 So, in the spirit of adapting to having a face-to-face meeting for the first time in a long time, one of the things that we've learned is that the HDMI and other outputs, video outputs from this computer no longer work, and who would have ever had a chance to test that out over the last year and a half or two years. So the good news for you all is that my slides were really terrible. So you won't have to suffer through the slides, so that's the happy outcome. I'm going to be talking today, I'm Roger Schoenfeld from Ithaca SNR. I'm going to be talking today about a paper that a colleague and I did over the summer and published just a few days ago on research cores. And I'll say a little bit more about what those are and how we understand them. It's really the purpose of this presentation is to just give an overview of this type of research infrastructure. I hope we'll have a chance for a little bit of questions and discussion as well. We only have 30 minutes, so I'm going to move a little bit quickly. I did want to kind of foreground that some of you know our work at Ithaca SNR for our, I think, deep work on libraries and users and publishers. But over the course of the last year or two, we've been doing a lot more work on the research enterprise. A series of publications on the impacts of COVID-19 on the research enterprise that I did with my colleague Jane Radicke, a project on the role of the senior research officer, what that role is like and where they fit into the university or sort of organizational structure, which I did with my colleague Oya Rieger, a paper on academic research budgets and how research enablement and support is actually budgeted for in terms of the complexities of higher education in the U.S. My colleague Jane did a paper on indirect costs specifically, so we've been really digging into some of the different elements of the research enterprise. And this summer I was fortunate to have an intern, a Ph.D. student from Princeton University with us for the summer named Ujo Bai. And Ujo and I dug into this idea of the research core. Ujo is a Ph.D. student at Princeton who's a humanist. He's been doing a lot of work in the East Asian library, and so it was great to have him as part of the team. And I'm sorry, I had a slide with a great picture of him and a chance to kind of introduce him virtually. I'm sorry he couldn't be here, but due to the fact that we added this session on a late breaking basis, we couldn't make that happen. So just want to really make sure to credit Ujo on the project. So what are research cores? Okay, research cores, basic overview. Research cores are provided by a university. It's facilities and equipment and services that enable usually scientific research that are provided by a university on a shared basis for the use of multiple research labs. So every research lab that needs a microscope has a certain kind of microscope, but then there's a level of microscopes that no individual research lab can afford or doesn't need to have pervasive availability to for, and that's an example of a resource that can be shared across the campus. Research cores today are vital for research competitiveness at the research universities. There's a whole category I'll get into them, and a moment there's a whole category of different kinds of research cores that have been at the leading edge of providing advanced equipment as well as other kinds of shared facilities that enable research. One of the things that characterizes research cores is that they typically require significant capital investment. So we're talking not about the $100,000 level of equipment, but the multi-million dollar level of equipment. And increasingly one of the interesting patterns that's been taking place is that originally research cores 15 years ago tended to be distributed. The biology department might have one, or the School of Medicine might have a few, but increasingly as the role of the senior research officers come into greater prominence, we've seen again and again choices that universities are making to centralize their research cores on a university-wide basis. And I'll talk a little bit more about some of the ways that that can play itself out. So in addition to the actual physical facilities and equipment, which of course are often at the heart of research cores, it's really important to emphasize that these are not just equipment. These are services, and especially for the librarians in the room, I think you'll immediately appreciate the distinction between the stuff and the services. And of course the key differentiating characteristic is the people that enable stuff to turn into services in the context of a research core. So there are technical staff and managers who staff these facilities and often provide extremely sophisticated and advanced services to researchers and research labs. This is the kind of thing that would have been better on a slide, but just to give you a sense of the kind of cores we're talking about. We're talking about things that I can barely pronounce, like confocal laser scanning microscopy and nuclear magnetic resonance and biomolecular magnetic resonance and genetically engineered murine models. And each of these at different institutions is a research core. And there's a whole long list of these. Sometimes 3D printing is considered to be an example of this. Often telescopes or observatories are considered to be research cores. In many institutions, especially biomedically focused ones, animal care facilities will be considered to be a research core. Sometimes research computing is considered to be a research core. The nature of the examples will vary. Research cores are found pervasively at the research one level of institutions. So both publics and privates have a very wide array of research cores and why they enable those institutions to recruit the leading faculty members in those fields because they have the most advanced, the newest equipment and thereby attract grants. So research cores are absolutely driven by competitiveness for faculty recruitment and grant seeking. There's a very tight connection between the logic that informs the growing attention to research cores at some of the major research universities. Research two institutions, some of which are probably represented in this room, and selective