FAQ  •  Register  •  Login

CyTOF comp beads-best practices questions

<<

mleipold

Guru

Posts: 5796

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Tue Mar 09, 2021 4:37 pm

CyTOF comp beads-best practices questions

Hi all,

We haven't yet adopted capture bead compensation (like in the initial Chevrier paper, or updated in the Os-labeled Budzinski paper). But I was recently talking with someone analyzing a dataset with comp data, and a couple questions came up.

1. In this dataset, there were 2 panels, so 2 sets of Comp bead files. These comp bead samples were run separately from the cells (no Os-labeling to distinguish). One file appeared to contain EQ beads (Ce140 signal, but no Ce140-labeled antibody), while the other did not.
- This seemed odd to me: shouldn't you include EQ beads to first norm your Comp before doing the Comp (just in case your machine was behaving differently)?
- looking back at the Chevrier paper ( https://doi.org/10.1016/j.cels.2018.02.010 ), there is no statement of including EQ norm beads: "Briefly, for each channel assessed in the panel, one full drop of BD Compbead was loaded in a well of a v-bottom 96 well plate and stained with 1 μg of the corresponding metal-labeled antibody. Beads were stained for 15 min at room temperature. After staining, beads were washed three times in CSM and then pooled in a single tube. Beads were then fixed in 1.6% PFA/PBS for 1 h at room temperature. After fixation, beads were washed twice in PBS and twice in water. Before acquisition, beads were resuspended in 500 μL of water. Bead and cell data were acquired on a Helios mass cytometer (Fluidigm) using instrument-based dual-count calibration, noise reduction, and randomization."

- in the Budzinski paper, because of the Os-labeling, the comp beads could be pooled with the cells and run together, so EQ beads were also present because of the cells: "For the experiment shown in Fig. 3, a large batch of beads was labeled with OsO4, aliquoted into cavities of a 96-well Deep Well Plate (Corning, Corning, NY), and incubated with individual Ab conjugates (Supplemental Table I). After loading of beads with Ab conjugates and washing, bead aliquots were pooled and either stored at −80°C (Supplemental Fig. 1B) or directly combined with cell suspensions previously stained with Intercalator-Ir and the same set of isotope-tagged Abs (1:4, v/v) for simultaneous acquisition."


2. Also in this dataset, 7 batches were run in a relatively short period of time (a few weeks), and then an 8th batch was run a few months later. As far as we can tell from Comp bead FCS file header timestamps, they only acquired comps once, during Batch 1 of the 7 batches.
- This also seems odd: it seems weird enough that they wouldn't acquire an aliquot of Comps with each day's acquisition, but it seems even weirder that months later they would reuse the same Comp file.

- So, that leads to my next question: for those people who are regularly performing CyTOF compensation, how often are you running a Comp bead sample (separate or spike-in)?
- I would have assumed that people are doing this *each day*, but I know even in the fluorescence flow world, people don't always do that.
- Note: "do you run comp beads with every experiment?" is different than "do you prep comp beads with every experiment?"; Budzinski et al state "Importantly, once prepared, compensation bead pools were storable at −80°C for at least 2 wk without affecting the signal intensity of the captured Ab conjugates (Supplemental Fig. 1B), simplifying its use as a routine spike-in compensation control in mass cytometry."


Mike
<<

GregBehbehani

Master

Posts: 85

Joined: Tue Apr 12, 2016 10:17 pm

Location: The Ohio State University, Columbus, Ohio

Post Tue Mar 09, 2021 7:14 pm

Re: CyTOF comp beads-best practices questions

Hi Mike,

We don't routinely perform compensation, but when we do, we typically perform it on the basis of the known isotopic contamination of the isotope in question and a manual adjustment to create an average of 0 counts into the channel receiving the spillover. I don't think running beads daily will help much, since they will only approximate your spillover for reasons I will discuss below. I will preface all of the following, by saying that we haven't tried to extensively test bead-based compensation, so it would be great if someone were to take the time run daily bead controls for a few weeks, to see if things really are consistent, but I bet they will all be closer to each other than the bead spillovers will be to spillovers in your actual cells. My reasons are as follows:

