Francis Collins: 'More significant than going to the moon'

Francis Collins, a physician, is director of the National Human Genome Research Institute, a division of the National Institutes of Health that is responsible for the Human Genome Project. His research laboratory was responsible for identifying the genes responsible for cystic fibrosis, neurofibromatosis, and Huntington's disease. Collins was interviewed in May 1999 by CNN's Stephen Frazier for the Newsstand program.

CNN: [In the reading I've done I've seen comparisons] to the Manhattan Project or the effort to put a man on the moon. How would you position this effort in terms of big science that we grew up with?

COLLINS: I would say the Human Genome Project is probably more significant than splitting the atom or going to the moon. That sounds really audacious, doesn't it? [INTERVIEWER LAUGHS] But I think history, when we look back in a hundred years, will agree with that conclusion. This is an adventure into ourselves, to read our own blueprint, the consequences of that, for our ability to understand health and disease, and a whole variety of other issues that relate to humanity are profound, so I don't think it's grandiose or overstated to claim this is the most significant organized scientific effort that humankind has ever mounted, bar none.

CNN: And when you say organized, coordinated effort, that was the thing about the moon shot and about the Manhattan Project. All the talent assembled or coordinated from different locations. Is this on a par with those in that sense too?

COLLINS: It certainly is a project that's required enormous organization. The project was initiated in 1990, so we're nine years into it, it has been characterized, as a large effort of this sort must be, by precise milestones, so that you could tell whether it was working or not, deliverables, things that you have to do, and a certain ... attitude that you ought to be ambitious, and not just coast, not just do what you think you can do easily. The milestones there ... have therefore always been a little beyond what the people setting them thought they could do over a five-year period, but in every instance we have met or exceeded those goals, and it's been done for a budget that's significantly less than the original predictions. So here we have a federally funded project which is ahead of schedule and under budget in every way.

CNN: Which has to be a first.

COLLINS: I don't know if it's a first, but it sure is nice to be part of one that has that characteristic.

CNN: What are the accomplishments of the project to date?

COLLINS: So at the beginning, the goals of the genome project were to focus on building maps. You can think of those as mile markers along the chromosomes to sort of get you oriented. Not the precise single letter code of A, C, G, and T, but the mile markers. They are various types of maps; they are genetic maps and physical maps. Those were achieved about a year and a half ahead of schedule in 1994 and 1996, so we were able to get those markers in place to begin to roll up our sleeves and go after the hard part of reading out the precise script. In addition, we had goals about technology development, when we started this project in 1990 there was no way we could afford it, it was much too expensive to do this sort of large-scale activity. We had goals about trying to understand the DNA sequences of simpler organisms, bacteria and yeast and the roundworm and the fruit fly, because those will inform our understanding of the human. All of those have also moved ahead, ahead of the schedule that was originally projected.

CNN: Did those happen in parallel with the mapping that you described?

COLLINS: Yes, so there were various parts of the project that were focused on one or the other, and they were all moving along in parallel with careful oversight by a group of advisers that have from the beginning paid close attention to whether we were meeting these goals or not.

CNN: Well, then how do you feel now having achieved all those accomplishments under budget, ahead of schedule, by the sudden ramping up of competition that's been created by Dr. Venter [at Celera Genomics] and his pronouncements from the podium that he's going to race you guys to the finish and win?

COLLINS: I think competition is a good thing, it gets the juices flowing. And certainly the involvement of the private sector in genomics is something we have wanted and hoped for from the beginning, and has played a significant role in some of the technology development, for instance. So I would say for the most part, it's a good thing, to have this kind of dialogue going on and the competition that's arisen recently about sequencing the human genome in more than one place has stirred the pot and gotten everybody really fired up to see how quickly one can go.

CNN: Although you did sound pretty stirred up and juices flowing pretty hard before his arrival.

COLLINS: Well, indeed, the project was already pretty revved. It's not as if everybody was sort of coasting along saying, Oh well, let's take our time, there's no hurry here. This is the Book of Life, this is the blueprint of human beings, this carries within it the clues of every disease which we as human beings want to see alleviated, how could we do anything less than work extremely hard at this, so we were already pretty flat out. What happened with this announcement of interest in the private sector was to perhaps stir that effort up even one notch further, but it was already very revved.

