Monthly Archives: January 2015

U.S. proposes effort to analyze DNA from 1 million people

The United States has proposed analyzing genetic information from more than 1 million American volunteers as part of a new initiative to understand human disease and develop medicines targeted to an individual’s genetic make-up.

At the heart of the “precision medicine” initiative, announced on Friday by President Barack Obama, is the creation of a pool of people – healthy and ill, men and women, old and young – who would be studied to learn how genetic variants affect health and disease.

Officials hope genetic data from several hundred thousand participants in ongoing genetic studies would be used and other volunteers recruited to reach the 1 million total.

“Precision medicine gives us one of the greatest opportunities for new medical breakthroughs we’ve ever seen,” Obama said, promising that it would “lay a foundation for a new era of life-saving discoveries.”

The near-term goal is to create more and better treatments for cancer, Dr. Francis Collins, director of the National Institutes of Health (NIH), told reporters on a conference call on Thursday. Longer term, he said, the project would provide information on how to individualize treatment for a range of diseases.

The initial focus on cancer, he said, reflects the lethality of the disease and the significant advances against cancer that precision medicine has already made, though more work is needed.

The president proposed $215 million in his 2016 budget for the initiative. Of that, $130 million would go to the NIH to fund the research cohort and $70 million to NIH’s National Cancer Institute to intensify efforts to identify molecular drivers of cancer and apply that knowledge to drug development.

A further $10 million would go to the Food and Drug Administration to develop databases on which to build an appropriate regulatory structure; $5 million would go to the Office of the National Coordinator for Health Information Technology to develop privacy standards and ensure the secure exchange of data.

The effort may raise alarm bells for privacy rights advocates who have questioned the government’s ability to guarantee that DNA information is kept anonymous.

Obama promised that “privacy will be built in from day one.”


The funding is not nearly enough to sequence 1 million genomes from scratch. Whole-genome sequencing, though plummeting in price, still costs about $1,000 per genome, Collins said, meaning this component alone would cost $1 billion.

Instead, he said, the national cohort would be assembled both from new volunteers interested in “an opportunity to take part in something historic,” and existing cohorts that are already linking genomic data to medical outcomes.

The most ambitious of these is the Million Veteran Program, launched in 2011 by the Department of Veterans Affairs. Aimed at making genomic discoveries and bringing personalized medicine to veterans, it has enrolled more than 300,000 veterans and determined DNA sequences of about 200,000.

The VA was a pioneer in electronic health records, which it will use to link the genotypes to vets’ medical histories.

Academic centers have, with NIH funding, also amassed thousands of genomes and linked them to the risk of disease and other health outcomes. The Electronic Medical Records and Genomics Network, announced by NIH in 2007, aims to combine DNA information on more than 300,000 people and look for connections to diseases as varied as autism, appendicitis, cataracts, diabetes and dementia.

In 2014, Regeneron Pharmaceuticals Inc launched a collaboration with Pennsylvania-based Geisinger Health System to sequence the DNA of 100,000 Geisinger patients and, using their anonymous medical records, look for correlations between genes and disease. The company is sequencing 50,000 samples per year, spokeswoman Hala Mirza said.


Perhaps the most audacious effort is by the non-profit Human Longevity Inc, headed by Craig Venter. In 2013 it launched a project to sequence 1 million genomes by 2020. Privately funded, it will be made available to pharmaceutical companies such as Roche Holding AG.

“We’re happy to work with them to help move the science,” Venter said in an interview, referring to the administration’s initiative.

But because of regulations surrounding medical privacy, he said, “we can’t just mingle databases. It sounds like a naive assumption” if the White House expects existing cohorts to merge into its 1 million-genomes project.

Venter raced the government-funded Human Genome Project to a draw in 2000, sequencing the entire human genome using private funding in less time than it took the public effort.

Collins conceded that mingling the databases would be a challenge but insisted it is doable.

“It is something that can be achieved but obviously there is a lot that needs to be done,” he said.

Collating, analyzing and applying the data to develop drugs will require changes to how products are reviewed and approved by health regulators.

Dr. Margaret Hamburg, the FDA’s commissioner, said precision medicine “presents a set of new issues for us at FDA.” The agency is discussing new ways to approach the review process for personalized medicines and tests, she added.

