Genomics brings together concepts and skills from several different disciplines. Succeeding in the field demands flexibility, skill in communications, and an understanding of informatics.
Created in the fervent effort to sequence the genomes of humans and other organisms, genomics has become a critical discipline in modern life science research. Genomic specialists contribute to everything from fundamental understanding of gene function to the development of diagnostics and drugs for polygenic diseases.
The pursuit is multidisciplinary by its very nature. “Genomic scientists are often the bridges that connect different research areas,” says Jason Johnson, senior investigator in the informatics group at Rosetta Inpharmatics, a wholly owned subsidiary of Merck & Co. Thus successful practitioners must possess the ability to process concepts and data from several fields. “The only way to get information out of genomic research is by integrating the data into another context,” says Shawn Burgess, head of the developmental genomics section in the National Human Genome Research Institute (NHGRI).
The most critical fields that contribute to genomics are molecular biology and bioinformatics: “I can’t overemphasize how important it is to understand those two fields,” says Janet Warrington, senior director, clinical genomics research and development at Affymetrix. “The need for people with those skills is tremendous.”
Range of projects
A project at Northwestern University’s Center for Functional Genomics illustrates the multidisciplinary nature of genomics. “We are looking at ethyl nitrosourea mutagenesis screens in the area of the nervous system and behavior,” says center director Joseph Takahashi. “It involves behavior, physiology, neuroscience, genetics, genomics, and informatics.”
Myriad Genetic Laboratories, a unit of Myriad Genetics, has just announced research projects on human obesity and depression. “These are diseases that people would term polygenic,” explains president Gregory Critchfield. “We’ve found some important genes in both those areas. Once we find an important gene, we ask what pathways it’s operative in and what known aspects of the disease it influences. Then we go on to generating important targets.”
Affymetrix is completing a study on reproductive health funded by the National Institutes of Health. It involves 50 Finnish women treated with hormones or surgery for endometriosis, a condition that appears to have a heritable component. “This gene expression study has identified some new genes and pathways never before identified with the disease,” says Warrington. “We have validated those findings at the RNA level and are now validating with immunohistochemistry. We have identified a number of genes that encode for proteins in the blood that could be used for early diagnosis.” The company also has a collaboration on lung cancer with Boston University Medical College and a study on prostate cancer with Stanford University Medical Center.
The branch in which Burgess works at the NHGRI, meanwhile, has a broad range of research projects that also illustrate the diversity of scientific opportunity in genomics. “Researchers in our branch are applying genomics to complex human genetics such as multifactorial diseases like diabetes or cancer,” he says. “Our branch chief is looking at the genomic structure of specific genes across many species to determine what information can be learned by a multispecies genomic analysis. My research involves developmental genomics – zebrafish genetics and embryology integrated with genomics. We also have a bioinformatic core here doing a lot of protein and DNA homology work as well as other forms of computational biology. The integration of these disciplines is what makes it a great place to work.”
The bioinformatic imperative
Bioinformatic skills are perhaps in greatest demand in genomics laboratories. “We prefer candidates who are comfortable working with databases, writing code to conduct statistical analyses, and interpreting analysis results in light of recent biological research publications,” says Johnson of Rosetta. “While many recent graduates have strong computational skills and a good biology background, few seem to have experience putting them together. The rarest are candidates who also have the biological insight to frame testable genomic hypotheses that can impact pharmaceutical research.” Burgess agrees. “Individuals who have a lot of biology experience and computer experience are unique,” he says. “It adds to their value to have both sides.”
Absent a sufficient supply of those candidates, says Myriad’s Critchfield, “The debate that occurs questions whether you are better off taking someone with a hard science background and teaching them life science or vice versa.” Burgess generally opts for computing skills. “It probably works better if you come in with real computer experience and learn biology rather than the other way round,” he says. “But it will be better to embrace both worlds.”
Scientists who enter genomics with traditional life science training must pick up several informatic skills along the way. “They need a fundamental understanding of programming, bioinformatics, and data mining,” says Warrington of Affymetrix. “It’s critical to be able to mine your data and to be conversant in the language of programming.”
Breadth of training
Warrington looks for other scientific abilities when she recruits genomic researchers. “They should have a solid background in molecular biology, molecular genetics, or biochemistry and have distinguished themselves through their Ph.D. or postdoctoral work,” she says. “I also look for people with statistics experience – biological statisticians or plain statisticians who have some biological background.”
Certainly the field attracts – and requires – scientists from several traditional disciplines. “We have curator-type positions for Ph.D.s in biology with a computer science background,” says Northwestern University’s Takahashi. “We even have engineering needs that involve creating and interfacing all our different data acquisition systems with the informatics platform.”
Myriad seeks scientists with several types of background. “In our commercial segment we look for people who have experience in pharmaceutical sales or a background in genetics and a desire to be involved in the sale of genetic testing,” Critchfield says. “On the scientific side we’re looking for people with experience in managing laboratories and developing new tests and capabilities. We want people with pharmacology backgrounds for our pharmaceutical group, as well as scientists with mammalian cell biology backgrounds and people with bioinformatics skills. We’re also seeking people at a lower level than the Ph.D. to evaluate sequence data.”
Like other members of the National Institutes of Health, NHGRI works in much the same way as universities in its hiring practices. In late 2001, however, it started to hire scientists for what it calls associate researcher positions. “They have some research independence, but they are also involved in running support programs such as a bioinformatic core or a microarray core,” explains Burgess. “We now have 16 associate investigators. They usually come out of a postdoctoral program, with experience in particular technologies that many groups can draw on.”
Collaboration and communications
Nonscientific criteria play an important role in decisions to hire scientists for work in genomics. Potential employers particularly value the related skills of collegiality and communications ability. “Almost all our research is done collaboratively as part of a team,” says Rosetta’s Johnson. “So we prefer applicants who have a record of successful team projects, preferably with previous industry experience. Good writing and oral presentation skills also make a significant impact on our evaluation of candidates.”
Northwestern’s Takahashi agrees. “The ability to work in a group is very important since our projects are all team projects,” he says. “And communication skill is a positive asset.” Indeed, the very nature of genomics mandates those requirements. “The ability to work in groups is especially important because of the interdisciplinary nature of the work and the fact that nobody can know everything,” says Burgess. “In NHGRI we also have to maintain good relations with various extramural groups.”
Critchfield of Myriad makes a similar point. “You need people who are able to communicate and learn the languages of other team members,” he points out. “People from good laboratories who have been mentored well generally have a better understanding of the value of teamwork and communications.” Warrington of Affymetrix has a pithier observation. “If you can’t communicate,” she says, “you won’t be able to do your job.”
The need for flexibility
Recruiters add one more piece of advice for scientists who want to make their careers in genomics: Be flexible. “Recognize that there’s not one thing that’s the right set of tools and always will be,” says Burgess. “By the time you’ve done your training, the techniques you’ve learned are often obsolete. So be ready to change quickly and cross disciplines.”
Communication skills and flexibility can play key roles in landing a first job in genomics. “Make sure that you can tell potential employers a story about the work you’re completing. Being able to communicate that story is very important,” says Warrington. “My advice is to get the best scientific training you can, establish expertise in the area in which you’re being trained, and try to understand what the companies you’re interested in are about,” adds Critchfield. “And ask a lot of questions to help you make good decisions.”