DAVID ERICKSON: Welcome, everyone. I'm David Erickson, the Director of Mechanical and Aerospace Engineering here at Cornell. And let me welcome you all here today for what will be a very exciting event on academic entrepreneurship. Cornell, of course, has been a very exciting place for entrepreneurship in general, and academic entrepreneurship specifically over the last few years, most notably with the establishment of facilities like the McGovern Center and Praxis Center, the growth of all of our classroom-based and experiential entrepreneurship programs, and of course, these sort of long-term efforts of entrepreneurship at Cornell in making these types of programs available across the university.
The collection of these efforts has really become a true Education to Enterprise program that's really been instrumental in allowing students and faculty to take the inventions that they make in the lab and the technologies they develop, and bring them out into the marketplace where they can be truly impactful. And I think it's in that spirit that we're very excited today to be able to host Professor Stephen Quake, who is our newest A. D. White Professor. And he's one of the most successful academic entrepreneurs in the country.
So Stephen is very much one of these people who has a CV that makes you feel bad about yourself. He is the Lee Otterson Professor of Bioengineering in Applied Physics at Stanford, as I said, and Co-president of the Chan Zuckerberg Biohub. He's received innumerable awards and honors that I could not possibly list off, so instead, I will list off two that I thought were very interesting, which was that way back in 2002, he was named by MIT Tech Review as "one of the young innovators that will create the future," and by Forbes as "one of the 15 people who will reinvent the future." So you both create and reinvent it, I guess, depending on which publication you read.
But I think that's quite telling. Because I think Steve is really one of the people that has been able to do that. And that's reflected in the fact that he's a Member of the National Academies of Sciences, Engineering, Medicine, and Inventors.
Of course, what we're going to talk about today is entrepreneurship, so let me brag on behalf of Steve of all his successes. So Fluidigm, which is a publicly traded company on the NASDAQ was, confirmed today, the first publicly traded microfluidics company in the country, at least to my knowledge. And I understand that was Steve's first company that he started. So I wonder how many people's first company ends up as publicly traded.
Verinata Health some you would have heard of, which is one of the early companies to offer non-invasive prenatal testing using cell-free DNA from the mother's blood. It was acquired by Illumina for $350 million. And then more recently, Karius Diagnostics, which again uses cell-free DNA for identifying and diagnosing infectious diseases. And I'm sure we're going to hear a lot about those companies, or perhaps some about those companies and all the kind of very cool stuff that Steve has done in terms of making those things happen. So welcome, Steve.
STEPHEN QUAKE: Thanks. It's a real pleasure to be here at Cornell. And so what I want to talk about today is creating impact beyond publishing papers. As academics, we do scholarly research, sometimes make discoveries. We share those with our colleagues in the world, by publishing them in the scientific literature.
And that is sort of the main part of what we do. But there are times when you'd like to see impact beyond just journal papers. And sometimes, what you do is you publish the paper with a great idea, or discovery, or invention, and hope that somebody recognizes it, and picks it up, and runs with it, and translates it into industry so it can help other people. And other times, that doesn't happen. You need to give it a nudge.
And that's what I'm going to talk about today, is how you give it a nudge, and cases where that can be helpful. And I'll tell you a couple of stories about that in my career. So I'm very honored to be the A. D. White Visiting Professor.
And I was doing a little homework trying to figure out who is A. D. White, first Founding President of Cornell. And I discovered a couple of interesting things that made me very proud to be the A. D. White Professor. One is in his inaugural address when Cornell was getting launch.
He said he wanted to create "an asylum for science, where truth shall be sought for truth's sake, not stretched or cut exactly to fit revealed religion." And he was, as a historian, very interested in the interplay between science and religion, and wanted to create an educational institution that really was focused on science and finding scientific truth, which is great. I'm proud to be associated with that.
The other interesting thing I learned about him was that when Leland and Jane Stanford decided to found a university out in California, the person they asked to be the first president was A. D. White. And he turned them down, and recommended one of his former students, David Starr Jordan, who then became the first president of Stanford. So there is a connection.
So coming to the topic of the talk, academic entrepreneurship, what's the deal? And I'm going to sort of talk about how these things work in general so we're kind of working from a common playbook. So suppose your research leads to an amazing discovery or invention.
And what happens then? The university owns the patent rights. You signed a document, probably all of you assigning all such rights to the university. And so then the university decides whether to file a patent or not, and how to license it for commercial development.
And what you get in return, often, is some fraction of the royalties that the university makes on your invention, for most. It varies from place to place. Typically around 25% to 33% of the royalties go to the inventors.
And if your invention is recognized as something awesome, it's done. You just wait and wait for those royalties to roll in, and for the products to hit the market. And the company does all the work.
But if the rest of the world doesn't yet appreciate the value of your invention, perhaps you need to nudge it, as I said. And one way to do that is to found a company. And then the company will need to license the patent from the university, raise money from investors, and build an organization to make and sell the products resulting from the invention.
And of course, probably you want to help that company along, which then raises the question, what about conflict of interest? Bill Brody, when he was President of John Hopkins, came out to Stanford and gave a great talk about this, where he said, "No conflict, no interest," which is to say, the person who is the inventor and most interested and conflicted is the one that's going to be in the best position to figure out how to bring it to realization, how to commercialize it, and understand how it works.
And so the goal of conflict of interest policies is not to eliminate conflicts. In fact, conflicts can be a very good and motivating thing. What's problematic is when conflicts are hidden and not discussed, and undisclosed.
And so if you went to a doctor, and the doctor prescribed you medicine, you'd like to know that the doctor is not getting paid by the pharmaceutical company to prescribe that medicine. That would be a hidden, undisclosed conflict. If your doctor told you, "I'm getting paid money from this company, and I'm going to prescribe this medicine for you," you at least can decide whether or not you think that's a good idea.