Toni Clarke and Sharon Begley, Reuters

CRG board member Robert Green in the news-The US has laws against genetic discrimination, but people still fear it

Genetic discrimination hasn’t yet become a problem in the US. Since the Genetic Information Nondiscrmination Act (GINA) was signed into law in 2008, very few GINA-related charges have been made. That doesn’t mean that Americans aren’t worried that this sort of discrimination might arise in their own lives, however, according to a perspective piece published yesterday in the New England Journal of Medicine.

GINA was signed into law almost seven years ago to prevent health insurance companies and employers from discriminating against individuals based on family history of illness or results from genetic testing. For example, the law prevents an employer from firing an employee because they have a family history of cancer. But GINA hasn’t been invoked very often since 2008. In 2013, for instance, the US only had 333 GINA-related charges of employment discrimination, compared with 90,000 charges relating to other areas of discrimination. Moreover, most of those charges were added following an investigation relating to an Americans with Disabilities Act claim.

“GINA has sent a powerful message throughout society that we won’t tolerate genetic discrimination,” says Robert Green, a geneticist at Harvard University and co-author of the perspective piece. “And I’m guessing that these companies realize that they might get tremendous public push-back if they were to do this.”

But the small number of charges may also be a sign that Americans aren’t aware that GINA exists. An online survey of 1,479 individuals performed in 2014 by Green and his colleagues showed that 79 percent of respondents didn’t know what GINA was or what protection it offered. And people who did know about GINA didn’t fully comprehend it. What’s even more striking, however, is that 30 percent of participants said that they became more worried about genetic discrimination after reading about GINA. “We noticed how present in the minds of our participants concerns about genetic discrimination were — particularly by employers and insurance,” Green says.

This fear is a big problem for researchers because some people refuse to participate in studies as a result, Green says. In order to implement genetic data in a research project, participants have to be willing to tolerate genetic information being included in their medical records and discussed with their doctors. This practice comes with an “omnipresent concern about genetic discrimination,” Green says. And yet, this sort of fear is exactly what GINA was trying to avoid in the first place. “What was fascinating to me was that the very language of GINA was attempting to reassure the public, so that they could take advantage of the fruits of genomic research,” Green says.

The fact that GINA doesn’t cover genetic discrimination relating to life insurance, long-term care insurance or disability insurance may be the reason for continued fear in the US. Yet Green says that he hasn’t seen many insurance companies request genetic test results. “I’m not aware of any recent incidents were anybody has been denied insurance on the basis of genetic testing,” he says.

This could still happen, of course. A 2010 study found that people who learn that they have a higher than average risk of developing Alzheimer’s disease are more likely to buy long-term care insurance. Once insurance companies clue into this, they may start increasing their long-term care insurance prices based on genetic risk factors — a change that would be perfectly legal under GINA.

That is one possible future, Green says. But even that might not last very long. Genomic science is moving quickly, and it’s bound to become more present in our lives. Eventually it may help us personalize medicine and tailor our diets and exercise regimens. Once that happens, the term “genetic discrimination” may become obsolete, the researchers write — genetic-based medicine will simply become all medicine, and any form of medical discrimination will be outlawed by extension.

In the meantime, however, researchers and government officials should try to find a way to overcome the public’s fears. “We’ve had so many stories and narratives of people losing health insurance that this is sort of an existential concern that permeates our society now,” Green says. If we want people to take full advantage of the medical knowledge we’ve gained recently, we have to overcome the “absolute fear of genetic discrimination,” he says. And like many other fear-based issues, education might be the only way forward

Arielle Duhaime-Ross, The Verge

DNA privacy bill strikes right balance

In 1997’s futuristic “Gattaca,” one’s DNA determined one’s destiny. Released a dozen years after the beginning of real-life genetic testing, the movie tracked a man “conceived by love rather than in a lab,” as Variety put it. Because his lousy genes limited his career, he went to “a DNA broker who sells false identities to the genetically inferior.”

As with all dystopian stories, “Gattaca” uses the future to explore trends in the present. Since the movie’s release, the genetic-data conundrum has become more pronounced. Testing offers great opportunities to cure disease and exonerate the innocent accused of crimes, but it also offers potential for government and private firms to invade privacy and misuse data.

A new bill, proposed by Assemblyman Mike Gatto, D-Glendale, attempts to create a balance by making it clear who owns the DNA samples already collected in California hospitals. “A government repository of the DNA of all children born? It sounds like something out of 20th century dystopian fiction,” Gatto’s office wrote, in a statement last week announcing the proposal.

