Addressing “The Dress”

I know you’ve seen it, whether from TV, social media or a friend pushing their phone into your face and asking “What color is this dress?”

(Disclosure: I only see it as blue and black.)

Don’t worry; there isn’t something wrong with your loved ones who see the dress differently than you. The biological explanation lies in the way in which human eyes have evolved to determine colors under various light conditions, and how those determinations vary from person to person.

Human beings evolved to see in daylight, but daylight changes color. That chromatic axis varies from the pinkish red of dawn, up through the blue-white of noontime, and then back down to reddish twilight. “What’s happening here is your visual system is looking at this thing, and you’re trying to discount the chromatic bias of the daylight axis,” says Bevil Conway, a neuroscientist who studies color and vision at Wellesley College. “So people either discount the blue side, in which case they end up seeing white and gold, or discount the gold side, in which case they end up with blue and black.” (Conway sees blue and orange, somehow.)

Screen Shot 2015-03-01 at 3.44.43 PMCheck out this excellent article from Wired analyzing the science behind why people can’t agree on the color of a dress in a poor quality photo.

23andMe and the Genomics Frontier


Last Thursday, the US Food and Drug Administration authorized the biotechnology company 23andMe to sell a consumer test for the genes that cause Bloom syndrome, an inherited recessive gene disorder that causes short stature, heightened skin sensitivity to sunlight and an elevated risk of developing cancer. The importance of this news comes not from its conclusion, but because it’s a stark reversal from the FDA’s previous dealings with this California-based start-up.

A New Age

Founded in 2006, 23andMe set out to sell humans a peek at the code that defines them. Simply spit into a vial, mail it away, and you’d soon be able to learn about details gleaned from your DNA, ranging from your ancestry to whether you’re destined to hate cilantro. But the tests’ most attention-grabbing feature was its analysis of one’s genetic predisposition to health problems like Parkinson’s or heart disease.

“The advent of retail genomics will make a once-rare experience commonplace,” wrote Thomas Goetz in a November 2007 article in Wired that declared 23andMe’s launch the start of the “Age of Genomics.”

“Simply by spitting into a vial, customers of these companies will become early adopters of personalized medicine. We will not live according to what has happened to us (that knee injury from high school or that 20 pounds we’ve gained since college) nor according to what happens to most Americans (the one-in-three chance men have of getting cancer, or women have of dying from heart disease, or anyone has for obesity). We will live according to what our own specific genetic risks predispose us toward.”

Though the test originally cost nearly $1,000, 23andMe soon cut the price to $399 in order to build up a database of genetic data for use in pharmaceutical and biotechnology research. Time named the test its Invention of Year in 2008, and continued buzz allowed the company to raise more than $50 million from investors and eventually drop its selling price to $99.

“You’re donating your genetic information,” 23andMe co-founder and CEO Anne Wojcicki told Time in 2008. “We could make great discoveries if we just had more information. We all carry this information, and if we bring it together and democratize it, we could really change health care.”

The Dark Years

2010 turned out to be a bad year for 23andMe. The FDA has long allowed testing kits, such as pregnancy tests, to be sold directly to consumers. 23andMe sold its “personal genome service” (PGS) under the assumption that it was allowable as a home testing kit, but the FDA had reservations, and sent the company a letter that year saying it considered the kits “medical devices,” and thus much more heavily regulated. The company was also dragged before a congressional committee to respond to damaging allegations from a sting operation conducted by the US Government Accountability Office. For a more thorough summary of this affair, check out this post from the excellent (and sadly inactive) science blog Genomes Unzipped.

After years of debate, a stern warning letter in November 2013 seemed to seal 23andMe’s fate – stop selling health tests until you get approval.

“The Food and Drug Administration (FDA) is sending you this letter because you are marketing the 23andMe Saliva Collection Kit and Personal Genome Service (PGS) without marketing clearance or approval in violation of the Federal Food, Drug and Cosmetic Act (the FD&C Act),” wrote Alberto Gutierrez of the Office of In vitro Diagnostics and Radiological Health. […] Most of the intended uses for PGS listed on your website, a list that has grown over time, are medical device uses under section 201(h) of the FD&C Act. Most of these uses have not been classified and thus require premarket approval or de novo classification, as FDA has explained to you on numerous occasions.”

23andMe was chastened, and quickly announced that it would discontinue all of its health-related testing, thus limiting its products only to ancestry information and raw genetic sequencing. Though the FDA intended to limit inaccurate diagnostic use of a person’s genome, its rebuke of 23andMe was widely viewed as overbearing and pointless. Science writer Razib Khan argued in Slate that this decision “highlights the tension between the paternalistic medical establishment that arose to deal with the dangers of 19th-century quack medicine, and a “techno-populist” element of American society pioneering personal health assessment and decision-making by leveraging new information technologies.”

“The glaring weakness in an aggressive strategy against interpretative services is that there will always be firms such as 23andMe, and there’s no reason that they need to be based out of the United States. Not only that, but there are open-source desktop applications, such as Promethease, that provide many of the same results by combining individual raw data with public peer-reviewed literature, if less slickly than 23andMe. To truly eliminate the public health threat that the FDA is concerned about, the U.S. government would have to constrict and regulate the whole information ecology, not just a strategic portion of it, from scientists distributing research about genetic variants, to international genome sequencing firms returning raw results on the cheap.”

On the other side, some argued that 23andMe was brazen in its defiance of the FDA, and it should have seen this coming. In a column for Forbes, health writer Matthew Herper called out the tech company for having gone more than six months at a time without communicating with the FDA.

“Either 23andMe is deliberately trying to force a battle with the FDA, which I think would potentially win points for the movement the company represents but kill the company itself, or it is simply guilty of the single dumbest regulatory strategy I have seen in 13 years of covering the Food and Drug Administration.”


In its announcement last week, the FDA not only authorized the one test for Bloom syndrome, but also exempt all future carrier tests from “premarket review” by the agency. In particular, the language of the announcement seems to express a newfound willingness to accept direct-to-consumer genetic testing as part of 21st-century healthcare.

“The FDA believes that in many circumstances it is not necessary for consumers to go through a licensed practitioner to have direct access to their personal genetic information, wrote Gutierrez. “Today’s authorization and accompanying classification, along with FDA’s intent to exempt these devices from FDA premarket review, supports innovation and will ultimately benefit consumers.”

“This is a major milestone for our company and for consumers who want direct access to genetic testing,” Wojcicki said in a press release. “We have more work to do, but we remain committed to pursuing a regulatory path for additional tests and bringing the health reports back to the US market.”

Now, it seems only time will tell when the Personal Genome Service will return to the United States. I know I’m very eager to try it when it does.