In my experience, when compensation is needed, the signal creating the spillover is out of the linear range of the detector (which only goes up to about 1,000 - 2,000 counts), so you end up having to manually compensate based on cell events that are known or expected to be negative for antigen on the channel receiving the spillover. This tends to be close to the expected isotopic contamination, but is rarely exactly the same. As a result, I don't think it's really necessary to run your beads every day, since they are still only going to only be an approximation. I think the only way to do this perfectly would be to have a bead mixture with different antibody binding capacities (e.g. beads of 5-10 three-fold steps in intensity), you would then need to pick the beads that were closest to the signal intensity of the cells you need to compensate.

I don't think the beads really need to be run daily or run with Eq beads. Since you're only compensating the relative spillover, that percentage will only be minimally affected by the absolute machine sensitivity (the machine sensitivity will affect all of the measured channels equally and cancel out when calculating a percentage). The only case where this would matter would be the non-linear spillovers, but the beads can't normalize a non-linear signal change, so they won't help with that anyway. As far as running daily, this is important in flow because there are multiple detectors, the sensitivity of which can be independently changed, and that in some cases, deteriorate over time in different ways. In addition, the fluorophores can also change on an almost daily basis. None of this really happens in mass cytometry. In mass cytometry, most of the spillover is due isotopic contamination, which is a property of the metal preparation used for labelling and is essentially fixed, not only for that reagent, but for that entire preparation of isotopically purified metal (Fluidigm hasn't ever disclosed how many lots of metal are out there, but I suspect there aren't many). There is also only one detector, the sensitivity of which decreases essentially equally across all measurement channels.

Thus, I think running beads once every few weeks or months would be fine, but is still only going to give an approximate compensation. The only exception might be spillover due to oxide. I still doubt that you can perfectly replicate this with beads, since oxide can vary based on the plasma conditions (which might differ slightly between cells and beads), but a daily bead sample will probably get you closer than one a few weeks old.

I generally just try very hard to avoid problematic spillovers in your panel design, and I don't think there is a great solution to deal with them once they occur. If you have a really critical experiment where you're stuck with a bad spillover, I think you will either need to manually adjust the compensation, or run single antibody-stained control cells for just the channels that are causing problems. Additionally, if you know in advance, you can use the "cold-competitor" antibody approach we published (Sekhri et al., Cytometry A, 2019, PMID: 31120628), to reduce the signal on the channel creating the spillover to a level that is either not problematic, or down below 1,000 counts so that you can perform an accurate linear compensation.

Best,

Greg
<<

EricHaas

Participant

Posts: 9

Joined: Tue Jun 30, 2020 7:43 pm

Post Wed Mar 10, 2021 8:13 pm

Re: CyTOF comp beads-best practices questions

Hi All,

To Mike's question about frequency of comp bead use, I haven't seen many people use it, but one of my former users would perform comp bead analysis with each batch of samples run, essentially being every week (but not always each day) they ran samples and I don't believe they spiked EQ beads into the comp bead samples.

That being said, I agree with Greg's point about avoiding problematic spillover through panel design. Additionally, ensuring proper titrations are performed goes a long way in preventing much of the spillover. I generally recommend people titrate in three rounds and if necessary/they have the time, money, and patience, to do single metal titrations. This along with practicing the protocol on non-critical cells seems to work well in most cases of preventing spillover, thus eliminating the need of running comp beads. In my experience the bead signal and cell signal for a given metal can be drastically different, even stained with proportional amounts of antibody.

In general, I'm not sold on the necessity and effectiveness of the use of comp beads for cytof, as it seems to be more of a band-aid for larger issues and may provide a sense of security to allow too much leeway for poor experimental design and/or sloppy technique. I would rather a user take the time to set their panel and protocols up properly and take time to validate both, then most signal overlap issues are negligible. (I know at this point, I'm preaching to the choir in this thread! :) )

Eric

Return to CyTOF general discussion

Who is online

Users browsing this forum: No registered users and 12 guests