And so the role of conflict of interest disclosure in institutions is not to eliminate conflicts, but to manage them and ensure that they're disclosed appropriately. And there are many complicated and subtle issues around this. Maybe we'll get into some of them in the panel discussion.
But one thought I'd like to leave you with is that there's no one-size-fits-all solution to all this. There's nuance, and they should be handled as such. And there's a lot of interesting examples that we can get into.
And it's sort of, I think, not uncommon for people to point to faculty who are involved with companies, and say, "These sort of conflicts are not good for universities. And it taints their work."
And my response to that is, who is more conflicted? Those of us who are involved with companies and have declared our conflicts and that's clear, or those who are maybe not connected to companies, but are totally dependent on grants to make their salaries and feed their families. And so how does that affect their approach to scholarship and research? That's a conflict that I feel is pretty substantial, maybe even bigger, and one that as a community we haven't really recognized to the extent it should be-- also a thought for future discussion.
So I was recently involved in founding a new institution called the Chan Zuckerberg Biohub. And we had a chance to think about how we wanted to handle all of these issues-- inventions, communications, disclosure, licensing. And I thought I'd just share that with you, again, as a point of future discussion.
We decided that in fact, universities, in many decades of being involved in technology transfer, had worked out a pretty nuanced sense of how to handle that sort of thing. And our sort of patenting and licensing guidelines sort of are very much based on university models. So we think they're pretty well thought out.
Where we decided to be very aggressive is in our policies around science communication. And our donors challenge us to try to accelerate science. And thinking about how to do that, we thought one way to do that is to accelerate communication of science.
Typically, when you submit a paper, it's 12 to 18 months on average before it gets published. And that sort of delay that the rest of the community doesn't know about your discovery is actually sort of substantial. And you can sort of calculate the knock-on effects of eliminating that delay, and it's pretty good.
And so we decided to do an experiment, and to ensure that all the research that we do and the research that we fund, we require all our grant recipients and employees to share their research via pre-prints on the day they submit it to peer-reviewed publication, to really do an experiment getting rid of that delay. And we've been doing that for about three years now. And the results are very interesting.
There's more than 100 papers that have been through this. And we're starting to sort of follow outcomes. A little early to see it all the way through, but it's been fun so far.
So rapid communication sharing of scientific results and data-- we have made that kind of a centerpiece. When someone makes an invention, our policy is to empower the inventors to decide whether or not to patent it, and ask for a patent, or to have it be put in the public domain. Many universities do that. Not all do. Some sort of say, well, if you make an invention, we're going to decide whether it's patented or not.
But we feel very strongly it's important to empower the inventors to make that decision. There are people who feel that their work is best served by putting it in the public domain, and we don't want to be the ones deciding that. And so that happens.
And then if the decision is to seek a patent or file a disclosure, we've adopted, as I said, sort of very university-like policies in how that's handled. We've sort of signed on to the licensing guidelines that have been put together by the AUTM which is the Association of University Technology Licensing Offices about best practices for ensuring that licenses are going to help inventions reach the largest possible group of people, including neglected populations and areas. We also, as a matter of policy, allow all non-profit research institutions to use our patents for education and research purposes, sort of full stop.
They don't need to ask for a license. That's part of everything we do. People just use it. And anytime we do a license to commercial entity, this term is included.
And finally, we've decided that-- and this is maybe a little bit of a twist-- that we're going to generally only seek patent protection in high-income countries as defined by the World Bank, which is about 16% of the world's population, those countries, plus the BRIC countries-- Brazil, Russia, India, and China. And everyone else in the world can practice our inventions without restriction, because we're not going to apply for patent protection there. And so that's sort of how we're going about it. And I'm happy to answer any questions about this in the discussion part.
And with that as kind of a general framing, I thought I'd tell you two stories of academic entrepreneurship that I've personally been involved with over the course of my career. One is creating automation tools in biology to solve something I like to call "the tyranny of pipetting." And I'll tell you what that is in a little bit.
And the other is an effort to try to replace invasive biopsies with blood tests, which is a way of creating health equity in the world because it's sort of a sad but true fact that 50% of all doctors graduate in the bottom half of their class. And if you're in an underserved area, I tell you what-- you're not getting the ones in the top half. And there's certain things that just require physician skill, and doing invasive procedures is one of them.
And if you can replace invasive procedures with blood tests that work better, that's great, because then everyone in the world can have the benefit of that. It's easy to draw blood anywhere. It can be shipped. And all of a sudden, your health care quality is not as much determined by your geography, and location or proximity to a tertiary care center. And I'll tell you a few stories about how we've tried to do that.
So the first story is one about creating research tools. And it's based on lessons from the electronics revolution. And you think about what happened over the course of that period, when computation became automated, going from vacuum tubes to transistors to integrated circuits. Integrated circuits solved an important problem that engineers called "the tyranny of numbers."
And they realized that any circuit they built out of discrete components, there was a limit to the complexity of what they could build. They could design things of arbitrary complexity on paper, but when it came to actually building them, whether it was vacuum tubes or transistors, inevitably there'd be some error somewhere-- a cold solder joint, defective component-- that created a practical limit to the number of components they could put in a circuit. And that was called "the tyranny of numbers."
And Noyce and Kilby solved that by inventing the integrated circuit, which by batch fabricating, allowed one to create enormous numbers of components, all in parallel in a small number of steps. And I'd gotten interested in trying to do the same thing for biology to solve the tyranny of pipetting. Pipetting is how you manipulate fluids in an experiment, and doing it by hand kind of limits the complexity of any given experiment you can do.
