working on all the behind-the-scenes work needed for such an effort. Standard operating procedures needed to be developed for bringing in samples, and slicing them up for staining in the lab. Informatics systems needed to built specially for the task, robots needed to be set up to automate as much of the process as possible, and user-friendly interfaces had to be developed to help scientists make use of the resource.
There were all-night marathon work sessions heading into December 2004. Jones recalls one terrifying moment, in which a hard disk failed, and another when the software development team came down with the flu.
“We knew we had a good plan in place, and knew we had risk points,” Jones says. “You always have your doubts, you have your dark days. But the team here was quite motivated to show we could do it.”
The team met its goal. That was an important milestone, and gave the team confidence it could scale up what was started, and keep hitting its deadlines.
Tessier-Lavigne recalls the scientific advisory board having its doubts about whether the team could truly deliver on the full mouse brain by its deadline in 2006, but Jones walked through the timeline step-by-step, insisting that the team could still meet that goal. They did. “The first 10 percent was the hardest. Then you’re scaling on your processes,” Jones says. The full adult mouse brain, the developing mouse brain, and the mouse spinal cord were built on similar templates, as the team honed its process.
The next big challenge was the human brain, which required most every piece of technology to be redeveloped, Jones says. While the mouse is a good model system because of its genetic similarity to human brains, a hemisphere of a human brain is far too big to fit on a standard 1-inch by 3-inch microscope slide. The Allen Institute needed to switch to bigger 6×8-inch slides and find some way to scan slides that big—which are too big for off-the-shelf scanners, Jones says.
Plus, since good human brain tissue is hard to find, the institute would have to be very judicious with every sample it got, getting the most information possible. That meant it couldn’t use the original gene-by-gene mapping technique from the mouse brain—known as in situ hybridization—because then it would probably only be able to show functional mapping of a couple hundred genes, not enough to be useful for scientists, Jones says.
So the decision was made to use another platform, a microarray technology from Agilent Technologies, in which the brain tissue was analyzed to see which genes in the full human brain were active and which weren’t. (For a more in-depth look at how the Allen Institute creates these brain maps, watch Jones give this TED talk).
Over the years, Jones and Allen have developed a close working relationship, as the billionaire personally remains engaged in the institute’s work and attends all of the scientific advisory board meetings. The founder isn’t a neuroscientist, but he works hard to educate himself, speak the language, and keep the scientists on their toes. “That’s why it’s great to work with Paul Allen,” Jones says. “When you’re just getting going with momentum on a project, he’s already thinking about the next thing.”
Jones says he’s enjoyed the past couple years, seeing scientists at conferences who now thank him and his team for creating the resource. “The perception has gone from ‘who are these guys?’ to ‘we love your data.’ That’s really rewarding to get from the field,” Jones says.
Now the Allen Institute is extending its ambitions even further. In March, Jones said the institute hopes to complete its sixth human brain map by the end of this year. The new $300 million commitment will go toward building a variety of research tools that Allen said will help explore what makes us human.
As chief scientific officer Christof Koch put it, one of the efforts will be to “exhaustively identify, catalog, record, and intervene in the cognitive networks underlying visual perception, visual behavior, and visual consciousness in the mouse.” And that’s only part of what’s being built.
As the institute has grown, Jones has a new set of management challenges to deal with. It’s expanded into three buildings in Fremont, which makes it tougher to collaborate, and has prompted the search for new space. There are cultural issues too, as academic scientists often revel in exploring their own ideas and getting glory in publications, not through toiling on big projects that put data in the open. Jones talks about the need to find top-notch talents who are willing to be team players.
No one can say what discoveries will come from the various brain maps, but the founder has made clear that he wants to see medical progress, at least partly because his recently deceased mother had Alzheimer’s.
It’s also too early to say what kind of medical progress might eventually be traced back to the resource created by the Allen Institute. But one thing is for sure—the credit will probably go to someone else, not the low-key scientific manager and his team who built the resource. And Jones says he’s fine with that. When I asked what he does outside of work, he laughed, saying he’ll have time for hobbies when he’s older—for now all he has time for is work and family activities.
“I want to look back 10 years from now and say ‘Look at the great team we built, and look at what we were able to accomplish.’” Jones says. “It’s always been an interesting thing, having me as CEO. I’m not a big ‘sales-y’ person, a get-out-there kind of person. I’m not that person. That’s fine. It’s about achieving the mission.”
