If you suspected that bioengineering is the new frontier of human endeavor, the experts agree. Genomics – with its proliferation of metaphors – scientists as pioneers and bioengineering as the home of innovation and rugged individualism – is the new Wild West. Consumer genomics, made possible by the plummeting costs of DNA testing and its growing popularity via the internet, is already a reality. Its current and future value in medical diagnostics, as well as a timetable within which it can become widely available as a healthcare tool for the general population, are the topic of much debate.

Dr. Gholson Lyon is an assistant professor of human genetics at Cold Spring Harbor Lab (http://lyonlab.cshl.edu). The Lyon Lab, where he works, is dedicated to finding the genetic markers of neuropsychiatric diseases. In our conversation, Lyons stated that he is much more pessimistic than some of his colleagues in the field when it comes to predicting a timeframe for the application of genomics research in everyday medical situations (IE: fully sequencing the genes of all newborns, as was brought up in my interview with one of the founders of Tute Genomics). He does not see it being widely used in the next 5 to 10 years and feels that it might be as long as 50 years before it can be widely implemented and integrated into today’s healthcare system.

He attributes this probable delay to stringent FDA requirements that consumer genomics companies prove the validity of the medical information provided to consumers through extensive clinical testing. The FDA thus plays very conservative role, to make sure products are absolutely safe before they can be marketed. Given these beaurocratic constraints, Lyon’s predicts that it will likely take a long time before companies like 23andMe and uBiome (http://ubiome.com) can market these technologies in the US.

From the point of view of medical research, this represents a catch-22 situation. Vast data streams of genetic information and medical conditions are needed before valid correlations can be made; at the same time must very hard to incentivize people to participate in the research out of altruism. Dr. Lyon feels that one way to get people to participate in genotype sequencing is by offering them a product as 23andMe did.

The FDA, however, has taken an very proactive role in the case of 23andMe, a privately owned personal genomics company. Their personal genome test kit was chosen as the 2008 Invention of the Year by Time magazine. By 2010, the FDA notified 23andMe along with other companies that it regarded the test kits as medical devices and federal approval was required to market them. FDA concerns centered on the interpretation of medical information being provided to customers, which could be used to make major decisions without a valid basis. In December 2013, 23andMe’s health-related tests were shut down and the website now reads:

“At this time, we have suspended our health-related genetic tests to comply with the U.S. Food and Drug Administration’s directive to discontinue new consumer access during our regulatory review process. We are continuing to provide you with both ancestry-related genetic tests and raw genetic data, without 23andMe’s interpretation.”(https://www.23andme.com/ancestry-only-notice/?redirect=f8A5PuhSB7l1NXhT7FNJ_zkUDt-QKyX6b1TSAIgoxRU=)

The FDA has been widely criticized for this decision (http://www.theguardian.com/commentisfree/2013/dec/04/23andme-consumer-genomics-fda-ban-regulation). Critics have argued that the FDA and regulatory mechanisms must recognize that the internet and consumer genomics have changed the way people look at healthcare. Some of the medical and ethical concerns attending the spread of consumer genomics are outlined at http://www.genome.gov/26524402.

Dr. Lyon viewed this as a problem of different political perspectives. The first, which is a more paternalistic approach, and wants to protect the American consumer from any kind of danger, would feel that 23andMe was making bold and un-validated medical interpretations. A libertarian outlook, to which he subscribes, holds that information should just be freely provided and people should be free to make their own interpretations and seek second opinions. 23andMe in his opinion were not making unfounded or invalidated claims, and the website had adequate disclaimers about the medical validity of the results provided. The American pioneer spirit, he feels, allows people to do things that others might consider unsafe, like skydiving, drinking alcohol, buying herbal supplements etc.

These are, Dr. Lyon emphasizes, very much the pioneer days in genome sequencing. Very large data streams are needed to accurately correlate genetic information with medical conditions. Yet it is precisely these data that are difficult to obtain for medical research. In the absence of data, targeted sequencing is sometimes considered an alternative. Dr. Lyon’s concern is that limited sequencing may not provide reliable correlations and opens the  possibility of errors.

Genealogy, always a popular concern for Americans, may become the point of entry into the widespread use of consumer genomics and the  creation of data streams which can yield information regarding health issues as well. Direct to consumer (DTC) companies like Ancestry.com and 23andMe do genotyping and link up families, building family trees which can provide an understanding of how diseases run in families. This field is moreover not as severely regulated at the moment. For this information to have reliable medical diagnostic value, the first step is globalize family trees and contextualize medical information about disease phenotypes within those.

Genealogy might thus serve two purposes – to familiarize people with genome technology, and to be able to place their own disease markers within specific contexts of close family and environment. In Lyon’s view – the forces in place won’t allow genetic data to do much else on a large scale (for now), but it’s a contribution he still sees as worthwhile.


TechEmergence conducts direct interviews and consensus analysis with leading experts in machine learning and artificial intelligence. Stay ahead with of the industry with charts, figures, and insights from our unparalleled network, including executives from Facebook, Google, Baidu, Yahoo!, MIT, Stanford and beyond: