Recently, DNA analysis companies have been all the rage. There has been a major urgency felt from the public to find out “who” they are. The popularity of this idea has blown up so dramatically that there are multiple companies, they demand is high enough that now these DNA analysis kits are under $100, and they even inspired a recent South Park episode.
These DNA kits originated in 2006, were incredibly costly (about $1000 per home kit) and were used by three main companies: 23andMe, DeCodeMe, and Navigenics. The latter two did not make it very far, but 23andMe is still very relevant. In fact, aside from being one of the main successful DNA kit companies.
Earlier this month, the company 23andMe announced tremendous news: through the company’s most recent fundraising round, led by their new investor, Sequoia Capital, the company has now raised $250 million in growth financing, totaling to a $1.75 billion reported valuation.
This major shift has now widened 23andMe’s horizons, which has pushed their attention in a new direction. It seems as though they can move on to DNA-data-mining to look for the next breakthrough in medical science.
For two years now, the 23andMe has been drafting lucrative research contracts with big pharmacy companies. They have also put a tremendous amount of money into their own research and development program where 23andMe workers look through the DNA samples from over one million customers. Out of all of the tests taken, about 85% of the DNA results consented to having their DNA analyzed for research purposes. To date, it is the largest consensual, re-contractable database for research on genetic issues.
23andMe with Parkinson's Disease
While some people may be unnerved by the Twilight Zone-esque premise of a big company having their DNA to use for research, the company is hoping to do some good for patients of a variety of different medical issues. The one disease that has received much of 23andMe’s focus is Parkinson’s. In the past month alone, the company has found over a dozen types of new mutations that they found to be linked to Parkinson’s.
Parkinson’s Disease was made a household name to people fortunate enough to not have to deal with the disease on a personal level when Michael J. Fox was diagnosed with it in 1991. Parkinson’s is a neurodegenerative disease, which means that the structure and functions of neurons in the brain begin to fade away, sometimes killing some neurons. It currently affects 1 million people in America and is currently incurable.
Because of the evasive workings of Parkinson’s, many scientists are unclear on how the illness actually works or even to what extent genetics impact its development and progression. Some other genetic diseases, like Huntington’s (a fatal genetic disease where the nerves in the brain are broken down), have a single genetic signal that enables the disease. Parkinson’s, on the other hand, continues to baffle as scientists realize that there is no one genetic signal for the disease. The overall evasive and confusing nature of the disease is a big reason as to why 23andMe is driven to continue to use their database to look for patterns, links, and specific genes in order to help to find a cure. They hope to attain this goal through pattern matching, which they feel can help to find the genetic causes, which will hopefully lead to finding cures quicker.
Within the past two weeks, 23andMe researchers met with Gentech and the National Institute of Aging to publish what is now the biggest meta-analysis of Parkinson’s disease. They went through DNA and data from more than 425,000 people, over 75% of whom had given their DNA as well as their consent to 23andMe when they submitted their kit. The study published was piggybacking on a significantly smaller study conducted in 2014. The study took 6,476 people with Parkinson’s Disease and used 302,042 controls, which was followed up by a meta-analysis, which looked at the information they gathered against over 13,000 peopler with Parkinson’s Disease, 95,000 controls, and 9,830 overlapping variants. Using this study, 23andMe looked at the data from the 6,476 people who were self-reported people living with Parkinson’s Disease and compared them to 300,000 disease-free clients from 23andMe who had submitted their kits. This study confirmed the genetic variants that were already thought to be connected to the disease, as well as 17 brand new patterns and links.
One of the lead scientists for the company, David Hinds, co-authored the aforementioned study. He came to 23andMe about eight years prior. His initial reasoning for joining the company was not to help people trace their ethnic routes as much as it was to help cure Parkinson’s. When he first started, Hinds and others were convinced they could possibly identify most of the Parkinson’s genres with samples from a few thousand people. This number was clearly very short of tremendous amount of samples they now have.
After some time in the lab, Hinds realize very quickly that they were not going to be able to do this easily. They discovered new layers of complexity with the Parkinson’s Disease genes, realizing this was going to be harder than they initially thought. “These much larger sample sizes are required to make reliable discoveries. You need a lot of findings to see how some of these associations line up among different cellular processes.” Hinds explains. He brings up a good point. Often people think of diseases as one dimensional, but Parkinson’s, as well as other degenerative diagnosis, need to have cures looked at regularly through different stages to see if there is any shift in different cellular processes and stages.
Initially, Hinds reporting found pretty regular news for people who are already living with or have experience with Parkinson’s Disease. One analyst who deals primarily with pharamacogenomics, Derek Lowe, is convinced that while the paper published shows promise, that no one should get their hopes up. “It is a pretty large data set, but GWAS [genome-wide trait association studies] have their limitations in suggesting new drug targets, particularly in well-worked-on areas like Parkinson’s,” Low explains. This is because GWAS will pick up common traits over mutations that are directly responsible for disorders. Often, these mutations that scientists look for are in possession of signals so faint that often they get glossed over.
This reasoning is why Genetech has decided to make their next step: to sequence the complete genomes of 3,00 of 23andMe’s patients who are living with Parkinson’s. All of these patients are volunteers. These volunteers answered questions about family history, the speed of their disease, and a list of treatments and how successful they were. The hope is that if they go through all three billion base pairs, Genetech will get past the more common and obvious traits o the disease. Rather, the company has decided to look for more rare variants, often more devastating ones which can ruin more of the biological workings of the body than average. They feel that working backwards by tracking the devastation to parts of the body backwards would help them find these variants.
The current agreement is this: Genetech will have two full years of exclusive access to this database after the last person’s DNA is totally processed. After two years, the database will open to all scientists. Rob Graham, a senior scientist for Genetech, and co-author of the Nature Genetics paper, says “Personalized medicine is more challenging in Parkinson’s, than in, say, cancer, because of the complex heritability of the traits.” While the road ahead will be bumpy, there is definitely hope for making some excellent findings to help cure Parkinson’s Disease.