 Good morning everyone and welcome to the opening session of our meeting on the future of bioscience graduate and professional training. I'm Janet Weiss, Dean of the Graduate School here at the University of Michigan and in that role one of the key stakeholders in the future of graduate and postdoctoral training. Like many of you I've been reading for at least ten years about the problems and the challenges sometimes called the crisis associated with graduate and postdoctoral training across the biological sciences. Like many of you I've been listening to our own students and trainees talk about their struggles. Like many of you I have attended countless panels and meetings talking with colleagues around the country bemoaning these struggles wishing that someone would do something about them. Many extremely smart people a number of them sitting in this room have thought long and hard about the challenges and a number of insightful analyses have been published in recent years. Like many of you I have now had enough of dwelling on the problems. We understand pretty clearly what the issues are and it's time to act. I am indebted to the organizers of today's session because their agenda is moving forward. What can we do as a community to provide effective and appropriate training for the next generation of PhD scientists and how can we ensure that the astonishingly successful bioscience enterprise will continue on its trajectory of discovery and innovation for future generations. To address this agenda many voices are needed and many voices are represented here today. More than 110 institutions are represented among those who have registered for our full two-day meeting. So welcome to all of you who have traveled from far and wide to join us today. And we also welcome the guests in the audience who are not registered for the full conference but are here just for this morning's plenary session. Before we begin I just want to note that your program has instructions for connecting to the internet here in the auditorium. To avoid overwhelming our Wi-Fi if you would connect only one device each we would appreciate that and your fellow participants will appreciate that. I do want to acknowledge those who have contributed to today's events. First of all I want to thank the generous sponsors of today's meeting from the University of Michigan, the offices of the president, the provost, the graduate school and the medical school and the medical school's endowment for the basic sciences. And the best program at Wayne State University is also a sponsor. Second, I really want to thank the hardworking and dedicated organizing committee of faculty, postdocs and students who have designed this two-day meeting. David Engelke, Prashasnika Gailot, Peter Hitchcock, Laurie Isom, Nathan Matheson, Mary O'Rearden, Egon Rangini and Michelle Swanson. They are all listed in your program but I encourage you to introduce yourselves to them. They have things on their name badge that say program committee and make sure to give them your ideas and suggestions and ask them any questions that you might have during the meeting. I'm now delighted to introduce to you Dr. Mark Schlissel, the 14th president of the University of Michigan. As a physician scientist himself, with an MD and PhD in physiological chemistry, President Schlissel is deeply knowledgeable about the issues facing principal investigators. As a former department chair and dean of biological sciences at the University of California at Berkeley, provost at Brown University and as University of Michigan president, he also understands the stakes for institutions of higher education in finding effective strategies going forward. Please welcome President Mark Schlissel. Thanks Dean Weiss for that kind introduction and for your work along with the medical school and the organizing committee to make this event possible. It's a pleasure to welcome you all here to the University of Michigan for this important nationwide summit. I very much appreciate your commitment to addressing the issues we're going to discuss today and tomorrow. The world we work in right now looks very different than the one in which I began my own career. My first experience in research was volunteering to work in a research lab when I was an undergraduate at Princeton in the late 1970s. I took photographs of chromosome preparations that my mentor Raju Kuturelapati used in his efforts to map genes onto human chromosomes using somatic cell hybrids. Now the entire human genome has been sequenced. The challenges to the decades-old training paradigms that many of us were educated in are well documented. For myself, I got my first NIH grant when I was 34, 35 years old and I felt incredibly old and my progression very delayed. I had done an MD and a PhD, a residency and a postdoc and I thought the world was ending since I was 35 and everybody was leaving me behind. And now that's somewhat of a youngsters age to break into the grant system. Today's summit will help us examine some of the approaches that are already in practice at local and national levels. As well as hearing interesting proposals and training models for the future. We have leaders here from more than 100 institutions, government, university, private sector institutions across the country. We've got an enormous opportunity to hear a spectacular breath of input that's needed to best inform consensus building and the creation of innovative practices and policies. I view this work as essential. There's so much at stake and it comes at a critical time for society. The biosciences have become a fundamental target for discovery in many fields once far distant from biology. Engineers, chemists, physicists, computer scientists, mathematicians and others are helping us unravel fundamental secrets of life and contributing to the development of treatment for disease. Social, environmental, epidemiological research are teasing apart the many and varied influences that have impacts on life science that range from the biology of living cells to the quality of human life. But despite the greatest potential for discovery and history, the system as we know it is no longer supportive of the best science. It's incredibly difficult for young investigators and I put quotations around the term young since we're talking about those in their later thirties and early forties to break into the system of federal funding. It's not just a question of how we're training the next generation of biomedical scientists and how many we should be training, but also how we're using increasingly constrained federal, foundation and university resources and whether we're producing the greatest amount of the most important science possible at this level of investment. Is it possible to make discovery more efficient? I've made the future of biosciences a focus of my own work here at Michigan as well. We've examining challenges at a symposium last fall that was part of my inauguration. I agree with Harold Varmus who led a panel at that event called the mismatch between supply and demand in the life sciences amalthusian dilemma. I've since convened a group of leading faculty here on campus to propose solutions at the level of our university, not just to issues surrounding the training of biomedical scientists, but to the optimal organization of the life sciences at a research intensive university such as ours in the current era. I'm delighted that momentum for conversations around these issues is swelling over the last couple of years. The AAU presidents recently took up this question last month in D.C. And I know the University of Wisconsin just did work on this topic earlier this year culminating in a workshop just a few weeks ago. We need to build on these conversations over the coming days and hear what you have to say. I look forward to learning your conclusions and ideas for the future. Thank you very much and I now turn it back over to Dean Weiss. So at our first session, we have two extremely distinguished academic scientists to set the stage for our conversations over the next two days. I'm going to introduce them both and encourage you to read fuller biographies in the program. They'll each speak for about 25 minutes and then we will open up to questions from the audience about both presentations. Gregory Petzko is the Arthur J. Mann Professor of Neurology and Neuroscience at Weill Cornell Medical College and Director of the Helen and Robert Appel Alzheimer's Disease Research Institute. He's been elected to the National Academy of Sciences, the Institute of Medicine, the American Academy of Arts and Sciences, and the American Philosophical Society. He's currently President of the International Union of Biochemistry and Molecular Biology and he has a TED Talk that's been downloaded almost 700,000 times on the coming epidemic of neurologic disease. And he'll speak today about the postdoc experience. Following Professor Petzko will be Keith Yamamoto. He's the Executive Vice Dean of the School of Medicine and Vice Chancellor for Research at UCSF in addition to his appointment as Professor of Cellular and Molecular Pharmacology. He has also been elected to the National Academy of Sciences, the Institute of Medicine, the American Academy of Arts and Sciences, and the American Academy of Microbiology. He chairs the Coalition for Life Sciences, which works to foster public policies that advance basic biological research and its applications. And he will speak today about Next Gen Nurturing for Biomeds Next Gen. So I will now turn it over to Mr. Petzko. Well, thank you very much. Can you all hear me OK? Good. I'm going to use the lavalier mic because I think better when I'm moving around. It's great to be back in Ann Arbor. Some of you may not know this, but I actually lived in Detroit for five years when I was starting my career. I was an assistant professor at Wayne State Medical School. And I used to come out here all the time for football weekends. They were beating Ohio State then. Anyway, that's another story for another day. It's great to be back because I know a lot of people here. One of my best friends in the world, George Kenyon, is here, for example. So any excuse will do. But in particular, this is a subject about which I care very deeply, and one which I'm hoping we will be able to have a productive dialogue about. With that in mind, I want to congratulate people at the University of Michigan for undertaking this particular event. When you do things like write the report I'm going to tell you about, and nothing happens, it's extremely frustrating. When something like this takes place, it's very gratifying. And I hope we'll serve as a catalyst for some really significant developments. About 15 years ago, the National Academy of Science issued a report on the postdoctoral experience. It suggested, among other things, that there was a significant set of problems that needed to be addressed with the way postdocs were trained, mentored, and the way their careers developed. The recommendations were published. A number of institutions, I think one of the leading ones is Keith's institution, UCSF, implemented some very significant steps towards improving the postdoctoral experience. But a lot of things didn't happen, and a lot of places did nothing. And so four years ago, the combined academies of the United States, the National Academy of Sciences, the Institute of Medicine now called the National Academy of Medicine, by the way, and the National Academy of Engineering. For the first time ever, these three academies jointly commissioned a study to find out what had happened since that 2000 report and to make additional recommendations and see if they couldn't be implemented. The study is here. It was published a little while ago. You can find it online here, as you can see. And it was called the Postdoctoral Experience Revisited, where ideality and reality, in fact, diverge. I was the chair of the committee. It was a terrific group of people that put this report together. It took us a lot longer than we expected it would. One of the main reasons for that, as you'll see, was that it was almost impossible to find all the information we needed in order to put the report together. We did the best we could. I had a couple of ideas that were not agreed with by the rest of the committee. The most significant of which was I wanted the report called something else. I wanted the report called the Invisible People. And the reason I wanted to do this was because one of the most striking things we found as a committee was that on many campuses throughout the United States, postdocs were, in fact, invisible. We set out to simply count the number of postdocs in various scientific subjects throughout the United States. And we found to an alarming degree that many institutions didn't know how many postdocs they had. In fact, a number of them couldn't even estimate two within a factor of two. There's a heck of a difference between 500 postdocs and 1,000 postdocs. There were a number of reasons for this. One, postdocs aren't always called the same thing. And a given campus, you might have postdoctoral research associate, postdoctoral fellows, research associates, et cetera, et cetera. The other thing was they're funded in a wide variety of ways. And of course, if you don't keep track of where the money's coming from, you don't keep track of who has it. So this is a significant problem. But it was even worse to find out that almost no institutions, again, UCSF being a notable exception, actually tracked the outcome of postdoctoral education. Actually made an effort to find out where their postdocs went and what happened to them and to make that information available. The paucity of data, as Mike Tidalbaum can demonstrate by, because he spent a good bit of time staring at this with us, was just appalling, just appalling. All right. So the first point I want to make, one of the first points we made as a committee, was there's nothing quite like the word postdoc. How many people in this auditorium have a doctoral degree? Please raise your hand. Congratulations, you are all postdocs. It turns out I've been a postdoc for 46 years. OK, admittedly, I should have published more. Nevertheless, what does this say? Look, if you're an assistant professor, you're an assistant professor. If you're a graduate student, you're a student. If you're a postdoc, you're a time interval. By assigning a name like that, we are tacitly admitting that we have no obligation to the people who have that title. The fact that we call them what we call them says an enormous amount about what we expect and what they should expect from us, which is nothing. So we decided we would define the term a bit differently. We'll get to that in a moment. As I said, our biggest problem in dealing with this report was the absence of data. If nothing else comes of this report, but that institutions are shamed into finding out how many postdocs they have, where they come from, how they're supported, and where they go when they're finished, I would be satisfied with the report. If that doesn't happen, then we're in a lot of trouble. Because if we don't even bother to count them and to make publicly available what happens to them, what does that say about our feeling of responsibilities for them? It is a cliche that the number of postdocs has grown exponentially. And it's really true, especially if you look at recent years. It's an amazing chart that we have here. This is the number of postdoctorates. Here you have the number of doctoral recipients, which, as you see, dropped for a while, interestingly enough, right around the time of the NIH doubling, then rose again and now has started to decline slightly. The number of foreign postdocs is increasing dramatically. The number of domestic postdocs is increasing at about the same rate, and the net result is a steep upward climb. But this goes all the way back to 1980. This is not a recent phenomenon. It's just gotten worse in recent years. There are fields where you wouldn't think this would be an issue. Computer science is one of them. You would think naively that computer scientists would have lots of options, and that therefore there would be very little pressure on computer science to do postdocs. In actual fact, looks what's happened in recent years. And this is before the financial crisis. Since around 2000, the number of computer scientists doing postdocs has risen steadily, and in recent years it's climbing at a rapid rate. The number joining tenure track faculty positions has actually dropped substantially. If this is happening in computer science for Christ's sake, imagine what it's like in other subjects. Why is this happening? Well, one reason it's happening, of course, is that there are lots of opportunities for people to get postdoctoral positions. The striking thing about a postdoc, and I think Keith will say more about this, is that it's actually often easier to get a postdoctoral position than to get a good graduate position. Normally, when you go higher up the ladder of a career path, it should be tougher each time, and only better people should be able to get positions. This is in fact exactly the opposite of the situation we have at the moment. And one of the main reasons that's driving this is the paucity of research funding and the difficulty of sustaining a research career. Young faculty, midterm faculty, old faculty, all need to publish lots of papers in order to get grants and keep grants, and with this grant success rate approaching zero, okay, I mean it really does seem to extrapolate to zero, doesn't it? Then with this situation, the pressure on labs to get bigger, to keep people longer, and to bring people in with more skills so that you can do more detailed elaborate studies which are needed to publish papers in boutique journals which you need to get the grant, you understand the kind of vicious cycle that's generated here. This is particularly true in the life sciences. It is true in others to some extent, but the life sciences as you will see is in fact the 500 pound gorilla that drives this entire conversation. Most postdocs are supported by federal grants. This is the number that's supported by federal grant. This is non-federal sources. Look at what's happened to traineeships since 1980. Flat as a pancake, okay? This is a disgrace. The fact that the federal support of science has paid so little attention to traineeships and fellowships is in fact catastrophic for our scientific labor