You may have heard rumblings about something called SNPedia. I finally got around to checking it out the other day.
Image credit: Stewart Butterfield (click image for link to original)
SNPedia has been called the ‘Wikipedia for DNA’ and to me it kind of seems like OMIM, but for SNPs. It is an online crowdsourced and publicly accessible database where you can find ‘information about the effects of variations in DNA, citing peer-reviewed scientific publications.’ SNPedia is totally independent of the companies that are selling DNA sequencing or microarray testing, and they are often consulted by people who are looking for a second opinion on a result reported in their DTC report. SNPedia is affiliated with a tool called ‘Promethease,’ which helps build a free report (in 3 hours!) for you based on your uploaded SNP-based data. Promethease can be used to pool the results for people who have data from multiple online SNP-based testing services (23andMe, Navigenics, deCODEme). (Side note: I wonder how many people have purchased SNP-based testing from multiple online companies?)
Interestingly, in browsing the SNPedia FAQ page I stumbled upon this question:
“Can you refer me to a physician or a genetic counselor to discuss my SNP testing results?”
Their response:
Not yet. If you are a qualified physician or genetic counselor interested in helping individuals interpret their genomic test results or Promethease report please email us at info@snpedia.com.
I wondered if they have had any takers, so I sent an email. I learned that they had heard from one interested genetic counselor in past, but due to issues with the GC’s institutional policies regarding referrals they were unable to make it work. In addition, they have had interactions with a handful of GCs who have contacted them to discuss results on a specific case they’ve been involved with.
I asked Greg Lennon, Co-Founder and Director of SNPedia how he envisions a genetic counselor might be able to collaborate with SNPedia. Here is his response:
We (SNPedia) welcome their input, especially in the form of edits to entries to improve their utility to other GCs and health care professionals (and of course, members of the public), but GCs should always also feel welcome to just email us (info@snpedia.com) with suggestions of any type, whether for edits they won’t or can’t do, or for features they’d like to see added to either SNPedia or it’s companion software, Promethease.
By now we all know that the cost of sequencing the genome isn’t going to be the major barrier in accessing our genomic data. It is going to be the interpretation of that data. SNPedia seems like a huge step forward in making sense of the wave of genomic info that is coming our way. And I think that our community has an opportunity here to help influence the way in which this information is delivered.
I’m interested to hear if any of you have experience with SNPedia, and what your thoughts are on their service, reporting etc. Also, if you have questions or suggestions about how GCs can collaborate with this service, please leave a comment below. As Bob Resta recently pointed out, ‘comments are what make blogs interesting.’
The DNA Exchange 
To some extent, SNPs remind me of the 1970s when HLA typing was in its prime. Every day a new paper came out claiming an association with just about every genetic and non-genetic disease you can think of, and the relative risks were roughly the same as those associated with most SNPs. HLA haplotypes were also used to estimate ancestry and population migration history. HLA typing was going to change the face of medicine, disease prediction, and the understanding of human evolution and migration. While HLA studies have certainly provided some key insights, especially in organ transplantation, eventually, many of those HLA/disease associations fell by the wayside, except the more significant associations, like HLA-B27 and ankylosing spondylitis. I wonder if SNPs will follow a similar path.
Thanks for your comment, Bob. I agree with you that SNPs and Genome Wide Association Studies could very well represent a passing genetics ‘fad.’ But I think the difference is the context. With the cost of sequencing being so low (why pay $2500 for sequencing a panel of 5 genes, when you can sequencing the entire genome for $1000?), SNP data is going to be generated anyway. And if you have a result of a particular SNP in front of you, you may as well know what the reported disease associations are.
I should mention that according to Greg Lennon, in the future SNPedia plans to include data about more classical mutations as well as structural variants (CNVs).