Re: 3D Graphics for flow data display

Howard Shapiro <hms@shapirolab.com> writes:

> I note that Tony Rossini, who has been playing with 3-D display for
> some time, seems to share my opinion.

Very nice song.  With caveats (i.e. without training), I definitely
share it.  It is hard to figure out what is present in marginal and
conditional projections, which is the fancy way of describing gating
algorithms.

One paper that might help people get a sense of what is there in high
dimensions, with the 2-d projections (scatter matrix, or pairs-plots)
is: 

    http://www.jstatsoft.org/v02/i06/paper.pdf

with corresponding materials:

     http://www.jstatsoft.org/v02/i06/paper/

(this is from the Journal of Statistical Software, 

     http://www.jstatsoft.org/

and the reference is:

 Dianne Cook, Calibrate Your Eyes to Recognize High-Dimensional Shapes
 from Their Low-Dimensional Projections, Volume 2, 1997, Issue 6.
 Dates: submitted: 1997-10-04 accepted: 1997-11-25


Note that this is more for those who havn't done much work with
multi-color flow data than for those with lots of experience.  It's a
nice review of what happens with projections of ordinary/common shapes
and distributions for univariate high-dimensional data visualization,
and there are always lessons to be learned.  It would be good to
construct a similar, flow-specific, paper, I think.

best,
-tony
-- 
rossini@u.washington.edu	    http://www.analytics.washington.edu/ 
Biomedical and Health Informatics   University of Washington
Biostatistics, SCHARP/HVTN	    Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC  (M/W): 206-667-7025 FAX=206-667-4812 | use Email

CONFIDENTIALITY NOTICE: This e-mail message and any attachments may be
confidential and privileged. If you received this message in error,
please destroy it and notify the sender. Thank you.