The New Manhattan Project: Q&A with NYGC Scientific Director Robert Darnell
November 27, 2012
November 28, 2012 | Rockefeller University physician scientist Robert Darnell, son of Lasker Award winner James Darnell, is the newly appointed scientific director of the New York Genome Center (NYGC). Darnell gives his first in-depth interview to Bio-IT World editor Kevin Davies on his reasons for taking the job and his initial goals for the high-profile center.
Bio-IT World: Bob, did you have much involvement with the New York Genome Center before your appointment as scientific director?
DARNELL: Yes. My involvement with the center came through my own lab’s struggles in bioinformatics. We’ve been successful on a do-it-yourself scale. Tom Maniatis moved to Columbia, we were both dealing with these issues. He introduced me to [NYGC Founding Executive Director] Nancy Kelley, who already had the center up and running to some degree.
The bottom line is: I’m a user of genomics and a physician, and I’ve spread my wings among these two fields. My group—about 25-30 people—started doing genomics fairly early, around 2006, when the first [next-generation sequencing (NGS)] machines were coming out. We had to devise our own ways of approaching it.
We were much more interested in the ‘dark matter’ of the genome—there’s a huge amount of the genome transcribed that we don’t know anything about. We were working in this area way before it was popularized. We did our own bioinformatics, and then we hired a few people. What I saw in my own group and at Rockefeller University, where I’m a professor, is an unsustainable status quo. You can do a bit [of NGS] on your own but it’s not adequate.
And so what appealed to me about this [opportunity] was the idea we could have both a core platform to do high-throughput sequencing and bioinformatics, standardized approaches for my lab, Rockefeller and New York in general. But superimposed on that is the larger idea that what we really need is computational science—using the value of machine learning, network analysis and so on. All of that is stuff we’ve done custom in our own laboratory… I’m an egalitarian in science—making a level playing field and an opportunity for everybody to harness the power of genomics is very appealing to me…
The bigger part of [this opportunity] is trying to harness this stuff for the benefit of mankind—for clinical medicine. The challenge there is several-fold: one is dealing with the mass of information, making collaborations happen, making things CLIA-certified etc. Once that data is together, there’s low hanging fruit to be plucked away—exomes, coding sequence—but there’s the potential of machine learning, identification of information in non-coding RNAs—largely unconserved and have things we don’t understand in them—that needs mathematicians and computational scientists and powerful machines.
That’s the real hope here—building enough bioinformatics, using the core to make an incredibly useful facility for doing personalized medicine and clinical genomics, but to also do it at a level that is not just the simple, straightforward, trivial, but do it in a way that is deep and sets the standards going forward for how this should be done. So this opportunity has a practical arm and a scientific discovery arm.
On the clinical side, how will this advance the treatment of your own patients?Well, I’m sort of one of the world’s ‘experts’ on a very rare group of diseases… These diseases are really interesting, rare, but provide the potential for cracking open insights into really big questions. These are diseases where people have elements of naturally occurring, successful tumor immunity. That’s triggered when they make an immune response against proteins in their cancer that are normally brain-specific proteins. That triggers effective suppression of the tumor and that immune response, similar to lupus, can cross into the brain and cause an auto-immune disease. We figured early on we needed to harness molecular biology to get at pathophysiology, and that led us to discover the target genes in these disorders. They turn out to be brain-specific gene regulatory proteins, but not things regulating transcription at the DNA level, but regulating RNA. So the brain has its own system for regulating this dark matter that no other cell in the body uses, and presumably it relates to normal brain function and brain disease.
So we’re interested in using high-throughput sequencing in these patients to ask what’s different about their tumor, their immune response, and maybe in some cases think about what’s different in their brain. It’s a very interesting issue, which we’ve broadened and generalized into a question of what makes the brain different from a liver, kidney or spleen cell? It’s really genomic information, but as far as we can tell, it’s not the machines that turn genes on or off at the DNA level but the processing and regulation at the RNA level. That’s partially what makes neurons so complicated and special.
We’ve developed technology to look at this, which turns out to be very relevant to fragile X mental retardation, Lou Gehrig’s disease, which has blossomed in recent years as an [accepted] RNA-mediated disease, neurodegenerative diseases, pediatric disease… all of these are things that our lab has touched on. So you can see why we’re interested in being a user of the NYGC, driving the science forward, how to handle all the data, how to discover things in the data and tie it back to human disease.
