Here are some comments that are from Don Riddle, I asked for his
opinion because he was involved in the CGC and gene name allocation
for almost 10 years.
Off the top of my head, I guess I agree there is a problem, but the
solution is pretty ugly. My immediate response is that different names
for the same gene happen enough as it is. They
should not be programmed into the system. The potential for the
converse is also scary.
Suppose two labs disagreed on which gene was an orthologues. The
requires the same name for the two different genes. It is just not
worth the hassle. Cb-tra-1
should be a C. briggsae tra gene that may or may not be an orthologues
of C. elegans tra-1. The
name should not be required to contain such information. Scrap the
I agree that assigning blocks of numbers to species is a bad idea.
I disagree with the view that three digit numbers are too difficult to
remember. There is
nothing wrong with them. Most genes have one or two digit numbers
anyway. Only let has
3 digit numbers."
I might add, that unc also has three digits now (has
had three digit names for two decades), and lin is moving
up quickly. I see no problem with three digit numbers and find them
infinitely easier to deal with than thousand's of "cute" names, that
eventually be come a burden. I agree with Don in all of the points he
David Baillie, Professor
Canadian Research Chair in Genomics http://www.chairs.gc.ca
Currently on leave at the BC Genome Sequence Centre, BC Cancer Agency
Department of Molecular Biology and Biochemistry
Simon Fraser University, 8888 University Drive
Burnaby, British Columbia, Canada, V5A 1S6
web site http://crick.mbb.sfu.ca