And so my students and I spent a number of years, as David mentioned, 20 years actually now, figuring out how to basically build miniaturized plumbing. And so what you want for biology is instead of transistors, and gates, and wires, you want valves, and pumps, and tubes, and pipes. And so we figured out how to do that, and spent quite a bit of time exploring the applications of those technologies for a variety of biological problems.
And there was a great deal of really fun basic research. But there's also a sort of academic entrepreneurship story hidden in that. And I'll walk you through why both are interesting.
So one view of this whole field of miniaturized plumbing is that you want to do what I just said-- you want to do for biology what integrated circuits did for computation, and just build miniaturized ways to do large-scale automation. And I'd call that the "engineering view." Sort of the science view of the whole field is that it turns out that the laws of fluid physics scale with length much more quickly than other laws of physics.
And so you can very quickly get into regimes where there's novel fluid behavior that's not available at the macro scale. And you can find ways to take advantage of that in clever device design. And there's been very interesting literature about that over the years, which Todd Squires and I reviewed some years ago.
And here's kind of a summary of various dimensional numbers. And you can see that many of them scale with length scale. And very quickly, as you get down to the tens of micron length scale, these numbers change quite dramatically from how they behave at the macro scale. And Reynolds number is only the most famous one. There are many others that matter.
I'll show you a little movie, maybe. Let's see if the movie is going to work. I didn't test this. Maybe not. I'll save the movie for another time, perhaps.
But my student Carl Hansen, who's now a Professor at University of British Columbia and I in collaboration with James Berger, spent a bunch of years developing ways to miniaturize protein crystallography. And what we realized was that this was one of those stories that kind of had the best of both worlds, in that there's great value in doing engineering economies of scale by miniaturizing the process, but also there were really fun ways to exploit the fluid physics to grow crystals in ways that couldn't be done in larger volumes.
And that led to serve a suite of publications and a commercial story. And this worked with all sorts of refractory proteins. The particular physics that mattered here is the Grashof Number.
It's an analog of the Reynolds number. This is the ratio of buoyant to viscous forces. And it measures how much density driven convection there is in the system. And it turns out convection is a bad thing when you're trying to grow crystals.
And so if you can get the Grashof Number to be small, you can manipulate the crystallization kinetics in interesting ways. And there's a couple of ways to make the Grashof number to be small. One is to go to very small volumes, which is what we did.
Another is to make gravity go away. And it turns out NASA spent many years trying to grow protein crystals in space for exactly this reason. Turns out it's not that cost effective to do so. This ended up being more successful.
So we published our papers. And people wanted to use our crystallization toys. And it's just not feasible to imagine my group supplying the world with-- structural biologists with chips to do these experiments.
And so I ended up founding a company called Fluidigm to make these chips commercially so that anybody could buy them. And you didn't need expertise in micro fabrication, microfluidics, or any of that to use them, which is sort of this gap I was talking about earlier. We published the papers on how to make the chips and how they worked, and put all the protocols there.
We'd share the mass designs and anything, anyone who wanted to reproduce our work, but only a relatively small number of people in the world had the skills to make the chips. And none of them were structural biologists. And if you really want to enable that group, you want to make a solution they can buy and use.
And so I ended up founding a company called Fluidigm to do that. They figured out how to make these chips at scale, how to manufacture them, actually much better than we could in the lab, at the end of the day, and went on to industrialize the whole process, and solved a lot of problems along the way. There were all sorts of obstacles.
People thought that the material we were using to make the chips, which is a rubber elastomer, wasn't something that would be amenable to industrial processes. Turns out it was, but they had to invent a whole fabrication facility around it. And after a certain point, they had successfully commercialized it.
And by 2011, they had manufactured and shipped more than one billion of these micro fabricated valves. That answered that question. Yes, they're manufacturable. And it allowed a very large number of people in the structural biology community to use the chips.
And you could see that in the literature. Here's a collection of some of the most important structures that were published by a variety of groups. And you can see that it would allow people to work on things like Ebola virus, glycoprotein, to look at PCSK9, which is now one of the important cholesterol-reducing drugs, inhibitors of that protein, other things like H5N1 influenza hemaglutinin. And having these commercially available really amplified the results of our research.
And this is an example, is one that I love. It is sort of a small one from the perspective of market, but a big one from the perspective of enabling science. And this is just a fraction of what was discovered. Many of the structures that were done by customers were deep in the heart of pharma companies and were never published, but played a role in the pharmaceutical industry.
Their second set of products revolved around single-celled genomics, and creating ways to capture single cells on chips, amplify their transcriptomes, and sequence them. And Fluidigm made the first commercial product for that field. They called it the C1.
And that, again, opened up a whole field. And I can't even list papers, hundreds of papers of people who were enabled by that sort of technology. And again, it allowed people to enter the field without needing the expertise of how to do the technology. They brought the biology expertise. And I'll spare you the details of this story.
I'll also say that the valves we developed represent one sort of corner of the microfluidic field. And we at one point began playing with other ways, such as making little emulsions on chips. And Todd Thorson, whose thesis that represented, was the first to figure out how to make little droplets on chips. And that created a whole other thriving subfield that has also become an important aspect, both, I'd say, in terms of scholarly research and commercial applications of microfluidics.
And we've seen the commercial field really blossom. And from a variety of technologies, whether it's valves, droplets, or micro fabricated well plates, there is a number of companies playing in the area with a variety of applications. And it's become a very healthy ecosystem. And it's been lovely to see all that happen.
