Best,
Santhosh

User avatar
votti
Posts: 21
Joined: Mon Nov 27, 2017 12:46 pm

Re: Compensation slide acquisition

Post by votti » Tue Nov 13, 2018 8:35 am

Hi,

I think my suggestion form the protocol (https://docs.google.com/document/d/195e ... sp=sharing) is still the one I would recommend:
2. Load the slide and generate panoramas
It can be easier to generate 1 panorama per row of the spillover array. These panoramas can be huge and take time to acquire, but it is often the fastest way to locate the right spots.

3. Set the acquisitions ROI ca 2000 pixels (10x200 ROI) should be enough
Unfortunately sometimes it is hard to see the spots, even with Trpyan blue. In such case it is often easier to look for the boundaries of the spots.
So I would do exactly the array layout you describe. Then I usually do 1 panorma overview per row, going through all the spots of a row. Then within each spot I select a Roi of 10x200 that is within each spot.

If you are short on IMC instrument time, doing one panorama per spot would be likely faster to generate the panorama, but would take much longer to setup.

Does this make sense?

User avatar
sbooth
Posts: 8
Joined: Wed Apr 25, 2018 9:40 pm

Re: Compensation slide acquisition

Post by sbooth » Thu Nov 15, 2018 11:03 pm

Thanks Vito, Santhosh. Your suggestions are great.

Fingers crossed I guess!

Steve

Post Reply