CNN: What was the difference that it made? Did it bring more private dollars into technology development? Did it put more good minds on the subject? What happened?

COLLINS: I wish I could sort of look at a parallel universe where these factors could be [separated] and I can't. I think having this intensity of the effort right now has added to the enthusiasm for getting the sequence out there as quickly as possible, in the hands of working scientists. And I should say, that's been a commitment of the public effort from the beginning, the Human Genome project funded by the U.S. government and our international partners, particularly in Britain, have decided some years ago that all of the data ought to go into the public domain, so all of the sequence that's being produced on human DNA is immediately available in 24 hours on the Internet to anybody that wants, and it's been that way for years.

CNN: Of course, now you're getting to the heart of the matter here, which is that the private sector would like to hold onto that information for private use for some time, I gather. How should I characterize their point of view?

COLLINS: Well there's a difference in the goal. The publicly funded genome project has as its goal putting the information into the hands of scientists all over the world, immediately so that they can try to figure out what it means. We think of this as building the periodic table of the elements for human biology. Would it have been a good idea for the periodic table of the elements for chemistry to have intellectual property attached to it so that people couldn't work on it without signing some secrecy agreement or paying some royalty ...

CNN: You mean, if I patented hydrogen ...

COLLINS: Right, would it have been good if hydrogen and neon and helium had all had licensing and royalty stipulations before you could go to the lab and do an experiment? It doesn't seem like a good idea. It's not that different here, what we're talking about is this list, this table. The elements in this case are the human genes. There are 80,000 of them. The public effort believes the best way to figure out what they do and how to do something with that information to improve human health is to give everybody immediately and free access, no strings attached, and that is our determined goal. Now if I'm in the private sector and I'm producing this kind of sequence, it's all well and good to say, I'd like to give it away, but are my investors going to smile on that? Of course not. They're expecting a return, which means that at some level, the information has to be controlled, kept in a private database, patented, something so that it has value that allows that private sector company to recover their investment. That's different model. That may be a model that works pretty well if you're talking about something that's far down the road to discovery almost to the point of building a product that's going to help the public. That's what patents really [are] intended for. But my view, and this is the view of everybody involved in the public effort, is that the genome sequence, the Book of Life, is very far away from that kind of product and should not be constrained by these strings and attachments of royalties and patents and licenses and secrecy agreements. That will get in the way of understanding how it works and helping people, and that's what we want to try and see happen.

CNN: Some of the people we've been talking to have been explaining that very quietly the courts have applied standards for ... the delivery of a patent that they've used on other matters until now. ...

COLLINS: Non-obviousness, yeah. Novelty, utility, and non-obviousness are the three. ...

CNN: And they said, some of the work being done with genes seems to meet that standard, so a patent may be okay. So there's a legal framework that disagrees with you, it's going to be very difficult to change that thinking I should think.

COLLINS: We are in an odd circumstance. The patent laws were written in advance of this revolution in genetics. The patent trademark office views their role as taking the existing laws and trying to figure out how they apply to genetics. Without necessarily worrying too much about whether a decision that they make one way or the other is going to be good for the public. They don't see that necessarily as their job. Their job is interpreting the law. And so what it does often come down to is [that] the courts try to decide was this going outside the boundaries of what our founding fathers had in mind when they put together patent laws. I mean, patents were designed for a particular purpose, and don't get me wrong, I think patents are actually a wonderful benefit to the public when applied to the right circumstances. If Genentech had not been able to file for a patent on that TPA gene, we probably would not now have this drug, which is very useful in heart attacks and strokes and probably has saved many lives, because they wouldn't have been assured that it was safe to invest $500 million in developing this drug, and not had their market taken away the next day by a competitor who invested nothing. So that's an appropriate utilization, most people would say of an intention of patent law to genetics. On the other hand, if you're talking about a stretch of DNA, maybe a whole chromosome's worth, whose sequence you do not understand, whose function you have not determined, to claim that as a topic for intellectual property, to file a patent on it or to put it on a secret database, that doesn't seem to me to fit with what patents were supposed to do. That doesn't provide an incentive to develop a pro ... a product, it actually prevents that development, and this is where we have a conflict between what the law says and the way the patent office interprets it, because they're just interpreting the law, and what most people view would be good for the public. And the courts will have to settle this, and unfortunately that's going to take years and years.

