In a departure from initial plans, Harvard Medical School's Personal Genome Project will choose participants for whole-genome sequencing based on their phenotype, according to project organizers.
The PGP, which will enroll 1,000 participants in the so-called PGP-1K cohort this month, said that it plans to make more genome data from current participants publicly available by the end of November, including complete genomes of the first 10 participants, known as the PGP-10, and four additional genomes from "recent enrollees."
Originally, the PGP had planned to sequence every participant's genome and record their phenotypic data at the same time. But in a newsletter sent to interested parties this week, PGP organizers said they "have learned from patient groups, disease foundations and research collaborators that, in many instances, the phenotype determines which genomes should be sequenced."
Over the next few months, the project, headed by Harvard Medical School professor George Church, plans to launch several new collaborations "where phenotyping is used to help identify PGP participants for whole-genome sequencing." The first example is a partnership with Peter Gregerson at the Center for Genomics and Human Genetics at the Feinstein Institute for Medical Research to study the genetics of absolute pitch. For that study, PGP participants with this trait will be selected for whole-genome analysis, with screening for the trait expected to begin later this year.
The PGP has also updated its study protocol and informed consent to allow participants to provide phenotypic data prior to sequencing. Participants can, for example, link their Google Health account with their PGP public profile. The project expects this change will "increase our ability to expand enrollment in the PGP more quickly" and will help to "identify organizations willing to sponsor specific groups of participants or aspects of the PGP."
The organizers said they are "actively pursuing relationships with financial sponsors, research collaborators, sequencing providers, and disease foundations" to help with the project. The Alan and Priscilla Oppenheimer Foundation has already provided funding to sequence one participant from the PGP-1K cohort, who will be chosen later this month. That participant will receive his or her genome sequence in early 2011, along with a report, and will be asked to publish the sequence on the PGP public repository.
A new analysis site for PGP genomes, called GET-Evidence, will provide annotations from sources such as the Online Mendelian Inheritance in Man database and serve as a "Wikipedia-style location where individuals can record their interpretations of individual variants and review relevant literature," according to the organizers, who said they hope "this tool helps make the analysis of whole genome data more efficient and productive over time."
In addition to exploring links between participants' genotypes and phenotypes, the PGP now also plans to include microbiome data. In a collaboration with Rob Knight and Noah Fierer at the Howard Hughes Medical Institute and the University of Colorado, the project will study the microbiomes of volunteers from the PGP-1K. The initial goal is to catalog microbes in five body sites and to test for associations between specific microbes and diseases. Longer term, as more PGP-1K genomes are sequenced, the goal is to study genome-microbiome interactions.