RE: DiVa-FlowJo axis events

From: Holmes, Kevin (NIH/NIAID) <KHOLMES@niaid.nih.gov>
Date: Thu Apr 22 2004 - 14:40:05 EST
You can try to use the "Display transformation" option in FlowJo.  See
http://www.flowjo.com/v4/html/displaytransform.html for more detailed
information.  Basically, the scale is modified so that it is linear near
zero, but maintains the log scale in the upper decades.  More importantly
for your question, it allows negative values to be displayed so that the
cells you were not seeing now can be visualized in a nice symmetrical
cluster.  The above link goes into more detail, including displays comparing
with and without transformation.
Make sure that you read all of the documentation regarding using this, or
you will get some 'interesting' results.  The most important tidbit that I
found was that you need to 'define transformation' using a data file that
has all of the fluorochromes in your experiment (i.e. don't use single color
controls for this).
I suspect that this kind of display will begin to be utilized more and more
in the future.


Kevin L. Holmes, Ph.D.
Chief, Flow Cytometry Section
Research Technologies Branch
Bldg. 4, Room B1-38
NIAID, NIH

Phone: 301-496-9071
FAX:  301-402-4532
Email: kholmes@niaid.nih.gov

 -----Original Message-----
From:	Andy Johnson [mailto:andy@brc.ubc.ca] 
Sent:	Wednesday, April 21, 2004 4:00 PM
To:	Cytometry Mailing List
Subject:	DiVa-FlowJo axis events

As ever I will get more "out of office" replies, than true responses, 
but that seems to be the price to pay.

I appreciate that the issue of DiVa data and events off the axis has 
been discussed thoroughly, but what are people doing about data 
presentation when they convert it to flow Jo ?

It is possible to produce some nice plots in Flow Jo, but when you 
put quadrants on the plots you find that what appears to be "visibly" 
an empty quadrant has in fact 50% of the cells....all stacked up off 
the axis.

The plots look fine in DiVa (some events on the axis), so what is the 
trick to getting flowJo and DiVa to look the same for publication 
purposes.  We have been using flowJo for a long time and tried most 
of the tricks, but this is a constant issue that we cannot solve.

Any ideas are much appreciated.

Andy
-- 
Andy Johnson
FACS Facility Manager
Biomedical Research Centre
UBC
Vancouver
604-822-7838
Received on Fri Apr 23 17:38:00 2004

This archive was generated by hypermail 2.1.8 : Sat Apr 24 2004 - 03:12:04 EST