liberal arts colleges will also have research cores but not nearly the same array of them. So they may have a few of them but not dozens and dozens of them. So it's a different level of investment and in those institutions it may be more about what is essential to the research program and especially the instructional program at a liberal arts college. So I wanted to walk through for me what was an extremely vivid example when we were interviewing senior research officers a little more than a year ago. We interviewed 40 some senior research officers and so many of them were talking about cryo electron microscopes. And this is a really, really, I'm going to take a second to explain it because I think it's exciting scientifically and it's exciting to people in this kind of role. So a cryo electron microscope is a device in which you freeze the sample, okay cryo, right? And you freeze it to, I don't know what, 10 degrees Kelvin or some, you know, not our freezers freezing, right? And the benefit of doing this, which is extraordinarily delicate and complicated work to freeze the sample, is that you can study a protein, the shape of a protein. So Clifford when you were talking yesterday about the innovations in protein folding and predicting and all of that, this is a piece of that whole need in proteinomics to be able to look at the shape of a protein and understand its function from its shape. And we've used X-ray crystallography for that purpose for decades but the cryo electron microscope provides the opportunity to do this at greater scale and in some cases with greater accuracy and if you are a biomedically focused institution, you need a cryo electron microscope and this is just developed in like the last five years. There was no such thing, it was invented, someone won the Nobel Prize for it and now everyone needs one. And I've got a beautiful picture on the slides which you can't see of some of these cryo electron microscopes so just so you can imagine that. What I do want to recognize as a cryo electron microscope is not a single piece of equipment. It's actually about half a dozen or a dozen pieces of equipment that each does a different thing, that one is involved in cutting something, one is involved in freezing it, one is involved in putting it on a backing and then there's the actual microscope which is like, I don't know the size of this table or something, that you actually use to shoot the electrons at it and get the imagery back. And so this is a delicate and sophisticated process. You can imagine there are people who are expert at using this facility, these are the service providers, these are the professionals inside of this core and so the scientist who has a lab who's studying a certain, has gotten funding let's say from NIH to study a certain part of, I don't know what a certain protein or category of proteins or function or whatever finds that they want to use a cryo electron microscope and they basically assign or rent the use of this facility, they pay a fee to use it and the people in the facility very often actually do the work of running the experiment so no one from your lab needs to actually run the experiment, it's run at the research core. So this is the hot, this has over the last several years been the hot new item, it's been the most sought after item by senior research officers, it costs about $7 million to outfit a one cryo EM machine, the set of machines that's necessary to have one cryo EM workflow, it often involves facilities, renovations to make it possible to, this is a facility, this isn't just a closet so there's load bearing, there's all sorts of renovations that can happen and then of course there's operating costs for it. So for those of us who understand the way budgeting works in many parts of higher education, the idea of saying I need $7 million to buy one of these pieces of equipment and I need maybe another half a million dollars to renovate a facility and I need to hire three or four people to run the facility, that starts to sound in many cases like a multi-year budgeting exercise. So how is it that a university senior research officer is more or less snapping their fingers and say I have to have one of these and I must have one of them now because this other institution is going to out-compete us for grants this fiscal year if I don't show that I've got one of these. So there's been a whole set of investments that have been made in this particular research core when research competitiveness is seen as being at stake or arguably is at stake and so the senior research officers shared with us that they have been cobbling together resources from many different funding sources. They'll go talk to a laboratory PI here and the Dean of the School of Medicine there and the Dean of Biological Sciences there and they get a buy-in of you give me half a million dollars you give me a quarter of a million dollars I've got a million dollars in a reserve fund and they are like moving with what I think is actually remarkable speed within the context of higher education budgeting and acquisitions practices to outfit these things. Now, so in just a few years seven out of ten R1 institutions have won or in some cases more than one of these facilities and to take an example of what I mean when I talk about centralization this is actually very interesting thing. So Yale is an example of an institution which has a pretty substantial biomedical research enterprise not one of the absolute largest ones but a relatively large biomedical research enterprise over the last several years, last five years Yale has now four cryo EM cores I think one at the med school one at the science complex one at a different science area and so they have four of them so that would be a capital of investment of let's say 28 or 30 million dollars and at a certain point they decided it doesn't make sense that the school of medicine is running professionally the same set of equipment that the arts and sciences school is so let's create a truly shared core where the four machines the four facilities are operated on a single basis not reporting to the school of medicine and the dean of arts and sciences but rather to the senior research office senior research office centrally so you can sort of understand by that example I hope the logic of why