force. It appears to be an irreversible trend. Something ought to be done about it. But as a result, of course, now increasingly postdocs are supported by grants. In all of this, the NIH is the 500 pound gorilla. If you look at what we've been able to find about the percent of graduates entering postdoctoral positions, you can find that in the social sciences today, about 35% of graduates enter a postdoctoral position and that's very unusual. It used to be less than 20%, but it's climbing in all fields. Engineering, it's gone up to around 35, used to be 20. The humanities, it's even starting to creep up. I mean, that's astounding right there. But look at the life sciences where it's never dropped below 60% and now is hovering right around 65 to 70%. So the postdoc is the default, especially in the life sciences. And this enormous life sciences research enterprise that we have created in this country is feeding people into the postdoc pool at an amazing rate. The end result is we don't have anything like a true labor market. A true labor market eventually tries to equalize supply and demand. We know that there's a relatively limited demand for postdocs in academic positions. This should have the effect of reducing the postdoctoral labor force. In fact, as I've shown you, postdoctoral labor force is increasing exponentially. Why don't we have a true labor market? We don't have a true labor market because the number of postdoctoral positions is not set by the availability of jobs after the postdoc. The number of postdoctoral positions is set essentially by the NIH minimum postdoctoral salary. Divide the total amount of NIH support by the minimum postdoctoral salary and you have a rough idea of how many postdocs the system can support. And as long as this is low and as long as there is no true labor market, you will be able to snap up a huge number of people with postdoctoral positions, low-paying postdoctoral positions, without regard to what happens to them when they finish. Let me show you, actually, I'll go back to the slide for one second, all right? This is what happens to five-year cohort of people. That is to say five, six years after they graduate. In 1973, in the life sciences, more than 50% of all of them went into tenure-track positions. Now, that figure is 15% and dropping, okay? You can see that in nearly all other subjects, it's remained close to flat, okay? But especially in the life sciences, there is no labor market at all. And this, despite the fact that the number of academic jobs, which was considered for many years in the primary purpose of a postdoc in the life sciences, has in fact been going down considerably. Okay, so where do they go after their PhD? Well, 37% of US citizens and permanent residents who have a PhD, regardless of subject, all right, end up doing a postdoc. That's pretty extraordinary. For temporary visa holders, it's an even larger number, 45%. So when I say the postdoc is the default, it really is. And doing this is actually incredibly expensive for the individual who does it. If we look at what happens to you, five years after you receive your PhD, and we compare the salaries earned by people who go directly from PhD into some other job with the salaries of people who went into a postdoc, well, of course, most postdocs now take five years or longer, so what you see is the result of that. People doing postdocs in pretty much every profession are earning under $50,000 a year, five years after their PhD, on average. Whereas people who don't are earning anywhere from $64,000 a year up to close to $100,000 a year. In other words, we're taking an enormous number of people who are extremely well trained, well enough trained to get decent positions that pay a living wage, and we're putting them into a position that doesn't pay them a living wage, and that sets them way behind their fellows in their salary attainment throughout their career. These were some of our findings. They suggested, and I think it's fair to say, right, Mike? The system is not working the way it should. I think you had more colorful language that you occasionally used. As I said, it was very hard to figure out what people ended up doing after they finished their postdoc. These were the best numbers we were able to come up with. They're real soft numbers, okay? 23% end up in subkind of tenure track faculty position, and another 20% end up in non-tenured academic research or teaching, all right? We were assuming, and we didn't know whether we should assume this or not, we're assuming that these teaching positions do involve mentoring research on the part of undergraduates or even in some cases, high school students, but we couldn't find that out, and another 24% roughly end up doing some kind of industrial or government research, but that still leaves a third of all postdoctoral recipients who are doing jobs that have no research component whatsoever. One of the things we did as our committee went about its job was to interview a large number of people who do a large number of things. This range from science journalism to patent law to policy and so forth. And every time we did this, I always asked the same question. I said, do you have a PhD? And they nearly all answered yes. And then I said, did you do a postdoc? And they nearly all answered yes. And then I said, was the postdoc useful to you in the job that you have now? And they nearly all said no. It became clear to us on the committee that the purpose of a postdoc should be for advanced training in research to prepare you for a career in research, ideally with some level of independence and that if that is not the career in which you want to go or in which you are likely to go, doing a postdoc isn't just buying time. It's actively, as I hope the numbers have shown you, probably a mistake because you're burning some of your peak creative years and peak earning years. John Maynard Keynes memorably said in the long run we're all dead, okay? But it turns out if you did a postdoc and are looking for an academic job, it happens sooner. What do I mean by that? I mean if you look at people one to five years since their PhD, from 1993 to now, about 50% were in some kind of non-tenure track position. Now it's approaching 40%, all right? But six to 10 years since their PhD, that number's dwindled to 25% and 11 to 20 years out, it's now down here and more than 20 years, this is me, right? Now you're down around 10%. And by the way, this number here, the light blue line, that's unemployed. So there's no question that in the long run outcomes from scientific research training very frequently involves something other than a scientific research career. So these were our findings basically. What did we decide should be done about it? Now I'm gonna go through the recommendations relatively quickly because Keith's gonna talk more eloquently and I think maybe with more depth as well about some of the things that really need to be done. But I'd like to set the table for him a little bit. Our first conclusion was we need to change what we call these people. Instead of just calling them postdocs, which is as I said, a time interval, we should call them postdoctoral researchers to indicate at least what their function is. All right, and we defined this, that a postdoctoral researcher is an individual who has received a doctoral degree or its equivalent and is engaged in a temporary and defined period. Note the word temporary please. Of mentored, note the word mentored, advanced training in research. And that if that doesn't satisfy the demands of the position you intend to go into, you should not do a postdoc, it's not simple. Maybe you need to go to law school, maybe you need to go to business school, maybe you need to do an internship, maybe you just need to get into the labor force, maybe you need to go to journalism school, there are a lot of possibilities. But many of those may be most of the non-academic, non-industrial research positions. Not only don't require a postdoc, it is a waste of five plus years of your life. All right, now the question that came up over and over again was are we training too many PhDs? And after thinking about this a lot, we came to perhaps a surprising conclusion. So if you look at it, these are the number of PhDs being produced. Again, this is the life sciences since 1966. You'll notice the extremely rapid growth in many fields, but particularly again the life sciences. This is math and computer science. If we have too many postdocs, an obvious conclusion to draw is when we must be having too many graduate students, therefore we need to reduce the number of graduate students. But the committee, after much thinking and debating, decided that that would be in fact the wrong conclusion to draw. We decided that there are not too many graduate students because a PhD in the sciences is actually fabulous preparation for a hundred different careers. These range from venture capital to public policy, from journalism to law, okay? I could list several dozen off the top of my head. Nearly all of these are superbly prepared for by the kind of in-depth research training under a good mentor that you get at places just like this one. The problem isn't that. The problem is that from that PhD, many, a third or