Public-Private Genome Debate Resurfaces

Smoldering differences between the Celera Genomics Group and the Human Genome Project erupted last month as three leaders of the international public consortium published an online article in the Proceedings of the National Academy of Sciences (PNAS)1 criticizing the results published last year by Celera.2 For the first time in scientific literature, Robert Waterston of Washington University, Eric Lander of the Whitehead Institute for Biomedical Research's Center for Genome Research, and John Sulston of the Sanger Institute published a long-held point of contention.3 They gripe that because scientists at Celera included DNA sequence information from GenBank, the public sequence repository, they neither published a true independent sequence of the human genome, nor provided a meaningful test of the whole genome shotgun method, currently being employed—at least in part—in the public effort to sequence the mouse. Some observers call the recent spat perplexing and speculate that it is spurred more by egos and Nobel Prize aspirations rather than genome integrity. Celera scientists published a commentary two weeks later saying that the simulations performed by Waterston and colleagues were flawed and based on false assumptions.4 Dispelling Myths? The simulations demonstrate that the fully sequenced chromosome 22, when shredded in the methods Celera described, would retain enough identifying information to anchor Celera's shotgun data to the assembly. They also show how a true simulation of whole genome shotgun reads, which would be inherently more randomized, could not produce the results Celera scientists published in February 2001. Celera scientists say that the scope of the simulations is misleading because it focuses on a fully sequenced chromosome accounting for only 1% of the genome instead of the entire unfinished sequence. Besides, Mark Adams, vice president of genome programs at Celera, adds, "Our goal was not to demonstrate a pure whole genome shotgun approach, it was to sequence the human genome." Philip Green, Howard Hughes Medical Institute investigator and professor of genome sciences at the University of Washington, argues that Adams is backpedaling. "The explicit claim is in Celera's [2001] paper that they did an assembly by the whole genome approach. You can't allow false claims to persist in the literature." Green also published a commentary dispelling, among other things, the widely held notion that the competition Celera brought to the human genome project was a good thing, saying instead that it "has the downside of encouraging shortcuts that may compromise the ultimate result."5 While some sequencing-center heads maintain these criticisms, others have ducked the fray. Richard Gibbs, director of the Baylor College of Medicine genome-sequencing center has had the help of Celera in the effort to sequence the rat genome since shortly after publication of the human genome. "We wouldn't write a paper like that," he says. "We understand that the early efforts were adventurous, and you could pick at them. There were shortcomings, sure, but it was all great science." Proof is in The Pudding Other onlookers find that the recent papers raise more questions than answers. Terry Gaasterland, assistant professor of computational genomics at Rockefeller University, says, "I would want to see this data on chromosome 8 or some other chromosome that's not finished." Her group compares the annotations of both versions of the genome, and she says quite frankly, "neither annotation is very accurate." Green points out what lacks in Celera's response. "They did not do the one analysis which could have laid this controversy to rest," namely, an assembly sans public data. This has been done, claims Gene Myers, vice president of research informatics at Celera, and with a resulting assembly better than the first. "One of the things we discovered after the fact is that the use of [public] data actually degraded the performance of our whole genome assembly," he says. But despite claims from former Celera CEO J. Craig Venter that these results would be published in scientific literature,6 they've yet to be seen. Adams points to the pudding of the Celera mouse genome where the whole genome shotgun was employed solely and has drawn many satisfied customers. He adds, "The world doesn't need another incomplete human genome sequence." For the public mouse genome, whole genome shotgun is being used in conjunction with the clone-by-clone approach. Gibbs quips, "I think the same guys [Waterston, Lander, and Sulston] will tell you, well, actually it does work—if they do it." And, proof of principal for whole genome analysis of a mammalian species is what some say this is all about. Says Gaasterland, "The Nobel Prize isn't going to whoever sequenced the human genome. That was an engineering exercise. [The prize] goes to discovery and invention. Who actually came up with something new here?" Brendan A. Maher can be contacted at bmaher@the-scientist.com. References 1. R. Waterston et al. "On the sequencing of the human genome," Proceedings of the National Academy of Sciences (PNAS), 99:3712, March 19, 2002. 2. J.C. Venter et al. "The sequence of the human genome," Science, 291:1304-51. Feb. 16, 2001. 3. E. Russo, "Behind the Sequence," The Scientist, 15[5]:1, March 5, 2001. 4. G. Myers et al. "On the sequencing and assembly of the human genome," PNAS, www.pnas.org/cgi/doi/10.1073/pnas.092136699. 5. P. Green, "Whole-genome disassembly," PNAS, www.pnas.org/cgi/content/full/082095999v1. 6. B.A. Maher, "The human genome—one year later," The Scientist, 16[3]:29 Feb. 4 2002.

Public-Private Genome Debate Resurfaces

Interested in reading more?

Become a Member of