As it further explains, newborns are tested — via a blood sample taken from the heel of about a half-million babies born here each year — and the information is sent to laboratories to screen for diseases and genetic disorders. Most parents don’t even know about the test. All states do this, but California is one of a handful where the state owns and retains the information, and even sells it to private research companies.

That’s the problem, according to Gatto. In an interview last week, Gatto described potentially troubling scenarios. When hackers stole people’s credit-card data from a chain store, that created problems for consumers — but imagine what can happen if hackers get into a database of genetic information? Right now, there are no real protections.

And then there’s the potential for official misuses of the data. Technology is developing rapidly. It’s not hard to envision a “Gattaca”-like scenario in which, as Gatto explains, eventually it’s possible to learn whether a person has a “violent gene” or is predisposed toward alcoholism. “It gets kind of disturbing,” he said. “I don’t mean to sound alarms, but it gets alarming.”

Supporters of the current system say the information used for research is not attached to any person’s name. But that may be a weak protection. “The state screening programs … de-identify newborn baby blood spots before loaning them out to research, but so far truly ‘anonymizing’ DNA has proved impossible,” according to a July Newsweek report that inspired the Gatto bill. Massachusetts researchers found the identities of anonymous DNA donors “by cross-referencing their data with publicly available information.”

AB 170 is a work in progress, but the parameters are simple: “The bill would authorize a parent or guardian of a minor child and the newborn child, once he or she is legally an adult, to request that the department destroy, not use for research purposes, or both, the blood sample, and the department would be required to do so.” It, in essence, gives the ownership right to its rightful owners. And it requires the state to provide various disclosures.

  • Reddit

Fortunately, this bill doesn’t take the Legislature’s usual “if in doubt, ban it” approach. The Newsweek article begins with a touching interview with a 36-year-old filmmaker who was identified as an infant with a rare genetic disorder. Had he not been tested and then treated, he would have faced a horrifically debilitating disease. Instead, he’s living a normal life. Gatto makes it clear that such testing “does a lot of good.”

But the program might need some constraints — and it certainly is worth a debate. Gatto is chairman of the new Assembly Committee on Privacy and Consumer Protection, which promises to be one of the most significant committees in the Assembly given myriad and growing threats to our privacy and civil liberties.

Maybe the committee will help rebuild a long-needed and bipartisan “civil liberties coalition.” After all, it’s in everyone’s interest to assure movies such as “Gattaca” remain nothing more than entertaining fiction.

Steven Greenhut-UT San Diego

Privacy is Dead, Invasive Technology is Here to Stay

Imagine a world where mosquito-sized robots fly around stealing samples of your DNA. Or where a department store knows from your buying habits that you’re pregnant even before your family does.

That is the terrifying dystopian world portrayed by a group of Harvard professors at the World Economic Forum in Davos on Thursday, where the assembled elite heard that the notion of individual privacy is effectively dead.

“Welcome to today. We’re already in that world,” said Margo Seltzer, a professor in computer science at Harvard University.

“Privacy as we knew it in the past is no longer feasible… How we conventionally think of privacy is dead,” she added.

Another Harvard researcher into genetics said it was “inevitable” that one’s personal genetic information would enter more and more into the public sphere.

Sophia Roosth said intelligence agents were already asked to collect genetic information on foreign leaders to determine things like susceptibility to disease and life expectancy.

“We are at the dawn of the age of genetic McCarthyism,” she said, referring to witch-hunts against Communists in 1950s America.

What’s more, Seltzer imagined a world in which tiny robot drones flew around, the size of mosquitoes, extracting a sample of your DNA for analysis by, say, the government or an insurance firm.

Invasions of privacy are “going to become more pervasive,” she predicted.

“It’s not whether this is going to happen, it’s already happening… We live in a surveillance state today.”

Political scientist Joseph Nye tackled the controversial subject of encrypted communications and the idea of regulating to ensure governments can always see even encrypted messages in the interests of national security.

“Governments are talking about putting in back doors for communication so that terrorists can’t communicate without being spied on. The problem is that if governments can do that, so can the bad guys,” Nye told the forum.

“Are you more worried about big brother or your nasty little cousin?”

However, despite the pessimistic Orwellian vision, the academics were at pains to stress that the positive aspects of technology still far outweigh the restrictions on privacy they entail.

In the same way we can send tiny drones to spy on people, we can send the same machine into an Ebola ward to “zap the germs,” Seltzer said.

