Irreverent Genetics

Credit: © andy reynolds / andyreynolds.com Eric Schadt doesn't care much about how he dresses. In fact, the executive scientific director of research genetics at Rosetta Inpharmatics has about 20 or 30 identical white polo shirts and about 10 pairs of identical khaki shorts hanging in his closet. He dons this recurring outfit - over white socks and Birkenstock sandals - every day before heading to work at Rosetta, a wholly-owned subsidiary of Merck. The 43-year-old says he has more important things to concentrate on than fashion - like using expression microarrays to construct complex gene interaction networks which identify novel targets to treat diseases. "What I have to do every day is push super hard on the thinking front and on the doing front," he says. "It's much easier to push and think hard if you don't have to think about other things that don't matter so much." Schadt has been championing the importance of modeling complicated networks of gene interaction to discover novel drug targets since he started at Rosetta in 1999. Stephen Friend, the president of Rosetta who hired Schadt, says that the company was willing to let Schadt explore this approach at a time when the drug industry, as a whole, wouldn't - instead, companies were primarily searching for new drug targets one gene at a time. "Eric is someone who took on a problem that others felt was not ripe," says Friend, now the senior vice president and franchise head of oncology at Merck. "We had a reverence for people who were irreverent." Schadt with Leila Shiraiwa, Merck Biologist Credit: © andy reynolds / andyreynolds.com It's paid off. Since joining Rosetta, which was acquired by Merck in 2001, Schadt has built a research genetics program from the ground up, and he's helped develop new technologies and partnerships within the company. He helped launch MegaCross, a target discovery engine based on mouse models. Later, he built another tool, called Human QuickChip, which elucidates linkages between genes and proteins and can predict the effects of perturbations to genetic networks. Nancy Thornberry, vice president and worldwide basic research head in Merck's obesity and diabetes division, says that about a third of the novel targets that are currently being pursued by the company in the obesity and diabetes arena were identified using Schadt and his group's genetic methods. "It's hard to know exactly who's responsible, but I think he, more than anyone else, has led the way. I can't think of anyone more influential in putting together complex traits with genetic diseases." -Jake Lusis According to geneticist and microbiologist Jake Lusis, who taught Schadt his first advanced biology class at UCLA, his former student's approach to genetics is likely to transform how drugs are developed and diseases are treated. "I think it's going to have a revolutionary impact on biology," says Lusis, who still collaborates with Schadt. "It's hard to know exactly who's responsible, but I think he, more than anyone else, has led the way. I can't think of anyone more influential in putting together complex traits with genetic diseases." Religous roots, scholarly pursuits As a child, Schadt's rise to the forefront of the genetics field might have been hard to predict. He grew up in the small town of Stevensville, Michigan, and says that academics were not stressed in his family, which subscribed to "a more extreme interpretation" of Christianity. "I grew up in a family that was very religious, to the point of being anti-education," he says. Because the young Schadt didn't see college as an option, he left Stevensville and joined the US Air Force in 1984. After a short time in the service, however, Schadt was injured and could no longer serve as an Air Force pararescueman. So he started college at age 20, and in 1991 graduated from California Polytechnic State University with a degree in applied mathematics and computer science. He went on to the University of California, Davis, where he earned a master's degree in pure mathematics. His "anti-education" family eventually came around, and they now enjoy a good relationship. Around this time, a new idea began stirring in Schadt. "Math was very intellectually challenging," he says, but he felt the urge to do more than "just sit in a room and be really smart and figure out hard math problems." Schadt says that because he wanted to "understand the fundamental complexity of life," he enrolled in UCLA's biomathematics graduate program, where he would earn a dual-PhD in both molecular biology and mathematics. Jake Lusis taught Schadt one of the first advanced biology classes the young graduate student had ever taken: human genetics. Lusis says that Schadt, who was wearing his polo shirt and shorts outfit even back then, was the top student in a class of about 200. "We tried to make exams really difficult so he wouldn't get 100 percent," Lusis says. Kenneth Lange, chair of human genetics at UCLA and Schadt's PhD advisor, agrees that Schadt showed a preternatural talent for absorbing and synthesizing information. "These hybrid fields tend to attract people who have inordinate curiosity and real intellectual talent," Lange says. "He's bursting with energy and he has lots of ideas he likes to pursue simultaneously." As Schadt worked on his PhD project (using mathematical models to improve the construction of evolutionary trees and modeling evolutionary dynamics all the way down to the codon level), he got married and started a family. This meant he needed a steady source of income. He began working at Roche, where he had access to the latest expression array technologies coming from Affymetrix. "Eric had access to Affymetrix's technologies before many academics did," Lange says. "They were very, very expensive, and only the companies with deep pockets were funding that kind of work." So Schadt got to work. A bigger pond His goal was to use this emerging technology to mine massive datasets for genetic keys to disease. As a graduate student studying evolutionary trees, he was struck by the narrow focus that researchers had in constructing scientific theories. Instead of considering the full complement of genes that might bear on a specific trait or characteristic, researchers were focused