Page 1 of 1

Tolerance in panel design

PostPosted: Mon Oct 23, 2017 3:14 pm
by ssivajothi
Hi all,

I have a basic question when it comes to panel design, especially with regards to 'tolerance'. When designing a panel with ~40 antibodies, I find it impossible to make a 'clean' panel. I want to ask the users here, based on your experience, how much spillover can you tolerate? Is it essential that the spillover does not cross the particular antibody's 'tolerance' value? The general guidelines seem to be to match marker abundance with appropriate bright/dim metal, try to place mutually exclusive markers in more dirtier channel etc. But following all these guidelines often leads to many custom conjugations (which seems to further deter users from adopting cytof, since we don't maintain an antibody bank). Is this just the way it rolls or am I missing something obvious?

Thanks for your input,


Re: Tolerance in panel design

PostPosted: Wed Oct 25, 2017 3:03 pm
by mleipold
Hi Santhosh,

Bluntly, if you want to do CyTOF, you have to accept the cost; it's not a cheap technology.

I try to keep my background (including spillover) to under Dual=10, ideally under Dual=5. This was discussed nicely in the Takahashi et al paper: viewtopic.php?f=10&t=557&p=1731&hilit=takahashi#p1731

If you're doing CyTOF, you're also going to be doing custom conjugations. Fluidigm conjugates are generally good, but there are a few that I find problematic for spill reasons (CD57-Nd142, CD45-Pr141, etc) or resolution. Even setting those aside, almost every customer I have has some markers they want that Fluidigm doesn't sell at all, or they sell the "wrong" clone, or it doesn't fit into the open slots I have in my standard panel.

I'm a big fan of MMO experiments (like FMOs in flow): to me, they're the *only* way to conclusively say what is undesired signal (isotopic spill, nonspecific background, etc) and what is true signal. But, they can be expensive in terms of time, money, and material.

At the end of the day, you (and/or your customers) have to make the best panel you can; evaluate where problem areas are; and then decide whether those problems are small enough to live with or large enough that you have to remove the markers.


Re: Tolerance in panel design

PostPosted: Mon Oct 08, 2018 4:35 am
by singh423
Hi Mike
what is MMO here you mentioned?

Re: Tolerance in panel design

PostPosted: Mon Oct 08, 2018 3:29 pm
by mleipold
Hi Amar,

MMO = Metal minus one. It's the CyTOF version of the fluorescent flow cytometry "FMO" (fluorescence minus one). Basically, you stain with everything *except* the channel you're concerned about spills in. You compare that with the full panel, and see how/if the signal intensity changes in that panel.

So, for example, let's say you have a 30 marker panel and you're concerned about spill vs true signal in, say, the PD-1 channel. In that case, you would run the full 30 marker panel, and *also* run a 29 marker panel (everything *but* the PD-1).