“The technology is there, it is up to us how to use it,” she added.

“By and large, tech has done more good than harm,” she said, pointing to “tremendous” advances in healthcare in some rural areas of the developing world that have been made possible by technology.

And at a separate session on artificial intelligence, panellists appeared to accept the limit on privacy as part of modern life.

Rodney Brooks, chairman of Rethink Robotics, an American tech firm, took the example of Google Maps guessing — usually correctly — where you want to go.

“At first, I found that spooky and kind of scary. Then I realised, actually, it’s kind of useful,” he told the forum.

Anthony Goldbloom, a young tech entrepreneur, told the same panel that what he termed the “Google generation” placed far less weight on their privacy than previous generations.

“I trade my privacy for the convenience. Privacy is not something that worries me,” he said.

“Anyway, people often behave better when they have the sense that their actions are being watched.”

The World Economic Forum in the swanky Swiss ski resort of Davos brings together some 2,500 of the global business and political elite for a meeting that ends Saturday.

 Agence France-Presse

UK moratorium on use of genetic tests by insurers extended

The UK Government and Association of British Insurers (ABI) have agreed to extend their voluntary moratorium on the use of predictive genetic test results by insurers by another two years, to 2019.

Originally established in 2001 and periodically reviewed and renewed ever since, the Concordat and Moratorium on Genetics and Insurance is a voluntary agreement intended to protect the interests of both individuals (by ensuring appropriate access to insurance) and of insurers (by ensuring access to information about risks).

Once again, there has been no significant change to the substance of the agreement; individuals must only disclose genetic test results to insurers in the case of Huntington’s disease tests and for life insurance policies worth in excess of £500,000. It is worth noting that anyone undertaking a genetic test for Huntington’s disease almost certainly does so in the face of a family history of the condition, which must be disclosed to insurers anyway.

However, the concordat contains two new elements. An additional Annex 2 explains the important distinction between predictive and diagnostic genetic tests. Predictive or presymptomatic tests are taken in the absence of clinical signs of disease – for example, to assess the risk of developing specific cancers. Even though some of these tests may reveal a very high risk of cancer (for example, the lifetime risk of breast cancer in BRCA1/2 mutation carriers is up to 85%), UK insurers have agreed that such test results need not be disclosed when applying for health or life insurance. Diagnostic tests confirm or rule out the presence of a genetic condition that is known or suspected on the basis of clinical symptoms or non-genetic test results.

The concordat also clarifies that the results of whole-genome sequencing performe in the context of ‘major research projects, like the 100,000 Genomes Project run by Genomics England’ need not be disclosed to insurers, because they are part of a research project.

In the US, where there is a largely insurance-based health system (often linked with employment), the Genetic Information Nondiscrimination Act (GINA) prohibits the use of genetic test results by insurers or employers. This is intended to ensure that people with high genetic risks of disease are not condemned to pay heavy insurance premiums or denied health insurance – although as in the UK, family history of disease is always likely to have a strong impact, such that people who receive negative genetic test results indicating a normal disease risk may actively wish to share this information with insurance providers.

The purpose of the UK moratorium is similarly to ensure that individuals are not debarred from health or life insurance as a result of genetic testing. However, it is a voluntary agreement as opposed to a law; the UK has a taxpayer-funded National Health Service (NHS), such that only a small proportion of people have additional, private health insurance. The Medical Director of NHS England, Professor Sir Bruce Keogh, recently opined that the advent of genomic medicine made the free at the point of care NHS “more important than ever”.

The Concordat and Moratorium will be reviewed again in 2016.

Dr Philippa Brice, PHG Foundation

New DNA technique may reveal face of killer in unsolved double-murder

There were no witnesses to the gruesome murder of a South Carolina mother and her 3-year-old daughter inside a busy apartment complex four years ago. But a new technology that can create an image of someone using DNA samples left at crime scenes might bring police closer to catching the killer.

Reston, Va.-based Parabon Nanolabs, with funding from the Department of Defense, has debuted a breakthrough type of analysis called DNA phenotyping which the company says can predict a person’s physical appearance from the tiniest DNA samples, like a speck of blood or strand of hair.

The DNA phenotyping service, commercially known as “Snapshot,” could put a face on millions of unsolved cases, including international ones, and generate investigative leads when the trail has gone cold.

