Interested in reading more?

Graphic: Leza BerardoneThis is the beginning," pronounced J. Craig Venter, president of Celera Genomics in Rockville, Md., at a press conference at the American Association for the Advancement of Science annual meeting in February. "There [are] 30,000 genes, and up to one-quarter of a million proteins [encoded by them] in 100 trillion different combinations--that's how many cells we have. That's why we don't think this is the blueprint for humanity; it's the starting point of understanding how the living cell modifies itself." The "it" Venter refers to is proteomics. Defined in a broad sense, this growing field characterizes proteins by defining when and in which cells they are expressed, how a cell's machinery changes the protein once produced, and ultimately how proteins interact with each other in biochemical pathways. With so many areas to investigate, proteomics, in a professional sense, spells job security. Photo: ©2000 Lisa DumontEnrique Dalmasso"The peak of demand for those skilled in this field is yet to come," says Enrique Dalmasso, manager of Fremont, Calif.-based Ciphergen Biosystems' biomarker discovery center, because companies and research labs are just starting to ramp up, as are many pharmaceutical firms that are developing their own internal proteomics groups. Three types of companies lay claim to being involved in proteomics: tools and instrumentation firms, contract service companies, and information database firms. Instrumentation firms like BioRad Laboratories of Hercules, Calif., supply two-dimensional gel electrophoresis, Applied Biosystems of Foster City, Calif., provides mass spectrometers, and Ciphergen offers protein arrays. These tools variously separate and classify proteins by their electrical charge, weight, and binding affinities. The contract service companies such as Large Scale Biology of Vacaville, Calif., and Genomic Solutions of Ann Arbor, Mich., offer services for separating and identifying proteins. And then there are the information database firms such as Celera and Incyte Genomics of Palo Alto, Calif. Celera staff members expect that construction on their new proteomics facility will be completed by early summer. It will occupy an entire 30,000-square-foot floor at the Rockville site, one floor down from the genomics floor. Some firms provide a combination of equipment and services, such as Genomic Solutions, which offers both automated equipment to prepare proteins for mass spectrometry and protein separation services for industry and academic labs. Beth Allen, director of proteomic research services at Genomic Solutions, comments that her company will be expanding similarly to most firms in the coming years. Currently, the small firm employs about 12 scientists--from Ph.D.s in protein chemistry to B.S.-level technicians--on the research service side. At Ciphergen, about two-thirds of the Ph.D. life scientists work as field representatives. This cadre of 30 to 40 scientists demonstrates the ProteinChip system (consisting of a protein array, a time-of-flight mass spectrometer, and data analysis software) in labs contemplating purchasing the system. Dalmasso says that these scientists generally have their Ph.D.s in biochemistry, cell biology, or related fields. He notes that it's important that they have lab experience, but also a strong interest in interacting with people. Information Age Currently, the number of gene products in databases is probably in the hundreds of thousands. Because of that, says David Smoller, senior vice president at Incyte Genomics, the key now is understanding the complexity of that data in terms of protein interactions, expression, structure, and modification in addition to the growing number of characterized proteins. Incyte is concentrating on making a database of full-length gene products. To this end, they recently acquired Proteome Inc. of Beverly Mass., which sorts protein information and enables searches of the data for specific protein functions, interactions, structures, expression, and links to the research papers describing them. Incyte is hiring scientists with backgrounds in bioinformatics, IT, molecular biology, and protein separation chemistry. Smoller says that Incyte needs people in all these areas--from Ph.D.s to lab techs right out of B.S. programs--because of the company growth. One positive change, according to Smoller, is that there are more openings for entry-level positions now than any time in the past five years. Proteomics is an area of constant demand in both the private sector and academia, agrees James Garrels, president of Incyte's proteome division based in Beverly, Mass, and former CEO of Proteome Inc. So what skills are companies really seeking? According to Garrels, it's computational skills. "Proteomics is 80 percent bioinformatics." Finding enough people with that experience is a limitation in many companies. With mass spectrometry experience "you can also write your own ticket," he adds, and although two-dimensional gel electrophoresis is an older technique, it's still in demand (See also, "2-D Glasses,"). Protein skills such as biochemistry and