on the behavior of single genes. "It just struck me as odd that people would be basing so much scientific theory on such small datasets," Schadt says. "Biology had become this purely reductionist discipline. What I was being told was, 'Just give me the single gene.'" Schadt knew that this way of thinking ignored all the compensatory changes that might happen in a perturbed network of genes. As microarray technology came to the fore around 1998, Schadt found the tool that allowed him to use his computational and mathematics background to understand and utilize large datasets. But funding was a different story. Schadt says that he applied for several NIH grants, through UCLA and Roche, which included large-scale expression array data mining. His applications were unsuccessful; he says that he received comments from reviewers such as, "This isn't hypothesis testing," and "This isn't really science." Schadt knew that to get the funding he needed to generate huge datasets, he would have to turn elsewhere. "No granting agency was willing to fund the work at the scale we wanted to do it. The avenue I saw to pursue this was to go to people who had a business incentive to pursue this kind of game." That's when he made the acquaintance of Stephen Friend through a Rosetta colleague who had seen Schadt deliver keynote addresses on his success interpreting the latest microarray technology. Friend says that while Roche was focused on short-term success using established methods, Rosetta's focus was to "change the paradigm of doing drug discovery." And when Schadt was hired, that's what he set about doing. He quickly began assembling a bioinformatics group at Rosetta, wrote key software, and essentially got Rosetta "into the bioinformatics and computational genetics game." (It may have been hard to forecast Schadt's success, though, on Christmas Eve, 1999. On that day, FBI agents came to Schadt's Seattle home, and said he was the target of a criminal investigation because he had published improvements to gene expression software developed by Affymetrix and used at Roche. Schadt says that he and his collaborator, UCLA statistician Wing Wong, simply wanted to make the software more useful to biologists. "We love the technology because it's the future, and we want to make it better" Schadt says. "Affymetrix took it differently." Investigators were led to believe that Schadt had taken some of Affymetrix's technology with him to Rosetta. But after an internal investigation by Roche and the criminal investigation, the case never went to court, and Schadt was cleared of all charges. Both Roche and Affymetrix declined comment on the matter, as did the FBI.) In 2001, Schadt was busy helping to annotate the recently published human genome when Merck came a-courting. "Initially we were very disappointed," Schadt says. "I was like, 'Aww, this is not good.'" According to Friend, people at Merck were equally unsure about the geneticist at Rosetta who wore a white polo shirt and khaki shorts every day. "People in the larger company said, 'Who is this Eric? This is a different type of guy,'" Friend recalls. Schadt's skepticism turned to optimism when he realized the resources that the pharmaceutical giant could bring to his work. "It became clear to me that all the experiments we wanted to do but couldn't afford to do at Rosetta, Merck was interested in doing," he remembers. John King, former COO at Rosetta and later senior vice president of research planning and integration at Merck, says that Merck was willing to take the risk of incorporating Rosetta because it was betting that the direction in which Schadt was steering the company was the future of drug development. Schadt continues to push the envelope of genetics research as a part of Merck. His work has helped form a new field in statistical genetics, called the genetics of gene expression, which essentially treats genes and the traits they confer as a complex and interwoven system. Schadt and collaborators used the MegaCross engine to study the genetics of obesity in mice, resulting in a 2003 Nature paper outlining the genetics of gene expression (Nature, 422:297-302). Human QuickChip, which arose out of the work on MegaCross, goes even further, Schadt says, to make the connection to drug reactions by tracking the effect of therapeutic compounds in human gene networks. Schadt published research using Human QuickChip to ferret out the relationship between thousands of SNP genotypes and human liver gene expression traits in a recent issue of PLoS Biology (6:1020-32, 2008). He now works on parsing out the genetic intricacies behind a suite of diseases including diabetes, Alzheimer's, and coronary disease. Schadt says that he's gone from getting his NIH grant proposals rejected to being sought out as an expert by the agency. "I get invited to participate in more study sections than I have time to do." He has continued to buck traditional work dress attire while he questions traditional approaches to the genetics of disease. "I get it in my reviews every year, 'iconoclastic,'" Schadt says. His group recently received Christmas presents of polo shirts - blue polo shirts - with their team logo printed on the breast. When Schadt wore the shirt during a site inspection of Rosetta by a Merck supervisor, the inspector said, in front of the entire group, "Oh my lord, Eric's wearing a blue shirt!" "Everybody was roaring," Schadt says.

Irreverent Genetics

Become a Member of

Meet the Author

Bob Grant

Published In

July 2008

Nergal Networks: Where Friendship Meets Infection

Research Resources

Podcasts

Webinars

Videos

Infographics

eBooks

Unraveling Complex Biology with Five-Dimensional Multiomics

Resurrecting Plant Defense Mechanisms to Avoid Crop Pathogens

Seeing and Sorting with Confidence

Streamlining Microbial Quality Control Testing

Products

Product News

Agilent Unveils the Next Generation in LC-Mass Detection: The InfinityLab Pro iQ Series

Pioneering Cancer Plasticity Atlas will help Predict Response to Cancer Therapies

Alamar Biosciences, Alzheimer's Disease Data Initiative, and Gates Ventures Unite in Global Partnership to Advance Translational Research in Alzheimer's Disease

How Alderley Analytical are Delivering eXtreme Robustness in Bioanalysis