“This is particularly useful when there are no witnesses, no hits in the DNA database and nothing to go on,” Dr. Ellen McRae Greytak, Parabon’s director of bioinformatics, told

“Traditional forensic analysis treats DNA as a fingerprint, whereas Snapshot treats it as a blueprint — a genetic description of a person from which physical appearance can be inferred,” Greytak said.

Parabon’s new technology reads the parts of the human genome that code for the differences in physical appearance between people. Snapshot is able to predict such critical traits as skin color, hair color, eye color and face shape. It can also predict the individual’s ancestry as well as highly-detailed traits, like freckles.

Using sophisticated computer algorithms that have been trained on thousands of reference samples, Snapshot translates this raw genetic code into predictions of physical traits. These are then combined to create a composite profile, or “digital mugshot” of an unknown suspect — with remarkable accuracy, according to the company.

“Traits are generally predicted with more than 80 percent confidence, and importantly, Snapshot also reports which phenotypes can be excluded with more than 95 percent confidence,” said Greytak.

While developing the technology, the company “made thousands of predictions on people who we knew” to ensure accuracy, she said.

The investigator or crime lab sends evidence or extracted DNA to a Snapshot partner lab, where the DNA is run on a genotyping machine to produce the genetic information, according to the company. This genetic information is then securely transferred to Parabon, where an analyst runs it through Snapshot’s predictive models to produce a prediction. A report of the results is then delivered to the agency that requested it.

The new DNA analysis, however, is not able to predict age and height — traits Greytak described as “very complex.”

For investigators in Columbia, S.C., a digital mugshot created by the company might provide a break in a 4-year-old double homicide that remains unsolved.

The bodies of 25-year-old Candra Alston and her 3-year-old daughter Malaysia Boykin were found inside their home at the Brook Pines Apartments in Columbia on Jan. 9, 2011. There were no signs of forced entry, leading police to believe Alston and her daughter knew the killer or killers.

Police have not disclosed how the mother and child died, only saying Alston and her daughter were killed by different means. A laptop, Gucci purse and trash can were stolen from the apartment, as well as gifts the 3-year-old had received for Christmas.

The one clue detectives had was an unspecified DNA sample left at the crime scene.

Throughout the course of the investigation, police interviewed close to 200 people, 150 of whom submitted their DNA to authorities. But none of the samples proved to be a match, according to Mark Vinson, a cold case investigator with the Columbia, S.C., Police Department.

The police department then turned to Parabon’s DNA phenotyping to create a facial image based on DNA left at the crime scene. Vinson said the computer-generated photo is a “person of interest” in the murders.

“This is DNA from just one person and it’s possible more than one person was involved,” Vinson told The person of interest is dark-skinned with brown hair and brown eyes. The picture does not indicate an exact age, so the individual could be older than he appears.

“We suspect the child also knew the person who did this, which might explain why she was killed,” Vinson noted.

“We’re very hopeful this composite could be the thing that prompts someone to come forward,” he said.

While several agencies are now using Snapshot to help solve cold cases, including international ones, the Columbia, S.C. Police Department is first in the nation to publicly release a digital image generated by the new DNA analysis.

Cristina Corbin, Fox News

How drug companies will mine your genes

Imagine a world where genetic sequencing is free, like Gmail. That’s where we’re headed. Genetic data is going the way the rest of our data has gone on the web. Companies will mine it, repackage it, and find a way to make money off it.

For eight years, personal genomics company 23andMe has been giving consumers access to their genes. It started off costing $999. Today it’s $99. In 2012, the company allowed consumers to share their data with third-party apps, the way you might link your Facebook profile to your Hulu account.

And this week, what could be seen as another step toward the full webification of our DNA went down, thanks to a new deal between 23andMe and Genentech, one of its early investors.

The Deal

To start, Genentech will pay to get access to the genetic and health information 23andMe has amassed from more than 10,000 of its customers with Parkinson’s disease, a disorder that affects movement. Genentech scientists will be able to see what medications they’re on, what other conditions they have, the symptoms they experience, along with whether they have tweaks in their DNA that put them at risk for developing the disease. (All these customers previously consented to be part of research by 23andMe and its partners.)

But the deal goes further. 23andMe will help Genentech identify approximately 3,000 customers who want to participate in more nuanced research.

23andMe only analyzes snippets of your DNA. Genentech needs the whole genome. So, it’s going to pay to sequence them, if they agree. Whether or not the patients will get that information back is still being worked out, 23andMe’s Emily Drabant Conley, who brokered the Genentech partnership, told me.

