Sage Bionetworks, a fledgling Seattle nonprofit attempting to launch an open-source biology movement, has nabbed a $6.7 million grant from the National Cancer Institute to train young scientists to learn to better use genomic data to help improve drug discovery and patient care.
The grant lasts for four years, and it may be supplemented and extended, says Sage Bionetworks founder and president Stephen Friend. Here’s how it will work: Each year, the plan is to identify two young postdoctoral physicists who have deep understanding of mathematical models and algorithms, and to pair them with two clinician/scientists at the same stage of their careers, but with deep knowledge of biology, Friend says. The vision is that these pairs will cross-fertilize each others’ thinking, so that together, they can use math and computing tools to build predictive models that can show the connections between genomic alterations, the faulty proteins that arise from them, and clinical symptoms of diseases that are the end result.
By finding the connections between underlying genotype and the phenotype, so-called network biologists will be able to create the first predictive models of disease—sort of like how engineers build predictive models of an airplane on a computer before it flies. This new kind of network approach could reduce the huge waste of time and money on drugs that fail in development, by weeding out duds early, and someday helping guide personalized treatment decisions for doctors.
But if this vision is going to become reality, it will require a culture change in biology, making it more open and less proprietary. The new training program is one key step toward making that happen.
“There will be a national competition to identify the winners of these spots in the training program. I can see this becoming like winning a MacArthur fellowship,” Friend says.
Sage made its public debut about a year ago when Friend announced he was leaving his high-profile job in charge of cancer research at Merck. Even at a company with as much money and brainpower as Merck, Friend said the problems of biology are too big for any company, and that his next quest would be to apply crowdsourcing or collective brainpower to the challenge. The organization secured $5 million in charitable commitments, and was seeded with data from Merck that represented about $150 million of R&D. Sage has since secured more support from Quintiles, the giant contract research organization, and Pfizer.
There are really three main themes Sage is pursuing in its early days, Friend says. It starts with building up its research capability through support from foundations and companies. Then comes the training component that’s being supported by the National Cancer Institute. The third theme is building up the Sage platform, in which it creates the public repository for data with common standards and annotations, proper governance, and a user interface that hopefully makes this whole thing user-friendly so a large number of scientists will actually use it. Sage hopes to make progress on that third theme at its first international congress on April 23 and 24 in San Francisco.
But as for today’s news, it’s another significant step for Sage as an organization. The nonprofit now has about 20 people, counting staff and postdocs, and the grant will allow it to grow to about 25 after it adds some more mentors for its aspiring network biologists, Friend says. Sage will now collaborate with the Fred Hutchinson Cancer Research Center in Seattle (where it is physically housed), as well as Johns Hopkins University, the Dana-Farber Cancer Institute, the University of Hong Kong, and the Netherlands Cancer Institute.
In the beginning, the research projects will focus on breast, colon, liver, and pancreatic cancer models. Researchers at Sage will build computational models of disease, and then run wet-lab experiments to see how accurate they are, and to refine their methods. The goal, over time, will be to extend this sort of training to many other centers beyond Sage, Friend says.
“We want to set up a paradigm for training. If after four years, this is the only place doing this training, it will be too bad,” Friend says.