So that's one story of how something that started as a research project in an academic lab turned into a story of academic entrepreneurship that's enabled many researchers to get access to technology they wouldn't otherwise have. My second story, the story around diagnostics, is one that's very directly connected to human health. And so it's more about enabling doctors.
And it started with work we did in my lab, again, sort of very basic research. We were interested in doing single molecule. We did single molecule biophysics, wanted to find ways to sequence DNA molecules.
Figured out how to do that. And it was sort of a very academic proof of principle, which we were able to publish, but got the attention of the venture capital community, who then funded the company to make a machine to do the sequencing, which turned into, at the time, the world's fastest, cheapest sequencer. And we had a lot of fun with it in the lab.
We got one of the first instruments, commercial instruments, back in my lab, used it to sequence my genome. There's a whole story beyond that that's fun to explore. But that sort of marked sort of a transition for me in my research, where I realized, we've been successful in making technology. There's other ones out there. Let's focus on how to use these technologies to do interesting science and clinical research.
And the problem I decided to turn to was one that was sort of personal. It was around prenatal diagnostics. I had just recently started a family, and my wife was of an age and the risk group where the doctor said, you should get amniocentesis.
And oh, by the way, you might lose the baby as a result of this. And it just seemed crazy to me that the doctor was suggesting that we take a test, a diagnostic test, and there'd be risk associated with a diagnostic. And that kind of stuck in my head as not a great situation.
And so I thought about that quite a bit. And it turns out that the thing people are testing for with these tests are aneuploidy, so when you have an extra copy of a chromosome. And it's the single largest source of genetic disorders for live births.
And is there some way to do this without having to create a health risk? And it turns out the answer lies in this phenomenon called "cell-free DNA," which was discovered back in the late '40s by two Frenchmen working in Strasbourg, Mandel and Metais, even before people realized that DNA was the molecule of inheritance. I mean, Oswald Avery figured it out right around the same time, and nobody believed it. And very few people even heard about his results.
This was their data from the paper. Amusingly, they actually looked at a case of pregnancy, and saw something funny there, [INAUDIBLE] pregnancy. Values seemed higher. And this result was kind of-- I mean, this paper was lost for many years. People forgot about it.
This field of cell-free DNA lived on. And the sort of phenomenon is that when your cells die in your body, they release their contents in your bloodstream. Nucleus gets out. Contents, the genome gets out in the blood.
The DNA gets chewed up into little pieces. And so you have that DNA that circulates in your blood all the time, from all the tissues in your body contributing to it as their cells die. That's what these guys discovered.
So we all have it all the time. All our tissues are contributing. If you're pregnant, guess what? You're getting some of your baby's DNA in the blood, too.
And that was really discovered and proved in a molecular way, I'd say rediscovered, proved in a molecular way in '97, in 1997. And people spent 10 years trying to figure how to harness that as a practical diagnostic, and tried a lot of different things, all of which failed. The challenges were sort of substantial, one of which is that there's not a lot of fetal DNA in the mom's blood.
It's only a couple percent of the DNA there comes from the baby. It increases over time with the pregnancy. But in the key window when you want to do this sort of test, it's only a few percent.
Another challenge is that you're not looking for mutations. Remember, the most common disorder you're looking for is an extra copy of a chromosome. Down's Syndrome has an extra copy of chromosome 21. That's the most common.
So you can't use tricks to look for sequence differences. The DNA is all the same. There's just an extra copy of one in the baby. And people tried all kinds of ways to work through that and tease through it, and tried ways to purify the fetal DNA, looked for biochemical differences in the baby's DNA versus the mom's.
None of that worked. Eventually, Christina Fan, who's a student of mine and I figured it out, how to solve this. And the trick is not to look for differences between the fetal and maternal DNA, whether biochemical or genetic. That's all sort of a red herring.
What you want to do is find a way to count molecules. Because if you can count molecules and assign them to a chromosome, you can figure out if one chromosome is overrepresented with respect to any other, even if it's a very small difference. And in fact, your error or your sensitivity in determining that is very easy to calculate.
It goes like square root of n statistics, so that the more you count, the better your precision is. It's a very unusual approach to a test. And so it comes down to being able to count enough molecules.
And it turns out these sequencers that I've been working on developing are great ways to count molecules, and to count lots and lots of molecules. And so we sort of put two and two together, and decided to try to count cell-free DNA using a sequencer. I couldn't use the sequencer I'd invented, because of conflict of interest limitations at my institution, so I had to use the competitor's sequencer.
I had to spend a lot of time to figure out how to work the errors out of their sequencer, because it wasn't as good as mine. We did all that and published it. And together with Yair Blumenfeld, who is a maternal/fetal medicine doctor, we were able to-- he recruited a patient cohort at the hospital, and was able to get their blood, and did independent invasive measurements, karyotyped the fetuses, and Christina purified the cell-free DNA, sequenced it, and counted the molecules by mapping them back to the human genome, and letting each molecule vote for a chromosome, and looked for overrepresentation.
Here, the example shows the relative representation of chromosome 21 sequence reads or molecules. The red data points all represent samples from women who were carrying babies with Down's Syndrome, fetuses with Down's Syndrome, and the blue, not. And you can see there's a very clear overrepresentation around chromosome 21, because all those fetuses are contributing a little bit of extra chromosome 21 to the cell-free DNA in the blood.
There's nothing particular to chromosome 21 in the nature of measurement. You can ask that question of overrepresentation of any chromosome. And so we did.
Here's all the chromosomes on the axis, the relative representation of chromosome 21 in the middle. Turned out we had two women in the cohort who were carrying fetuses with trisomy 18, extra copy of chromosome 18. Those are those two black ones that are overrepresented. And on the far left, chromosome 13, we had one women with a baby with trisomy 13. So you can read out any of the major aneuploidies that one cares about.