more perhaps, of all of those PhD students should be doing something other than going automatically into a postdoc in research. It's not that we're training too many PhD students. You could argue we can't train too many PhD students. This country needs more scientifically aware literate people. But we are producing too many postdocs. And doing a PhD actually does you quite a bit of good. This is the unemployment rate among scientists and engineers. It is quite high since degrees with people with a master's or bachelor's degree. But although it's climbed up a little with people with a postdoc degree, basically it's stayed in the three percentage range for a long, long time, for 35 years basically. And the postdoc doesn't change this, okay, at all. This is the PhD. PhD is a good thing for you in life. It does really improve your job stability and your financial and other outcomes. All right, now all good things must come to an end. Although a lot of postdoctoral research supervisors don't seem to recognize that fact. So we did state, I hope, with some justification that it was very important that the postdoc experience should total no more than about five years. Although obviously this should be elastic depending on childcare and other issues that might come up. But nevertheless, after this time people need to transition into something more like the old research faculty positions or research staff positions that universities used to have in droves when we were younger scientists. And these were fabulous people who ran instrumentation, who helped train people in a variety of techniques who were the backbone of large multi-research group projects. Most of those positions have disappeared. Why I'm not entirely sure, but I am entirely sure we need to bring them back. All right, Keith will talk much more than I'm going to about the importance of getting people the right information at the right time of their career so that they don't automatically do something like going into a postdoc that they shouldn't do or don't need to do. This is a crucial point, but more on that in a moment. We spent a lot of time debating the issue of salary. We decided that postdoctoral salaries were simply too low. But we couldn't agree on what they should be. Finally, with much debate and some dissent, we came up with a rough figure that the minimum postdoctoral salary should be raised from where it is now, which is down around 40,000, depending on the subject, to up around 50,000, again, depending on the subject. I wanted it higher. Some people didn't want us to make any recommendation at all. But I think it's pretty clear that if the people we have as postdocs are as good as we say they are, we should be paying them a living wage. The argument, of course, is that we don't need to do this because there's a training component to what they're doing. I think this argument is spurious. And finally, and this was our most heartfelt recommendation, please, let's at least find out what the numbers really are. Most of the numbers that I've shown you are soft. We did the best we could to come up with them. But if we don't even count them, if we don't worry about what happens to them when they leave, what does that say about what we care or don't? So with that, I'll stop here, turn the floor over to Keith. I have to go fly to the Middle East for reasons that are unimportant, so I have to leave after the break. But happily, another member of the committee sits here in the audience and will be delighted to participate, I'm sure, in discussing the report in more detail. I also hope that you will take what we said to heart because this is a big problem. Okay, Keith. This is Gary. Great. I'll wander on the other side of the room for you. So let's get started. Put the first slide up. Good, that's almost the last slide. Quick review article. Here we go. Listening to Greg's always fun, unless you're the speaker after him. And he makes comment after comment, foreshadowing what I'm assigned now to be saying and giving the impression that we plan this thing in a tightly coordinated way. Greg got in late last night, I saw him as he was loading his slides, so we'll see how well we can get done here. We have a next generation of bioscientists who we have for a long time throughout this training, long extended training period that Greg referred to. And what everyone in the room is here struggling to do is to try to think about whether we're doing it right and changes that we might make to do it better. So I'll discuss that. There's been a plethora of reports and studies, hand-wringing, accusations and so forth that's been made on this topic. The last two of which that at least I think are beginning to show some impact are the 2012 study from the advisory community of the director of working group on the biomedical research workforce on which I served. Data slides that you saw from Greg that were labeled BMW 2012 were not things that he found in his car, but were from this report. And of course the very fine report that Greg shared that he's just reviewed with you. In a way though, I think that we need to sort of back up and ask what are we talking about? Greg talked about the definition or lack of definition of postdoc. I really like that. Another way to look at it is what's evolved over time in the definition of postdoc and this was actually drawn from with a historical section of his report. In the 1920s and 30s when this whole thing, the postdoc experience began really as a way to try to keep up with German science in this country, it was defined this way. An individual who manages day to day of research operations of a laboratory allowing faculty time to supervise other research, teach, obtain research, funding and attend to other administrative tasks. Sounds fantastic if you're a faculty member. Today, for a version of the definition that Greg put up, which is really agreed upon by the consortium of NIH, NSF and the National Postdoctoral Association, an individual who was engaged in a temporary and defined period of mentored advanced training to enhance the professional skills and research independence needed to pursue his or her chosen career path. Pretty different. Now what's a little bit distressing about this is that one could take this same slide and relabel the temporal periods this way. It's a little bit overstated, but there are faculty who would say that a postdoc individual who serves the faculty member to free up time to be able to do other stuff. And a postdoc view of course is, please help me get trained so I can take your place in the research community and be able to carry on work which has really never been as an exciting potential that we're all enjoying right now. So there's that issue. And then there's the place that we probably, I think many of us would agree, that we need to go. That this, that there should be separate career tracks right now which are conflated, both in definition and in title, that need to be separated and distinguished. And that there should be professional research staff whose job it is to manage the data of any research operations in a laboratory, et cetera. And there are trainees, postdocs who are engaged in advanced mentored training to be able to move in as the next generation of practitioners in our profession. So what we have here as the chain gang foreman, the uncool hand Luke said, is a failure to communicate. Although he said it with a much better accent than I did. So in a way this is a struggle that we're all up against and we need to try to come to terms with. So we can start by asking, what is it that we're trying to accomplish? And if you read the report and engage in discussions on this topic as I know that you all have, some would say, well, we need to right size the workforce. And I know that some of the people on the NIH committee that I served on and probably some of the people on Greg's committee entered that with that, maybe even as a primary goal. We need to improve training for postdocs. And as you just heard and as you all know, that's pretty low bar. You could call postdoc education an oxymoron. We need to really define what it is that what postdoc education is, is there a need for it? How do we accomplish it? And so others might say, we need to define and address this wonderful array of career paths that we now enjoy in the biomedical sciences, biosciences, educate students, trainees to what those options look like in an effective way. Again, right now a pretty low bar compared to what we're doing and not doing. Is that it? Or is there other stuff? And I would argue that there is. What about these two? Could we actually define and address educational missions and goals for graduate students and postdocs? Are they the same? Is it a continuous track? Are they different? And I'm not aware of any programs that have really stepped back to those sort of first principles to ask those questions when we think about the postdoctoral experience, what happens in the duration of training and so forth. And we really stepped back to ask, what is it that students and postdocs need to know to have in their pockets to be successful in this endeavor, whichever directions that they choose to practice it? And then what about this? Can we reestablish the excitement