Skip to main content
Premium Trial:

Request an Annual Quote

23andMe Publishes Web-Based GWAS Using Self-Reported Trait Data

By a GenomeWeb staff reporter

NEW YORK (GenomeWeb News) – In a paper appearing online last night in PLoS Genetics, researchers from 23andMe and Columbia University reported the first genome-wide association findings stemming from 23andMe's web-based, participant-driven research program.

The study, which involved more than 9,000 individuals genotyped through 23andMe's direct-to-consumer testing service, looked for genetic associations related to nearly two-dozen common traits. Using this web-based survey approach, the researchers verified previously reported associations for five traits and identified new SNPs linked to four of the traits, including hair curl, freckling, sneezing in response to light, and the ability to detect asparagus metabolite odors in urine.

"[W]e confirm that self-reported data from our customers has the potential to yield data of comparable quality as data gathered using traditional research methods," co-author Anne Wojcicki, president and co-founder of 23andMe, said in a statement.

For the study, the team drew from 23andMe's direct-to-consumer genetic testing community, bringing together SNP and survey data for thousands of customers. As such, they explained, the researchers were able to put together "a single, continually expanding cohort, containing a self-selected set of individuals who participate in multiple studies in parallel."

"Our ability to contact individuals multiple times and ask follow-up questions puts us in a position to zero in on associations that could be the building blocks for future research aimed at prevention, better treatments, and potentially cures for a multitude of diseases and conditions," lead author Nicholas Eriksson, a statistical geneticist at 23andMe, said in a statement.

Each of the 9,126 participants provided information on at least one of the 22 common traits. These traits were selected, in part, based on heritability and the feasibility of collecting related phenotype data easily in a web-based setting. The traits included everything from hair and eye color to "photic sneezing," a predisposition for sneezing when looking at the sun or other bright light.

The team then integrated this data with genotype data for 535,076 SNPs assessed using the Illumina HumanHap550+ BeadChip. At least 1,500 unrelated individuals of northern European ancestry were evaluated for each of the 22 traits.

Using this approach, the researchers found associations for eight of the 22 traits. Among them: previously reported associations for traits such as eye color, hair color, and freckling and new associations for four of the traits.

For instance, a new freckling-associated SNP turned up in an intron of the zinc finger gene BNC2, while hair curl was associated with SNPs near the TCHH, LCE3E, WNT10A, and OFCC1 genes.

In another new association, the team found that the tendency to sneeze when looking into light was linked to two SNPs: one near the ZEB2 gene and the PABPCP2 pseudogene and another near the NR2F2 gene.

The researchers also found that individuals who can smell an asparagus metabolite called methanethiol in urine tend to carry SNPs in a linkage disequilibrium block in and around 10 olfactory receptor genes. Of these, the most significant SNP fell upstream of the olfactory receptor gene OR2M7.

In an editorial appearing in the same issue of PLoS Genetics, the journal's deputy editor-in-chief Gregory Copenhaver and its gene expression profiling and natural variation section editor Greg Gibson, who are affiliated with the University of North Carolina at Chapel Hill and the Georgia Institute of Technology, respectively, addressed ethics, consent, and data access concerns related to the 23andMe study.

The pair noted that the study's publication was delayed because the journal wanted to deal with several such issues — for instance, ensuring that individuals included in the study were not pressured into partaking in the study and understood that they were participating in genetic research. The editorial also addressed institutional review board questions arising from the study.

"After considering all of the evidence, we decided that publication, accompanied by an editorial providing transparent documentation of the process of consideration was the most appropriate course," Gibson and Copenhaver wrote. "[W]e have had extensive discussion with the authors of this study to address our concerns and to update their processes, but we anticipate broad evolution of GWAS consent and review in the near future."

Meanwhile, the 23andMe team is setting its sights on additional web-based GWAS studies. In their current paper, they noted that the approach makes it possible to ask research questions using data from an ever-expanding group of individuals.

"Our research model makes possible studies that might be infeasible otherwise due to the low marginal cost of asking additional questions over the web and the speed of broadcasting recruitment messages in parallel online," the team wrote, adding that "providing participants with well-explained descriptions of their genetic data can substantially benefit genetic research as a whole."

In addition, 23andMe said yesterday that it has now secured IRB approval for its web-based research protocol. According to the company's blog, The Spittoon, 23andMe customers will now be given more leeway over how their genetic information is used and must "explicitly choose to allow their genetic and survey data … to be used in published research."