centralization especially at some of the larger institutions is coming to have a very strong logic that way a laboratory that wants to schedule the use of one of these cryo EM machines can do so based on do I want the one that's just in the next building or am I willing to go to the next campus the medical school campus so that I can get access this week rather than next week so there's a little bit of that kind of scheduling that actually becomes very interesting as you start to centralize so let me just say a word or two about the business model of these services because as you can see just from the way that I described the capital investment I think this is a little bit of a I don't want to say exactly a challenge but I think it shows an alternative model for how research enablement services can be developed and offered in a university that I don't want to suggest I want to be clear I don't want to suggest that this is a model for the library but I do want to suggest that those of you who are responsible for libraries there may be some things to take away from this as an alternative way of funding research enablement and support so at a basic level this is if you can imagine this is the pie chart of the operating expenses of a shared research core and this is from a national study that was done a couple years ago labor is 50% of the operating expenses so those professionals who are staffing these services and doing many of the experiments inside of them are fully half of the operating expenditures the equipment maintenance, service contracts, etc maintaining the equipment is maybe another 20% and then the consumables the things that need to get used up in the process of the research those are about another 20% there's another 10% of others so it's sort of like it is 50% labor 20% equipment maintenance 20% consumables so that just gives you a sense of what they need to pay for and as you can see with labor the costs are fairly fixed so this is an important element of the business model you need to have the professionals in place whether or not there's demand regardless of the level of use of the facilities so that's an important element of the business model which tends to work on a cost recovery basis so not the capital costs the capital costs are rarely there's rarely an effort to recover those costs directly but the operating costs are often are typically run on a cost recovery basis and the approach is usually fee for service so let's say just to take the cryo EM as an example maybe it's an hour of the use of this facility is $100 or maybe it's you know the scanning five samples is $500 or whatever the service basis may be it's often subsidized to some extent so it may be that that $100 an hour pays only 90% of the operating costs or 80% of the operating costs but it is intended to pay for a substantial amount of the operating costs and the way that that is funded this should be interesting when we think about the APC environment the way that it's often funded is on the basis of an internal charge from the principal investigator from the lab that has a grant so you have a grant from NIH and one of the line items might be the use of certain kinds of shared equipment shared facilities, these research cores which is works pretty well in normal times during COVID this was a catastrophically bad business model and you know just to say that so just to go back to the history of the last two years right in spring 2020 a lot of the research activity at our universities was suspended right at many of your universities was suspended for a couple of months maybe for as much as six months and during that period the labs themselves were allowed to continue to build the federal funding agencies in the US for their the direct costs of their employees so the PI continued to be paid from the grant budget but the research cores didn't have nearly as much work in some cases maybe no work at all as the actual scientific research paused for a while and so without any work to do the revenue that this fee for service model evaporated and so universities found themselves basically taking losses on many not all of their research cores during that period of time the costs largely remained but the revenue disappeared right the universities have actually asked the federal government for emergency recovery funding specifically to support research cores as a result of this interruption in funding so I think that's an interesting example of the risk of this model I think the model has some interesting perhaps benefits or features but also some risks as well so I'll just say a word or two about the staffing there's a totally new set of professionals that have been developing inside of universities around research enablement and support inside of these research cores they're not librarians obviously they're not IT professionals they're not scientists on a track of grad student post doc assistant professor etc they're not on that kind of track they're typically PhD level scientists who are I think it's fair to say struggling to see how their professional identity will develop so there's been some real benefits but also some challenges in the way that universities aren't always very good at providing for these other kinds of professional research support and enablement tracks there can be a lack of job security there can be a lack of well established career paths without any question sometimes there's a lack of professional development like the university isn't really investing in them as professionals in the same way that they might in some of their other employees and there can be a lack of established mechanisms to recognize their contributions so if the members of a research core are essential to a paper are they listed as co-authors of it or is it only members of the laboratory or the co-authors of the paper as an example of where recognition can work in some different ways one of the things that we became incredibly interested in when we were conducting this project but didn't dig into nearly as much as I wish we had was cross institutional collaborations and so up until this point I've been talking only about things that are taking place inside a single university but actually there are a number of research facilities that shared research facilities that exist on a cross institutional