CNN: And as a voice for the public, as you said a minute ago, you think it's better for the public that this information not be fenced off or closed up.

COLLINS: Well think of research as a road. You start on the road with very basic information, and way down there, far down the road you get to a discovery that's going to rise to a product that's going to benefit the public. The word has to be maintained, so one can understand the need for a tollbooth here and there in order to be sure that happens. But to put all the toll booths right at the beginning is going to make people not want to travel on the road, and that's what we're facing with all of this effort to try to fence off information about the human genome, whether it's the sequence or information about variations between your DNA and my DNA, which is another area that many people are trying to patent or render secret ... how genes function in the very basic way before you really understand it well enough to apply for patent protection. All of those things, it seems to me, have been overly considered as intellectual property, which we would have dreamed of doing 20 years ago, and in the long run I think we need to think carefully about whether that's helping or not. I suspect it's not. The goal of the publicly funded genome project is to move swiftly to get all of the information into the public domain. Once the sequence is in the public domain, there it is, and we can get past this silly chapter of trying to claim large amounts of it as if they belong to somebody and then really reward people who want to go in and find out how it works, as opposed to just cranking out the basic sequence in a language we largely don't understand yet.

CNN: Well, that, that's a question that the public would ask. When you've completed the map, your work is done?

COLLINS: [LAUGHS] I think there is a tendency when a gene gets discovered for breast cancer or diabetes or heart disease, to say, oh, the disease is now cured. There's even more global tendency to say once we have read out the whole human sequence, all three billion of these letters that make up our instruction book, then we'll understand how it all works and we'll know exactly what to do for every disease. That isn't true. I wish that were true, because we'd be very close to having the answers to everything, but it will take years, decades of intense research to understand what this book is telling us. We'll understand some parts of it sooner than others, we already understand some parts of it pretty well, but to really unravel all the mysteries is going to be years of very hard work all over the world, it will require a lot of research funds and it won't happen if people can't get access to the basic information that underlies it all which is the sequence.

CNN: Because the sequence is patented or large chunks of it are patented.

COLLINS: Right.

CNN: Okay.

COLLINS: Or it's in some private database, where unless you pay an exorbitant subscription fee or make some deal about future intellectual property rights, you can't see it. That just doesn't make sense.

CNN: To go back to the time issue and the work that lies ahead, in an earlier life you discovered the gene associated with cystic fibrosis. Is there a cure, and how many years ago was that, and what has happened since then?

COLLINS: It's been almost 10 years since the cystic fibrosis gene was found in a collaboration between my research group and one in Canada, and a lot has happened in those 10 years. We had no clue in 1989 what the basis of this disease was. We knew it ran in families, but we had no idea what was wrong. Now we know very precisely what's wrong, there is an error in a molecule that transports water and salt outside of cells in the lungs and in the intestinal track. And out of that have come more than a hundred patients who have been tried on various gene therapy protocols, with some evidence of chemical improvement in their circumstances, but so far no clinical benefit, and we now see ourselves with three new drugs this year that are in clinical trials that are based on understanding how the gene works, which we never would have come up with if we didn't know how the gene works. That being said, we haven't cured the disease yet. Ten years of very hard work have not yet led us to the outcome we all hoped and prayed for, which was to eradicate this disease because we'll know exactly what to do for it. I think there is considerable optimism that time is coming, I have that optimism perhaps in the next decade we will cure this disease. But one shouldn't underestimate how challenging that is, and every disease is going to have to be attacked a somewhat different way, it's not one of those things where when you figure out how it works with cystic fibrosis, you'll immediately know how to cure Alzheimer's disease too.

CNN: Right. Which just underscores the point you made a moment ago, that discovery of the gene, creation of the map is not the completion of the therapy.

COLLINS: Exactly. Finding the gene, hooking it up with a particular disease gives you immediate insight into what the actual molecular problem is. It gives you almost immediately the ability to predict who's at risk for that disease, and in some instances that itself can be life saving, if you know for instance you're at high risk for colon cancer, well, you go in and you get screened for that disease and you pick that up while it's still easily treatable, and that's a home run. That's terrific, that's what we hope for. But not all diseases allow you that kind of intervention, there are many steps that you have to follow then before you can harvest from this wonderful information about the gene, how to put that into practice in the medical arena, but you can't do that harvest if the gene information's not available to you.