“We’re now starting to do [studies] at this really unprecedented scale,” she added. “Our database has hit a critical mass. These people can be recontacted. They’re engaged. The other thing that’s happening is pharma is seeing the value of bringing in genetics into the R&D pipeline earlier.”

You can see how this might evolve.

“With 23andMe,” wrote Lisa Miller in New York magazine last April. “[CEO Anne Wojcicki] wants to do with DNA what Google did for data—because, after all, DNA is data.” And what Google — and other web giants have done — is to create powerful platforms through which other companies can sell us stuff.

This vision of Googlized genetics thrives only in a world with minimal hurdles to getting data. It might take a decade or more, but eventually the cost will get low enough to allow companies to offer full genome sequencing for free, at massive scales.

“I would love that. It would be great…We want to have a product that’s accessible,” said 23andMe’s Drabant Conley. “I don’t foresee that happening in the near future. There are real costs involved.”

But when cost does hit that critical minimum, the true democratization of genetics will unfurl. It will just be something very different from what we were sold on.

The Promise

Eight years ago, when 23andMe first launched, it wooed technophiles with the idea that genetic information shouldn’t be locked away in a lab. It belonged to you. We had the right to spit in a tube and find out what diseases we might be at risk for, bureaucracy be damned.

Geneticist Misha Angrist, a senior fellow in Science & Policy at Duke University’s Social Science Research Institute, was one of the people who bought into that vision. He sent his DNA off to be analyzed by 23andMe and other companies that offered direct-to-consumer (DTC) genetic tests. This was the future.

Until it wasn’t.

Reports started coming out that genetic tests offered by different companies served up different interpretations of people’s genes. Sometimes, they were misleading. Genome wide association studies — the science upon which DTC genetics companies were built — started coming under fire. How much could you really tell from looking at individual mutations without analyzing the whole genome? Regulators started to notice. Then last year, the U.S. government slapped 23andMe with a cease-and-desist letter ordering it to stop marketing its spit-box genetic test as a health report. That reduced 23andMe, if only temporarily, to a glorified ancestry service since. The over-the-counter genetic testing industry, as we imagined it, seemed in jeopardy.

And then, this Genentech deal.

“I kind of think of this as a sad day. I think it’s sort of the final confirmation that direct-to-consumer genetics is not a viable business,” said Angrist. “I think this day was probably inevitable. It’s hard for me to feel betrayed. And you now if my genotype contributes to a blockbuster drug…and it improves people’s lives, I can’t be an asshole and say, ‘Boy that’s terrible!’, but given the rhetoric and the marketing of DTC genetics, which again, I admittedly bought into and wanted to buy into — I wanted it to be true — it’s hard for me to stand on the sidelines and say, ‘Yay! Rah-rah! Go Genentech.’”

That rhetoric harkened back a bit to the days of the early internet, when we learned to equate the sound of beeping modems to a superhighway of information that would empower us to be better. The information was seemingly accessible to all, for a fraction of the price it might have cost to access in the real world. That democratization morphed into something beyond our expectations, too.

“Just as it became clear that people didn’t want to pay money to AOL or Prodigy for email, Google found a way to give email away and derive ad revenue from somewhere else,” said Angrist. “It’s all the same business model,” a model in which you are the product. This internet-age adage is true, especially for DNA.

The new Genentech deal is only the beginning. Reset Therapeutics put out a press release saying it was going to use 23andMe’s massive database to study diseases related to circadian rhythms, the internal clocks that dictate when we sleep and eat. (Whether Reset will also be sponsoring its own genetic sequencing wasn’t clear.) And there are eight other deals coming, the details of which will trickle out in coming weeks. All the company would say for now is that they span a wide range of diseases and conditions.

As disappointing as the news may be to some, the strategy isn’t all bad. It could really open up new areas of research, in ways that current methods simply can’t.

“This is very exciting.  An integrated, enthusiastic cohort of 800,000 is far beyond academic capacity,” said George Church, one of the leaders of the Human Genome Project, in an email. (He’s also an advisor to 23andMe and was on the advisory board of a company recently acquired by Roche, Genentech’s parent company.) “I expect that the 23andMe cohort will continue to grow and be increasingly targeted to diseases of interest to pharma.”