and joy in bioscience careers? I would argue that we have managed somehow to take one of the most exciting, important, enduring, intellectually rich, societally important endeavors, human endeavors, and make it into this ponderous, onerous, endless track that just seems incredibly unattractive. It wasn't easy, but we did it. So what can we do to maybe rebuild that? Because I would argue that some of the problem that we have is that we are losing at the front end some of the very people that are most needed to make this endeavor continue to thrive, to flourish in new ways is gonna be able to move forward. So let's see if there's anything that can be done there. Now what we just heard about in Greg's report is that there were proposed solutions to these three guys in white, but not to the last two. So I'm gonna sort of take these in turn and try not to over, just to give you a straight review article of what you just heard from Greg and looking at the proposed solutions. And I'll do that by coloring them with my own biases and proposals that just sort of sneak them in there like they were all along. And I'll tell you what was different about what I'll say. And then at the end, we'll try to come back to see if there's anything that can be done here. Because I think those two points are really important as well. So you just heard a bunch of this stuff but maybe you'll hear it in a slightly different way or colored with my own biases or suggestions. Increase postdoc stipends. Column stipends, not salaries. The postdoc is not a career. And so in the same, in keeping with that and a little bit differing from what the report said, I would argue that there should be benefits that should come with those stipends. Health insurance, parental leave and so forth, but not retirement. Retirement really does make it seem like it's a career. So those may be small differences but I think those nuances turn out to be important in the way that the establishment, academia, funding agencies and so forth, think about what the postdoc is and about how the postdocs themselves think about it. Greg urged us, I don't know what number of report this is that urges the NIH to be able to be more responsible and effective in his data gathering and reporting and so forth. And Greg's report does a very effective job of trying to not only urge that but provides a pathway. I'm gonna put a sharp point on one aspect of this that I think could be important and could be effective. And that would be sort of for trainee tracking and outcomes reporting. I would suggest, and I know it's been suggested in numerous committees, including the workforce committee, to assign a unique identifier at grad school entry or even before, in fact, an increasing number of trainees come out of their bachelor's degrees and go off and do something else that's relevant to work in graduate school or work in a company and so forth. So it'd be great to have those data if they could be collected in that way. Anyway, assign a unique identifier, let's say your genome sequence, at graduate school entry or undergraduate completion and then use that identifier to track you for their entire career. And so essentially that number would then, or DNA sequence would track you and you'd use it to log in and log out of every one of your transitions, whether it's within the institution, going from graduate student to postdoc, associate professor to professor, lab to lab or moving to a new institution. So then the NIH and other interested institutions would have those data that would just follow the number and not worry about title codes, all the things that have confounded the NIH's data gathering efforts here to four and led, if you looked at the big table in the workforce report, the number of postdocs in the U.S. was between 38 and 68,000. It's not so good. I mean, twofold differences in molecular biology don't count for much, but if it's the number of postdocs in the country or your weight or your salary, it's a different story. And then use the information, not only in the ways to be able to test questions of great urgency for being able to establish where we are in this field and what kinds of experiments are working and not working, but also to report outcomes data in a transparent way at the level of institutions, at the level of individual laboratories so that prospective trainees who are shopping for an institution or prospective faculty members, prospective postdocs who are shopping for an institution or shopping for an individual lab can ask where the people have gone that have come out of there. It'd be great to have an obviously longitudinal, so you could say, what have people gone in the last 10 years and in the last 10 years before that? But right now, almost anything is better than what we, better than the nothing that we currently have. And those kinds of data would be enormously useful as students look at laboratories to go to, to say, oh, here's a lab where sort of an equal number of people are going to academic research and to other areas that sounds good for me. Other people would say, no, I'm not interested. I just wanted to do academic research. I'm gonna just find a laboratory where everybody succeeds in going in that direction and so forth. Faculty mentoring, that's something that Greg's report addresses, not something he's spent a lot of time on, so let me spend a couple of minutes on it. Mentoring is not the same as supervising. Point that's made nicely in the report. A supervisor could be viewed as readily simply defined, maybe oversimplistically defined, but it's not unfair to say that a supervisor's job is to make sure that you understand what his or her needs and expectations are in ways that you'll meet them. And a mentor is someone whose goals are to understand what your needs and expectations are so that he or she can meet them. Pretty big difference. It's not that one person can't be both, definitely can. There are terrific supervisors who are terrific mentors, but they're not the same thing. The report points out accurately that most faculty have not received any formal training in mentoring. Maybe there should be a requirement, not just a suggestion, that faculty members are taught how to mentor. This is a field that celebrates the fact that we operate on sort of a word-of-mouth folklore tradition because it means that different people can practice their science in the ways that they want. But it also means that getting started can be strenuous. The transition from postdoctoral system professor is very strenuous, and a part of that strenuousness is just trying to find your way to becoming a mentor, what does it mean? Something I tried for in the workforce report and failed to muscle through resistance from certain members of the committee require a statement of mentoring plan. This is something that the National Science Foundation requires, on all research grant applications that include payment of trainees. The reason that was given, astoundingly enough, for not doing this at the NIH, is that this is almost a quote, the NIH regards research and training as completely separate endeavors, so there shouldn't be any mixture. Despite the fact that the research is done by trainees and we're trying to train people to do research. And then finally, mentoring is not the same as cloning. A lot of people when they begin think they're doing good mentors, it's the people come out looking just like them. The same values, same experimental approaches, wear the same clothes, tell the same jokes. And with that comes a stigma that gets attached to anybody who wanders away from that, towing that line. And that stigma, of course, is warping the way that the workforce is coming out because students are concerned about the judgments from the faculty, their own advisors or others on the faculty, if they even act like they're gonna go do something else other than what the faculty look like. It's human nature that if you think like I think, I think you're smart, but we're paying a price for it. Here are options, I can discuss this nicely. There's a set of tools that have been developed that I think are effective and are recently being used, will be discussed extensively in the breakout sessions and the sessions in this meeting. And I suspect that other ideas will emerge. The report suggests that it's important to begin to familiarize trainees with the range of options at the beginning in the first year of graduate school. And the goal of that is, in my view at least, to produce a training endeavor that has students confident about choosing a career track on the day that they receive their PhD. So why is this important in the context of some of the things about the workforce that Greg was discussing? The current situation is this, and that is that there is a terminal from which career options emerge at the level of a postdoc. So that many PhDs are told that there's other stuff to do besides being an academic researcher, but they're not told enough to be able to decide to do any of them. There's a big space there. And so many, as the report says, and I totally agree, default from a PhD to a postdoc. Because of the way the workforce operates in laboratories, as Greg said, and I totally agree, it's actually easier to get a postdoc