In his view, the answer is no. Church said that there is a role for a so-called "open source Personal Genome Project" in which participants consent to "full disclosure" of their genome sequence and phenotypic data, including medical records, imaging data, and other information.
Such a project would run in parallel with similar efforts underway in which participants contribute de-identified DNA samples under strict conditions of anonymity, and would provide "a richer data set" for researchers to mine for links between genotype and phenotype, he said.
Church said that a Harvard Institutional Review Board recently approved this effort after a year of careful consideration, and that he is currently seeking collaborators from academia and industry, as well as "highly informed" individuals to serve as initial donors. Further details are available here.
Church told GenomeWeb News after his talk that he is in discussions with a number of companies developing next-generation sequencing methods, as well as other firms "at the pharmacogenomics interface" who have expressed interest in participating in the project. Church declined to name specific collaborators, but said that he is interested in getting together "as many partners as we can."
As for the ethical, legal, and social implications of the project, Church said that the idea for the effort actually sprang out of the ELSI component of another grant, and noted that ELSI groups at
As the price of sequencing comes down, it is becoming increasingly more likely that individual will be able to have their genomes sequenced. But as more and more of this kind of information becomes available, it will become more difficult to guarantee that it remains secure, said Church. Not only are "hackers getting better at breaking through security," he said, but many researchers who gain permission to access de-identified data sets may unwittingly "put it somewhere that Google can get to it."
Rather than fight a losing battle against releasing this data into the public domain, "we may need to start consenting to full disclosure" for all such efforts, Church said.
Church noted that the project will require a better informed public, and called for the scientific community to help "demystify and destigmatize" personal genomics.
To help spur the effort, Church said that his lab is holding a series of workshops to share its so-called polony sequencing method with the research community. The method currently offers a 70-fold reduction in price compared to Sanger sequencing, mainly due to the low price of the primary equipment required: a $140,000 standard microscope that many labs already use for cell biology.
Church said that his group has rewritten the microscope's image-processing software for polony sequencing, and has released this software through its website.
Church's lab has licensed some of its polony sequencing technology to firms developing next-generation sequencing technologies, but noted that the methodologies are also available in the public domain, and that he's "not exclusive" when it comes to working with commercial entities. "I'd like to see them all do well," he said.
So far, the lab has licensed its multiplex tagging method to Lynx (now part of Solexa), its polony method to Agencourt (now part of Beckman-Coulter), and its fluorescent molecule technology to Pyrosequencing (now Biotage). In addition, Church sits on the scientific advisory board of Helicos BioSciences.
"I want to enable as many of them as possible," Church told GenomeWeb News, adding that he's not even sure if these companies are using his methods in "their front-runner stuff."
Church said his primary goal in sharing the polony technology "is to move the field forward."
The polony sequencing technology currently costs around $0.07 per kilobase with an error rate of one in every 10 million bases, he said. For the human genome, this would translate into 1,800 errors for 6X coverage. Church said his goal for 2006 is to bring the error rate down to one in every hundred million bases, or around 60 for the whole human genome. As for the price, his lab is still shooting for the $1,000 genome, but he said that $10,000 might actually be "the tipping point" for personal genomics to take off.