Staining Beads

[CRStevens]

Hi all,

We have been playing with staining beads (i.e. One Comp beads), with our metal conjugated Abs. I was curious what peoples experience with this was and whether some beads worked better than others. We've noticed some beads give cleaner results than others. Also, what is the best way to "clean up" your sample when gating on beads?

-Chad

[mleipold]

HI Chad,

I have used BD's anti-mouse Ig K capture beads (cat #552843) pretty successfully for single-metal antibody bead controls.

This has worked with the vast majority of my MAXPAR-labeled antibodies. Out of 33 or so tested, I think I've only had 2 consistently give issues: very dim or no signal on the beads, while still staining cells properly. I haven't bothered to track down the issue (eg, no kappa chain).

When I do these experiments, I run one bead-antibody at a time. Therefore, just gating on a high-M-metal signal is usually sufficient. I do make sure to do a number of washes, though.


I should mention: even with this bead data, we haven't been able to make a "compensation" matrix in FlowJo. Whether this is a math problem because of all the zeroes in the negative data, we don't know. But FlowJo doesn't make much, if any adjustment, with compensation on or off. This includes cases where we *know* there's an isotopic spillover (eg, Dy163 impurity in both Dy162 and Dy164 stocks).

If anyone has suggestions or comments regarding that, please chime in.


Mike

[Evan]

Hi Mike,

I'd also tried using flowjo to setup compensations without much luck. Instead I had little bit better success by arbitrarily fitting the spillover from one channel to another using a polynomial (using data where spillover was isolated). Then I used the fitted curve to subtract crosstalk in experimental data. Although there was significant of 'spread' in the compensated data, which I think is expected, this did seem to help in an instance of significant crosstalk. It might be interesting to try this kind of curve-fitting approach using the beads data you've collected.

Evan

[maciej]

Hi, I'd be curious to look at these data in the flowjo debugger if anyone could share a zip with me.

Maciej/FlowJo