Big tech companies have been in the spotlight with news coverage of how our personal data has been used or abused–and for many people, the lack of privacy is an unavoidable reality. But tech companies aren’t the only ones interested in obtaining our personal information. Health researchers and data scientists are looking to the widespread sharing of personal data as an opportunity to learn more about genetics, diseases, and overall personal health.
“Big Human Data,” the first taste of science event of the year, welcomed two experts on the topic: Hannah Bayer, Ph.D., chief scientific officer at Data Cubed, and Wendy Chung, Ph.D., director of clinical research at the Simons Foundation.
Bayer compared her view of big human data to the laborious, weather-dependent approaches early astronomers used to gain a base understanding of the stars. The practice was revolutionized about 25 years ago, she said, when scientists discovered that bolting a telescope to the ground allowed them to create a massive library of images while the earth was turning. A database including all the black holes in our universe made it easier for scientists to “go in, and just pick out all the black holes, and do your research that way,” she said. This is what turned astronomy into a data science.
“What if we could do that for humanity?” she asked the audience. “What if we could understand what makes us ill, what makes us healthy, what makes us successful … What if we could create a catalogue in just the same way?”
The Human Genome Project, a map of all the genes of the human genome, was completed in 2003, but Bayer argued that this is not enough because “genotype is not fate.” Genetics and environment both play crucial roles in determining a person’s quality of life, and scientists have since learned that different lifestyles can trigger different genes.
By comparing the likelihood of a shared trait for several different health conditions among both fraternal and identical twins, Bayer said that certain ones, such as breast cancer, are not as reliant on genetics alone. Environmental and dietary factors can also play a role. However, for neurological conditions schizophrenia and Alzheimer’s disease, researchers have come to believe that genetic influences are a dominant factor. “The percentage of identical twin pairs who both have schizophrenia or Alzheimer’s is around 60 percent; so that’s a pretty high correlation,” she said.
Researchers are still figuring out exactly what kinds of questions they need to be asking people to collect better data. “We need better phenotyping, which is really going to help us unlock what the relationship is between the environment, lifestyle, and biology,” said Bayer.
She called particular attention to deep phenotyping, which could inform outcomes for health, education, and financial success through data categories such as diet, sleep, and profession. The catalogue, as Bayer imagines it, would include a trajectory of a person’s entire life (starting from early childhood and schooling) and financial information (to help inform social determinants of health). The goal is to give researchers at 360-degree view of what shapes peoples’ lives, she said.
When asked where scientists would get this data, Bayer’s reply was not a complete shock but still carried a cautionary tone: “For better or worse, we live in a world where that’s not so hard anymore. Much of this data already exists in places where you strew your data,” she said. Location devices in smartphones make it incredibly easy for phone companies—and any company interested in buying that data—to know exactly where a person has been and when, for example.
Chung’s research for SPARK (Simons Foundation Powering Autism Research for Knowledge) is a perfect example of how large datasets of personal information can be used for good. As principal investigator of the largest genetic study of autism in history, she stressed the importance of having access to this kind of data to better understand the disorder. The mission of SPARKS is to speed up research and advance the understanding of autism to help improve lives.
The study’s initial goal was to recruit 50,000 individuals with autism and their family members, and after two-and-a-half years, they now have 150,000 participants from around the world. By collecting participants’ DNA, researchers can identify whether they have the genetic diagnosis of autism. Chung made clear that they want to make the research and results behind the study as transparent and accessible to the public as possible.
There is currently no FDA-approved drug to treat the core symptoms of autism, she said, and the challenge for neuroscientists, psychologists, and neurologists is the heterogeneity in individual cases. If participants of SPARKS take part in the genetic portion of the study (it’s not required) and Chung’s team finds a genetic diagnosis for autism, they return a full diagnosis to the participant. Through this work, Chung’s team has identified the one genetic major risk factor believed to cause autism. (They are currently partnering with a clinical diagnostics laboratory to corroborate these findings.)
The accomplishments of SPARKS extend beyond the lab as well. Chung said that through this mass study, they’ve created communities of people who received their diagnoses. Facebook has allowed people with rare genetic conditions to find each other, she said. Through these online communities, people are continuously learning from each other and can share information with each other and with the research community.
Both Bayer and Chung represent the vast potential that personal information, or big human data, has in medical research. It’s one of the most powerful forms of data, said Bayer. “It can be used for ill will, but I would argue that it could be good as well.”
taste of science works to address the widening gulf between what scientists do and what public perception is. The volunteer-based organization holds events all over the country with guest speakers who are experts in their field. For more information, visit their website.
– Seimi Rurup