Say Genentech wants to screen compounds for a new heart-disease drug they suspect might only work on a subset of patients with a certain genetic makeup. They can query 23andMe’s database, come up with potential candidates, and recontact them to sequence them. What’s more using new technologies like organoid chips, they can take skin cells from these patients, reprogram them, and build up tiny tissue-specific systems on which they can test their compounds. The results could give them a good indication of which ones might be toxic or work in individual patients. They could run multiple experiments at once, potentially speeding R&D up significantly.

For pharma companies, it’s a no brainer, especially with 23andMe customers at the ready, willing to supply more data in the hopes of finding a cure for the condition that ails them or their relatives.

As 23andMe co-founder Linda Avey put it on Twitter:

That economic value, though, will accrue to 23andMe.

The company’s terms of service clearly state that “by providing any sample…you acquire no rights in any research or commercial products that may be developed by 23and­Me or its collaborating partners.”

Daniella Hernandez, FUSION

23andMe turns spit into DNA data sales to Pfizer

23andMe Inc., the genetic-testing startup backed by Google, is sharing DNA data on about 650,000 individuals with Pfizer to help find new targets to treat disease and to design clinical trials.

The collaboration with Pfizer is the broadest announced so far in 23andMe’s ambitious plan to become a repository for humanity’s genetic makeup, and to turn data gathered from $99 saliva tests sold to consumers into multimillion-dollar deals with drugmakers.

The agreement unveiled Monday gives the U.S.’s largest drugmaker access to anonymous, aggregated information from consumers who bought 23andMe’s test over the past seven years to learn about their own genetic histories. It includes only people who agreed to let their data be used in research. Pfizer and 23andMe declined to give the deal’s value.

The Silicon Valley startup, named for the 23 pairs of chromosomes in human cells, is betting its growing troves of genetic data will prove essential to drug companies, medical researchers and even health and wellness companies.

Even as it seeks to expand its consumer tests around the world, the company is repairing relations with the Food and Drug Administration. An agency ruling in late 2013 left 23andMe unable to sell health analyses from the saliva tests.

While about two-thirds of 23andMe’s 800,000 customers agreed to let their test data be used in research, data-sharing agreements with drugmakers are likely to raise the hackles of privacy advocates who have questioned the wisdom of compiling highly personal information.

The deal gives Pfizer access to a broad cross-section of data, the first agreement in which a drugmaker has access 23andMe’s newly created research portal. 23andMe Chief Executive Officer Anne Wojcicki plans to pitch the service to other health companies this week at the JPMorgan Healthcare Conference in San Francisco. The company plans to announce a total of 10 similar deals with drugmakers and biotechnology companies this year.

23andMe, based in Mountain View, California, is near the Bay Area’s biggest technology companies, including Google, co- founded by Wojcicki’s husband Sergey Brin. The two are separated, though still legally married.

Google’s venture-capital arm and investors such as Russian billionaire Yuri Milner, Johnson & Johnson and venture-capital firm New Enterprise Associates have contributed $126 million in funding to date, according to 23andMe.

The company isn’t yet profitable, and it’s too early to consider an initial public offering, Wojcicki said.

While Pfizer has already worked with 23andMe to enroll 10,000 patients for irritable bowel disease research, the deal broadens their collaboration. Pfizer and 23andMe will also enroll 5,000 patients to conduct a study on the genetics of autoimmune disease lupus.

The company has also signed agreements on specific diseases, including one with Roche Holding’s Genentech unit announced last week, to study Parkinson’s disease patients.

Genentech will use the data to find the connection between patients’ symptoms and other personal traits, and their genetics,

said Alex Schuth, head of technology innovation and diagnostics in business development at Genentech, in a telephone interview.

Genentech is paying 23andMe $10 million upfront and as much as $50 million if the deal hits certain milestones.

The appeal to Pfizer and Genentech isn’t just the size of 23andMe’s data set – it’s the additional information that the company collects on users’ personal lives.

Every time consumers who bought the kit return to 23andMe’s website, they are prompted to answer more questions from an “infinite question box,” which quizzes them on everything from hair color to bra size, said Patrick Chung, a 23andMe board member and a partner in the Cambridge, Massachusetts-based venture capital firm Xfund. That additional information can help researchers make more connections about people’s characteristics and their health.

23andMe plans to debut in other countries this year, said President Andy Page in a telephone interview, though he declined to name the locations. The consumer spit kits, which

include reports on known risk factors for Alzheimer’s, Parkinson’s and cystic fibrosis, started sales in Canada and the U.K. last year.

