|
|  |
|
in-cites,
February 2004
Citing URL: http://www.in-cites.com/scientists/DrTerrySpeed.html
|
|
|
|
An
interview with:
Dr. Terry Speed |
|
|
|
 ccording
to a recent analysis of the ISI
Essential
Science Indicators
Web product, Dr. Terrence Speed’s work in the field of
Mathematics garnered the highest percent increase in total
citations in the latest bimonthly update. His current record
in this field includes 19 papers cited 173 times to date. Last
year, a paper he co-authored, "Statistical methods for
identifying differentially expressed genes in replicated cDNA
microarray experiments," (Stat. Sinica 12[1]:
111-39, January 2002), was named as a "Fast-Breaking
Paper" by ISI Essential Science Indicators. Dr. Speed
divides his time between the Department of Statistics at the
University of California, Berkeley, and the Walter and Eliza
Hall Institute of Medical Research in Melbourne, Australia,
where he is the Head of the Division of Genetics and
Bioinformatics.
 Read
by Terry Speed
|
Why do you think your work is highly cited?
Simply because the field of microarray data analysis is expanding
rapidly, and many people like me are getting involved.
What are the circumstances which led you to your work?
|
“…the field of microarray data analysis is expanding rapidly, and many people like me are getting involved.”
|
|
Lots of scientists I know are starting to use microarray data,
and because I am a statistician somewhat fluent in the language of
modern biology, some turn to me for assistance. This is great for
me, as I am very interested in these issues.
Would you describe the significance of this work for your
field?
The significance is quite modest. My statistical collaborators
and I think it important to try to do a good job with what we call
"low-level" and "basic" analyses, and that's
what it is. It is not "big science," rather, taking a bit
more care pre-processing one's data and carrying out initial
inferences before using it to cure cancer or understand some other
fundamental scientific issue.
Where do you see this research going 10 years from now?
I hope we will be beyond "low-level" and
"basic" analyses well before 10 years are up. Of course
there will be continuing "low-level" challenges as novel
data-generating technologies come along, and the occasional new
"basic" analysis, but we would like to do some rocket
science or brain surgery as well. By this I mean challenging,
specific problems, which are far from routine.
What lessons would you draw from your work to share with the
next generation of researchers?
All of the cited papers are about trying to understand one's data
from the ground up, and to take this into account in any
pre-processing done, and then do as well as we can with existing
statistical tools, only doing something new when it is required.
This is neither novel nor surprising, but it does seem that some
people need to be told this each time a new data-generating
technology comes along. Of course we like it when something new is
required, but that's far less often than you would think.
Terrence P. Speed, Ph.D.
Division of Genetics and Bioinformatics
Walter and Eliza Hall Institute of Medical Research
Melbourne, Australia
And
Department of Statistics
University of California, Berkeley
Berkeley, CA, USA
Read
by Terry Speed
|
|
|
in-cites, February 2004
Citing URL: http://www.in-cites.com/scientists/DrTerrySpeed.html
|
|
cites
