AT A GLANCE: Has worked in the IT industry for over 20 years. Prior to joining AxCell, served as manager, data products, for PE Biosystems and director of bioinformatics development at Gene Logic. Spends personal time golfing and enjoys spending as much time as possible with his six- and eight-year-old boys.
QWhere will bioinformatics be in two years? Five years?
AThe next two years will bring further integration of bioinformation. This will be a shift in the way research is conducted, allowing researchers access to data from other research groups in a central electronic locale. A focus on knowledge is evolving rather than concentration on isolated data sets.
In five to ten years we will begin to see the lines blurred between what we now call genomics, proteomics, bioinformatics, and drug discovery in general. This systems view of research will allow us to gain intricate knowledge on how minute changes in DNA or amino acids impact everything from protein interactions all the way to patient outcomes.
QWhat are the biggest challenges bioinformatics must overcome?
AWe must get the computational and integration tools in the hands of all researchers. Solutions need to account for the disparate research interests as well as the disparate data sources. At the same time universities focus on training bioinformaticists, the industry needs to focus on solutions linking researchers across an organization.
QWhich databases do you use?
AWe access the publicly available genomics data, however, most of our focus is on proteins, so we currently license Swiss-Prot. Primarily our work is with our own proprietary protein-protein interaction database, which links interaction and pathway information to other available protein information.
QWhat bioinformatics software do you use?
AWe use a combination of in-house development and third-party software. Our LIMS is in-house development due to the novel techniques used in our laboratories to generate our domain/ligand-based protein-protein interaction data. We use GenoMax as our main research solution but also use GCG.
QWhat hardware do you use?
AWe use both NT and Compaq Alpha systems. As we grow, we are looking into Linux as a solution to handle some repetitive computational needs.
QHow do you integrate your data?
ABecause we are producing data as a product, the integration occurs as the end function of our quality check process. Once data from our laboratories have been through our QC steps, we link the data to publicly available data.
QHow large is your bioinformatics staff? Is the company hiring additional bioinformatics staff?
AOur informatics staff is roughly 50% of the company. During our first year in our facility, we assembled a team of seven dedicated bioinformatics scientists and five bioinformatics software developers. We have two open positions for bioinformatics scientists and anticipate hiring additional scientists and developers as the year progresses.
QWhat non-existing technology do you most wish you had?
AWe are working with a partner to investigate applying our novel protein-protein interaction assays to chip-based technologies. If successful, this could increase our throughput by an order of magnitude.
QWhat projects are you working on now?
AOur first database product will be available shortly and we are currently preparing our data for delivery to our partner InforMax. We are also mining our data for novel proteins, novel pathways, and novel interactions with disease relevance. Our bioinformatics researchers have also developed domain and ligand characterization models.
We have begun a project with publicly available SNPs to find those that cause an amino acid change that could disrupt an existing interaction or cause a new interaction. This narrows the millions of SNPs to a select few that could be candidates for further research.