SAN DIEGO -- It is fitting that James Watson, co-discoverer of the double helix structure of DNA in 1953, should become the first person to have his genome fully sequenced.
At least, that is the claim of 454, the Connecticut next-generation sequencing company that says it has essentially completed sequencing the genome of the 1962 Nobel Laureate, who famously solved the structure of DNA with the late Francis Crick. Opening CHI's inaugural Next-Generation Sequencing meeting this week, 454 Vice President Michael Egholm said that "Project Jim" is almost complete.
The idea to sequence an individual human genome occurred to 454 management in 2005, around the time the company launched the first commercial next-generation platform, the GS 20. Egholm recalled chatting with Baylor Genome Center director Richard Gibbs and 454 founder and Chairman Jonathan Rothberg after an advisory board meeting, when "we came up with idea that the idea that the first human individual to be sequenced must be Jim Watson."
Rothberg duly asked Watson, who is President Emeritus at Cold Springs Harbor Laboratory on Long Island, whether he would donate his blood. He readily agreed, though as Egholm said, "what we hadn't counted on was Jim turning around the next day and telling a journalist at the New York Times!"
Egholm acknowledged that the idea was a little premature. "We had gone a bit ahead, we weren't ready to sequence humans," he said. Aside from throughput issues, the error rate of 454's first instrument was unacceptably high. But with the launch of the GS FLX this year, Project Jim has gathered momentum. For the past few months, 454 has been "cranking out sequence," producing a total of 10 billion bases from 40 million sequence reads of average read length 250 bases. Data analysis is being performed by the Gibbs and colleagues at the Baylor genome center.
Egholm noted that 3% of the Watson DNA sequence -- 1.3 million reads -- do not assemble onto the human genome reference sequence, suggesting that, "the human reference genome is about 97% complete," Egholm said. Interestingly, 20% of the reads that don't match are found in the Celera assembly, but Egholm stressed they had not done the reverse analysis.
In total, 454 has identified 1.9 million substitutions in Watson's DNA from the reference sequence. 68% of those substitutions -- 1.31 million variants -- are found in the SNP database, dbSNP. "That's actually pretty reassuring," said Egholm. That leaves 600,000 novel DNA variants. 80% of these are probably real, whereas a minority could be artifacts.
Of those nearly 2 million SNPs, "50 are found in database listing phenotypes of human polymorphisms." When informed, Watson reportedly joked, " 'What, only 50 things wrong with me?' He was kind of disappointed," Egholm quipped. In addition to SNPs, Egholm said the sequence revealed a "mind-boggling 68,000 indels [insertions/deletion] from 3 bases to 7000 bases." It's about 1:25
Who's on First?
Egholm's assertion that Watson would be the first individual to be fully sequenced is open to question however. J. Craig Venter donated his own DNA for the Celera first draft genome assembly in 2000-01. That sequencing effort has continued at the Venter Institute. Venter even disclosed some of the findings from analyzing his own sequence to the Wall Street Journal last year. Moreover, Illumina, through its acquisition of Solexa, is also reportedly sequencing the genome of an anonymous donor to the HapMap project.
Egholm later qualified his statement to say that Watson is the first individual to be sequenced on a next-generation sequencing platform. "Why would one want to sequence a known individual?" Egholm asked rhetorically. "Why not? Someone's got to be the first. This will make the issue much more personable for the public."
Watson's sequence will be presented in more detail at a scientific conference in May. It will likely be deposited on the Cold Spring Harbor Laboratory website for everyone to peruse. "Clearly there are a number of legal and ethical issues to be resolved along the way," said Egholm.
*Next-Generation Sequencing and Analysis (CHI); San Diego, 20-21 March 2007.