CNN: I think on a cocktail party level people resist the idea of anyone patenting genes because it's special, it's sacred, really, and can't belong to any kind of private entity that might hold onto the information for private gain. But your explanation is actually one based on intellectual precedent and other examples of shared knowledge.

COLLINS: Yes. I think it probably doesn't get us very far to take patents and try to argue this as a moral issue. It's really a legal construct and its intention is to benefit the public. So it seems to me when you're arguing whether or not a particular patent is appropriate, you ought to ask, Is this going to benefit the public or not? And if you take a gene that is already connected with a pharmaceutical product, like TPA, for instance, or Arithropolytin, you can, I think, argue that a patent in that circumstance provides an appropriate incentive to develop a product the public wants, and therefore it's a defensible way to go. If on the other hand, you're talking about something that is miles away, you know, from a pathway that even leads you to product that you could identify, then it doesn't seem to fit that general principle of what patents were all about. I think we do better when we argue the concept on that particular grounds then if we get down this pathway of saying these belong to all of us, how can somebody own my genes. Well, you know, people have filed patents on natural products for a long time. Things they get out of the soil have been patented, in fact that's part of the process, because it allows somebody to be rewarded for having developed the ability to purify something that may have been there all along, but in such trace quantities that you couldn't use it. patents applied to DNA, you can think of in some of the same breath. At the same time, you've got to really think about how much utility and novelty and non-obviousness, the big three for a patent, ought to be considered necessary for the patent on a gene, and my sense is we haven't been nearly stringent enough in the last few years in the way those criteria have been applied when it comes to genetics. ... It seems to get protection for something that really doesn't involve that much novelty or non-obviousness or utility and the consequence of that is this tangled mess of intellectual property rules, of royalties and patents and licenses, that you can't do an experiment until you've hired 10 lawyers and done a year's worth of work to make sure you're not breaking somebody's rules.

CNN: Infringing a patent. I don't want to get too terribly theoretical here, but I read the article that you drew us to, talking about the tragedy of the anti-commons, and we've done some work on CNN about the tragedy of the commons. I thought that was an interesting pun that she [the author] came up with there, and maybe only economists knew what that term was that she [COLLINS LAUGHS] sort of subverted and reversed ...

COLLINS: Yeah, it was a bit new to many scientists, this whole terminology.

CNN: Were you clear on what the tragedy of the commons was? What did she mean the anti-commons?

COLLINS: Well, Becky Eisenberg, who's a very smart legal scholar at the University of Michigan, in considering where we're headed with this effort to attach intellectual property to very basic information has raised this as a tragedy of the anti-commons, because there is so little shared property in a circumstance where everybody needs access to it. So it's not a commons, where you have a collection of basic information that everybody would like to tap into, which can create its own set of abuses, and that with the original tragedy of the commons, but it's an anti-commons, where you have for instance, the human genome, owned in various pieces and various parts by 20 or 30 different groups. The most interesting experiments are not going to be the ones where you study one gene at a time, they're going to be the ones where you study all the genes at once. But if in order to do such an experiment you have to come to some legal agreement ...

CNN: And financial agreement.

COLLINS: And financial agreement, with 20 or 30 different entities, there's a profound disincentive to even bothering. That's the tragedy of the anti-commons that she's pointing to. And she gave as an example what's happened currently in Soviet Russia, where you would have thought with a thawing of the marketplace environment, that you would see bursting forth in all the storefronts, a lot of new activity, and in fact the storefronts are all shut up, and the only activity is happening at kiosks on the sidewalk. The reason being is that each one of those storefronts has about 20 different legal agreements about who's allowed to make decisions about how they're to be used, and nobody can get together and agree on that, so nothing happens. Is that what we want for human biology? Is that what we want for the future of medicine? That kind of tangled entrapment of a whole host of inhibitors that keep people from doing really interesting experiments to understand how the human genome works. I don't think that's a place we ought to go.

CNN: Do you think that because you're interested in the basic science, or do you too feel this sense of urgency that there are diseases out there that need treating, that there are people with conditions out there who need therapies?

