CAMBRIDGE, Mass.--Ken Fasman, a public-sector bioinformatics practitioner since 1990, last month was appointed director of bioinformatics for the new Astra Bioinformatics Center here. In part two of his exclusive interview with BioInform, Fasman discusses the challenges of coordinating bioinformatics needs for a multinational company.
Before joining the Swedish pharmaceutical firm Astra, Fasman spent 18 months heading sequencing informatics at the Whitehead Institute/MIT Center for Genome Research. For the preceding seven years he ran informatics at the Genome Database at the Johns Hopkins University.
BioInform spoke with Fasman in May, shortly after he took his new post.
BioInform: Can you outline your priorities after only eight days on the job?
Fasman: My first and foremost priority is recruiting this initial core group. Without them I can't get anything substantial done. Currently there are three of us, all people like myself, PhD's with a mixed background in biology and computing. A fourth just accepted an offer and will join us August 1.
BioInform:What will happen to the group in Sweden?
Fasman: All of the Sweden group is staying put. There's no way that a central group based in Boston can deliver the day-to-day informatics needs to 10 research units spread across four continents, so the existing informatics groups within the company that support each of the local research departments will be working with us.
Our job will be to develop standards, the infrastructure, but also more importantly to find common solutions to research problems that exist in multiple molecular biology, genetics, pharmacology groups across the company. We're looking at projects where, if we help to provide a new software system or a new database for one such group, we know that the work will be reusable by other groups in the company.
The local informatics groups will be the front line of support and the people who will customize solutions for the needs of an individual group within Astra. Then our job is to coordinate the efforts of these different groups and provide them with core solutions that are reusable. What's happened is the original Astra Bioinformatics Group in Sweden is now the local informatics support group for that particular Swedish research unit.
BioInform:What about other priorities?
Fasman: At the same time as I'm hiring the core group, one of our big jobs now is to survey the research computing needs across the company. We are trying to build relationships with molecular biology, structural biology, and chemistry groups throughout the company to understand what are the common needs, what are the most pressing informatics requirements of these groups, and particularly, we're looking for the common ground where solving one group's problem will pay off and help several other groups as well. In a company this size we've got to do a bit of homework to get things going.
The bioinformatics center is one of six competence centers within Astra--four in Sweden, one here in Boston, one in Australia--that provide specialized services for the company. Research groups throughout Astra can come to these groups for special services.
For example, the competence centers would be called on for solving the crystal structure of a newly isolated protein, or doing very high-throughput screening studies, or, in our case, developing bioinformatics tools and datasets. We are trying to coordinate our activities with these other competence centers and develop a model by which these services can best be deployed throughout the company. Each of us who runs one of the competence centers reports directly to the head of preclinical affairs.
BioInform:You'll be doing a lot of traveling?
Fasman: It's pretty unavoidable. I shouldn't be going to Australia that often, but travel to Sweden, UK, and other sites in Europe is certain to be the case. But it won't be constant. If I'm on the road all the time I obviously can't get anything done here. The company, being multinational, already has a well established system for e-mail and teleconferencing, so I'll take advantage of the best of modern telecommunications.
One of the nice things about informatics is that it's network-based. You can do quite a bit of your work remotely. Everybody I know who's heavily involved in this field lives in front of their mail client all day long. In fact I'll more quickly answer my e-mail than my telephone.
BioInform:How else will your work life be different from the way it was at Whitehead?
Fasman: Whitehead was a real opportunity for me to get back deeply into experimentation. The Genome Database was a central collection and distribution site for human genome data that were being gathered all over the world, but we weren't doing genome research there. We weren't doing any wet-lab work. At Whitehead, for the last year and a half I was involved directly with the people who are mapping and sequencing the genome, so I had a real chance to work with the day-to-day lab problems. Here I'm going to be back in a role more like the role at Genome Database--helping to facilitate research across the entire company, but not being directly involved in any significant number of these projects.
BioInform:Does Astra do any of its own sequencing?
Fasman: Not at the scale of Whitehead, certainly, but we have groups doing cDNA sequencing in particular, EST's, and also directed sequencing of various putative drug targets.
Also, Astra has had several major genomics collaborations with significant sequencing components: Genome Therapeutics for H. pylori; Millennium for asthma genes; and another which we will embark on shortly. We expect our total collection of genomics data to be greatly expanded this year. And we are actively testing DNA chip and array technologies at multiple Astra research sites.
BioInform:How would you compare Astra's bioinformatics efforts with those of other pharmaceutical companies?
Fasman: In refocusing this unit we are absolutely striving to be at the forefront of this work. We have a program that is already several years old and still in its growing stages. It started out a little more slowly than was hoped, but now we're seeing a pretty rapid acceleration.
I expect that my group will be making an impact on the company even within its first year. I expect that a lot of people who may not be aware of Astra today will certainly know about us within a year's time. The fact of the matter is that Astra has the best selling drug in the entire world right now. Prilosec, known in Europe as Losec, is the drug of choice for treating peptic ulcer disease. Yet many people still are only learning the name.
If you think of Tagamet as the drug of choice in the seventies and early eighties that helped launch SmithKline Beecham's rapid growth, and Zantac as the follow-on drug from Glaxo Wellcome that helped launch its phenomenal growth, Prilosec has now eclipsed Zantac as the best-selling drug worldwide and its sales are in turn fueling Astra's growth. But Astra is definitely not a one-product company. We also have the best-selling asthma medication in Europe--Pulmicort--and a number of other very promising medications in the pipeline.
BioInform:When do you think we'll start seeing genomics research directly affecting drug sales?
Fasman: Astra has seen a real acceleration of target identification and other early steps in the drug discovery process, but it's too soon to predict genomics' full impact on the development of saleable drugs.
BioInform:Where do you see your career heading?
Fasman: I'm expecting my work here at Astra to be all-consuming for the foreseeable future. This is a very big, very exciting challenge for me, and getting research informatics established in a robust way across Astra is as big a challenge as I could have hoped for in my career. I'm quite happy with the way things have worked out.