Skip to content Skip to navigation

Linking Genomics and Clinical Data: A Glimpse of Tomorrow

December 4, 2013
| Reprints

The 1964 World's Fair                           credit: Wikipedia Commons

Not far from where I live is Corona Park, the main site of the 1939 and 1964 World's Fairs. Both fairs are well known as far as World Fairs go, specifically the 1939 rendition, which offered spectators a "glimpse of tomorrow."

While these kinds of international expo events still occur today, the cultural impact is significantly less than what it was back then. The 1939 World's Fair was one of those events that had a tremendous impact on an entire generation. I've often read about the "The World of Tomorrow," a showcase at that fair which previewed the futuristic world of speeding cars, transcontinental flights, personalized video screens, and much more! Twenty-five years after that, exhibitors at the 1964 World's Fair held similar previews for "tomorrow" involving space travel and the use of computers.

I almost wish these idealistic (albeit corporate-sponsored) sneak previews still existed in a grander scale. Can you imagine the thrill of seeing this world of promise so neatly laid out for you like that? Of course, we're still exposed to futuristic innovation, but not on that kind of level or without cynicism to accompany it. 

When it comes to advancements in medical technology, there are innovations that deserve the kind of buzz that TVs received at the 1939's World Fair. There's Watson, Google Glass, and perhaps most importantly, the increased linking of clinical and genomic data.

This past week, a team of researchers across the country, led by a group from the Vanderbilt University School of Medicine, revealed the results of the first ever large-scale phenome-wide association study (PheWAS) in the journal, Nature Biotechnology. This kind of study has researchers looking at the various diseases associated with one kind of genetic variant (typically it's done inversely through genome-wide association studies), using electronic medical record (EMR) data.

The collective, known as the Electronic Medical Records and Genomics (eMERGE) Network, was able to link gene variants to various diseases. The network has built a database that includes the DNA samples from about 51,000 individuals linked to medical records, which can be extracted for analysis. In total, the researchers discovered 63 new links by digging through the available genomic data within the EMRs of 13,835 individuals with European ancestry and categorizing the data into 1,358 EMR-derived phenotypes.

The researchers also replicated findings of 66 percent (51/77) of sufficiently powered prior genome-wide associations. In an interview with The New York Times, Joshua C. Denney, M.D., one of the lead authors of the study and an associate professor of biomedical informatics and medicine at Vanderbilt, said that it was impossible for the researchers to do that well only by chance. In the same article, Robert C. Green, a geneticist at Harvard Medical School, said it was a "phenomenal proof of concept," and Daniel MacArthur, a geneticist at Massachusetts General Hospital, predicted that large volumes of EMR data would help link diseases and rare gene variants.

The possibilities for what research teams like eMERGE can discover with this kind of data at their fingertips are endless, especially as EMRs begin to contain more sets of genetic information. Dan Roden, M.D., an assistant vice chancellor for Personalized Medicine and principal investigator for the Vanderbilt eMERGE site, said that this kind of research can help determine why certain diseases are presented differently in various people, or why drugs might produce unpredicted effects in some patients.

Certainly, there are challenges that must be addressed before these kinds of research discoveries are mainstream and widespread. For one thing, an article published in Genetics in Medicine, described the obstacles in integrating genetic and clinical information. Those include the size and complexity of genetic test results, inadequate use of standards for clinical and genetic data, and limitations in EMR capacity to store and analyze genetic data. As many researchers have pointed out, current EMRs aren't made to store this kind of data.

“EHRs are designed to facilitate day-to-day patient care,” said Justin Starren, Ph.D., chief of the Division of Health and Biomedical Informatics in the Department of Preventive Medicine at Northwestern University Feinberg School of Medicine. “EHRs are not designed to store large blocks of data that do not require rapid access, nor are they currently capable of integrating genomics clinical decision support.”

Dr. Starren and his colleagues argued in a March 2013 study, appearing in the Journal of the American Medical Association, that current EHRs are not able to handle the amount of information, including genomics, created by currently available medical technology.

Despite the present day obstacles, it's hard not to get excited about what is viable and within reach. A few weeks ago I wrote about Google Glass, and while that technology is enthralling, I sincerely believe these kinds of genomic/clinical connections will be the most interesting, important development of the next 10 years or so.