FAQ  •  Register  •  Login

2014-Yao et al-J Immunol Methods

<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Wed Sep 21, 2016 8:05 pm

Re: 2014-Yao et al-J Immunol Methods

Hi VIncius,

Length of acquisition depends slightly on the instrument version:
1. CyTOFv1 and CyTOFv2 with standard sample loops have a flow rate of 45uL/min and a cell transmission efficiency of ~20%.
2. CyTOFv1 and CyTOFv2 with a SuperSampler have a flow rate of 45uL/min and a cell transmission efficiency of ~40%. So, in theory, it would take half the amount of time to acquire the same number of cells. In practice, you often need to dilute your sample slightly more because of background, etc, so your time savings might not be half but maybe only 2/3 of the regular sample loop time.
3. At least currently, Helios (CyTOFv3) have a flow rate of 30uL/min and a transmission efficiency of ~40%. You get similar numbers if you use a SuperSampler on a Helios. Again, because of additional dilution (background, clogging, etc), you actually don't save much time relative to the v1 or v2 at the faster flow rate.

In short: to acquire 200-300K Ungated (found) cells, you're typically looking at ~20min acquisition time. Of that, you'll wind up with ~150K Live Intact Singlets (ie, all the garbage gated out and you're ready to do all your marker-based gating).


Regarding antigen-specific T cells: as I said in a previous post in this chain, their frequency is often reported as per 1 million CD4+. Therefore, you might need to collect several million cells in order to get your tens to hundreds of antigen-specific cells. Of course, this depends on the frequency in the sample: a normal healthy person might have a very low level of antigen-specific cells, whereas someone with an active infection or an active autoimmune condition might be much higher. You can often find approximate numbers in the literature to help you calculate your target number of cells, and whether enrichment might be necessary.


Mike
<<

cguidos

Contributor

Posts: 28

Joined: Tue Nov 18, 2014 3:10 am

Post Wed Sep 21, 2016 9:02 pm

Re: 2014-Yao et al-J Immunol Methods

HI Vinicius

For PBMC we typically suspend at ~800,000/ml to acquire on our Helios (lower than Fluidigm recommendation to minimize clogging). At their recommended flow rate of 30uL/min, it would take about 10' to acquire ~240,000 events for a non-bar-coded sample, and our capture efficiency ranges from ~55-65%. Thus we will end up with ~156,000 events in the FCS file, of which ~90% are single cells in a good sample. So typically we will acquire 10'/sample but 5' will suffice for titrations and Ab tests where >20% of the cells should be positive.

The cell counting statistics (which I think I sent you?) still apply to SPADE or Visne as they are just visualization tools. So in my example from above, we would measure 132,000-156,000 cells total (live singlets) during 10' of collection time. You can be highly confident that you can measure a 1% subset (=1560 events in your gate if efficiency is 65%) with a high degree of confidence (CV=2.5%). However, you would need to collect 10X more total events (1.6M) to detect a 0.1% subset with the same precision. If you collect for only 5' in this example, you can still measure a 1% subset (800 total events) with a CV<5% which is acceptable. So generally I advise collecting each sample for at least 5' but 10' is safer if sample quality unknown.

You also have to account for cleaning needed between each sample to prevent clogging. So for 10'/sample collection times we would run ~4-5 samples/hr, and about double that for 5' collection times.

Hope that helps

Cindy
<<

vmotta

Participant

Posts: 13

Joined: Mon Jan 26, 2015 10:34 pm

Post Thu Sep 22, 2016 9:00 pm

Re: 2014-Yao et al-J Immunol Methods

Hello Cindy and Mike,

Thank you so much for your answers.

I appreciate very much your help.

Vinicius
<<

vmotta

Participant

Posts: 13

Joined: Mon Jan 26, 2015 10:34 pm

Post Fri Sep 23, 2016 5:26 pm

Re: 2014-Yao et al-J Immunol Methods

Hello Cindy and Mike,

I did an excel sheet with the information you shared with me in order to automate the calculation for time of acquisition and number of events to run in CyTOF.

In case you may find it useful.

Please see the excel file attached and let me know if you have any comments, suggestions or any bugs.

Thanks again for your help.

Vinicius
Attachments
EstimatingNumberCellsvsTimeCyTOF.xlsx
(22.29 KiB) Downloaded 238 times
Previous

Return to Literature

Who is online

Users browsing this forum: No registered users and 46 guests