You’re a physician scientist running a big lab. When did you start to think seriously about becoming the scientific director of NYGC?
I fell in love with the place early, through Nancy, because I loved the vision of the place. Selfishly I wanted it to happen—I was an unbiased observer for a while, volunteering my time trying to make the center happen. The way the search worked initially was that we originally thought [the scientific director] should be someone outside of New York…
It could have worked out, but in some ways that was the wrong idea. We’re talking about a consortium that’s never been done before. This is going to be the new Manhattan Project! This will involve a new generation of clinical medicine, with science and computers all fitting together. It’s never been done, and here it’s being done as a consortium of all the major institutions in New York. Trying to pull that together might in fact need a New Yorker!
I had been comfortable watching this from afar. As Nancy likes to point out, I’ve been associated with more than half the founding institutions. That’s a lot of fun, I’m very comfortable interacting with these people, and I collaborate scientifically with a large number of people at many of these institutions. And I’m a New Yorker—I grew up in New York.
So frankly I wasn’t thinking about [the position] until they approached me. My initial instinct was, wow! It clicked right away. NYGC is interested in clinical medicine, computational science and biology. That’s exactly what my lab does—and that’s exactly where my lab is stuck. The struggles in my lab are a microcosm of the struggles of all labs in New York and science in general, dealing with the revolution in genomics.
Will you move your lab into NYGC?Yes and no. I’m remaining a Rockefeller professor and a Howard Hughes Medical Institute Investigator. My wet lab does a lot of mouse genetics, and we’re not going to set that up at the NYGC. But the genomics there will be very powerful, so that part of my lab will move.
Will there be other senior investigators conducting research at the center?Definitely. There’s the intention to make this an academic tower of power. If we’re really going to figure out how to do the next generation of genomics and apply it to clinical problems, it’s a research problem. It can’t be done by a computational scientist alone, or a clinician, or a classical genomicist alone. It demands this sort of consortium. I’m the nidus for recruiting future investigators.
Do you envision NYGC leaning more towards basic or clinical research?We need to bring clinicians in who have access to patients and want to sequence their DNA. We have to educate them about why they might want to do that, how to do it. Once it’s brought into the center, we want to do more than just sequence and spit it out. We want to help them learn. It’s not as trivial as it sounds. Even when you hit a single-gene defect, it’s not easy to figure out how that translates into a treatment. It requires a mixture of interested scientists—you can call that basic [research]—and computational people to deal with the data. I see a large amount of bioinformatics and computational science, with some interested users, some of whom will interface with clinicians, and some (like me) will be clinicians.
Apparently you’re quite the triathlete. Will you be jogging the five miles from your Rockefeller lab to NYGC?Yes, I could swim too! I’m very excited… My history is a little peculiar, because I’ve done two careers. An MD/PhD, I see these patients with rare disease, I teach at Memorial Sloan Kettering as an attending physician, and I run a very serious basic science lab that’s out of the box—that’s why I’m an HHMI investigator. One of the questions that comes up is, how will I do this additional thing? I’m good at multi-tasking. If I wasn’t, I wouldn’t be here. The triathlon is the same genre.
Your father is a Lasker Award winner, and your daughter seems to be following in the family business. I read she come second in the Westinghouse Science competition a few years ago?Yes, they almost got it right! She finished at Yale undergrad and is now a brand new baby graduate student at Harvard in Molecular and Cellular Biology, with David Liu.
A fun personal story: My dad was a left-leaning scientist when I grew up, we used to march against the Vietnam War. I became very interested in science, like my dad, but wanted to be a humanist. As a 14-year-old, I saw him as an ivory tower scientists working on RNA, so ironically, I went to great efforts to distinguish myself in science and stayed committed to idea of doing clinical medicine, trying to do basic science on human disease. I ended up working on one of the world’s rarest diseases—perineta degenerations—figured out a way to discover the basic patho-physiology and the genetics, and it turns out I discovered these multi-gene families of brain-specific RNA proteins! Even I could not believe the way fate set this up. I ended up at Rockefeller, where my dad is, and we’re doing RNA genomics collaborations.
