Hello David and Barbara, As far as your initial question is concerned, I agree that there is most probably a compensation artifact playing a role. I have the impression that also your PMT4 (is ECD) versus PMT2 (is FITC) compensation is erroneous. In your PMT2-histogram we see essentially three clusters : "true negatives", "dulls" (these essentially are CD4 Tcells), and "real positives". By increasing PMT4 vs PMT2 compensation you would probably "push" the "dulls" into the "negatives". When we do these experiments, we observe the following : We normally observe antigen ebb, when wee induce apoptosis by in vitro irradiation at these doses. This means that you will see a decrease of CD8 and CD4 labeling intensity in apoptotic (annexinV positive) cells. By the way, you should include a permeability marker ... We get more apoptotic cells in the CD8 Tcells than in the CD4 population. This proportion will not exceed 40% or so, for CD8, and 20% for CD4, with 2 Gy of gamma. A lot of this depends on timing of the apoptosis test with respect to the moment of irradiation, but also on the quality of the PBL preparation, especially displayed by the proportion of apoptotic cells at zero irradiation. Jan David.C.McFarland@gsk.com wrote: > > Barbara, > > In reference to you original posting about the origins of the tail > seen in your AnV-FITC vs CD8-PE plot I think that it is a compensation > issue. The FITC signal looks undercompensated from the PE, but I > wouldn't go so far as to say it is "really bad". In fact, it doesn't > really affect your quadrant statistics at all since the entire tail is > in a single quadrant. It just doesn't "look" right. You mention > below that it might be overcompensated. If this was an example of > overcompensation, you wouldn't see the tail. Everything would be > squished against the axis. Some might say that the events to the left > of the tail ARE squished against the axis. They are a little > squished, but I think this is because the PE PMT voltage is a little > low. Running with a little higher voltage in all detectors to pull > the negatives a little further from the origin may make suitable > compensation a little bit easier to visualize. &n bsp;I also think you > can rule out autofluorescence. In my experience, problematic > autofluorescence typically shows up as a double positive population > that streaks out from the origin of a two parameter plot. That little > curl of a tail at the high end of the FITC scale just doesn't fit the > pattern I'm used to seeing. > > On another note, what about dead cell discrimination? I find it > imperative to use a dead cell discriminator along with AnV to gate out > false positives. What does the rest of the flow world have to say > about this? > > > David McFarland > GlaxoSmithKline > ----- Forwarded by David C McFarland/DEV/PHRD/SB_PLC on 12-Aug-2002 > 14:46 ----- > > Barbara_Kutzner@hc-sc.gc.ca > > 09-Aug-2002 08:39 > > > > > To: "Cytometry Mailing List" > > cc: > Subject: data question - omitted info > > > > > > Sorry about that folks, it seems I forgot to attach the axis labels to > the plot. > PMT2 is Annexin V -FITC, PMT3 is CD8 - PE. The gate was placed on > lymphocytes > only not all cells. > As it has been asked, I have included all the data plots I have from > that sample > (1.0 Gy) analysis (suggestions were that it may be bleed over from another > channel, that it is over compensated, that the population is too close > to the > axis thereby throwing off compensation) > > Original plot 1.0 Gy sample data > (See attached file: flow plot.JPG) (See attached file: 1Gy flow.jpg) > > > > Thanks again, > Barb > > ____________________ > Barbara Kutzner > Technologist > Radiation Protection Bureau > Health Canada > 775 Brookfield Road, PL 6303B > Ottawa, ON K1A 1C1 > (613) 952-9069 > FAX (613) 941-1734 > -- Jan Bayer, Ph.D. Fondation Jean Dausset - CEPH Email : bayer@cephb.fr 27, Rue Juliette Dodu Tel : +33 1 53 72 51 14 75010 Paris, France Fax : +33 1 53 72 51 28
This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:26:18 EST