The company has also given its kits away to get a more diverse set of genes for its database, including giving away 10,000 kits to black Americans one year, Wojcicki said. “You’ll see us in the future sponsoring those kinds of programs,” she said.

Caroline Chen Bloomberg News

23andMe’s New Formula: Patient Consent = $

Facebook generates about $8 a year in revenue from each of its users. But what if you offered a company not just your photos and updates, but your entire genome?

Then you could be worth as much as $20,000.

That’s my rough calculation for what Genentech could pay direct-to-consumer gene testing company 23andMe for the chance to trawl the DNA of each of several thousand of its customers for genetic clues to Parkinson’s disease.

The deal between the two companies, announced today, provides some fascinating insights into the evolving DNA business and the commercial prospects for 23andMe, a high-flying company that’s had some problems in the U.S. with regulators. According to detailed coverage over at Forbes, Genentech will pay as much as $60 million for access to 3,000 Parkinson’s patients in 23andMe’s database.

The backstory is that 23andMe pioneered direct-to-consumer genetic tests starting in 2006. It asked consumers to spit in a tube and send it in, and sent back a detailed summary of their risks for common diseases like macular degeneration. But then in 2013 the U.S. Food & Drug Administration banned the test out of concern that the information wasn’t accurate.

That put a big crimp in 23andMe’s business, but it didn’t end it. As Forbes points out, the real business here is mining this data:

Such deals, which make use of the database created by customers who have bought 23andMe’s DNA test kits and donated their genetic and health data for research, could be a far more significant opportunity than 23andMe’s primary business of selling the DNA kits to consumers. Since it was founded in 2006, 23andMe has collected data from 800,000 customers and it sells its tests for $99 each. That means this single deal with one large drug company could generate almost as much revenue as doubling 23andMe’s customer base.

The company hasn’t stopped gathering DNA data either. It still sells its Personal Genome Service health kits in countries like Canada. In the U.S. it continues to offer a more limited genealogy test to people who want to learn what their DNA says about their ancestry and relatives.

The result is that 23andMe may have the largest DNA database anywhere that’s open for medical studies. Of its 820,000 customers, the company says, about 600,000 have also agreed to donate their DNA data for research purposes. According to Forbes:

“I think that this illustrates how pharma companies are interested in the fact that we have a massive amount of information,” says Anne Wojcicki, 23andMe’s chief executive and co-founder. “We have a very engaged consumer population, and these people want to participate in research.”

It’s also a reminder that 23andMe’s real business isn’t selling $99 tests, but selling access to data that it has managed to crowdsource as cleverly as Facebook has gathered other personal details. To some observers, that’s pretty worrisome. In 2013, journalist Charles Seife, writing in Scientific American, called 23andMe intentions “terrifying.”

As the FDA frets about the accuracy of 23andMe’s tests, it is missing their true function, and consequently the agency has no clue about the real dangers they pose. The Personal Genome Service isn’t primarily intended to be a medical device. It is a mechanism meant to be a front end for a massive information-gathering operation against an unwitting public.

Seife’s worry is that the consents customers agree to when they donate their DNA could turn out to be meaningless. Once you are hooked, companies like Google and Facebook often change their privacy policies to expose more and more of your data. Why should DNA be any different?

So far, 23andMe seems more sincere than sinister. Parkinson’s disease is personally important to Wojcicki and her husband, Google founder Sergey Brin: his mother was diagnosed with the disease, which runs in families. Todd Sherer, the head of the Michael J. Fox Foundation for Parksinon’s Research, told me that the couple (now separated) has been the organization’s biggest donor, giving more than $150 million.

But Seife is right about the economics of DNA. It’s collecting free-and-clear data and amassing willing users that counts. According to the Fox Foundation, 23andMe actually gave its testing service away to Parkinson’s patients. That helped it assemble enough of them to create a useful resource it could sell to Genentech to start mining.

As part of its research, Genentech will gain access to the stored spit samples of 3,000 Parkinson’s patients in order to access their full genomic information. That is something that is allowed by the agreements customers signed. But to make sense of the DNA data, Genenetech will also need a lot of extra information about people’s health situations and medical records.

In this case, Forbes reports, the company will be reaching out to them to craft new agreements to access that, too.

Antonio Regalado, MIT Technology Review