Human Genome Project vs. Celera: The Rematch

NEW YORK, March 13 - The human genome debate teams recently lined up for another round of point-counterpoint, with the Human Genome Project's Robert Waterston, Eric Lander, and John Sulston on one side, and Celera genome sequencers Mark Adams, Granger Sutton, Hamilton Smith, Eugene Myers, and Craig Venter on the other.

In the latest salvo in the debate over the use of publicly-available data in Celera's sequencing of the human genome, both groups have once again argued the merits of their positions in articles published in the March 18, 2003edition of Proceedings of the National Academy of Sciences.

"We believe that this second round of the debate has clarified the issues surrounding this epoched contribution to science," Nicholas Cozzarelli, editor-in-chief of PNAS, wrote in an editorial column for the issue, which is available online.

The acrid debate between Celera and the Human Genome Project entered scientific literature last year in March as Waterston of Washington University, Lander of the Whitehead Institute, and Sulston of the Sanger Institute, published their claim that public data played a vital role in Celera's work. Perhaps like someone using the corner and edge pieces to decipher the order of a jigsaw puzzle, they argued that Celera did not really present a true assembly.

Myers, the lead author of the March commentary, et al., demurred, saying Waterston, et al., based their argument on faulty reasoning.

The major issues in the disagreement were not even clear, said PNAS. "The controversy is not about the validity of the WGS method of sequencing very large genomes, but about whether Celera used this method to assemble the human genome in 2001," Cozzarelli writes.

The journal decided to seek another round of commentary, under the ground rule that the public data advocates would first write their paper, then Celera would get a chance to read it prior to publication in order to make its rebuttal.

In this round, Waterson, et al., say:  Adams, the lead author of this go-round, et al., "underestimate the role of the HGP genome assembly in their work, because they focus on only one of the ways in which the HGP data were used. Morever, we note that the major role of the HGP sequence can be directly seen from the properties of the Celera assembly"

The data, methodology, and the properties of the Celera assembly are appropriately viewed as a refinement built on the HGP assemblies, Waterson, et al., conclude.

Waterston, et al., are incorrect, Adams, et al., reply. "The Celera assemblies were constructed on the basis of mate-pair information from Celera sequence data, and the HGSC contribution to the structure and content were minimal.

"In summary, Celera did produce an independent assembly of the human genome."