His are some of the giant shoulders upon which genomics will forever stand. Some of the concepts Aravinda Chakravarti developed while working alongside Francis Collins and others to map and clone the CFTR gene in the 1980s are still in use today. "Twenty years later we're using the same damn [approach], we now have companies, we make millions of markers, we have fancy computer programs," says Chakravarti, currently a professor of medicine, pediatrics, and molecular biology and genetics at the Johns Hopkins University School of Medicine. "But the fundamentals, the logic of it, we established way back then."
Since those early days, Chakravarti has gone on to map the major genes and the key variations involved in Hirschsprung's disease, a rare intestinal disorder; developed an algorithm for unraveling multigenic diseases; and helped establish the basis for the HapMap project. But despite these notches in his belt, his quest to fully define the rules of the double helix continues to be a life-long pursuit. "We still don't know the logic of the genome, and understanding that is important," he says. "So I haven't quite gotten to the end of the road and I might never get there, but it's sort of the path we follow."
While he's been an advisor to many, Chakravarti himself did not undertake a postdoc position — at that time in the early 1980s, he viewed the step as a delay to his own pursuit of where genetic research was headed. "I think I had an unusual career in that I never quite had a postdoctoral fellowship and I think it says something because it was a clash between the old ideas and what I thought genetics needed to do, which is to map," he says. "I took a teaching position and at the same time I knew that the recombinant DNA thing was already there. That gave me time to rethink, so I did a lot of theoretical work and methods development because I could do quantitative work and mathematics sitting on my own because you didn't need much."
When it comes to being a mentor, Chakravarti says he cannot spot the rising stars right off the bat. Instead his only real concern is working with young investigators who harbor a love of science and the willingness to get their hands dirty. "When people come into my lab, what I want from them is the ability to work hard," he says. "I've never been good at figuring out who's going to be successful at that stage … so I give them a lot of independence because that's exactly what they will need," he says.
Times have definitely changed as the model of research has evolved along with scientists' understanding of the genome and their ability to do high-throughput studies, says Chakravarti. "As we've gone more [into] genomics, it requires a certain kind of discipline and a style of doing research that didn't exist in the more classical, -hypothesis-oriented, -single-investigator research, and that's sometimes a conflict," he says. "You sometimes have very bright people who are very great thinkers but they have no interest in running a thousand experiments, so I try to do my best to meld these two interests into the person."
Concepts first
One of the big-picture themes Chakravarti espouses is the idea of concepts over data. For him, ideas are the currency of choice. "Most of my mentors emphasized that there's no way as human beings we're going to be able to amass and store and recognize all the … information we need," he says. "What we need to do is extract concepts that we can repeatedly apply, and so it's that conceptual aspect of genetics that is probably the most interesting thing to learn and to convey."
One not-so-obvious challenge facing graduate students and postdocs today is the endless distractions diverting students' attention away from the work at hand. "There are many more research papers, true, but there are many more bad papers," Chakravarti says. "So if I read 100 papers in the 1970s, they were not very advanced, but believe me, the batting record of learning something was probably 80 out of that 100," he says. "Right now, there are lots of papers with great titles, and I read them and say, 'So what?'" He says the real challenge is in learning how to navigate so much literature. "My frustration is that because people are inundated with so many things, I find that honing their ability to focus on a problem is much more difficult," Chakravarti adds.
Misha Angrist, an assistant professor at the Institute for Genome Sciences and Policy at Duke University, completed his graduate studies with Chakravarti working on the genetic basis of Hirschsprung's disease. Angrist remembers his mentor as an idea man above all else. "In the 1990s, I used to walk into Aravinda's office and he'd be sitting there chain smoking and he'd have a yellow pad of paper in front of him and a pencil and he would sit there and think — using only his own gray matter, solve a problem on population genetics, with no computer and no massive datasets," Angrist says. "So he comes from this very pure intellectual tradition in genetics."
A great example of Chakravarti's dedication to research is the way he managed to avoid a pitfall that commonly threatens leading researchers upon their appointment to a high-level administrative position. Angrist remembers that during Chakravarti's stint as director of Hopkins' Institute of Genetic Medicine, his mentor was forced to make a tough decision. "At some point he just realized he had had enough and that this was not where his heart was and he wanted to go back to be able to devote more time to science," Angrist says. "Typically what happens is that principal investigators get administrative jobs and they become deans or program officers and their science often withers away, but I think Aravinda was just incapable of letting that happen."
Angrist credits Chakravarti with instilling in him an appreciation for the lineage of genetic research, but always with a sense of humor. "I remember sitting in a bar with Aravinda and Hunt Willard when I was in graduate school and they started quizzing all of us on who discovered X-inactivation and what was the most famous mutation that Thomas Hunt Morgan identified. It was genetics trivial pursuit," he says. "But of course they didn't see this as trivial at all. They saw it as 'you guys should know this, you should know whose shoulders you're standing on.'"
Naming Names
Here are just a handful of the many postdocs and graduate students who have passed through Chakravarti's lab.
Judith Badner
Badner took part in the genetic analysis of Hirschsprung's disease and the development of non-parametric linkage tests during her graduate days in the lab. Badner is now a psychiatrist and a statistical geneticist at the University of Chicago.
Sherri Bale
Before serving as president and clinical director of GeneDx, a company she cofounded, Bale was a graduate student in Chakravarti's lab, where she worked on segregation and linkage analysis of familial colon and endometrial cancer. She completed a postdoc at the NIH and stayed there for 16 years to design and run gene mapping studies.
Marc Halushka
Halushka was part of a team that found 874 SNPs in hypertension candidate genes and described their patterns of polymorphism in 75 individuals. Currently, he is a cardiovascular pathologist focusing mainly on studying diabetic vascular disease using tissue microarray technology and endothelial cell cultures.
Tara Matise
During the mid-1980s and early 1990s, Matise was involved with the effort to identify the CFTR gene for cystic fibrosis. She maintains her focus on linkage mapping as head of the coordinating center for the NIH collaborative initiative on population architecture using genomics and epidemiology.
Sarah Shaw Murray
Murray, who was in Chakravarti's lab during both grad school and her postdoc, is currently the director of genetics at Scripps Genomic Medicine. She sums up her time in his lab as seeking out and implementing state-of-the-art technologies to apply to various disease gene mapping projects. Murray was with Illumina before joining Scripps.