AT A GLANCE
Holds a PhD in biochemistry from the University of Darmstadt in Germany.
Prior to Biogen, led bioinformatics and scientific computing programs at Glaxo Wellcome, Ariad Pharmaceuticals, and Aventis Pharma.
Hobbies include showing his Labrador retriever in dog obedience and agility events.
Q Where will bioinformatics be in two years? Five years?
A Bioinformatics will have to shift rapidly from a focus on algorithms to a focus on information management and knowledge discovery. Researchers will be inundated by data, arriving from disparate sources in many different formats. Our real interest is in gene or protein function, which will initially be described in the form of large data sets — for example expression profiles — and complex phenotypic readouts. Additionally, we need to link in the scientific literature more effectively. Truly intelligent systems are needed to separate the wheat from the chaff. Somewhere in the five to 10 year time frame, I can also see the first computational simulations of complex biological interactions with real predictive power, which for me constitutes the Holy Grail of bioinformatics – at least from a drug discovery point of view.
Q What are the biggest challenges bioinformatics must overcome?
A Perhaps the main challenge is to provide scientists with the means to cope with the onslaught of information that high-throughput biology in the 21st century will generate. I also believe we need to work harder to make bioinformatics more accessible. The Web has certainly made it easier to access bioinformatics tools and databases, but how to use them most efficiently is still a mystery to many scientists.
Q What hardware do you use?
A For largely historical reasons, Biogen has a very heterogeneous computing environment. On the server side, we use both Sun Microsystems and Silicon Graphics hardware. And on the desktop we have Apple Macintosh machines and Microsoft Windows NT-based systems. A Linux farm provides high-performance computing capabilities.
Q Which databases do you use? Public, proprietary, or third party.
A We maintain local copies of all important public databases and update them on a nightly basis. Biogen is also a subscriber to Incyte Genomics’ LifeSeq database. More recently, we got very interested in the Ensembl project, which promises to provide a powerful view of the human genome.
Q What bioinformatics software do you use? Do you use in-house developed or third-party software?
A Our basic approach is buy if we can, build if we must. Therefore, our scientists have access to a wide range of commercial bioinformatics applications, including software from GCG, Incyte, Spotfire, and Affymetrix, plus internally developed custom solutions. A focus of those has been pipelines for large-scale sequence data mining and novel approaches for analyzing gene expression data.
Q How do you integrate your data?
A Currently, data integration happens mostly at the application level. Web pages provide scientists with integrated views on novel targets. A sophisticated desktop application allows our researchers to browse through a large set of specific information on genes of interest and to filter and select interesting ones for further experimental workup.
Q Whose DNA chips do you use?
A We use Affymetrix GeneChips as well as commercially available spotters and scanners to produce custom oligonucleotide and cDNA microarrays.
Q How is your bioinformatics unit organized within the framework of the organization? Is it part of the IT department, the biological research department, or some other department?
A Bioinformatics is part of research informatics, which also includes structural informatics (molecular modeling and cheminformatics), and discovery support. Research informatics reports directly to the head of research.