AT A GLANCE
Created the Biotech Rumor Mill Web portal and the CuraTools bioinformatics products.
Hobbies include taking electronic products apart. Unfulfilled dream project is to make a ray gun from an old microwave oven. Also likes to hack code and get it to “bend to my will.”
Q Where will bioinformatics be in five years? Ten years?
A All the proteins will be out, people will be hard at work detecting protein variants, and (as they are now) trying to assign function to proteins through inference of other proteins. There will be more three-dimensional structural data and more people will be using this as an inference source.
As the whole genome will be out, people will have developed tools to be more successful at promoter recognition and begin the hard work of understanding gene regulation pathways.
Ten years from now, given a gene we will have maps on how the gene affects the expression of other genes and how it interacts directly and indirectly with other genes. We will be able to model cells. On a graphical interface, we will be able to modify the expression of one protein and predict how it perturbs the function of the cell. We will be able to create a “cellular automaton” bringing the “Game of Life” version from simple X and O to proteins.
Q What are the biggest challenges bioinformatics must overcome?
A Training the new generation of scientists and hiring them. Generating common formats of data for easier data exchange. The “latest” appears to be XML, which is being increasingly used internally and bridges many gaps.
Q What hardware do you use?
A Sun, TimeLogic, and a small Linux server.
Q How do you integrate your data?
AThrough automated sequence analyses linking databases and spidering through existing annotated databases and extracting the cross-references. We also use manual annotation and curation tools developed in-house.
Q Whose DNA chips do you use?
A We do not use DNA chips. In fact we prefer to use our own open-architecture GeneCalling gene expression profiling system. You are not limited to knowing the identity of the genes you are working with and it allows exquisite resolution of splice variants and other nucleotide polymorphisms — insertions, deletions, RFLPs — of virtually any organism.
Q How large is your bioinformatics staff? Is the company hiring additional bioinformatics staff?
A Bioinformatics at CuraGen is almost entirely software development. We have 50 people in bioinformatics of which 45-47 perform software development. In other organizations the term “bioinformatics analyst” is used where the people are biologists that are adept at analyzing large sets of biological data. If we include some of our discovery scientists that perform this task, I would say we have 100 people in a bioinformatics discipline. We are always hiring as bioinformatics is critical for the direction of our company.
Q How is your bioinformatics unit organized within the framework of the company?
A Bioinformatics is a separate group that performs the laboratory information management systems work as well as the discovery informatics. It also puts all of all this information into databases. Our database administration group is part of bioinformatics.
Bioinformatics is organized into sequence informatics (SeqCalling), gene expression informatics (GeneCalling), protein networks/pathway informatics (PathCalling), drug development informatics, technical support, LIMS, and data base administration.
Q What non-existing technology do you most wish you had?
A A technology that would allow you to put in a DNA sequence and then spit out the specific protein function, disease association, known pharmacophores that could be developed into small molecules, and market value of small molecule or protein therapeutic (antibody) drugs generated from that gene.