This story is over 5 years old.

It Takes 100 Hours to Analyze a Human Genome for Disease

Personalized medicine will remain a (mostly unreliable) luxury item until analysis tactics improve.
March 12, 2014, 1:30pm
Image: Nerds for Nature/Flickr

Despite the promise of the $1,000 genome sequence, personalized medicine is likely to remain a luxury product—and not a very accurate one at that. For the foreseeable future, researchers will continue to struggle to overcome significant hurdles that make predicting someone’s propensity for disease expensive, time consuming, and potentially unreliable.

A new study on the present-day feasibility of whole-genome sequencing for clinical use by researchers at Stanford University found that it will cost at least $17,000 per person to sequence a genome and interpret the results, and it’ll take roughly 100 man-hours to perform any sort of meaningful analysis.


“The gist of it is we found that the results are generally not clinically acceptable,” said Frederick Dewey, lead author of the analysis, published in the Journal of the American Medical Association. “It’s a relatively sobering thought, and there are tough hurdles to get over before this is common.”

Dewey and his team completely sequenced the genomes of 12 people and analyzed them to predict their propensity for genetic diseases and other health concerns. The study was designed to test out some of the leading genome sequencing techniques and analysis methods. The expensive part, Dewey says, isn’t necessarily the sequencing of the genome (a cost that is constantly coming down), but the analysis after the fact.

“The main challenge at this point, assuming you have technically valid data, is from a manpower perspective,” he said. “What we have learned is that a sequence doesn’t equal interpretation, there’s a significant manual interpretation job afterward.”

But that’s not all. The reason the results were not deemed clinically acceptable in most cases is because, even with the most advanced sequencing techniques, the researchers found that, on average, between 10 and 19 percent of inherited disease genes “were not consistently covered at a read depth that was sufficient for a comprehensive survey of genetic variants.” In general, whole genome sequencing wasn’t great at picking up nucleotide insertion and deletion mutations—with one commonly-used sequencing method, just one third were picked up. That’s important because those mutations are commonly medically relevant.

The study wasn’t all bad news, however. Of the dozen subjects whose genomes were sequenced and analyzed, the team found one, a woman with no family history of breast cancer, who had a mutation in the BRCA1 gene that makes her much more likely to develop the disease.

Dewey says that, though the results of this initial study suggest that personalized genomic medicine isn’t quite ready for primetime, that all of the problems the team found are ones that researchers are actively working on.

“It’s remarkable how far we’ve come and I would expect the pace of discovery and innovation will continue,” he said. “There’s a long way to go, but what we’ve done so far is far beyond anything we would have anticipated years ago.”