At A Glance
- Professor, Plant Pathology, Director, Fungal Genomics Laboratory, North Carolina State University.
- Age 44, PhD in plant pathology from the University of Kentucky (1986). Thesis: plant immune mechanisms. Postdoc work in fungal developmental biology at the University of Georgia. Left Clemson University, where he was associate director of the Clemson genomics institute, in 1999 for N.C. State, a move clinched with a new $1 million laboratory. Since then, he has brought in $11 million in grants.
- Co-authored research paper that describes the sequencing of the rice genome, published in the April 5, 2002, issue of Science. Work provided the framework for the genome’s sequencing and assembly.
- In July, along with a team of scientists at the Whitehead Institute, released a draft sequence of the rice blast fungus (Magnaporthe grisea).
You’ve been very successful in raising research funding, going back to your time at Clemson in the early 1990s. How do you do it?
When I got to Clemson, I was given a free hand. As long as I could get the funds, I could do anything I wanted to do. And, I keep looking for funds.
But, I’ve been pretty focused too. I’ve been working on rice blast (Magnaporthe grisea) for 12 years or more now. I’ve operated by looking for partnerships. I look for people who can help me and I put teams together.
Genomics is a multifaceted interdisciplinary activity and no one person can do it all. I’ve strived to pull various communities together and then convince the funding agencies to give us several million dollars.
I bring together a team and show that is has strength in different areas and that we are strong enough, with leaders to coordinate activities and make sure things get done.
What is your research interest?
For my lab, it is the appressorium, the special infection structure that’s unique to plant pathogens. Not all plant pathogens make them; it’s a structure that basically attaches itself to the side of the plant and then — literally — the fungus drives itself, forces itself, jackhammers itself, into the plant underneath. It’s a very special mechanism that fungi have evolved to infect plants.
Now that we have a genome for rice blast and various genomes for rice, we have a unique opportunity to dissect the interaction, and to do that, we have to look at what genes are being expressed by the host and by the pathogen in this complicated interaction of the fungus for rice blast when it’s on the plant. One of the ways we can do that, suddenly, is through microarrays.
What kind of equipment do you use?
We are using an Affymetrix arrayer, a 417, and a HP scanner. We can make about 40 slides a day, putting down about 4,000 elements.
That’s the main equipment we have and we also have available to us an Affymetrix [GeneChip] system. We have made lots of pilot arrays and done some pilot experiments, looking at which genes are expressed during the formation of this structure, to understand what genes are involved, and how they become targets for various chemical interventions. We have been playing with the arrays for about a year.
What we have been waiting for is to get the whole genome. We just decoded the whole genome and are currently annotating it. We know there are about 12,000 genes. Now, we will have to step up from arrays containing 4,000 to 5,000 arrays to making whole genome arrays. We want to do it once only. It would be so much more efficient if we made one complete array set that was available for the research community to use. Clearly, an issue is standardization, the minimum information that is required [for a microarray experiment].
Is this standardization forming around the Affy platform?
I don’t know if it is or not. We are moving now towards oligo arrays and we have been looking around for a commercial vendor that can make them. The one way we have thought about looking at this is using ink jet technology to actually in situ create oligos. We need 12,000 individual elements at the minimum. The problem with an Affymetrix array is the cost in making the mask. Genome annotation is a moving target. So, if after a certain amount of experimentation, we discover that a bunch of additional genes that were missed, or a whole bunch on the array are incorrect, using the Affy platform doesn’t give the plasticity to make changes rapidly or affordably. There is concern on some people’s part that perhaps we need to look at a more flexible technology. Certainly, a technology that some companies use is ink jet technology. We have talked to Amersham and Agilent and we are in the process of determining if they can meet our needs. At some point, something like an Affy system would be of the greatest utility. The upfront expense, in some situations, is affordable and in others, it’s just not cost effective.
Do you have any patents?
I do have some patents that relate to genomic science but we don’t have anything in mind for microarrays. Right now, my goal is not in technology development; it’s to use technology to answer important questions. If I feel that the technology has reached a point where I’m better served by buying it than making it myself, then I need to look at the option. We are very close to that with microarrays.
I still think there is going to be a shakeout coming. In this, it’s like computers and big screen TVs, once it reaches a certain price, everyone will start buying them. Once computers got below $1000, everybody went out and bought them. It’s the same thing for microarrays. When the prices come down, we’ll see a lot more use of the microarrays and a lot more companies producing them.
How do you handle data analysis?
We look at it on several levels. There is the standard pipeline stuff, processing raw information and getting it into a database. We have our own in-house team that creates a data pipeline so that the routine stuff can be done in an automated way. We have a team of scientists — three professionals and a various grad students — that then start to ask questions. We are interested in comparisons between whole genomes, or between two different microarrays. There is not a whole lot of software out there for whole genome comparison. We have looked at Rosetta Resolver and software that is publicly available for microarray analysis. We at N.C. State do pride ourselves on statistical analysis, and there is a [statistics] unit just down the hall from me. We do have SAS down the road, and we have looked at SAS software for analysis of variance for microarray data and we have custom software packages being developed by students. We are also developing Ensembl, a large whole genome analysis package from the European Bioinformatics Group. We are very much believe in open source. Opposite me on the Centennial Campus [of N.C. State] is Red Hat [Linux OS provider]. And, we have something here called the Biogrid, which is supported by the North Carolina Biotechnology Council. It’s a huge undertaking that involves some massive supercomputers maintained by IBM and others. We have taken advantage of it to do some data analysis. Just the fact that it’s there is extremely encouraging — that we can build some sophisticated tools and have the horsepower to run them.
China recently announced that it has developed a rice gene expression array that it will make available to researchers. Does that interest you?
The China stuff does interest us. The question is how good is that going to be and how much information will be available. There will be a number of issues that will need to be clearly addressed. The Chinese are clearly earnest about joining the international community and [want] to be a major player in the arena. The question really is, is the technology — the microarray platforms, the analysis, the design — at a point where it is robust and stable enough to do mass production? Those are issues that I struggle with and one of the reasons that I haven’t approached Affymetrix. I feel that we need to have some confidence in the gene predictions in the sequences we are dealing with.