A New Approach to Research: The 23andMe Parkinson’s Disease Initiative

By 23andMe Co-Founders Linda Avey and Anne Wojcicki

We started 23andMe with a simple, yet expansive, vision: to take DNA into the mainstream.  

In order to demystify genetics, we thought the best approach was to give individuals access to their genomes and help them gain personalized insight into their own unique code.   This was our premise when we launched a year and a half ago.   We now have

an expanding community of individuals, armed with their genetic profiles, who are the early adopters of what we believe will become standard practice – having ready access to vitally important information. We’re now moving into the next phase of our mission:   to provide a wholly new research platform that enables our online community to voluntarily participate in unprecedented genetic studies.  

Our approach is new because it leverages the web to bring people together from all over the globe who are willing to share information about their own health experiences (phenotype), which is then combined with their genetic profile (genotype).   This combination, genotype + phenotype, is  the same formula that drives genome-wide association studies (GWAS).  But ours is a community-centric model that also delivers on-going, valuable feedback to each member. Scientists have only recently had the ability to conduct GWAS, owing to tremendous advances in the technology platforms that generate the data, as well as spectacular decreases in the cost.   These studies are now yielding compelling results in the examination of many common diseases and are chipping away at the elusive genetic components of  conditions such as type 2 diabetes, heart disease and many of the cancers.  

These diseases aren’t like single gene disorders (such as Tay-Sachs, cystic fibrosis and sickle cell anemia) in which the genetic story is straightforward.   Most common diseases have complicated genetic as well as environmental components, and teasing out these factors is painstakingly difficult.   This also holds true for the study of genes and drug response (pharmacogenetics), the holy grail of personalized medicine.

One of the biggest challenges in conducting GWAS is identifying large enough cohorts of people with a disease or trait, and then being able to categorize the details of their symptoms and progression.   Combine this with the complexity of multiple genetic factors, each with a relatively small effect but somehow working in concert, and it becomes maddeningly difficult to put two and two together.  

This is why these studies require very large numbers of enrolled individuals, to achieve the statistical power required to make any headway.   If researchers are limited to a defined geographic region in which to recruit patients, they often can’t reach the bar.   This often leads to consortia-based projects, where multiple clinical centers combine resources.   The problem with this model is the lack of continuity between the groups, not to mention the power struggles that often ensue: Who writes the grant? Which lab runs the samples? Who controls the rights to the data?   Which institute files and maintains the patents? Who is the lead author on the publication? Our goal is to greatly simplify the entire process.  

By centralizing the recruitment of individuals, the lab work and the collection of phenotypic data, we believe we’ll be able to move beyond traditional hurdles and take GWAS to a whole new level that we’re calling Research 2.0.   We think the study of human disease and drug response deserves the application of 21st century technology, including the use of social networking tools proving so effective in web-based sharing of information a  la Facebook and YouTube.

So today we are announcing the first of many studies we plan to undertake.   Parkinson’s disease has all the elements described above:   complicated genetics, hints of environmental triggers, varying rates of progression, differing drug response.   We’re excited, and gratified, to have Sergey Brin’s support, as well as the cooperation of The Parkinson’s Institute and The Michael J. Fox Foundation, in jump-starting the world’s largest study of this disease – involving 10,000 individuals with PD.  

Please visit our Parkinson’s Community for more information on this ground-breaking project. This is just the beginning of our mission to establish a new paradigm that changes the face of research, and focuses on the people rather than the process.

  • cosmiquemuffin

    How does one get involved in helping out with this new research platform? There are so many diseases/afflictions out there that do not have the numbers to make it profitable for Big Pharma to fund research. I love the idea of being able to democratize research and lower the bar to share and process knowledge necessary to find cures.

    My interest is in a particular affliction (Spasmodic Dysphonia) but am willing to contribute even if the application to this condition is far off. I know there are a lot of diseases in your docket — Autism, Alzheimer’s and more — and would love to be a part of the work to make this new research platform happen.


  • 23andMe is developing a way for people to start their own communities around specific conditions or traits. In the meantime, if you send an email to help@23andme.com it will be directed to the proper person. Thanks for your interest!

  • Kent Kyle

    I am not really sure how to direct this within the group. Please direct to the appropriate party.

    I have lost two uncles, a father, and a step-father to parkinson’s and I as such have lived with this “disease” most of my adult life in one way or another.

    I have followed the research carfefully for the last 30 years and have concluded that 99% of the analysis and attempts to treat such symptons are worthless. Today we have the ability to prescribe the destructive sinemet, which is exactly where we were 40 years ago. Nothing has improved, and the general knowledge of the average nuerologist in terms of dosage is rather low. Often they are quick to prescribe with little understanding of the consequences. No research has been done to my knowledge that looks into the commonly known fact (among parkinsons victims) that dopamine needs diminish greatly in face of emotional stimulation — music, laughter, other positive stimulation.

    But enough about the symtoms — I am convinced that the key to findng a cause lies in data mining. While exciting, the LRRK2 gene finding I think shows a susceptabily rather than a cause: such as finding that a large percent of people with no coats that go out in the elements come down with a cold.

    I propose a data research initiative designed to investigate the family history and relationships not just of the victims, but much more importantly of their surviving family members.

    My own feeling is that Parkinsons has nothing to do with genetics: It is caused by something that is a) passed along at birth and b) passed along via sexual relations. This is just anecdotal of course based on my on observations for 30 years, but I wonder if any data mining type research has been done that could validate or dismiss this hypothethis? Of course, a sex-linked gene could mimic this.

    Parkinsons support groups exists in every town and if would be rather simple to target a very large pool of Parkinsons family member survivors with a large historical survey. Again, I do not know exactly what we are looking for, but I feel this is a path that could shed light on an area that I feel has fallen victim to focus on the tree and not the forest.

    Kent Kyle