all the information, none of the junk | biotech • healthcare • life sciences

Open Source Biology Deserves a Shot

Xconomy National — 

Gene sequencing has gotten incredibly fast and cheap, and researchers around the world are pouring huge volumes of genomic data onto their private servers, in the hope they will sift through it all to make groundbreaking discoveries. Should so much genomic data be so closely guarded, or should it be poured into a free and open database that all scientists share?

The idea sounds utopian in a high-risk, high-reward industry that protects intellectual property like Fort Knox. But no one disagrees that today’s approach to drug development takes too long, costs too much, and is too unpredictable.

Stephen Friend thinks shared data would change all of that—and allow researchers to see patterns they wouldn’t otherwise see, and make insights that would never emerge any other way. So he did something two years ago that most people would consider quixotic: He quit his high-powered job as a senior vice president of cancer research at Merck to go on a mission to disrupt biology.

The founding idea, at a nonprofit called Sage Bionetworks, was to spark an online movement like the one we’ve seen with open-source software or Wikipedia, in which thousands of loosely affiliated people around the world pool their brainpower to do something great. In this case, the wisdom of the crowd could improve drug development and personalize medicine.

The Sage Commons is built on the notion that the genomic symphony is too bewildering for any individual or team—even at a place with as many dollars and brainiacs as Merck—to figure out. That seems pretty obvious.

But two years into this endeavor, Friend has learned how many people still resist any change to business as usual. Academic institutions still cling to their intellectual property out of a hope it will someday make them money. Scientists, by and large, keep their experimental data close to the vest in the hope they can get career-making papers published in Science or Nature. Many drugmakers can’t really imagine sharing anything valuable outside their corporate firewalls, lest it undermine their competitive standing. So while Facebook and Twitter may have proven that humans have a deep-seated desire for sharing, this impulse is still widely suppressed in biomedicine.

Stephen Friend

When I visited Friend at his office in Seattle on Friday morning, he was brimming with his usual enthusiasm, but was also sober about how hard it is to get people to change their ways, and to get them to share for the greater good. When I asked him what the biggest obstacles are to this mission, he didn’t hesitate: “Apathy, and ignorance.”

But if anybody can spark a new way of thinking, Friend is the guy. Once described by Forbes as “one of the last great dreamers” in the pharmaceutical business, Friend has a rare blend of scientific chops, entrepreneurial success, Fortune 500 leadership experience, high-level connections in biology and IT, and the infectious energy and charisma that makes smart people want to climb mountains for him.

Which is why Sage has scooped up more than $20 million in support so far. It has built a team of 30 employees. Companies like Merck (NYSE: MRK), Pfizer (NYSE: PFE), AstraZeneca (NYSE: AZN), and Quintiles have contributed some combination of money and data to the cause. Amazon Web Services, the cloud-computing unit of the e-commerce giant, is hosting the massive amounts of genomic data that Sage wants to put in its public repository.

Some world-class scientists at Stanford University, Columbia University, UC San Francisco, and UC San Diego have agreed to pool their experimental genomic data. The FDA has become curious about how an open data repository could be used to track adverse events with drugs. Scientists from other disciplines, like physics and astronomy, have been vocal advocates for … Next Page »

Single PageCurrently on Page: 1 2

By posting a comment, you agree to our terms and conditions.

  • Dylan

    Great article, the potential that exists in “open source biology” is hard to even comprehend. Databases linking SNPs to disease to treatments to outcome…all of which can then be used for future drug development. As time goes on the volume and usefulness of the data will expand exponentially. If scientists and institutions don’t want to open up their databases straight away, perhaps some sort of common ground can be found. Let us hope!

  • There’s R&D, there’s r&D, and there’s R&d —

    In his book “Science Business”, Harvard Business School professor Gary Pisano makes a great argument for far more collaboration between biotechnology researchers and businesses such as being attempted by Sage.

    In the book, Pisano argues that the biotechnology industry has been organized on the “one patent, one company” high-tech model in which one organization tries to go it alone to create the blockbuster drug.

    The problem is that with an electronic widget the challenge is primarily development not research. The basic science is usually known. The specs–the required input and output–for the device are known. Little research is needed; the work is development. Pisano refers to this as little “r”, big “D” — as in r&D.

    But in biotechnology there are just to many possible complications, too many unknowns, so the task of product development remains research heavy: “R&d”.

    Furthermore with a widget you know relatively quickly if it’s going to work. If things don’t look promising, you can quit while you’re ahead. A promising biotech product, on the other hand, can fail in a phase III clinical trials after decades of work.

    Pisano concludes that what biotechnology companies, investors and academic researchers need to do is find a way to breakdown the silos so they can share the costs and risks of the research phase of biotech R&D.

    It’s a good read: recommended.

  • Dylan—thanks for the comment. One thing Stephen mentioned which I didn’t include in the article is how he foresees the Sage Commons being used to create “dynamic drug labels.” The idea is to put all the data in a public commons on how a patient responds (or doesn’t), and the kind of adverse events he or she experiences. It would create a much deeper body of knowledge about a new pharmaceutical that could be continuously updated in real-time, instead of waiting years for a pattern to emerge like the ones that led to the downfall of Vioxx and Avandia.

    Michael–good comment as well on Pisano’s book. Few in biotech have taken his point to heart. Collaborations between pharma/biotech/academia are still happening the old way—they take months, if not years, to set up and get thoroughly lawyered on things like IP ownership. The model is nowhere near what most people would consider an efficient way of sharing cost and risk. Given how poor the drug development model works today, I think the industry owes it to itself to seriously consider what Sage and others are doing to change things.

  • I have always supported the idea of an open network for sharing all the vast amounts of biologic data that has been collected since the early days of microarrays and genomic sequencing. It is frustrating that there have been so many groups that have developed proprietary platforms that have high barriers to entry (primarily cost) or are closed networks (much of big pharma) and I think Stephen is going in the right direction by trying to open this all up.

    The value for users is to increase the “connectivity” of the data so that new therapeutic approaches can be developed and the costs can be greatly lowered for people and organizations to find new methods to treat disease. You can still derive a proprietary approach to drug development from this sort of data so the incentives will still remain for investors and no one can complain that it will devalue IP because the sheer volume of information will be so enormous that there will be plenty to go around. Just the ability to take all the disparate data out there and present it in a digestible and useful fashion would lead to explosive growth in drug development in my opinion.

  • This was covered as well in Wikinomics by Don Tapscott. He refers to this as Precompetitive Knowledge Commons and specifically cites the case for this approach to drug development.

  • – t swell –

    Nice to have the financial links in the story. Why is there no clear link to Sage?