So we published this in 2008. And it got people really excited. Within a couple of months, our results were independently replicated by another lab, and it was sort of off to the races.
A number of large-scale clinical trials were launched, both in academia and in industry. And they read out over a course about three years. So by 2012, the first commercial diagnostic of this approach hit the market, and it took off like crazy.
In 2013, something like half a million women received a version of the test. Next year, it was a million women. And by 2016, it was 3 million women a year. Now it's up to probably 4.
And importantly, not only are people using this, but they're actually not doing amniocentesis. The rates began to decline dramatically. In fact, in academic studies in controlled situations where women were given the choice, you can see that in blue is number of non-invasive tests, and red is number of invasive procedures. And the reduction is really quite impressive.
So that was tremendous from my perspective. This is a case where we were able to do the proof-of-principle academic work. Company licensed the technology. I helped them understand how to use it, but they're the ones doing the clinical trials, getting the product launched. And it was something that very quickly was able to have a positive impact on human health.
We went on to do other academic work using this approach, figured out how to measure any aspect of the genome we wanted, including mutations. And there's interesting applications of that in the world. We realized that the cell-free DNA was good in many other situations, including organ transplantation.
And that's another story that has now made it all the way from the academic work we've done into the clinic. And [? Ewen ?] [? Devlamic ?] here played an important part in that when he was a postdoc in my lab at Stanford. And in doing that, he made another really interesting discovery-- that not all the cell-free DNA is human.
And in fact, the sequencing we're doing gives one a very detailed picture of all of the microbiome in our body, and can be used to study the microbiome and to study infection in interesting ways. And that's led, as David mentioned, to a company called Karius, which is now offering a hypothesis-free test for infectious disease. How am I doing on time?
What's that? You want one more quick summary? All right, so I'll bring you guys up to date on sort of the most recent chapter of this. It's not all greatest hits sort of talk.
This original paper from 1948, Mandel and Metais discovered not only that there's cell-free DNA in the blood. They also discovered the cell-free RNA. That's one of the other columns here, the middle one.
So it turns out RNA, even though it's a very labile molecule, when it's protected in vesicles around particles, it survives in the blood for a while, too. And so there's not only DNA from the cells. There's RNA, which means you can ask questions not just about genetics and genotype, but also about phenotype. What's going on in terms of gene expression?
And we got interested in trying to use cell-free RNA as a way of monitoring human development, and to look at transcripts from the placenta and the fetus, and studying the gene expression of the developing human just by drawing a little bit of blood from the mom's arm. And did our first proof of principle there in 2014, and showed that you could do this. And you could see a number of important developmental events.
And then we got very ambitious, and we decided we wanted to try to use this to monitor human development at very high resolution, and to use it as kind of a molecular clock of development to try to follow the fetus's development, understand is it on track, off track. As a clock, is it potentially a molecular clock when the baby is to be due, and in particular, could we predict when the baby would come early for preterm birth, which, as I'll discuss, is a really outstanding health challenge.
And this involved a large number of collaborators. Dave Stevenson and Gary Shaw, who are running a Center of Prematurity Research at Stanford, helped us organize the clinical aspects of it. We also had a great collaboration with Mads Melbye, who runs the Danish National Biobank. And in my lab, Thuy Ngo and Mira Moufarrej did the work that I'm going to show you.
So Mads was able to recruit an amazing cohort of very public-spirited women in Denmark, who were willing to come in and give blood every week during pregnancy. It's amazing. And so we were able to do these very high time-resolution measurements and track different genes.
So each graph here is a different gene, and this is over the course of pregnancy. We've averaged time together a little bit, so you don't see the full precision of it. But you can see that there are very stereotyped patterns for different genes, whether they come from the placenta, the immune system, or fetal tissues.
And based on that, we've figured out a way to build a clock. And we could take those changes, and predict gestational age of the baby, and do it pretty much as good as ultrasound, without requiring the very expensive equipment or skilled operator to do it. And so that was kind of the first molecular clock.
And then we got interested in asking what happens-- could we figure out if the baby comes really early? Because this question of premature birth ends up being the largest cause of neonatal death and complications later in life. And it affects a really large number of pregnancies.
The aneuploidies and genetic disorders I was talking about earlier are a percent or less of pregnancies. This is closer to 10%. And there's no meaningful diagnostic to understand who is at risk.
So we began to look at that, and looking at cohorts of women who had very early preterm birth. And to make a long story short, we were able to discover seven genes whose expression levels were correlated with preterm birth. And we're able to do this in two different cohorts, and sort of training a testing cohort, and were able to show that there's sort of potentially reasonable test performance.
And we could, in fact, predict spontaneous preterm birth up to two months in advance. So very excited about that. We published it last year, and this could be a whole other generation of diagnostics for maternal health.
But it also needs to be caveated, of course. This was an academic study. It was a very small number of women. The cohort was ethnically homogeneous, and there's a lot of questions that need to be answered before we understand if this is something that will really have impact in the clinic.
So we're at the beginning of the story, and we'll see what happens. There's a company called Mirvie, which I've helped found, which is going to explore all that, and really scale us up beyond, again, what we could do in an academic lab, and do it in a hopefully clinically definitive large-scale study. We should know the answer in another couple of years.
So to wind up this part of the talk, I hope I've given you a sense of these sort of simple blood tests which can be performed anywhere are actually saving lives. In the case of the amniocentesis one, millions of tests a year, if they're all avoiding amniocentesis, equates to thousands of lives saved. And you hopefully see how they're replacing invasive biopsies, which require the presence of a skilled physician.
