RE: Bad Data

From: Fred Menendez (fmenende@jhsph.edu)
Date: Thu Oct 25 2001 - 13:33:57 EST


This has been a very interesting thread.  It seems to me that in the final
analysis what is being discussed and flirted with here is the need for
standardization in flow cytometry, especially as it is used in research
applications.

Of course, some standardization currently exists in the form of Proficiency
Testing (PT) programs for clinical cytometry.  The College of American
Pathologists (CAP), among others, administers a flow PT program for CAP
certified clinical labs.  It includes challenges in the areas of
phenotyping (e.g.HIV/AIDS CD4 monitoring), leukemia/lymphoma and DNA
analysis.  Similarly, NIAID/DAIDS requires that flow labs participating in
and reporting results for the ACTG and MACS be part of the flow PT program
currently run by the folks at the New Jersey Medical School.  These
programs, while certainly adding to the cost of doing business and
generally being a pain in the ..., do serve a practical and important
purpose;  they reassure clinicians and researchers, respectively, that the
quality of the flow data meets some minimum standard!

As such, we need not re-invent the wheel.  A model already exists for a
flow PT program that could be used in the research community as well.  All
things being equal, the methods and techniques employed in the clinical and
clinical research flow community to produce quality flow data (PT programs
being one of those) are no different than those that should be employed in
the basic research community.  Compensation is compensation, and how one
does compensation is less important than that one's experiment is  properly
compensated (my apologies to MR).

I know this opens a can of worms but its a can that needs to be
opened.  Flow cytometry is an arcane field.  It is also a very powerful
tool that can be easily manipulated and misunderstood. It is just such
manipulations and misunderstandings that have given rise to this
discussion.  Standardization is not a panacea; it can never totally
eliminate bad data or protect the uninitiated from producing bad data.  But
it allows basic questions to be asked and answered about the quality of
data produced, both within an individual flow lab and by those outside that
same lab.

A final commentary and I'll get off my soap box:
We in the trenches - operators, grad students, post docs, etc. - look to
those of you with the most influence and knowledge in the field for
guidance.  Whose idea about standardization ultimately prevails is less
important than that consensus be found and agreed upon-----for the benefit
of the entire scientific community!

To paraphrase Steely Dan:  Bad data and a Pina Colada my friends......gives
one a headache!

Fred Menendez
Baltimore, Maryland



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:36 EST