basis probably the best example of this is some of the telescopes and observatories that occasionally a group of universities will collectively invest in to build you know in the some mountainous area where the universities are in the top of a mountain in Hawaii or something like that that they'll build an observatory and there's often very sophisticated ways of sharing the time of that observatory among the astronomers and astrophysicists at that group of member institutions but there are other examples as well you know both both in some of the very advanced nuclear physics realms is another example but also I think we're seeing an increasing interest in sharing research facilities at a metropolitan level or regional level I was talking to some architects a few months ago who were pointing out that there are actually this is actually happening among hospitals where there are cases where there are certain kinds treatment facilities that no single hospital could afford and so a consortium of let's say hospitals in the Washington DC area might build a shared treatment facility for certain kind of certain kind of disease so I think that those are an interesting example I think we will over time see the development of other kinds of consortia. Clifford I think you mentioned one or two of them in your remarks that have already come into being so I think that just to kind of reflect on this topic for one moment there's maybe three major issues or trends that I would close with one is the way that prioritization works here the institutional priorities in terms of do you care about astronomy if your institution cares about astronomy you have to periodically reinvest in the most up to date observatory facilities that are possible and so if you do that that is an not only an indication but as a lived example of your commitment to fostering the greatness fostering excellence in the field of astronomy if you care about biomedical research your willingness to provide these cryo EM machines today is an illustration of your commitment but how often your university refreshes these investments and which fields it's refreshing those investments in with which level of urgency and priority is one of the most vivid ways that universities actually illustrate which fields they're committed to and there are really interesting ways that that of course signals to folks in those fields that's a university that I want to be at because that's a university that has the equipment that I need to do the most advanced research and I think that feedback loop is a very very interesting one that we should learn some things from now also because of the way that business models and market dynamics actually work for these facilities at a certain point one of these research course may no longer be as needed on a campus as it once was or may not be up to date with the equipment or the services that it provides and what's going to happen when that takes place is that the scholars at that university will use it less and that core will be taking more of a loss and will require more subsidy and over time universities make decisions or should to decide to wind one of these core facilities down or refresh it right so there's that kind of fork in the road for them also and finally this pattern towards centralizing these research course at a university level you know first of all at least conceptually it increases efficiency it increases the resources that can be devoted to them but it also is a key enabler of cross institutional collaboration because once the universities have decided that this is an institutional priority they can then begin to collaborate institution to institution rather than department to department school to school to consider collaboration opportunity so I think in a way we're seeing that logical evolution of the path towards centralization so let me stop there I know we're just about out of time but if there's questions or discussion I'd be happy to stay for a few minutes and I'll do that. Thank you. Do you think there are models here that apply outside of the people who run research course? I guess that's my key question for the CNI audience. Hi Roger, Simeon Water from Cornell. The thing that's been going through my mind throughout your talk is what's the parallel between this and the idea of special collections as a research core within libraries and where do you think that parallel is useful or breaks down? That's a really interesting question so I think as many of our research libraries have been doubling down their investments in special collections distinctive collections I think that there is a parallel on the capital side we understand the expense side the expenses of developing and maintaining special collections is not just in terms of the acquisition expenditure but similarly there's a labor expenditure around processing them and making them discoverable and available so I think that there's a way in which there's some things to learn on the expense side I don't know that I've seen the level of, I've seen a few examples you may know more than I do where a university does feel an urgency to acquire a specific special collection that becomes available and is willing to make an unusually substantial financial outlay to do so like outside of the normal budgeting process so I think maybe there's at least in some cases a parallel there I don't think I've seen terribly many cases where on the business model side libraries have been seeing a kind of cost recovery model for special collections nor am I suggesting that that's necessarily a good idea but I don't think I've seen that I don't think I've seen that take place but I have wondered and I think your question really calls into relief the extent to which the library special collections or other parts of the library are a research core or are maybe a series of different research cores I don't think I think it's interesting that libraries have been obviously a much longer standing element of campus research enablement and support and I see this new model developing in parallel and I don't know how the two will fit together so I would really sort of lay that as a question for everyone okay well I think we're right at 11 o'clock so let's go get a cup of coffee thanks everyone really appreciate the chance to share this