At A Glance
Name: Roger Bumgarner
Title: Director of Center for Expression Arrays, University of Washington, Seattle
Faculty Positions: 2002-present: Assistant Professor, Dept of Microbiology, University of Washington, Seattle. 1998-2002: Research Assistant Professor, Dept of Microbiology, University of Washington, Seattle 1992-1995: Acting Assistant Professor, Dept of Molecular Biotechnology, University of Washington, Seattle.
Education: 1988: PhD, chemistry, University of Arizona; 1984: MS, chemistry, Eastern Illinois University; 1982: BS, chemistry, Eastern Illinois University.
SEATTLE As the founder and principal organizer of the Northwest Gene Expression Conference held annually at the University of Washington in Seattle since 2001, Roger Bumgarner has witnessed firsthand the rapid maturation of array technology in the academic and commercial settings.
At UW he serves as director of the Center for Expression Arrays, as well an associate professor of microbiology, and he admits that he often has more collaborative work than he can keep up with.
Following the 5th annual Northwest Gene Expression Conference held here last week, BioArray News interviewed Bumgarner to find out more about the evolution of the conference, his National Institutes of Health-funded research with pathway analysis company Teranode, and why there are so many biotech companies popping up in the Emerald City.
This year is the fifth conference. Have you been in charge of it the whole time?
And what prompted you to organize a Northwest Gene Expression Conference?
Well, it started out as just a way to learn from others about the technology and what they were doing. When we first started doing this five, almost six years ago, the meeting was set up so that the people would come up and talk about the technology, and they would say, "Here's what I'm going to do if we can ever make this stuff work." Now, we are seeing lots of great applications, and it's really rewarding to see how things have changed over time.
Has the content matter of the conference changed as well?
In the beginning there were so few people doing microarrays that anybody who had any kind of application or research wanted to get together with anybody else. So we had everything from cancer to infectious disease to other things at the time. Now, that the technology has moved on so much, it makes more sense to focus it more around microbiology than it did before. The past couple times that I've run it we've actually had a day of virology and a day of bacteriology. And then I always do one day of technology development and bioinformatics in the middle because I always have my selfish motivations to get the microbiologists educated further along.
Do you feel they need to be kept up to date on the latest technologies?
Yes, [and also because] the best and most interesting work that is being done, all the new developments that take place, are usually on the human biology and the cancer side, because, frankly, the cancer guys the National Cancer Institute have been much better at positioning the technology for research than [the National Institute of Allergy and Infectious Diseases]. So that if you want to see what's really going to happen next, you don't look at what's going on in bacteriology, you look at what's going on in cancer research. So I always try to bring in some people from these other areas to show new things.
[For example], the idea of using genetics and treating expression levels as quantitative traits, and using that to derive pathway information, is not something that the microbiologists are going to latch onto right away, but it's clearly the way to go. So I really like bringing in those kinds of speakers so that people who are doing bacteriology can say, "Maybe this is a new thing I should try."
What is your relationship with other universities in the area? To what level do you collaborate with the University of British Columbia, for example?
They have a good facility. I haven't got any collaborators up there. Kurt Gustin's [microbiology] group at [the University of Idaho in Moscow] has been a good collaborator on the rhinovirus work. But most of the work we do is really Seattle-centered. UW is the second-largest federally funded academic institution in the country, so there are plenty of people here to collaborate with.
Maybe you can give me a little overview what are all these different people working on? I know a number of them presented.
We have a really strong program at the University of Washington on cystic fibrosis and Pseudomonas. We have a lot of good virologists at UW, and good HIV programs, so there are a number of HIV researchers. We have a great program here in biodefense. [Our] biodefense center [is] run by Sam Miller. Some of the attendees are all heart, lung, and blood guys we've got a good program there, we've got a big National Institute of Diabetes and Digestive and Kidney Diseases group good diabetes researchers. It's all over the map, so for me, for the technology, it's difficult to keep up with everybody's biology that I am collaborating with.
Are you at the pinnacle of this pyramid of research, with your position?
Oh, I wouldn't call it the pinnacle. I'm one hub in a big network. There are many, many, many people. But we did set the array facility up years ago and made it available to anybody on campus who wants to use it.
How does that work with regards to cost to the researcher? Is that provided through general funding for academic research?
