AT A GLANCE
PhD in mechanical engineering from Vanderbilt University.
Served as head of bioinformatics at Pioneer Hi-Bred before joining CuraGen.
Avid stamp collector: “Stamp collecting works well with genomics because I love to organize and catalogue things.”
QWhere will bioinformatics be in two years? Five years?
AFor the next two years it won’t be vastly different from where it is today. We’re going to continue to wrestle with the problem of data integration, so that will continue to be the big focus, but over the next two years there’s going to be a lot more emphasis on proteomics and comparative genomics.
In five years I think things are going to be quite different. The emphasis is going to shift to identifying the toxicogenomic and pharmacogenomic pathways. There will be a bigger role for bioinformatics in determining how we set up and conduct clinical trials and identify the effects of drugs on patients. In the next five years I also think the systems biology paradigm will really come into its own.
QWhat are the biggest challenges bioinformatics must overcome?
AThe biggest challenge at the moment is establishing meaningful relationships between different kinds of data. How do we establish a relationship between expression, protein, sequence, and SNPs that is meaningful and can be used in doing drug discovery.
I think another challenge is also going to be the level of expectation. Biologists have really come to depend on bioinformatics to do many different and important things, so I think that level of expectation is going to have to be continually examined over the next few years.
QWhat do you see as the most important task for bioinformatics to address beyond genome sequence analysis?
AClearly understanding disease pathways and building models of cellular processes. The important question now is how do you take 30,000-plus genes and get 100,000 proteins, and how do those affect disease?
QWhat bioinformatics software do you use?
AFor the most part we’ve built everything completely in house. When we started building our bioinformatics five years ago, we took a look around and at the time there really wasn’t much commercially available that met our needs. We are just now starting to explore some third-party software packages. We do use Spotfire for some of our visualization.
QHow do you integrate your data?
ABasically we integrate it by building a relational database system and indexing what flat files are necessary. We have a project here called CuraBase, where we’re building a comprehensive relational database that will serve as a complete bibliography of genes.
QWhose microarrays do you use?
A As we continue to map out what we call the pharmaceutically tractable genome, we found that commercially available chips only contain roughly 25 percent of the genes that we’re interested in, so we’re building our own microarray called CuraChip for our own internal use.
QWhat non-existing technology do you most wish you had?
AI wish I had an algorithm that could give me the complete ADME-T (absorption, distribution, metabolism, excretion, toxicity) genome.
QWhat projects are you working on now?
ATrying to determine the ADME-T genome!
QWhat made you decide to enter a career in bioinformatics?
AVery easy: My original PhD project ran out of funding. I was working in fluid dynamics as part of a group that was funded by NASA, but NASA cut that particular project. I got a call from Clark Tibbets who was at the Vanderbilt School of Medicine and needed somebody with an understanding of mechanical engineering and artificial intelligence. He had just bought one of the first ABI 373s. So I got into instrumentation and then became much more interested not just in data acquisition, but in doing the analysis and understanding how this data works in complete genomes.