COLLINS: I'm a physician. The reason I got interested in genetics was because of my sense that it is involved in virtually every disease and it gives us a window to understand a long list of conditions that we currently describe but don't understand how they come about. I spent too many days in a clinic talking to families about diseases that I can't do anything for, not to be motivated to try to move us, if I can in some small way, into a better world. And if that revolution gets trapped [CHUCKLES] in its own mistakes, when it comes to making this information which is so fundamental and necessary unavailable to the bright scientist somewhere, somehow who has an idea, then what have we done here? This ought to be about encouraging that, not getting in its way. And while I celebrate the contributions that can be made from the public sector and the private sector, working together to make this happen, the pharmaceutical industry is a big, important part of this as well, I really do worry about a scenario where something as fundamentally important as the sequence of the human genome, the Book of Life, might end up tied up in a private database by a single company, whose data release policies change from week to week. Is that a scenario that we, the public, not just even in the U.S., the whole world, are going to be comfortable with? I'm not comfortable with that.

CNN: Now one last question about methodology. Since there has been some legal framework to establish the right for patents, my question is whether the methodology that they're using up there in this huge rush, this speed, is employing a conceptual framework where they're only really getting incomplete information. I guess the question I'm gonna ask is, If patents are being awarded by the patent office, are they doing work up there that deserves a patent? Are they learning enough about the gene, or are they just finding out those little EST tags, and are they just getting partial data, and getting a patent on incomplete information?

COLLINS: Well, again, we come to the question of how much novelty, utility and non-obviousness ought one to require? And the National Institute's of Health has studied this issue and has just issued a set of guidelines about patenting and research tools that argues you ought to set a high bar, that sketchy information derived from incomplete sequence that's been subjected to a computer search and nothing else, ought not to be sufficient to grant patent protection on that particular stretch of DNA, it's just not enough. However, it's not an international standard, and the patent trademark office has not been acting as though that's their standard, and so there is a bit of a gold rush going on where companies that think they may be able to get themselves in a competitive position by filing patents are doing so, and I can understand that is probably in their best business interest, although it may be in the long term a bit of a deterrent for medical research from their own perspective, it's a way to stabilize their own economy and that's an understandable circumstance. It should be pointed out there are several ways that one could sequence the human genome. The way that is being pursued by the public effort is one that we know works. We've already done 17 percent of the human genome sequence, 10 percent is absolutely finished, another 7 percent is in the last final steps. When we break it down into bite size pieces, sequence those, sort of like a page out of the book, and then go on to the next page. The strategy which [---] is producing is a different one. It basically takes the whole book at one time and puts it into the shredder, and then tries to reconstruct what the book must have looked like. Those who have studied this from the point of view of modeling, whether it will work or not, for the human genome, are almost universal in their conclusion that it will not, that it will leave a lot of gaps and misassemblies, that you'll end up with a lot of information, but it won't be connected together the way that it actually is in the human genome itself, and that is another significant difference between the way the effort is being carried out. The publicly funded effort is not interested in ending up with a sloppy draft. We're interested in ending up getting it right. This is the Book of Life. Do we want to settle for something that's got a lot of mistakes, gaps, and screw-ups in it? No, this ought to be something we'll use for decades or centuries. If we do it right, we only have to do it once. So that is a very significant commitment of the publicly funded effort, to be sure that we end up with that kind of product, even though the final stages of cleaning it up and getting the gaps closed may be tiresome and may not be very profitable, in the long run that's what we have to have, and we will not back away from that goal.

CNN: If I heard you correctly, you're making it sound as if it is more efficient, too, because efficiency is the word that comes up when businesses start talking.

COLLINS: I think the method which we're using is proven, we've tried it for three years in a pilot effort, it is efficient, we know it will work. The other strategy is risky, it may turn out that it works better than some predictions, it probably won't work perfectly. I'll be interested to see what happens when the dust settles, but it is certainly likely that on an efficiency basis, if what you're talking about is trying to get the final product, that it will be less efficient than the method that we've been following for the last three years.

CNN: You're speaking in very rational terms here and, and very creatively expressing thoughts, but I'm surprised you're not kind of howling at the injustice, if it strikes you that these are guys racing down to the patent office with kind of an untested methodology, and incomplete results, but they're getting, they're getting a stamp on the application for a patent.