As I look back, the theme that connects the things I've done has been to try to find measurement technologies and use them to change biology. And these sequencing technologies turned out to be much more useful than just for sequencing genomes. Their incredible parallelism enables them to be used as molecular counters, and I've given you several examples of how that has practical utility.
And the microfluidic tools, again, also enable unprecedented sensitivity and parallelism. And this ability to do a very unglamorous thing of plumbing at the nanoliter scale also has had a really significant impact on science. And it's been fun to be a part of that field, as well.
And I'll leave you with one of my favorite quotes from Louis Pasteur about the connection between science and applied science. And his claim is that "There is no such special category of science called applied science. There are science and its applications, which are as related to one another as the fruit is related to the tree that has borne it." And I find that very motivating. Thank you.
DAVID ERICKSON: So that was really inspiring, and I very much appreciate the kind of way that you've driven these innovations, both from the engineering side and the biology side, and really drove a lot of the parallelization of biology and brought it to market. But there's lots to dive into there. So to help us do that, I'm going to invite our two illustrious panelists up here.
So allow me to introduce them. So first, we have Andrea Ippolito is the Executive Director of Engineering Management program here at Cornell. Prior to joining Cornell, Andrea served as the Director of the Department of Veterans Affairs Innovators Network, and which was preceded by her service as Presidential Innovation Fellow, which I'd love to hear more about some other time, in the White House Office of Science and Technology Policy.
In 2012, she co-founded a company called Smart Scheduling that helped to reduce variability in provider schedules, and improve patient access. And the company was acquired by Athena Health in 2016. And I'll say that in addition to her role in the Engineering Management program, Andrea has been, I think, incredible and really been a great contributor to the entrepreneurial ecosystem here at Cornell, helping to run a number of programs developing the Women's Entrepreneurs and STEM program, and doing things like contributing to I-Corps, and so forth. So welcome, Andrea.
ANDREA IPPOLITO: It's nice to be here.
DAVID ERICKSON: And David Putnam is a Professor in the Departments of Biomedical Engineering and Chemical and Biomolecular Engineering here at Cornell, where his research focuses on the design and synthesis of functional biomaterials for vaccines and drug delivery, among other applications. David joins us here in 2002, I believe, following his postdoctoral studies at MIT, during which time he co-founded Transform Pharmaceuticals, which was acquired by Johnson and Johnson for $230 million in 2005.
Among his many entrepreneurial pursuits since joining Cornell, he recently co-founded, along with Professor Matt DeLisa, Versatope Therapeutics, which is an immunotherapy company, which last month was awarded $18 million from the NIH. Congratulations on that.
So as I say, there's lots to talk about. But first, I want to ask maybe the obvious question, which is-- or maybe the one that maybe some would think would have an obvious answer, but maybe it doesn't. And that is, why should universities and educational institutions encourage academic entrepreneurship?
And the reason is because I think when you see a fantastic talk like this, and have someone extremely successful come, it's a no-brainer, because I'm sure there's all kinds of things going on. But there's a broader ecosystem, and not everything ends up like this. And it could take time away from faculty, if they're pursuing these kinds of pursuits instead of, say, basic science, which one might consider the role of the university. So I have my own thoughts, but I'd love to hear, maybe Steve, if you want to give me your thoughts on why this should be a priority of the university.
STEPHEN QUAKE: Yeah, well, I mean, as a vehicle to help faculty's research have impact, it turns out to be a pretty important channel. Because it's actually pretty rare that when you publish a paper, people start beating the path to your door, and saying, do no more. We'll take it from here.
DAVID ERICKSON: Yeah, very rare.
STEPHEN QUAKE: I mean, it's much more common that you need to be out there, evangelizing and helping people understand what the real implications are, and what the potential is. And if your goal is to have impact on human health, then I think it's very appropriate.
DAVID PUTNAM: So I think for three reasons, one what Stephen said, is it's our obligation to do good for society. And if we don't do it, who's going to? Two is the students learn tremendously from going from the bench, and bringing it through the process of commercialization. It's great for their development as students and future leaders. And three, for the economy-- so here we are in rural, upstate New York. And if we can translate successful technologies, and build the economy of New York State, it's good for the state as a vibrant institution.
ANDREA IPPOLITO: Well, I'll just agree with that. And as a former student of David's, seeing the impact of his work, we were talking-- I took his class the day he found out his company was acquired. And having that type of role model as a student, seeing that, we can learn those lessons.
So just to build upon that, I think this is very important for the brand of the university as well, whether that's recruiting students, whether that's recruiting faculty. I think some of the more junior faculty are expecting this as part of the culture at a university they join. So that's critical to have that culture built up at a university.
I also think this opens up new funding sources, whether that's through Small Business Innovation Research, SBIR funds, which a certain percentage of funds all government agencies need to devote to these SBIR grants. So this opens up tons more funding opportunities as well. And of course, for the brand and the impact that you can create.
DAVID ERICKSON: So this is being recorded. So I'll just forward all that to the provost afterwards, if that's OK.
ANDREA IPPOLITO: Perfect.
DAVID ERICKSON: All right. So we're going to take questions from the audience in a minute, but there's a few things I was going to-- I want to just make sure I get through. The one thing that I have seen been a big change, and I'd be very curious to hear if it's the same at Stanford, but the sort of sophistication of the entrepreneurial programs we have here at Cornell has increased dramatically in the last 10 years.
When I look at what was offered 10 years ago, it struck me as, looking back, very rudimentary. And I think that has to do with more experiential things and these kinds of things. So you've done this at Caltech, which I consider a very fundamental institution, Stanford, which has always had a reputation of doing this. So I guess one, I'd be curious to get your interpretation of how things have changed over the last 10, 20 years in terms of this.
