From: Geoffrey Osborne (geoff.osborne@anu.edu.au)
Date: Tue Feb 25 2003 - 00:55:57 EST
Fellow BD DiVa users,
Just out of interest you may wish to consider a little experiment to
the verify for yourself the "accuracy" of the sorting of rare events based on
whether the said rare event falls inside or outside a drawn region.
What am I talking about? Have a look at the attached jpeg file
which shows a zoomed dotplot display in which a region has been drawn
which has at least one vertex on an angle, in this plot from the lower right to
the upper left. Dots within the region (P6) should be coloured green. Note
that numerous dots (green) are drawn outside (below) the region, and more
worrying still that the occasional dot coloured pink (unfortunately) falls within
the "green" region. Note also the that statistics for P6 region remain
constant at 287 events. Now do we really want to sort those rebel "green"
events? And what about that stray pink one?
I understand the bitmapping, memory and speed constraints that
require this approach to drawing lines on plots, but I'd suggest when trying
to sort that 1 in 1 million rare event you carefully consider the accuracy of
the region drawn at displayed resolution on 18 bit data.
Just some food for thought...
Geoff--
Geoffrey Osborne
Specialist, Flow Cytometry,
John Curtin School of Medical Research,
The Australian National University,
Canberra, 0200, ACT. AUSTRALIA
email: geoff.osborne@anu.edu.au
http://jcsmr.anu.edu.au/facslab/facshome.html
The following section of this message contains a file attachment
prepared for transmission using the Internet MIME message format.
If you are using Pegasus Mail, or any another MIME-compliant system,
you should be able to save it or view it from within your mailer.
If you cannot, please ask your system administrator for assistance.
---- File information -----------
File: Diva_reg.jpg
Date: 25 Feb 2003, 15:05
Size: 54355 bytes.
Type: JPEG-image
This archive was generated by hypermail 2.1.6 : Thu Jan 01 2004 - 17:43:33 EST
