Skip to main content
Premium Trial:

Request an Annual Quote

Intermountain Healthcare Combining Consumer Genomics Data With EHRs to Propel Research


NEW YORK (GenomeWeb) – Intermountain Healthcare recently established a new online registry to collect raw array data from consumer genomics customers.

The Salt Lake City-based health system hopes to pair the data with nearly three decades worth of electronic health records as well as medical histories provided by contributors to create a new resource for genetic research.

According to Stacey Knight, a cardiovascular and genetic epidemiologist at Intermountain Medical Center Heart Institute, she was inspired to establish the registry, called GeneRosity, after taking a consumer genomics test offered by AncestryDNA.

Most consumer genomics providers, such as AncestryDNA and 23andMe, genotype their customers using microarrays manufactured by San Diego-based Illumina. Once they are genotyped and their results are reported back to them, customers also have the option to download their raw array data. They can upload this data to third-party sites to run their own analyses or to other providers, such as Family Tree DNA, to take advantage of their tools and services.

Established in 1975, Intermountain runs 22 hospitals in Utah and Idaho, and provides insurance to roughly a quarter of the population of Utah. While the system has EHRs dating back to the early 1990s, it lacks the amount of genetic data necessary to power studies. Therefore, the healthcare system is looking to get customers of consumer genomics firms to voluntarily donate their data to the GeneRosity registry.

"We have here at Intermountain a genealogy database that we had linked to our patient records, and we do have some DNA samples, but we didn't really have a lot of genetic test results for our patients," said Knight. "After seeing the raw genetic results [from AncestryDNA], I realized I could use it in research if I had several orders of patients who provided the same raw data result," she said.

Knight added that the data could be used in future genetic studies, medical studies, and for Intermountain to better understand how to treat patients by gauging genetic responses to different treatments.

Intermountain last month introduced the GeneRosity registry at Rootstech, an annual family history conference held in Salt Lake City, and is now accepting raw genetic data uploads. People who choose to donate are also asked to provide their name and contact information and to fill out a personal and family health history using a health-related family tree application.

The registry is open to anyone who wishes to submit his or her data for research purposes, although Intermountain patients are being encouraged to donate. By reaching out beyond its coverage area, Intermountain hopes to work with disease advocacy groups who might upload data from members specific to a particular rare disease, as well as expand the diversity of the database.

They could potentially draw on a large number of genotyped donors. AncestryDNA currently has about 7 million people in its database, while 23andMe has data from about 4 million people.

"Anyone from any area can upload," said Knight. "Obviously, for those who are not Intermountain patients, we won't have the full medical record, but we are asking [people who donate] to provide us with some basic disease and family health history," she said. Knight also said that using genealogical records the registry might be able to link people outside of Utah with those covered in the Intermountain Healthcare system.

Another issue is that different consumer genomics services rely on various custom arrays, meaning there is only a partial overlap between raw data generated by those services. Intermountain is requesting that people who have been tested by multiple companies upload all of their raw data files, so that they can impute the missing genotypes and then generate data  on a larger scale.

Should the registry gain enough data, Knight said that Intermountain's researchers would focus their initial research efforts on identifying genes that are protective against disease, and then leverage those findings to help better treat those with the disease.

"I would also love to see us be able to develop a better understanding of an individual's [genetic] risk and what they should do in terms of early screening, treatment, and lifestyle improvement, based on their genetic and family health histories," Knight said. "If we can empower the individual patient, that helps them better treat themselves."

Knight noted that Intermountain's research team includes investigators from the Heart Institute, as well as the system's oncology-focused precision genomics group. Intermountain's IT team also developed the registry's family health history tool in addition to tools for storage and processing. The operation is being supported via the Intermountain Research Foundation, which provides seed grant money for "worthy research projects that lead to clinical applications, such as this one," Knight said.

Intermountain will not return any analyses back to contributors based on their data. Rather, the healthcare system sees uploading one's data as a donation for medical research. "The benefit is that we can better understand genetics and how it relates to disease and the treatment of disease," said Knight. "The benefit goes to future populations."

That approach differs from DNA.Land, another effort to crowdsource genomic data from consumer genomics customers. Initiated by researchers at Columbia University in 2015, DNA.Land has collected nearly 90,000 genomes for research purposes and also gathers family tree information and phenotypic information from contributors. DNA.Land also provides back a number of reports, such as on ancestral origins, matching contributors to genetic relatives in the database.

The developers of the biobank described DNA.Land in a paper in Nature Genetics last month.

Yaniv Erlich, an associate professor of computer science and computational biology at Columbia and co-developer of DNA.Land, said that he is "very satisfied with the accomplishments" of the site, but noted that crowdsourcing genomic data has a few downsides, including the lack of representation of minorities.

"This is mainly due to self-ascertainment," Erlich said, "but as the price of ... tests keep declining, I think the bias will be smaller."

He said another challenge is reaching out to people who have taken consumer genomic tests to make them aware of resources like DNA.Land or GeneRosity, though he said that DNA.Land had been "successful" in that regard.

Erlich last year took a leave of absence from Columbia to join MyHeritage, an Israel-based online genealogy platform, as CSO. MyHeritage also offers an ancestry testing service similar to those offered by AncestryDNA, 23andMe, Family Tree DNA, and others. Its array is also manufactured by Illumina.

Knight noted that Intermountain differs from DNA.Land in other ways, given its focus on linking its existent medical records and genealogical data with genomic data drawn from consumer genomics services.

"What sets us apart are the massive amounts of EHRs that we have on our patients and our family health histories," she said. "It's true that we are asking people to donate something they have paid for," Knight added, "but we are asking them to be generous and provide it to us for research purposes."