COLLINS: What I would like to see is instead of this concept of a race, which I think has been played up heavily in most of the descriptions of this, is that we could figure out a way in which each of these enterprises has something to contribute. Celera has certainly indicated a willingness to release some of their data. Exactly under what conditions hasn't been worked out yet. And it may be that it won't be done in a fashion where it could be merged with the public data, and if it were, and they're not going to patent all of it, of course, they're going to patent blocks that they think are particularly interesting ...

CNN: Profitable.

COLLINS: ... and profitable. Then by merging the public and the private data sets, we might end up further along a year from now then either would be on its own. That's certainly going to be happening in one direction. You can think of this as a one way mirror, after all, all the data the public effort is producing goes immediately into Celera's computers, of course it is, it goes in anybody's computers who wants it, it's free. [LAUGHTER] They'd be crazy not to use it. So that will be happening. But there's some chance of some information coming back the other way, and my hope is that will happen. We have a pilot effort underway right now to try that out for the sequence of the fruit fly, Drosophila, where there is a memorandum understanding, a legal document about how the public effort and Celera are working together on that, and I'm looking forward to seeing how that goes. Obviously that sequence also has some intellectual property value. And it's clear that sequence is not being released by Celera until they've shared it with their private partners for a matter of a few months, but ultimately it will end up according to this agreement, in a public database where everybody can use it. Let's see how that goes, I hope that goes well. But, ultimately again, I do share what you're saying. Some sense of anxiety and concern about hundreds of millions of dollars going into an effort which is attempting to tie up really critical basic information about human DNA, information that most of us think ought to be accessible to anyone.

CNN: Would I be correct in fearing, as a layman, that all of your work could be hijacked in a sense?

COLLINS: Hijacked is a strong word. I do think one of the motivations for the public effort in getting all of the sequence information into the public domain and information about human variation as well, is once it's there, it's there. It can't be hijacked after that. That is a strong reason to move as swiftly as we can. I should say right now that most of the people I talk to in the private sector feel that same way. Pharmaceutical companies who are depending on the sequence of the genome and this information about variation between individuals to help them develop the new designer drugs of the future, are also quite alarmed about a circumstance where they might not be able to use that without a complicated arrangement of patents, licenses, and secrecy agreements. And the strongest evidence for that is this recent unprecedented development, where 10 pharmaceutical companies got together and put up their own money to speed up the process of building this catalog about human variation, which the genome project had already started, but which they thought could go even faster with their contribution. And so it is going to speed up dramatically in the next two years. And you know what they decided? They decided they didn't even want to see that data until everybody else did. So they're giving their money, but they get no private peek, no access, and all that information is being put free and clear in the public domain where nobody can patent it ... to be available to any scientist anywhere. Put in that context, this is not a battle between the ivory tower academics who are being unrealistic about this and the people in the private sector who are being entrepreneurial and trying to claim it all. It's really much bigger than that, and if you look across both those sectors, what you will find is the vast majority coming to a strong agreement that this information ought to be out there without obstructions, and that a small entrepreneurial group, hoping to make some money off of it, trying to move swiftly in a different way. it is a small part of the sector, to be sure, but obviously one that's gotten a lot of attention.

CNN: Dr. Collins, thank you very much. Have I failed to ask anything that's important in this conversation?

COLLINS: We haven't talked about what happens after the sequence is done and is this whole thing over. We could say a word about other organisms, so for instance, if we're going to understand how the human genome works, if we're going to be able to look at this periodic table and figure out what it's telling us, we will want to have the sequence of other organisms to compare to, and already that's happened with bacteria and yeast and the roundworm and soon the fruit fly, but the one we really want next is the mouse. The mouse sequence will also, if it's going to have value, will need to be accessible to everyone. The human genome project has already started that project, and we will be scaling it up in earnest in a couple of months, and that sequence will also be immediately released. Celera has announced their intention to also do some mouse sequencing, but it made it very clear that sequence will kept in a secret database that nobody will get to see unless they make some sort of a deal. So once again we have a circumstance where information of profound scientific value, if it's going to be available, will basically have to come from public funding.

CNN: Thank you. Nice chat. Thank you.

COLLINS: Well, I appreciate you asking questions that are fun to answer.

Back to the top