RE: Advice about QC protocol

Fredrik,

I will share with you how we do our QC on the FACSAria.  I tried a couple
different ways before I found one I was most comfortable with.	We do not
use the 8 peak Rainbow beads.  We use the Rainbow Fluorescent Particles, 3.0
- 3.4 um, catalog # 556291.  These particles are dyed to a single
fluorescent intensity.	They contain a mixture of fluorophores that are
excited from 365 - 650 nm, so they pick up a good representation of the
colors that can be used on the instrument.

I attached a short ppt that has pictures of the plots I use.  I use a
different global acquisition sheet for each laser.  I adjust the voltages
daily to put all the peaks at 100,000 +/- 1,000.  I adjust the area scaling
for FSC (which I have been told is not necessary on the Aria, but it seems
to give me better overall CV's.)  I also adjust area scaling for each of the
three lasers by comparing one color for each laser with Area and Height.
For example on the Blue Laser I look at the Mean of FITC-A and the Mean of
FITC-H.  If they don't match within about 1,000, I adjust the ASF on the
Laser tab until they do.  Then I proceed to adjust all the other voltages to
get them as close to 100,000 as I can.	Then I record 10,000 events.  I then
record the voltage I used, the CV for each peak and the area scaling factor
for each laser, as well as the laser delay for each laser.  I have included
a copy of my log sheet as well.  I kind of have to squeeze the ASF and laser
delay for all three lasers in one box.	I only check the  laser delay about
once a month, although I record it daily, in case some one inadverantly
changes it.  Early on our instrument had a lot of alignment issues and I had
to change this number frequently.  Since that issue has been resolved, that
number stays very stable.

>From my log sheet, I watch for trends with how much the voltages have to be
adjusted.  If they continually have to be cranked up, I suspect a laser
going out.  If they jump all over the place, you might suspect a laser
alignment issue or power fluctuations.	I find that extremes in room
temperature can affect the voltages needed.  I also monitor the CV's.  As a
general rule, the CV's of most of the peaks should be <3.0.  Occasionally it
is hard to get FSC & SSC to be that low, generally, one is a little higher.
Also the Far red channesl (PE-Cy7, APC-Cy7) have bigger CV's - 5.0 or under
is pretty good for our instrument.  If you see the CV's getting bigger, the
first thing to suspect is a dirty flow cell.  If cleaning the flow cell
doesn't resolve the problem, suspect laser alignment.

Other than just reviewing my log sheet, the other way of scrutinizing the
data would be to create Levy-Jennings charts for each channel, recording the
daily voltage and plotting it, to watch for trends.  You would mark 2 SD on
either side of the mean and if you were above 2 SD, you would need to take
action before proceeding.  In the clinical world, this is a given.  In the
research world, maybe not as necessary.  Someday I may get motivated and
find the time to do it this way.  It would be a little more critical way to
monitor.

I hope I wasn't too long winded, but specific enough to help.  Don't
hesitate to ask me for more specifics - I'd be glad to help.

-----Original Message-----
From: J.F. Wallberg [mailto:j.wallberg@erasmusmc.nl]
Sent: Monday, July 11, 2005 1:36 AM
To: Cytometry Mailing List
Subject: Advice about QC protocol


Dear all
We recently purchased a FACSAria and I am now trying to set up a 
protocol to check the instrument daily and monitor the performance over 
longer time periods. On our FACSVantage we always use different beads to 
align our three lasers but for the FACSAria I have been advised to use 
8-peak rainbow beads for all the lasers. I will use the beads to monitor 
CV for the highest fluorescent peak from all detectors, Forward scatter 
and side scatter. Furthermore I want to monitor mean value for the same 
peaks. I have been trying to add plots to the experiment to see if the 
area scaling looks good. I also tried to draw plots to monitor the 
resolution of the different peaks of the beads in the different 
detectors, this was shown to me during the course given by BD.
I would be very grateful to any input about this daily QC protocol. Is 
there any other parameters I should monitor or should I use a different 
approach? Any input would be helpful for me. I am also wondering how 
people set up an easy system to monitor and evaluate changes in the 
instrument over longer periods of time. Is there for example excel 
sheets available were I could enter my numbers and detect a change in 
performance.

Best regards
Fredrik Wallberg

Department of Cell Biology and Genetics
Erasmus MC
P.O Box 1738
3000 DR Rotterdam
The Netherlands


This attachment - 'QC Presentation.ppt' -  131.58 KBytes - can be viewed at
http://www.cyto.purdue.edu/MD-parts/fc6d8a60fc485a7a39e659f2f257b64325654c28.ppt