STEPHEN QUAKE: Yeah, well, there's definitely been an evolution. As I say, universities have reached equilibrium about what the relationship should be through the whole thing, that I think is generally pretty well thought out. And there are always things that need to be tweaked here and there.
But I was at Caltech for eight years on the faculty. And they were not in very good shape, the tech transfer just before I got there. And they brought in a guy named Larry Gilbert who took over the tech transfer office, and over the course of a decade, by the time I left, that year, Caltech had more issued patents per faculty than any other institution.
DAVID ERICKSON: Is that right?
STEPHEN QUAKE: Except UC as a whole, yeah, that's per faculty, not total.
DAVID ERICKSON: Per faculty, OK so you got--
STEPHEN QUAKE: No, sorry, total, not per faculty, total.
DAVID ERICKSON: Oh.
STEPHEN QUAKE: Yes, total, not per faculty. It was amazing. So he created a sea change at the institution.
And he told me once how he did it. He had a very simple method. He just went around to all the faculty members and made appointments with each of them, sat down, said, what are you working on?
What are you most excited about? What was your last paper? Oh, did you think about applying invention disclosure?
And many of them said, no. And he kind of talked to them through the process. And through that, he really created a thriving system there.
DAVID ERICKSON: And that was at Caltech, so the kind of evolution of the program there. What about at Stanford?
STEPHEN QUAKE: Yeah, at Stanford, it's always-- as you say, a much longer history of entrepreneurship there, going back to Hewlett-Packard. And I think because the Bay Area ecosystem has been so supportive of entrepreneurship that the university hasn't had to do anything. And it generally had very entrepreneurial faculty, and capital has been there, and the university hasn't had to push it at all.
DAVID ERICKSON: So it kind of enabled. I don't know, David, do you want to maybe talk about what the difference is between maybe here, when you see your students that might be interested in this, and maybe compared to your experience as you were going through the system.
DAVID PUTNAM: So going through the system as a graduate student myself, I was at Utah, which is a very, very different place than MIT--
DAVID ERICKSON: Well, as a postdoc.
DAVID PUTNAM: Yeah, so as a postdoc, it was just part of the fabric. So when I came here in 2002, Mike Shuler asked me to sort of push the entrepreneur side of things, which was not the prominent focus of junior faculty at that time. And I was just embedded, and Andrea knows the same, just embedded within [INAUDIBLE]. It's just a part of the fabric.
But I think one thing we struggle with here, and I hope you can actually help me understand this a little bit, is we have almost an embarrassment of riches on entrepreneur efforts on campus. It's everywhere, which is terrific and horrible at the same time, because it's very difficult to know what one hand is doing with the other. So just how do they organize it at the administrative level at Caltech in such an effective way?
STEPHEN QUAKE: Well, Caltech is just a very small place, only 200 faculty members. And so everyone sort of knows each other.
DAVID PUTNAM: [INAUDIBLE]
STEPHEN QUAKE: And so that's easy, small [INAUDIBLE]. Stanford, it's not organized, and I think that's OK. There's a lot of different ways to be successful entrepreneurial.
I mean, I've told you my story here about how this stuff flows out of twists and turns of research and things like that. And sometimes, I didn't really anticipate where I was going to go. Whereas Paul Yock, who runs the Biodesign program, takes a very kind of prescriptive, directed approach when he's finding all that. And I think both are perfectly valid. And it's fine for students to know about both ways of doing entrepreneurship.
DAVID ERICKSON: Andrea, you deliver a lot of these programs. So can you tell David where to get started?
ANDREA IPPOLITO: Well, it's always an evolution. So there is an education layer. So there is classes that hopefully, many of you are taking advantage of in the room. But if not, whether they're in Engineering or Business or across the university, whether that be eLab, which is a dedicated four-credit course, where you also get $5,000 to launch your venture. So that's one example of educational layers that help support this infrastructure.
There is actual infrastructure that supports the infrastructure, things like the McGovern Center or Praxis or Rev, actual space. I think there are some opportunities for growth there, with having [INAUDIBLE] or maybe shared materials list where people can access it. Also, there's legal infrastructure, like the Gateways to Partnership program, and also the FastTrack was to help give some prescribed royalty agreements and legal agreements that you, as a faculty member, and outside entities can work with.
And then the last but not least, which is the most nebulous, is just having role models. This is something that there were role models when I was a student here, but there's now a whole lot more. And so whether you're a student, or a faculty member, or a postdoc, I would say turn to the role models that have done this prior to you. And then the call to action, frankly, for the room is if you don't know where to start, just start. Because the best way you'll learn this process going through and navigating Cornell's ecosystem is just to do it yourself in collaboration with the resources that I've mentioned.
DAVID ERICKSON: So we'll take some questions from the audience right after this. So you can see there's a couple of microphones there, so I'll ask one more and then we'll turn it over to the audience.
But the one that I'm most curious about is so you showed a whole bunch of fantastic successes here. But I don't know how many papers you've published, but it's 100s, and not all of those ended up as a company.
And so when in that development process does it look like-- that you decide that this might be something that you want to commercialize or that there's potential? And do you consider things like market opportunities early in the development stage? Or these sorts of things.
STEPHEN QUAKE: Yeah, so I've got a couple of points to make. I mean, one is that one of the things I liked about being an academic is I don't need to work on things that have applications. I can work on things that are just basic science, no real application, but I find it interesting.
And so not every paper is meant to lead to something commercial or some kind of application. And I enjoy being able do both things. And then for the things that I do think are going to have some kind of application, there's a lot of value to having a peer-reviewed publication of the proof of principle.
