Job Market Elixir
- 03 April, 2002 11:57
Still yearning for the dot-com days, when IT skills were not only critical but also sexy? Wishing for a time when you could come to work with a credible chance of helping to change the world every day?
Take heart. There's still at least one industry in which advanced IT is so vital that without it, companies literally can't compete. With the right combination of technology, science and luck, such firms stand to be not only competitive, but also incredibly profitable while indeed changing the world-or at least curing disease and improving the human condition.
Pretty high concept, but that's the whole point behind biotechnology, an industry that's using the recently decoded human genome as a set of instructions for how to design drugs to treat cancer, diabetes, depression and dozens of other physical and mental maladies.
"A lot of it involves straight IT skills, but in the end, what you're trying to do is pretty significant," says Paul Dupuis, who has a background in mainstream IT and a talent for understatement. He is director of IT at Vertex Pharmaceuticals Inc. in Cambridge, Mass.
Leveraging IT to help read the human genome and design compounds according to its instructions could cut two to three years off the 15 it currently takes to develop a single new drug. It could also slice the average $880 million price tag for drug development in half, according to a report from The Boston Consulting Group Inc.
Biotechnology and pharmaceutical companies will spend increasingly large amounts of money on bioinformatics during the next few years, but the bulk of their spending will continue to go toward systems and applications they build in-house.
Advances in homegrown and commercial informatics are making the decoding process easier, but the data is growing faster than the technology used to analyze it.
"A year or two ago, we were all very focused on sequencing genes, figuring out how they assemble together into chromosomes, which 10 years ago seemed like a phenomenal project," says George Morris, director of bioinformatics at drug development firm Zycos Inc. in Lexington, Mass.
"Now we're interested in proteomics-investigating the proteins that genes code for-and they're much more complicated," Morris says. "There's a many-to-one relationship between a protein and a gene, and many different forms of it. There's a huge order of magnitude [of] greater complexity in this."
That level of complexity could flummox many mainstream IT professionals, but the skills that made them successful in retail, manufacturing, engineering and other industries are still applicable in biotech and pharmaceutical companies. There's just an additional layer of complexity into which they can grow-if that's the direction they choose, Dupuis says.
Because they have to maintain not only a global scientific research operation but also an immense communications infrastructure, large pharmaceutical companies provide the greatest opportunity for IT professionals without a medical or scientific background.
That range of needs means that IT workers with good technical skills may be able to plug into a research operation, according to David Pioli, head of global informatics in lead generation at Aventis SA in Frankfurt. The Schiltigheim, France-based pharmaceutical company was formed two years ago by the merger of Germany-based Hoechst Marion Roussel AG and France-based Rhone-Poulenc Rorer SA. Project management, application development, systems management and personnel management skills are always applicable, Pioli says.
Bonding IT and Scientists
But if IT professionals want to be successful in supporting or becoming part of Aventis' research teams, he says, they have to learn enough about the science and research activities to form tight bonds among the three groups that drive the process: scientists, informaticians and a global IT group that supports both research and mainstream corporate activities.
The focus has to be on the science and the tools that make scientific research faster and more accurate, Pioli says. Delivering those tools-algorithms or analytic applications that answer specific biological or chemical questions-is primarily the job of informaticians, who in turn rely on IT professionals for support.
"An informatics guy is one half a frustrated IT developer and one half a frustrated scientist," Pioli says. "It's the healthy tension between the two that makes a successful informatician."
Informaticians at Aventis must work with the IT department to make sure the applications are scalable, build and maintain the high-end servers that run them, and make sure the systems are transferrable to other research groups so they aren't reinventing the same tools over and over.
"It's all about synergies, avoiding duplication of effort," says Pioli. "If you just let informatics run away with itself, you end up building a mini-IS group in informatics and another to support the rest of the business."
The informatics and information systems teams at Aventis are still separate, but the company has created a semiformal scientific computing group to give concentrated IT support to the informatics groups, which have about 200 staffers worldwide.
Middlesex, England-based GlaxoSmithKline PLC (GSK) and many other pharmaceutical companies have similar mixed teams whose goal is to give the scientists as much research firepower as possible while maximizing the money and time spent on IT, says Terry Francis, head of operations and infrastructure for the cheminformatics group at GSK.
"You have a Grand Canyon chasm between a bench scientist and a classic IT person," Francis says. "One group doesn't understand that when the network goes down, people get fired. The other group doesn't know much about what the scientists do. You need an educated interface between the two."
For both GSK and Aventis, that educated interface consists of a mixed group of science-savvy IT people and informaticians.
IT people working in these groups have to be agile enough to pick up on scientific concepts, willing to prototype new systems rapidly to keep up with the pace of research and adaptable enough to keep up with rapid change, according to Simon Dear, who heads a 170-person global team as director of bioinformatics engineering for genetics research at GSK.
"We really are looking for a broad background and the ability to adapt," Dear says.
Application development skills in areas such as Java, C, database programming and administration, storage and data management, are also keenly sought after, Francis says.
But success in this field doesn't necessarily depend on a scientific background. Dear's background is in engineering.
"There's a lot of opportunity in informatics in GSK to prove that you're effective and to learn on the job," Dear says. "Everybody learns the appropriate level of the science. They just have to be willing to get in there and get their hands dirty."
At smaller biotech companies, the administrative integration problem is less severe because most firms' efforts go into the science itself and not into supporting a global administrative infrastructure, according to Mark Adams, vice president of bioinformatics at drug discovery company Variagenics Inc. in Cambridge, Mass.
Biotech companies still need database administrators, project managers and IT managers who can negotiate a contract, maintain a budget and build a data center where the compute-power needs are high, Dupuis says.
But anyone who can build a high-performance cluster or develop a complex analytical application that runs efficiently on a parallel-processing architecture will always be welcome in a biotech shop, says Adams, a bioinformatics manager who also has an engineering background.
"You just have to have people with varied interests who want to learn how to apply stuff in a different area," Adams says. "They have to be not just tolerant of change but excited by it and be able to recognize that the science is evolving and the challenge for IT is to not slow that down but to respond to that appropriately."