chromatography are also in demand. Instrumental to the Process The high-throughput nature of most proteomics efforts is limited only by the machines that do the work, so the range of disciplines in demand expands beyond the life sciences, notes David Hicks, a protein biochemist by training and now director of marketing at the Applied Biosystems proteomics research center in Framingham, Mass. Open fields at Applied Biosystems include electrical and mechanical engineers and physicists with experience in mass spectrometry. "We also have a real need for software engineers who can craft the instrument control and data acquisition instrumentation as well as data analysis tools, including database-searching tools," he says. "Our instruments are more like complex integrated systems that include the mass spectrometer, software, reagents, training, and support." Organic and protein chemists and biochemists are needed, as are bioinformatics specialists for data analysis. Often overlooked, adds Hicks, are those skilled in project management, organizational, and people expertise to coordinate the instrumentation, software, and data analysis research and development at different locations, at different stages of development, and with different technical challenges and budgets. By the Numbers Three years ago Richard Simon, chief of the biometric research branch at the National Cancer Institute, dedicated a section of his molecular statistics and bioinformatics branch to meet the demand for people with analytical methods to interpret the vast amount of data coming from microarraying labs and other high-throughput proteomics projects. "Biostatistics is very much needed in the proteomics world," he remarks. There is a large number of molecular biology Ph.D.s that want to go into bioinformatics because of the career opportunities, adds Simon. What's needed is the expertise to interpret results. "There are lots of golden nuggets in the vast amount of data out there, but there's also much risk of misinterpretation," he warns. Simon currently has four postdocs--one from math, one from statistics, and two with Ph.D.s in biology; and has had three previously. Of the postdocs who have moved on, two have gone onto permanent NCI positions--one a biophysicist and one a biomathematician--and the third left after a year to work at Large Scale Biology. The market for people looking to hire biostatisticians is tight, says Simon. There are tremendous job opportunities already for biostatisticians in pharmaceutical firms doing traditional research in clinical trials, where they can make a lot more money than as a postdoc. Training the Next Generation Philip Andrews, a professor of biological chemistry at the University of Michigan, fresh from the Association for Biomolecular Resource Facilities and PittCon meetings, did find some likely candidates for a new proteomics consortium he is codirecting. But the "pool of experienced people is small," especially when competing with business, he notes. Andrews is looking to fill about 15 positions for the new Michigan Proteome Consortium at the experienced B.S., M.S., and Ph.D. levels for supervisors, computer programmers, software engineers, and mass spectrometry specialists. The consortium is made up of three labs to provide cutting-edge services for universities and business. Members of the consortium are Russ Finley at Wayne State University, Andrews at the University of Michigan, and Doug Gage at Michigan State University. "We need people with practical experience in proteomics and people whose skills could be easily adapted to proteomics," Andrews comments. Those skills include proteome mapping, yeast-2 hybrid mapping, and protein microarrays. He says the essential skill is adaptability for solving problems in proteomics. The key limitations to advancing knowledge is at two levels. Conceptually, researchers need to know what kinds of questions to ask and then match the appropriate technology to obtain results and then know how to interpret the data. And, technically, the field needs a certain number of people who intimately understand the technology because it is changing so rapidly. "What we're seeing now in proteomics in terms of the job market is a recapitulation of bioinformatics," says Andrews. "Industry has stripped academia, so there is no one left to train the next generation of workers." A proteomics postdoc program that Andrews has proposed aims to be up and running by the end of this year. It's intended to be associated with the consortium rather than individual labs. The goal is to train Ph.D.s to work in industry, so they are approximating an industry type of job. Postdocs will work as part of a team, working out of a central lab, which is a fundamentally different set-up than postdocs in academia. The emphasis is on flexibility and adaptability to different projects. Andrews concludes, "Our biggest challenge isn't amassing state-of-the-art instruments, it's finding the right personnel." Karen Young Kreeger (kykreeger@aol.com) is a contributing editor for The Scientist.

Career Paths in Proteomics

Apr 1, 2001

Interested in reading more?

Become a Member of