[Cytometry] Compensation question: Substitution of Fluorochromes

Michele Black mblack2 at u.washington.edu
Thu Mar 10 11:56:46 EST 2011

I completely understand and agree with Mario. Not to go off on a tangent but
I do have researchers that try very hard to follow these compensation
"rules" and tend to get great data. The only issue is when we are trying to
identify a rare population usually dimly expressed just to make it more
interesting. The comp control using the specific antibody on the
experimental cells is not really possible, <.01% positive population does
not make a good comp control. 
Several possible options that I suggest; 1. using Comp beads with the
specific antibody 2. the same fluorochrome conjugated to an antibody that is
prevalent in the sample 3. A different cell type that has more of the
antigen of interest expressed (a cell line perhaps). All of these options
introduce flaws in setting up compensation. What are some thoughts on this? 

Michele Black 
Director | Cell Analysis Facility 
Department of Immunology
University of Washington

(206) 685-3014 
(206) 543-1013 
mblack2 at u.washington.edu 

-----Original Message-----
From: Mario Roederer [mailto:roederer at drmr.com] 
Sent: Wednesday, March 09, 2011 8:07 PM
To: cytometry at lists.purdue.edu
Subject: Re: [Cytometry] Compensation question: Substitution of


No no no no.  Even two different TR-PE reagents from the same manufacturer
could require different compensations.  (Rare nowadays, with good QC, but I
can show you examples).  This is true for EVERY tandem.  Each tandem might
be made with slightly different donor:acceptor ratios, which changes their

I don't understand the need to conserve on compensation samples.  It's a
trivial extension to do one more single color stain for your second TR-PE
(or PE-Alexa 610) panel and use a separate compensation matrix.  It's a
complete screw up if you don't, and end up unable to interpret the data from
your valuable samples when the compensation is off.

I insist that people do separate compensations for FITC and Alexa488 as
well.  Why risk your valuable experiment just to be lazy?


On Mar 8, 2011, at 5:51 PM, Diana Hamilton, PhD wrote:

> I have a researcher here with a question I don't know how to answer. 
> Alexa488 and FITC can be used interchangeably because the emission
> spectra are nearly identical. How about PE-Alexa610 and PE-TxRed? Or APC
> and AlexaFluor647? The emission spectra are not exact for either of
> these last two pairs, but how close is close enough? It appears that the
> project they have been working on have three "panels": one set has BOTH
> APC and AlexaFluor647 used in the mix, and the other two "panels" have
> either PE-AF610 or PE-TxRed (again with BOTH APC and AF647). The
> compensation set includes AlexaFluor488, PE, PE-TxRed, PE-Cy5,
> PerCP-Cy5.5, PE-Cy7, AlexaFluor647, AlexaFluor700, and APC-Cy7; the
> samples are being run using a four-laser LSRII (blue, green, red,
> violet).
> I told the researcher that as long as the APC/AF647 combination are both
> positive or both negative, it shouldn't affect things very much. I would
> hope that they are not looking for a positive for one and a negative for
> the other after all the discussions we had in the past, but this
> researcher is just running the samples, not doing the analysis, so he
> couldn't tell me much more than what I've relayed above.
> Thanks for your help.
> Diana Hamilton, Ph.D.
> Flow Cytometry Core Facility
> Oklahoma Medical Research Foundation
> 825 NE 13th Street
> Oklahoma City OK 73104
> _______________________________________________
> Cytometry mailing list
> Cytometry at lists.purdue.edu
> https://lists.purdue.edu/mailman/listinfo/cytometry
> Search the list archive at  http://tinyurl.com/cytometry

More information about the Cytometry mailing list