On the sequencing and assembly of the human genome

On June 26, 2000, Celera Genomics and the International Human Genome Sequencing Consortium (HGSC) announced at the White House the completion of the first assembly of the human genome and the completion of a rough draft, respectively. In February of 2001, the two teams simultaneously published their analyses of the genome sequences generated (1, 2). The joint announcement and subsequent publications were a result of long discussions among Celera and HGSC scientists on reducing the negative rhetoric and demonstrating to the public that both teams were working for the public good. Now three laboratory leaders from the public consortium, Waterston, Lander, and Sulston (WLS), argue that Celera did not produce an independent sequence of the human genome or meaningfully demonstrate the whole-genome shotgun (WGS) technique (3). This conclusion is based on incorrect assumptions and flawed reasoning. The key assertion of WLS is that by using information from the HGSC, Celera's method implicitly retained the full assembly structure produced by the HGSC. This is incorrect. As described in table 2 of ref. 1, we combined our data with a uniformly spaced 2× shredding of 677,708 individual bactigs, contigs of bacterial artificial chromosomes (BAC) clones shotgun sequenced by the HGSC, not the genome assembly reported in ref. 2. The goal of including this sequence was to take advantage (with attribution) of the work of the HGSC to the extent that it would contribute additional sequence coverage. The global order and the overall sequence of the genome were determined by using the set of 27 million mate-paired reads generated at Celera. Mate-pairs are sets of reads that are adjacent to one another in the genome and serve to link together nearby segments to promote assembly. The 38.7-fold genome coverage spanned by these mate-pairs provided the long-range order (over millions of …