Do you have any concerns about the challenges you face at this stage?I would just say that it’s going to be a grand adventure in bringing clinical science forward to something that is more than just intellectually interesting, but clinically powerfully important.
Bio-IT World: Bob, did you have much involvement with the New York Genome Center before your appointment as scientific director?
DARNELL: Yes. My involvement with the center came through my own lab’s struggles in bioinformatics. We’ve been successful on a do-it-yourself scale. Tom Maniatis moved to Columbia, we were both dealing with these issues. He introduced me to [NYGC Founding Executive Director] Nancy Kelley, who already had the center up and running to some degree.The bottom line is: I’m a user of genomics and a physician, and I’ve spread my wings among these two fields. My group—about 25-30 people—started doing genomics fairly early, around 2006, when the first [next-generation sequencing (NGS)] machines were coming out. We had to devise our own ways of approaching it.
We were much more interested in the ‘dark matter’ of the genome—there’s a huge amount of the genome transcribed that we don’t know anything about. We were working in this area way before it was popularized. We did our own bioinformatics, and then we hired a few people. What I saw in my own group and at Rockefeller University, where I’m a professor, is an unsustainable status quo. You can do a bit [of NGS] on your own but it’s not adequate.
And so what appealed to me about this [opportunity] was the idea we could have both a core platform to do high-throughput sequencing and bioinformatics, standardized approaches for my lab, Rockefeller and New York in general. But superimposed on that is the larger idea that what we really need is computational science—using the value of machine learning, network analysis and so on. All of that is stuff we’ve done custom in our own laboratory… I’m an egalitarian in science—making a level playing field and an opportunity for everybody to harness the power of genomics is very appealing to me…
The bigger part of [this opportunity] is trying to harness this stuff for the benefit of mankind—for clinical medicine. The challenge there is several-fold: one is dealing with the mass of information, making collaborations happen, making things CLIA-certified etc. Once that data is together, there’s low hanging fruit to be plucked away—exomes, coding sequence—but there’s the potential of machine learning, identification of information in non-coding RNAs—largely unconserved and have things we don’t understand in them—that needs mathematicians and computational scientists and powerful machines.
That’s the real hope here—building enough bioinformatics, using the core to make an incredibly useful facility for doing personalized medicine and clinical genomics, but to also do it at a level that is not just the simple, straightforward, trivial, but do it in a way that is deep and sets the standards going forward for how this should be done. So this opportunity has a practical arm and a scientific discovery arm.
On the clinical side, how will this advance the treatment of your own patients?Well, I’m sort of one of the world’s ‘experts’ on a very rare group of diseases… These diseases are really interesting, rare, but provide the potential for cracking open insights into really big questions. These are diseases where people have elements of naturally occurring, successful tumor immunity. That’s triggered when they make an immune response against proteins in their cancer that are normally brain-specific proteins. That triggers effective suppression of the tumor and that immune response, similar to lupus, can cross into the brain and cause an auto-immune disease. We figured early on we needed to harness molecular biology to get at pathophysiology, and that led us to discover the target genes in these disorders. They turn out to be brain-specific gene regulatory proteins, but not things regulating transcription at the DNA level, but regulating RNA. So the brain has its own system for regulating this dark matter that no other cell in the body uses, and presumably it relates to normal brain function and brain disease.
So we’re interested in using high-throughput sequencing in these patients to ask what’s different about their tumor, their immune response, and maybe in some cases think about what’s different in their brain. It’s a very interesting issue, which we’ve broadened and generalized into a question of what makes the brain different from a liver, kidney or spleen cell? It’s really genomic information, but as far as we can tell, it’s not the machines that turn genes on or off at the DNA level but the processing and regulation at the RNA level. That’s partially what makes neurons so complicated and special.
We’ve developed technology to look at this, which turns out to be very relevant to fragile X mental retardation, Lou Gehrig’s disease, which has blossomed in recent years as an [accepted] RNA-mediated disease, neurodegenerative diseases, pediatric disease… all of these are things that our lab has touched on. So you can see why we’re interested in being a user of the NYGC, driving the science forward, how to handle all the data, how to discover things in the data and tie it back to human disease.