than it is to get into graduate school. And so there's a very heavily traveled pathway here. And then the postdocs hope that they then finally acquire enough information to be able to make this career decision and then learn that the postdoc can be substantially more isolating and endeavor than his graduate school. And so their great danger, of course, is that the postdoc terminal becomes a terminal postdoc. So this is a major problem and it's fueling this very large number of postdocs that Greg described and that are in the report. So this would be the goal. The goal would be a PhD hub with career options spokes. That's just a few, and Greg said, and I agree, there are dozens of directions that one can use PhD training in the biosciences to be very effective in these other careers, effective in a way that helps the endeavor move forward. So this is not a jobs program. It's essential that we be able to populate each of those careers and others as well with people trained at the PhD level in the biosciences, not all of them, but some, in order to be able to move forward in this endeavor. So if this were to work and people were equipped, that isn't sufficiently informed to make these decisions on the day they got their PhD, there would be some important outcomes. One is that we would have an informed and empowered graduate student body that would be able to make decisions based on information that they really own. It would require, of course, that we undo the stigma that I mentioned a while ago, and there are several programs, including the one at UCSF, that has a specific focus on altering that kind of faculty culture. But if that were true, then the students would be able to make these decisions and would go and occupy these various career options. It would, for the first time, define a postdoc education imperative because as Greg said, it would mean that the spoke, the career spoke, that extends from the PhD to one of these career options that are becoming an independent investigator would be called a postdoc. So now there would be a specific explicit mandate to do something for postdocs beyond the PhD that would help enable them to become independent investigators just as law school would help to enable trainees to choose that spoke to move into these various areas that require a lot of degree. So the postdoc market would then be scaled, sized according to the perceptions of the people who choose that spoke. And I'm interested in the academic career or the bio med research director career, PI, the PI alternative. And these are the number of postdocs would decline. Now this wouldn't be dramatic, it wouldn't be overnight, right? I mean, I'm tremendously powerful, but I doubt that this would happen overnight. So it would take a long time. But we're instituting a program like this at UCSF, but we know that postdocs will continue to arrive at UCSF for a long time from other institutions that haven't had the benefit of being provided with enough information to make the decision. And so we will not only have to be defining that postdoc educational imperative, but we will also have to include the postdocs in any programs that provide this information about career tracks. But the postdoc market would then begin to adjust to the appropriate perceptions of those going down that spoke of what the career options are in that one defined career alternative. Similarly, the other options would be market driven, the perception not only the interest of the students to move into those areas, but their perception of what the need is. And the need will change as the PhDs begin to go into those markets with more bigger and bigger numbers. We're finding this already in a program that we have at UCSF that has internship periods for graduate students, 10 week internships to go spend every day showing up in a law office or communications at the public television station to talk about how to communicate science to non-scientists and so forth. And some of those companies that agreed to sponsor internships, I'm sure, said, ah, it's a service. What the heck? We like UCSF, they're nice people, and so we'll throw in some money and have some intern come by and the intern gets there and they're fantastic and they're getting job offers during their internships. So the market will adjust as they see what value comes from having people with this kind of training on board. It needs to be pointed out that as this succeeds, lab demographics will change and the traditional view of who it is that does the experiments in laboratories just become increasingly addicted to postdocs and it swayed the views even unconsciously of faculty members who think, oh, postdocs are the people who do our experiments. And when people talk about the workforce problem, the faculty, what faculty is thinking is who's gonna do my experiments? And the rest of the world is thinking, ah, do we have this right? So if we can make this change, the impacts, the outcomes would be substantial. A predict. So that's the addressing of these guys. But could we also define and address education missions and goals for graduate students in postdocs? And if we could, would it make a difference? Would they look different than what we do now and would that matter? And could we do something here? Could we do something to re-establish, re-establish the excitement and joy in bioscience careers? So let's have a shot at it. Here's a shot at it. Let's step back and think, what is it that PhDs from University of Michigan really need to have in their pockets to go off and succeed? And I would argue that it's a simple set of tools. Really simple. You still need specialization. And as the Pibbs program here, I know emphasizes, you want that specialization, but you want it immersed in a broad spectrum of good high quality science so that it establishes or demands of students a general literacy that as their problem or their interest move from one area to another, they're not just completely going into a cold. So specialized expertise, broad literacy, good. Biosciences are undergoing a transition, some would say have undergone a transition from being a descriptive endeavor to a quantitative one and increasingly everyone is going to need math skills, certainly an understanding of statistics and the capacity to ask and answer questions with computers. That wasn't the case early on, but it is distinctly the case now. And then what I would call tools of the trade, three simple ones. Really simple. Being able to identify an important problem, design an experiment to begin to chip away at that problem and decide which results to follow, which one's not to. And you learn those skills by iteration. So it's just like going to the gym and taking on a set of exercises that the first time you try it seem impossible. And in a couple of weeks you're doing three sets. And ask any graduate student here how they think that their thinking is improved from the first day in graduate school until the last day of the first year, just the first year. And if they haven't already thought about it themselves, they'll be knocked out by the change. So we live in a setting where iterating on those three things is all around us. Every time we go to a seminar, every time we read a paper, sit down at lunch to talk with a colleague about their experiments or yours, you have the opportunity to practice those skills. And pretty soon you think, I can do this. And then the working familiarity with career options. Now you look at those four bullets, turns out that they're not very complicated. There's nothing in there about becoming a world beater in some field about publishing cover stories and single word journals. What about postdocs? The first one is easy. There are various programs probably here as well that do drill down on management skills for the practice of science. The second one is hard. How do you train someone to be independent? So there are experiments that are being done. And we carry out both of these at UCSF. Kind of this bold and owned project development. This is something I started many years ago in my own lab. We now have a grant program for postdocs to develop an idea that is separate from the central focus of their laboratory mentors, which clearly won't be identified as Petsco, even though the person work in Petsco's lab, in which we encourage postdocs to think of some idea that's separate and begin to develop it into a research plan. Very often they do it by getting help from someone else at UCSF in our case who knows that field well, but it's really theirs. They really own it. They're really excited about it. Two of my postdocs, in fact, got jobs based on their central focus in my lab, but started their labs based on their owned projects. So there are ways, in fact, to work with trainees, to really show them what it is to develop an independent plan and begin to carry it out. And of course, as you know, there are fellows programs. UCSF fellows program, the Whitehead, Coltsman Harbor, other places have fellows programs in which PhDs are invited, fresh out of their degree to come to UCSF in this case, occupy a lab with four to six benches in it, given salary and startup money, equipment, and are permitted to write grants if they wish and carry out their own research. I