We run on a cost-recovery model. So essentially we have to charge researchers what it costs us to run arrays. But we've been lucky to get a large number of grants that have helped subsidize researchers. [For example], I just got a shared instrumentation grant to buy new Affymetrix equipment to update our stuff. I just got [another] shared equipment grant to buy a good computing facility. But the vast majority of stuff is really cost recovery.
What would you do for the programs and the center if you had unlimited funding?
If it was unlimited, I'd be spending most of my time doing research and really trying to develop networks and pathways. Probably we'd spend that time in bacterial systems where there's good genetics available to us, because it's a lot simpler than doing it in humans. I think it would be a better model. As far as what I would like to put together for other researchers to make use of, it would be great to go back and start making arrays, but do antibody arrays, which, frankly, I am kind of surprised hadn't developed to the level I had anticipated three or four years ago. I thought that we'd have larger collections of antibodies and antibody arrays commercially available by this point in time.
Why do you think they aren't available?
I think some of it is that it's a complicated problem to get good, specific antibodies for many, many genes. But I think also some of it is fear on the part of the industry. It is going to take a significant investment to do that properly. And, when they go out and talk to the customers, the customers say, "Well, I'm really only interested in kinases so just give me the kinase antibody." Somebody else is only interested in cancer-related genes so they only want antibodies for those. So they get this impression that what they should really be doing is making these custom subset arrays, and I think they waste a lot of effort and time on that when they could be making and moving harder towards whole-genome arrays for antibodies.
I think it's a combination of things. I think its bad market advice. I think that the customers who talk a lot sometimes aren't the ones who are the most forward looking. It's a number of factors. But somebody will make antibody or antibody-like arrays that cover a reasonable fraction of the genome. But since that hasn't happened, at least yet, I'd like to spend some time on that.
You also have an NIH grant to work with Teranode. Can you describe what that NIH grant is for?
They have got a Small Business Innovation Research Grant, and part of what we're doing with them is helping to define features and utilities that they should incorporate into their software. One of the things that we want to do is make it easier for people to map array data to the pathways and networks that they already have their hands on, and display that data appropriately. But I am really excited about their capability to do reaction models. There's a lot of pathway and network-displaying packages out there. There are a lot of competitors in that space. But there's really nobody that I've seen that's building complete kinetic models of pathways to the point where you can actually put reaction nodes in, put rate constants in, define the reaction types and the type of kinetics for each node, and then model the chemistry of all the nodes simultaneously. And that's something that you can do with the Teranode software.
What will that mean for the average researcher?
Well, in the long run if you work out at least little subnetworks, you'll get complete predictive ability of the concentrations of all the proteins, and all the substrates, and all the metabolites that you can model. Effectively, if you know the reactions you should be able to model all those things, directly. Of course if you have a good model you have predictive ability. I think for drug companies too, this kind of software is going to be great because once they work out all the networks and pathways that they are interested in, you can predict in advance which nodes to inhibit that will give you the biggest effect. You'll be able to see what happens downstream if you have good models.
Yesterday there was a presenter talking about rainbow trout (see Lab Report). There are also a number of companies operating out of the area in general. Is there anything coming out of the Pacific Northwest that you feel might be unique to the region?
The CombiMatrix technology is particularly cool essentially using electrochemistry to drive the construction of arrays there's certainly a lot of promise in that. Teranode's stuff we've already talked about. But there's about 150 or so biotech companies in the Seattle area. If you look at it on a per-company basis, Seattle is about the fifth or sixth largest biotech region in the country. A thing that's interesting that happened here is that we've had a new governor [Gov. Christine Gregoire] get put in not too long ago that really has a nice vision. She was the attorney general in Washington State who was the lead on the tobacco lawsuit. And as the lead, Washington got a little bit larger share of money than other states, and when she became governor, one of the things she pushed forward was a life science discovery fund. Where $350 million of this tobacco money is coming back to us in the form of research funding to help improve human health, which I think is a bit ironic almost. And that's going to be advanced from a number of other sources, and she's hoping that they can do about a billion dollars total in the next ten years, which is a pretty significant influx to the area. That's about $100 million total if we get the matching [funding], and about $35 million if we don't see the matching funds in the long run, per year. I think that's going to change the research landscape here.