FAQ  •  Register  •  Login

Design for Longitudinal Sample Acquisition

<<

rhannan

Participant

Posts: 2

Joined: Tue Feb 15, 2022 5:05 pm

Post Mon Sep 12, 2022 4:58 pm

Design for Longitudinal Sample Acquisition

Hello isotope friends!

I've discussed with my peers locally, but want to poll the depth and breadth of experience on the forums.

I am planning an experiment with six timepoints across seven weeks. The cells will be stained live, so staining will happen (from a frozen master mix) on each of the harvest days, then fixed and barcoded (intracellular pd barcode).

The two options:
1. Run each sample as quickly as possible post-staining+fixing (no more than a 1-2 nights in intercalator in the fridge). This would stretch acquisition dates across a two month period.
2. Cryopreserve stained+fixed sample until completion of the experiment, then run all the samples sequentially on the instrument within the shortest possible timeframe (1-3 days of acquisition). This would involve a freeze-thaw of fixed sample.

In the past I have performed both of these methods in various experiments, and haven't noticed any differences in normalization or variance between samples. I ask you all, then, which is the greater evil? A freeze-thaw cycle of stained/fixed cells, or spreading an acquisition out across two months?

I appreciate your perspectives!

~Riley
<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Mon Sep 12, 2022 7:59 pm

Re: Design for Longitudinal Sample Acquisition

Hi Riley,

I'm not sure I understand your experimental layout. You said "The cells will be stained live, so staining will happen (from a frozen master mix) on each of the harvest days, then fixed and barcoded (intracellular pd barcode)."

Does that mean that you're barcoding by Week (ie, Week 1 is a pool of all Donor samples from Week 1, Week 2 is a pool of all Donor samples from Week 2, etc)? Or are you barcoding samples individually and then planning to thaw and pool before running (ie, Donor A will be its own pool of 6 barcodes stained across 7 weeks, and run separately from Donor B's pool)?


We haven't seen problems with FBS+DMSO-frozen samples in any testing that we've done so far, but we do formally test every experiment during development if we're planning to do that.....we don't automatically assume that it will be fine. Same with the frozen cocktails: every cocktail goes through fresh vs frozen testing, and while 90+% of all antibodies seem to be fine, we have found a few that do have diminished response after freezing.

So, in general, whether you run them by Week or wait til the end and run them all in a row, shouldn't make a ton of difference in data quality.


That said: whenever possible, we try to put all timepoints for a given donor into the same experiment. In most cases, we try to *stain* all donor-timepoints together *and* run all timepoints together to minimize overall variability. This variability has at least two components: staining/reagents, and then machine performance. In your case, your staining/reagents would still have some variability even with frozen cocktails because you're performing the staining by week. Pooling all of Donor A's samples and running as a single sample at the end would technically only address the machine performance part.


Mike
<<

rhannan

Participant

Posts: 2

Joined: Tue Feb 15, 2022 5:05 pm

Post Tue Sep 13, 2022 2:12 pm

Re: Design for Longitudinal Sample Acquisition

Mike, appreciate the reply!

Apologies for the ambiguity - each 'donor' is actually comprised of 15 individual experimental samples and will be run by week, as you described in your first example. I have yet to see issues with any frozen master mix or cryopreserved stained cells, so I appreciate hearing that not all antibodies behave well post-freeze. I'm happy to hear you don't have a preference one way or the other! I often worry there are standard procedures in the field of which I am not aware.

I am hoping we will be able to manage some spike-in control samples which will assuage my concerns somewhat.

Thanks!
Riley
<<

cguidos

Contributor

Posts: 28

Joined: Tue Nov 18, 2014 3:10 am

Post Tue Sep 13, 2022 2:57 pm

Re: Design for Longitudinal Sample Acquisition

Hi Mike and other CyTOFers

I have related question on this topic. To avoid compromising fix-sensitive markers we tend to stain>fix/perm>Pd barcode. In other words we barcode only to allow multiplexed acquisition on the Helios.

So wondering if anyone has Pd-barcoded stained samples after freeze/thaw and if so how did it work? Or do you only Pd barcode (before or after Ab staining) prior to freezing the stained cells? This would present some logistical challenges but is do-able I guess

Cynthia
<<

CRStevens

Master

Posts: 60

Joined: Thu Jul 17, 2014 5:07 pm

Post Tue Sep 13, 2022 3:12 pm

Re: Design for Longitudinal Sample Acquisition

Hi All,

We have run a few longitudinal studies as well as large pre-clinical studies.

Our protocols are all based on staining live cells. We do our surface staining, fix, then freeze with 90% FBS + 10% DMSO (at -80). Once we have all of our samples I thaw all the samples, do a series of washes in stain buffer, then resuspend in the Maxpar Fix/perm to do Ir intercalation and I also simultaneously add my barcodes. Usually let that stain for 1h, then continue to wash the cells in PBS a couple times, combine all the samples then do a final water wash prior to acquiring. We have had good success with this protocol and it does not help any issues with barcoding for sample staining batch effects, but does effectively allow for running all samples at the same time to reduce batch effects from the instrument. For really large studies, I also will stain my samples with a different CD45 ever 20 samples so I can combine all samples still and use my CD45 as an additional barcoding.

-Chad
<<

AlexSchmitz

Participant

Posts: 2

Joined: Tue Jun 29, 2021 8:18 am

Post Wed Sep 14, 2022 9:06 am

Re: Design for Longitudinal Sample Acquisition

Hi Chad,

Barcoding during the intercalation step. - Very interesting.

Is this published somewhere ?
Are you using the commercial barcodes from the kit or do you have your homemade ones ?

Best
Alex




CRStevens wrote:Hi All,

We have run a few longitudinal studies as well as large pre-clinical studies.

Our protocols are all based on staining live cells. We do our surface staining, fix, then freeze with 90% FBS + 10% DMSO (at -80). Once we have all of our samples I thaw all the samples, do a series of washes in stain buffer, then resuspend in the Maxpar Fix/perm to do Ir intercalation and I also simultaneously add my barcodes. Usually let that stain for 1h, then continue to wash the cells in PBS a couple times, combine all the samples then do a final water wash prior to acquiring. We have had good success with this protocol and it does not help any issues with barcoding for sample staining batch effects, but does effectively allow for running all samples at the same time to reduce batch effects from the instrument. For really large studies, I also will stain my samples with a different CD45 ever 20 samples so I can combine all samples still and use my CD45 as an additional barcoding.

-Chad
<<

CRStevens

Master

Posts: 60

Joined: Thu Jul 17, 2014 5:07 pm

Post Thu Sep 15, 2022 3:08 pm

Re: Design for Longitudinal Sample Acquisition

Alex,

I use the commercial Pd barcodes. They are both stained in maxpar fix/perm, so just assumed they will work together. No issues on my end so far. If I do an overnight Ir intercalation I will leave barcodes out until the next day.
No publication to my knowledge.

-Chad
<<

jobagins

Participant

Posts: 13

Joined: Thu Oct 22, 2020 6:45 pm

Post Thu Sep 15, 2022 3:22 pm

Re: Design for Longitudinal Sample Acquisition

Hi all

we have few things published on this

one is

https://star-protocols.cell.com/protocols/1624

cheers

joanna

Return to CyTOF general discussion

Who is online

Users browsing this forum: No registered users and 14 guests