think that these things are essential. I think they're not very complicated. In fact, I think they're so non-complicated that the potential outcomes could include a tremendous reduction in the training period. So the total training period from first day in graduate school to completion of training will be four to eight years, four, we know works because there are UCSF fellows and so forth that come right out of their PhDs and start. And I can't think of any reason why a list as simple as this should actually take longer than four years. So do we need to step back and rethink what it is that we're presenting to students and trainees and postdocs and realign, or maybe align for the first time, those training modalities with what the students and postdocs really need. So they really feel that they're getting served by this endeavor and not exploited by it. And they become happier. They look at the field in a different way. They finish their training earlier. They get into the workforces and the pet investigators. And this would be then potentially appropriate training for this next generation of bioscience leaders. I don't know of any other field that doesn't do this. I mean, can you think of any other profession that doesn't train its next generation of practitioners? You wouldn't want that of a neurosurgeon. Okay, law school. Okay, I wouldn't, I mean, I've had my own experience with lawyers the last couple of years. So this is the last slide. Why does all this matter? And if we look at, if I think about this in a really big picture way, I say it matters for this reason. And that is that here in the 21st century, the biosciences are sitting at an inflection point in which we've enjoyed for the last three or four decades, I'd say, a tremendous run up in our capacities to gather data, to describe biological phenomena and processes and molecules and organisms in detail that we never expected. And what it's revealed is that those molecules and processes and organisms are way more complicated than we thought. But the capacity to describe, the capacity to collect data is enormous. And this is true of course in society in general. Everyone's overrun with information, but it's been a good run. The problem is that we're at an inflection point. We're stuck, we're saying, who are we gonna do with all this data? And students and professors and citizens in general are saying increasingly, I give up. I don't want, I don't want any more stuff. I'm forgetting stuff at a way faster rate than I'm learning it. So we need to work our way through this somehow. The good news is that on the horizon we can see ways to get there. That we can achieve mechanistic understanding of those molecules and phenomena and processes and organisms in ways that allow us to be predictive to manipulate them, right? In ways that demonstrate that we understand them. We know there's a great distance between data and understanding. And everybody's fretting about the fact that the distance is increasing rather than decreasing. But there are ways to do it as scientists work together, come together in teams of people who know very different things, but decide to work on a common, a problem of common interest together and so forth. But to do that's going to require that we recapture the interests and excitement of this endeavor. We're able to demonstrate to people that it can actually make a difference. A difference not only to the kind of intellectual richness of our society as the knowledge base increases, but in ways that allow us to take on increasingly urgent problems in society. So we need biosciences trainees and we don't know how many, exactly what Greg said. So this is not the time to shut off the training of PhD bioscientists. We need biosciences trainees, prepared and excited to make fundamental discoveries, researchers, to communicate what it is that they're doing that's important, whether it's journalists, policy makers, politicians, God help us. And to implement this societal changes that will make a difference, whether it's in health and medicine, in the environment, in energy, in food and agriculture, all of those, all four of them, are facing issues that right now, we don't have the ways to be able to meet the crises that are coming at us. Not only that, we can't meet them one at a time. We have to meet them together. If we grow all the food in Michigan and Iowa and then use fossil fuels to ship them to places where people are hungry, we're dead. So we have to solve them together. And the good news is that in this 21st century of biology, which many people have argued, that the tools and concepts and technologies, methodologies of the biological sciences could provide the fuel to be able to do that. And so the impact of being able to step back and align our training with ways that we'll be able to accomplish these ends, will at the same time take on some of the problems that we're discussing of the workforce and all these other things, logistic things that we're talking about. So I think that's what we're up against. And it's the reason that this is an exciting, important endeavor. And it's fantastic that the University of Michigan has pulled this group together and this set of topics to discuss. And I look forward to a couple of really exciting days. So I have a question. Can you hear me okay? Yeah. I'm Vicki Prince from the University of Chicago. And I'd like to thank both of you for your excellent talks and thoughtful suggestions. My question surrounds international postdocs. So I came here 21 years ago as an international postdoc and the US system has been very kind to me. How long did it take you to learn the language? Oh, I'm still working on that. So there's still unprecedented opportunities for training in the US. So I'm a little concerned that by moving the postdoc hub to a PhD hub and better educating our US-trained PhDs so that they fully understand the range of career options open to them, that's gonna drive an even greater influx of international postdocs for whom maybe $40,000 a year is actually quite appealing. But many of them end up in that interminable postdoc cycle or eventually because they don't have sponsorship for a green card going back to their home country, perhaps to much more limited opportunities. And I'd just be interested to hear your thoughts about that. Yeah, Keith's passing the buck because that's a tough problem. It's partly what drove us to make a specific recommendation for the increase in the postdoctoral stipend was the hope that if that minimum was high enough it might tend to mitigate somewhat against this potential exploitation of cheap foreign labor like yourself. I'm not that cheap anymore. Well, you know, when it was a third world country, which it was when I lived in it. We'll tell it later. Yeah, but I think to some extent it may be that we worry about this too much because one of the great things about the United States is that every time we get fat, lazy, and stupid an influx of people from the rest of the world helps give us a good swift kick in the posterior. And this has been particularly true in the sciences in the last 20 years. And I'm not sure that we want to discourage it entirely. And I think the issue is always one of striking the appropriate balance. So I realize that's a fuzzy answer to a very important question. But that's about as far as we've gotten in our thinking on this subject. I think my answer such as it is, and I'd like Keith to chime in please if he has something, is that I don't want to treat this as a crisis because even a crisis in the making because I'm not sure that it is. I agree completely. Thank you. So but you probably disagree. Oh, okay. I'm Noko Tanesa. I'm Dean of the Graduate School at NYU School of Medicine. I have a question about one of the topics you mentioned. So there are a number of, so I guess we really want to encourage PhDs to not just go into default postdoc mode but start exploring career paths. But I'm wondering, so even those PhDs who got into the postdocs and got jobs, you asked them, did postdoc help? And they said no. But on the other hand, does the job market expect PhDs to have some postdoc experience? Because if their mindset doesn't change, even though that might not have helped them in any way, then postdocs will be competing for jobs with PhDs or PhDs who will be competing for jobs with postdocs. I mean, how would you take care of that problem? Like how would you convince the job market to say, look, PhDs are appropriate for you to hire? Well, I mean, I only have two answers to that. Let me see what Greg says. The first is that we're already seeing in our internship program that we're beginning to gain the interest and confidence in career areas that didn't even know they wanted PhDs and that they find the PhD level of people to be very exciting. And so I think with time in the areas where the postdocs are now expected, that can begin to adjust. And so I think that both things, you'll see a flex in the opposite directions that will reach some appropriate equilibrium. It'll take time though, you're right. Right now those expectations are there. We've got to deal with the real world as it exists right now and need to respond to it. Yeah, I agree. I think if you look at those fields that