The uneven growth of science and technology leads inevitably to shortages of trained scientists in fields that are racing rapidly ahead and surpluses in others that are stagnating, because the educational and training systems cannot adjust quickly enough.
One of the biggest manpower shortages today is in bioinformatics, the area in which biology meets computer science.The discovery of new genes - linked to a myriad of conditions from obesity to breast cancer, schizophrenia to criminal behavior - receives a lot of publicity.
People write and broadcast about many aspects of the genetic revolution, from ethics to science, but they hardly ever look at the computing skills required to make sense of the deluge of information pouring out of the world's gene research labs.
Genetics is generating thousands of times more data than biologists have had to handle before. And bioinformatics is the key to making sense of it all and turning it into medical knowledge.
"Biology is becoming a data-intensive science in the same way that physics did almost 50 years ago when it became clear that computers would be needed not only to store information but also to process it," says David Searls, recruited from the University of Pennsylvania to become bioinformatics director of SmithKline Beecham, the Anglo-American pharmaceuticals giant.
SB has led the rush by the drugs industry to build up expertise in bioinformatics. More than any of its competitors, SB has staked the future of its research on genomics - the study of genes and the way they interact with one another and with the environment to cause disease. It made a pre-emptive strike in 1993 with a $125 million deal to acquire a stake in Human Genome Sciences, a Maryland biotechnology company that controls the world's largest human genetic database.
The bioinformatics department at SB already contains 33 scientists and engineers, and Searls plans to double its size within the next year. In an impressive coup, he's just brought in three of the leading figures in the field: Chris Rawlings, former head of informatics at the Imperial Cancer Research Fund in London; Jim Fickett, a senior computational biologist at the U.S. government's Los Alamos National Laboratory; and Randy Smith of Baylor College of Medicine.
In the public sector, the European Union-funded European Bioinformatics Institute in Cambridgeshire, England - an offshoot of the Heidelberg, Germany-based European Molecular Biology Laboratory - has managed to build up its specialist staff to about 70.
"There's a great shortage of people, partly because a lot of organizations are simultaneously seeing the need for bioinformatics and partly because the skills required are changing so fast," says Graham Cameron, head of services at the EBI. "Being an international organization, we can at least half-way compete."
From Cameron's perspective, "the pharmaceutical companies all know that they want bioinformatics and they know they want a lot of it, but they don't quite know what they want to do with it. I think some of the uncertainties will crystalize out over the next 18 months or so."
At SB, Searls knows he wants to proceed on a broad front. His department will concentrate on:
- Search and analysis, including new mathematical techniques for finding patterns in data;
- Knowledge management, including ways to integrate information from different databases;
- Mapping and genomics, including approaches to identifying the genetic components of complex traits;
- Sequence/ structure/ function, including rapid methods to predict the biological function of a gene from its DNA.
The mainstream computer and information technology companies are only just beginning to take an interest in bioinformatics. "There's a huge vacuum there," Searls says.
Therefore, SB cannot "outsource" bioinformatics to specialist information technology suppliers, as companies do in more mature industries such as financial services where the requirements are better defined.
Bioinformatics and genomics help drug discovery by giving researchers a huge number of new biological targets, such as enzymes whose over-activity causes disease. The companion activity is combinatorial chemistry - a new technology for creating a vast diversity of new molecules as drug candidates for testing against the targets.
The next step may be to combine the two approaches, so that pharmaceutical researchers can test many thousands of drug candidates at the same time for their activity against several targets. Keeping track of such an operation would, of course, require yet more computing power.
All manpower shortages correct themselves in the end. But it seems safe to predict that people who combine computing and information technology skills with biology or chemistry are going to remain in demand for a long while.
(Distributed by Scripps Howard News Service.)