You’re a physician scientist running a big lab. When did you start to think seriously about becoming the scientific director of NYGC?
I fell in love with the place early, through Nancy, because I loved the vision of the place. Selfishly I wanted it to happen—I was an unbiased observer for a while, volunteering my time trying to make the center happen. The way the search worked initially was that we originally thought [the scientific director] should be someone outside of New York…
It could have worked out, but in some ways that was the wrong idea. We’re talking about a consortium that’s never been done before. This is going to be the new Manhattan Project! This will involve a new generation of clinical medicine, with science and computers all fitting together. It’s never been done, and here it’s being done as a consortium of all the major institutions in New York. Trying to pull that together might in fact need a New Yorker!
I had been comfortable watching this from afar. As Nancy likes to point out, I’ve been associated with more than half the founding institutions. That’s a lot of fun, I’m very comfortable interacting with these people, and I collaborate scientifically with a large number of people at many of these institutions. And I’m a New Yorker—I grew up in New York.
So frankly I wasn’t thinking about [the position] until they approached me. My initial instinct was, wow! It clicked right away. NYGC is interested in clinical medicine, computational science and biology. That’s exactly what my lab does—and that’s exactly where my lab is stuck. The struggles in my lab are a microcosm of the struggles of all labs in New York and science in general, dealing with the revolution in genomics.
Will you move your lab into NYGC?Yes and no. I’m remaining a Rockefeller professor and a Howard Hughes Medical Institute Investigator. My wet lab does a lot of mouse genetics, and we’re not going to set that up at the NYGC. But the genomics there will be very powerful, so that part of my lab will move.
Will there be other senior investigators conducting research at the center?Definitely. There’s the intention to make this an academic tower of power. If we’re really going to figure out how to do the next generation of genomics and apply it to clinical problems, it’s a research problem. It can’t be done by a computational scientist alone, or a clinician, or a classical genomicist alone. It demands this sort of consortium. I’m the nidus for recruiting future investigators.
Do you envision NYGC leaning more towards basic or clinical research?We need to bring clinicians in who have access to patients and want to sequence their DNA. We have to educate them about why they might want to do that, how to do it. Once it’s brought into the center, we want to do more than just sequence and spit it out. We want to help them learn. It’s not as trivial as it sounds. Even when you hit a single-gene defect, it’s not easy to figure out how that translates into a treatment. It requires a mixture of interested scientists—you can call that basic [research]—and computational people to deal with the data. I see a large amount of bioinformatics and computational science, with some interested users, some of whom will interface with clinicians, and some (like me) will be clinicians.
Apparently you’re quite the triathlete. Will you be jogging the five miles from your Rockefeller lab to NYGC?Yes, I could swim too! I’m very excited… My history is a little peculiar, because I’ve done two careers. An MD/PhD, I see these patients with rare disease, I teach at Memorial Sloan Kettering as an attending physician, and I run a very serious basic science lab that’s out of the box—that’s why I’m an HHMI investigator. One of the questions that comes up is, how will I do this additional thing? I’m good at multi-tasking. If I wasn’t, I wouldn’t be here. The triathlon is the same genre.
Your father is a Lasker Award winner, and your daughter seems to be following in the family business. I read she come second in the Westinghouse Science competition a few years ago?Yes, they almost got it right! She finished at Yale undergrad and is now a brand new baby graduate student at Harvard in Molecular and Cellular Biology, with David Liu.
A fun personal story: My dad was a left-leaning scientist when I grew up, we used to march against the Vietnam War. I became very interested in science, like my dad, but wanted to be a humanist. As a 14-year-old, I saw him as an ivory tower scientists working on RNA, so ironically, I went to great efforts to distinguish myself in science and stayed committed to idea of doing clinical medicine, trying to do basic science on human disease. I ended up working on one of the world’s rarest diseases—perineta degenerations—figured out a way to discover the basic patho-physiology and the genetics, and it turns out I discovered these multi-gene families of brain-specific RNA proteins! Even I could not believe the way fate set this up. I ended up at Rockefeller, where my dad is, and we’re doing RNA genomics collaborations.
Do you have any concerns about the challenges you face at this stage?I would just say that it’s going to be a grand adventure in bringing clinical science forward to something that is more than just intellectually interesting, but clinically powerfully important.