Jump to content


Photo

Compensation in Kaluza


  • Please log in to reply
1 reply to this topic

#1 svirden

svirden

    Member

  • Members
  • PipPip
  • 1 posts

Posted 07 November 2013 - 07:18 AM

I have a question regarding the use of the adjust compensation feature of Kaluza. I completely understand the basic concepts of compensation and why it’s necessary, but putting it into use as a practical matter is another story.

 

Regarding the compensation matrix visible within Kaluza, it has been my understanding that modifying a cell within the matrix ONLY affects the interaction of two channels; e.g., FL4’s interference with FL3. What I’m seeing in Kaluza makes me question my basic understanding. I should note for reference that our lab is in the process of validating our new Navios, and we’re utilizing all 10 color channels on the vast majority of our panels.

 

I would expect that when I move the compensation sliders on a single dot plot and I watch the percentage value in a single cell of the compensation matrix change, that the change to the single matrix cell I’m making would ONLY affect that single dot plot. What I’m seeing though is that a compensation change made to a dot plot causes changes to appear in many other, seemingly disconnected dot plots in the same specimen tube.

 

For example, in a dot plot displaying CD24 PC5.5 (FL4) on the X axis, CD19 ECD (FL3) on the Y axis, when I move the vertical slider bar up and down, I can see the FL4/FL3 compensation percentage change and I see the corresponding change in this dot plot as I would expect. What I’m seeing that’s confusing is that cell populations in all the other dot plots in the same tube are also changing. If I’m only changing the compensation to the interaction between ECD and PC5.5, how can that affect other channels? Obviously, there’s a fundamental problem with my understanding of the compensation matrix and how it works. I need some help here. Thanks.

 

 



#2 Jessica

Jessica

    Member

  • Members
  • PipPip
  • 1 posts

Posted 07 November 2013 - 12:18 PM

The application of the spillover matrix to the data in Kaluza is based on the following publication:

BAGWELL, C. B. and ADAMS, E. G. (1993), Fluorescence Spectral Overlap Compensation for Any Number of Flow Cytometry Parameters. Annals of the New York Academy of Sciences, 677: 167–184.

 

The spillover matrix you see on the screen is inverted, then multiplied by the data to obtain compensated data.  When you move a slider on a plot, you are adjusting one value in the spillover matrix.  However, in the process of matrix inversion a single value's change will end up affecting multiple values.  Effectively, the spillover values are coefficients in a system of equations, and the system of equations is being solved for your data.  If you modify a single coefficient, the overall solution to the system is affected.

 

I hope that helps you!  The referenced publication goes into far more detail.