have actually had a bit of this influx in the past like venture capital, they're clamoring for more all the time. I'm constantly being asked by people in the private sector in places where PhDs in the life sciences have gone to work. Give us more, more people like this. But I've had two bioinformatics students who were recruited before they got their degrees. So there's a recognition. Hi, I'm Jill Schumacher from MD Anderson in the University of Texas Health Science Center. One thing that strikes me and I love both of your presentations is that we have to change the culture if any of this is gonna work that a single lab PI does not own their postdoc or their students. Have you had faculty or be interested in other places where they have internships? How you have the faculty to buy in to their student leaving the lab and not getting work done in order to do what they need to do for their own career? Yes, definitely. So UCSF, I'm known pretty much as a guy that's been running around talking about the crisis in the time to degree problem for a long time. And so then suddenly I'm running around saying, oh, what about 10 weeks off working at Genentech? And they say, what's going on? You can't talk out of both sides of your mouth but it's completely untrue, I can. He's known for other stuff too but we can't talk about that. And it's because I don't believe that those are inconsistent and that as students define what their interests are, it's human nature. If you know where you're going next and you're really motivated and excited about it, then you move more quickly through what you're doing. If you don't know where you're going next, why hurry? It's too scary out there. So I think it's really true, I think it really works. By the way, getting the faculty buy in on the internship program, the strategy that we took at UCSF and I don't know how it'll work other places is that we went to the program directors of the graduate programs. And my initial view is that we would just see if we could get two or more of them to agree for their programs to participate and that we wouldn't demand that everybody switch over at the same time or embrace this at the same time. And we were surprised that they all went but we didn't require that. We didn't think that was a requirement to get started. Hi, I'm Lois Wiseman. I'm a professor here at the University of Michigan. And I'm glad about how you articulated the problem and I totally agree. I'm concerned about the five year postdoc rule, squelching diversity and being too elitist. So in my case, I graduated from UC Berkeley in 1982. I was born in this country. I got my first faculty position at University of Iowa in 1993. What happened short and very brief was I had a three year fabulous postdoc of a owned project. It fell on its face and after three years I realized this project can't be solved yet. And in fact, it's a still interesting question somebody could start now. So I did a second postdoc and I developed a second owned project. It took me eight years and I was really a postdoc and I was being mentored. And so I hope the system can keep this going also. Not everyone's brilliant the day that they graduate. Oh, it's a very important point that you make and we were very careful not to make it a rule, right? I mean, it's a recommendation, but it's a fuzzy one and we didn't want it to fit people like a straight jacket. We wanted it to fit like a comfortable suit of clothes, you know, a bit loose and easy to move around in. And I think there'll always be cases like that where it's completely appropriate. I think the key- I didn't look extraordinary, honestly. I didn't, you know, after the first postdoc I think someone would have said to me, you know what, you're not cut out for this, go to law school. No, sure. Well, here's the thing, okay? I think one of the best recommendations that was made both in the original report and in our report was that postdocs need something equivalent to a graduate student committee to look after them. That can be an ombuds person or an office person but ideally it should be a couple of other faculty members who care about the postdoc but who are not that person's supervisor slash mentor. And they can always say to the institution, look, this person didn't have a good experience but they have what it takes, they should do another postdoc and it's just fine. I mean, these decisions can get made on an individual basis very easily, I think. But they're hard decisions and I think that you're correct that right now our enterprise kind of defaults to not making the decision at all and so that can go in the wrong direction, right? And that you're now concerned about building in something that'll be sort of viewed as the floor for what we are willing to do. I think it's something we just need to be very conscious of that people are different, that they do succeed and bloom at different rates. And so it'll be hard, but it's a good reminder. Yeah, it's a marathon lot of spring. Hi, I'm Gary McDowell, I'm a postdoc at Tufts University with Future of Research. A key assumption that everyone seems to be making in training for non-academic careers and also in not discussing reducing the number of graduate students is that the labor market is going to be able to absorb all of these extra people and that we're not training for already saturated markets and that also that a starting graduate student is going to be able to predict what the labor market will be for their chosen career six years down the line. And as someone who started a PhD in the UK in 2007, I knew very well that things changed very dramatically over that period. And so I'd really like to know what the data is that everyone is using to justify this assumption and that we shouldn't be talking about reducing graduate student numbers. I would say that we don't have data of that sort. The predictions of that sort are notoriously inaccurate in all fields, not just this one, but certainly in a field that is as dynamic as this one. It's especially problematic. And I guess all I was saying, and I don't want to see what Greg says about this, is that we don't know enough. I mean, my contention, but it's only that, is that we're just the beginning of a slope of increased needs by the sciences of PhDs. And the ways that they will occupy our society, both to enable the continuation of the endeavor and to expand it into doing all sorts of other things outside of society and application, is unknown. And so if we're really at the beginning of that trajectory, it means that we don't know what the final slope will be or when it will plateau. We're just saying it'll be embarrassing if in 10 years we have a tremendous shortage because in 2015 we decided to shut off the tap. And so that's the bet that's being made, but... Yeah, I mean, I'm no expert on this either, but Paula Steffen, a labor economist from Georgia State who is the smartest economist I ever met, and I've spent my life around those because my father was one, I think believes that at the lower levels, bachelor's degree, master's degree, PhD, the market can take as many of these as we can put out. But when you get much beyond that, that's where things get tricky. I see no reason to disagree with her, but it's not my conclusion. It's a conclusion of some pretty smart people like her. Thank you. We have time for one more question. So I'm Wendy Bolag, I'm also from Georgia. I'm from Georgia Regents University in Augusta. And I like your idea of a mentoring committee for postdocs, and we actually tried to bring that up at one point and got a pushback from some of the postdoc advisors saying basically we pay them, we should get to mentor them and basically tell them what to do. And I also like the idea of the postdoc fellows in the grants program, but again, it all comes down to funding. And I know that we do have a workshop on funding, but I'm just interested to know where that money's coming from. So it's coming from philanthropy. The grants are really small. They're $15,000. The postdoc is the PI. The postdoc applies, there's nobody else on the grant. And the first year we did it, we gave them $10,000 for their experiments and 5,000 supplement to their stipends. And they appealed to eliminate the stipend. And just to let them spend the money on their stuff. Several of them did that anyway. And they've reported to us that that is enough to actually get something started. And that it's been a big payoff. So the amount of money, I mean any amount of money these days in all of our programs is a lot. But it's not a terribly big load. So that's something to think about. And I think the experience across the labs that have had fellows who have gotten these grants has been that these other kinds of experiments that are being done actually do enrich the things that are done in the main line of the lab. So they're discovering, even those that were skeptical, that it can actually add. I do much the same thing in my lab, guess where I got the idea from. And we, in my case, I have a donor who's given me a little private money to support this. But you know what, it works so well. I mean it was such a good idea that if I had to pay for it out of my own pocket, I would, it'd be worth it. So please join me in saying thank you.