 |
 |
 |
 |
Of course, the proper way to evaluate the transfer function of a log
amp would be with a voltage pulse generator calibrated by the
National Bureau of Standards. But the point of the poster by Dave
Parks, Marty Bigos and Wayne Moore (Cytometry, Supplement 2,
1988, abstract #155) was that you could characterize the log amp
with fluorescent beads without the need for any additional equipment.
The handout had detailed instructions on how to characterize the
beads and the flow cytometer measuring linear fluorescence.
The initial linear procedure helped identified any non-linearity
due to ADC or PMT voltages. So by the time you started looking
at the log amp, observed variations could be attributed to the log amp only.
By following the handout and doing the calibrations, I learned a
tremendous amount about my instrument and it gave me an
intuitive feel for log measurements in general. I've looked at
some other instruments using D.Parks et. al. procedures and
there is substantial variation in log amp transfer functions.
Another issue is graphics software available for analysis. As Robb
Habbersett points out - most of the packages are 64x64 for bivariate
analysis with some at 128x128. But histogram analysis and sorting
could be improved if there were less fluctation in the log amp
transfer function.
I wanted to throw this issue out on the net so people can reexamine
their policies on data collection. Nobody wants to sacrifice resolution
in their data, but if the resolution is lost in the log amp there's no
reason to collect 1024 channel data. Sure data storage is getting
cheaper everyday, but the TIME spent acquiring, analyzing and
transfering data around is a bigger issue.
tom d.
-- ============================================================================== Tom Delohery | Internet: tdeloher@bigmac.mskcc.org Manager, Flow Cytometry Core Facility | Phone: (212) 639-8729 Memorial Sloan-Kettering Cancer Center | Fax: (212) 794-4019 ==============================================================================
|
|
|
|