I discuss the problem with 3-color compensation on one of my
pages,<http://www.drmr.com/compensation/3color.html>. (part of the
"detailed discussion" on compensation). The fact is that one can
never guarantee correct 3 color compensation using only pair-wise
compensators (i.e.., FL1<->FL2 and FL2<->FL3) because they don't
necessarily correct for FL1<->FL3 spillovers. Historically, this
hasn't been too much of a problem because of the coincidental
spectral properties of fluorescein and PE.
There is no good solution to your problem other than to use software
compensation (e.g., FlowJo on Macintosh or WinList on PC). If you
use this approach, make sure that no pair of colors is
overcompensated ("Squooshed"), because software can never correct
over-compensation--it can only correct incomplete undercompensation.
And, no, having less sensitive instrumentation will not fix this
problem.
By the way, I want to remind everyone that compensation settings DO
NOT DEPEND on the brightness of a reagent! Therefore, you should
always use the brightest reagent you have in any given channel to set
compensation!
mr
This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:54:23 EST