Q Where will bioinformatics be in five years? Ten years?
A In the coming years, we anticipate a continuation of trends that are already visible, many of which we helped pioneer. First, gene sequence information is going to be the foundation for many types of biological and medical research programs. So, we expect to see enhancements in the ability to acquire, store, retrieve, analyze, and interpret sequence information to become widely available and increasingly sophisticated. We believe the significant potential for bioinformatics in coming years is to streamline efforts to characterize and understand the function of genes.
Q What are the biggest challenges facing bioinformatics?
A The biggest challenge we face in bioinformatics is handling and interpreting the many different types of data that are generated by modern biological research. Much of the bioinformatics literature has focused on the amount of information generated by genomics research. However, this emphasis on quantity tends to downplay the complexity inherent in biological data and the difficulty of using computational methods to support biological research. Deriving meaningful interpretations from the information is quite difficult and initial analysis often leads to apparent contradictions.
Q What hardware do you use?
A We host most of our bioinformatics systems on industry-standard equipment. Systems typically run in a client-server configuration. The server equipment is predominantly made by Sun Microsystems (for database management, Web service, and general applications). We also use special-purpose servers such as Network Appliance Filers and Compugen Biocellerators. Our users interact with our bioinformatics systems via Macintosh, PC, and Unix workstations.
Q Which databases do you use?
A Our significant database systems have been developed in-house using commercially available database management packages (primarily Sybase and Oracle). We also subscribe to a variety of public and commercial databases, such as GenBank, Swiss-Prot, and the Derwent GeneSeq patent publication database. These subscriptions, when combined with our internally-generated databases, support our various research efforts.
Q What bioinformatics software do you use?
A A large proportion of our bioinformatics software has been built in-house. When we launched our genomics program in 1992, there were no commercially available systems designed for the bioinformatics market. So, we developed our own systems to track biological samples, maintain inventories, capture sequence data and automate sequence analysis. We frequently integrate public and commercial tools into our bioinformatics framework for special computational or analytical functions.
Q Whose DNA chips do you use?
A Since we can predict gene expression from database analysis, we have designed a gene expression program that confirms and improves the resolution of our database-predicted expression data. We developed our own technique for gene expression arrays by optimizing well-known hybridization techniques and arranging selected cDNA clones from our gene collection on nylon filters. These filters provide us with a good deal of flexibility in selecting and analyzing genes for expression patterns.
Q How large is your bioinformatics staff?
A Nearly everyone in our research organization participates in bioinformatics work. Our bioinformatics group currently has 24 employees and a small number of contractors who are used on an as-needed basis.
Q What’ s lacking in the bioinformatics toolbox?
A Any technology that gives us higher resolution into the processes by which proteins and cells interact will be eagerly adopted by us (and by our competitors!). For instance, protein structure prediction tools and tools for predicting function from structure could be used to direct research efforts on individual genes.