First, many thanks for the terrific April issue. I loved the articles about Jim Kent, Marcie McClure, and Lynx. Your timely coverage of the people and companies integral to today’s genomics is always infused with common sense and good humor. I have learned to look forward to every issue of GT!
I would like to point out a couple of drawbacks in other articles, in the hope that I am helping to make future issues even better.
“Where Robots Rule,” while replete with praise for Whitehead, fails to mention what all users of the public draft genome sequence know: While Whitehead has indeed sequenced more of our genome than any other center, it also has the poorest overall sequence quality in terms of contig fragmentation and clone update frequencies. There’s room to improve before attempting the more ambitious projects mentioned at the end of the article.
Second, Nat Goodman’s otherwise excellent take on the state of human transcriptome analysis (“Now What?”) does not mention a recent result contrary to the 30,000-gene estimate (http://genomebiology.com /2001/2/3/preprint/0001/).
The human genome contains at least one example of a “chain” in which “the first transcript overlaps the second which overlaps the third, but the first and third don’t overlap.” The human DCTN1-NBC4-MTHFD2 gene group fits that description, as reported in my recent paper (http://genomebiology .com/2001/2/4/research/0011/).
This may not be the only such example; “all bets are off,” indeed, in the business of counting human genes.
Leonard Lipovich, Department of Molecular Biotechnology, University of Washington
E-mail letters to the editor to [email protected] .com. Include your name, title, and company. Letters may be edited for clarity and length.