I mean, it helps you be confident that it's going to work, for one thing. And it helps investors to be confident, perhaps overly so. I mean, it's a very interesting phenomenon I've noticed, that you can show investors the same data before publication and after publication.
Before publication, I don't know. After publication, they put too much weight on it. They believe it to be the gospel truth.
And it's like peer review has blessed it magically, and they're more comfortable deferring to the judgment of valuation rather than using their own judgment, in a way that sometimes worries me. But yes, it does carry a lot of weight with investors, if you've gotten that peer-reviewed paper. And it does smooth the way.
Also, what is venture capital good for? I mean, it's good when you can go faster by spending more money. And they like that, because they want to see a return on their investment. They've all got business models, and they want to see a return in some period.
Whereas sometimes, more money doesn't make it go faster. You need that inspiration. You've got to sort of do the slogging, and then more people doesn't help. It's just got to get things sorted out.
And that's often what happens before the publication. And so it's a good time to think about spinning it out, when more money will make it go faster, and understanding that difference.
DAVID ERICKSON: Yeah, what is that difference? How do you tell when that happens? Is it--
STEPHEN QUAKE: I don't know. It's situational, a little bit. Situational, I would say.
DAVID ERICKSON: Dave, Andrea, anything?
ANDREA IPPOLITO: Well, one thing I'll add to that is, I think there is this misconception that venture capital is the only funding you have available. There was a writer for Forbes, [? Bedap ?] [? Roe, ?] who actually looked at this. And it turns out that 76% of billion dollar-plus companies have not raised any venture capital.
And so there is a ton of different funding sources for academic entrepreneurs, whether it's the SBIR, as I mentioned, whether it's Rev-sharing agreements, or whether it is venture. And so there's tons of opportunities to fund and propel your work here. So you don't have to lean on venture.
DAVID ERICKSON: Great. So I was going to go to you, but now I have my notes here. I remember I am going to allow-- I'm going to bite on the COI question.
So there's complicated issues around COI. And this is something, I used to be the COI police in engineering here, for a bit, which was not popular.
STEPHEN QUAKE: I hope you had nuance when you served in that role.
DAVID ERICKSON: So I think it's a very good point made, that there's a lot of time and effort put together around financial conflicts of interest in particular, in particular startups. And my sense of that, the reason for that is because those are the easiest to measure. It's very easy to tell how many shares I have in blah, blah, blah, whereas the other, more nuanced ones, which can also lead to the same problems, are much more difficult to measure. So I guess I'll just maybe turn it over-- make that comment, turn it over you, and ask what would you do differently? Or how would you look at it differently?
STEPHEN QUAKE: Yeah, well, I think the nature journals have adopted a new disclosure policy that is much, much broader than just commercial conflicts of interest-- anything that could influence your scholarship. And if people were being really honest, pretty much everyone in medical school would have to say, my salary depends on my grant. My grant depends on this Nature paper.
DAVID ERICKSON: Yeah, that's true.
STEPHEN QUAKE: And therefore, there's a bit of a conflict there. And it's maybe so pervasive that people don't think it should be declared, but I think it is something that should be acknowledged at some level. What are some of the other subtleties? Other subtleties center around what you have your students work on. And you want to be really careful that that's all aboveboard, and sort of be careful where you draw the line there.
And I think a good standard for that is not, is it commercially interesting or not? That's not really the question. The question is, will it lead to a very high-profile paper if it's successful, whether it's commercially useful or not? Because that's really what your job as an educator is. And if everyone agrees this is got good, scholarly potential, have at it.
DAVID ERICKSON: Dave, Andrea?
ANDREA IPPOLITO: Well, I was at a conference two weeks ago. It's called the Global Consortium of Engineering Centers [INAUDIBLE] and the head of MIT Innovation Initiative spoke. And he said that universities are mini COI engines, and how they're actually starting to create many, many committees to go through these, scenario by scenario. Because it's so hard to set a university-wide policy on a lot of these issues.
So I think the key is, as he said, is just to disclose it, and then work through it. But don't be afraid of the repercussions. Because we are COI engines here at Cornell, whether we like it or not. So I think it's just good to put it out in the open, and there is a healthy culture here that helps work through these issues.
DAVID PUTNAM: Let's go to the students.
DAVID ERICKSON: Please.
AUDIENCE: Hi. Hi. I'm a PhD student here studying system engineering. So I'm just really interested in, besides kind of the credit you get from venture capitalists, what else are the pros of starting my company or any other company here as a PhD student here in an academic institution, rather than by ourselves, outside of this institution?
STEPHEN QUAKE: Did you get that?
DAVID PUTNAM: No, I did not.
DAVID ERICKSON: So I understood the question to be, what are the pros of starting a company here, while you're sort of attached to the academic institution and have the resources available to that, as opposed to doing it, maybe after-- you're a PhD student, you said, so maybe after you graduate, thinking about it when you're on your own a little bit.
DAVID ERICKSON: So--
DAVID PUTNAM: So what are the advantages of starting during your PhD versus afterwards? I think you should always be thinking about it during your PhD. If you have technology that you think is going to be useful, the research that you're doing, you want to publish the best paper that you can to have the validation of the idea. But do the work so that when you need to translate it, you can.
And you should do it at the time when it's most appropriate for the company to be successful. If you are finishing your PhD, and things are coming to fruition, you're ready to go, then go. And there's a great program in the College of Engineering where you can take six months.
They'll pay you during that six months to learn the process of going through that before you graduate. It's fantastic. And so that's a real opportunity where you can get it done before you graduate.
The drawback is, you might be doing your PhD for six more months. Well, if that's beneficial to you and to the company, why not? But if you want to cut ties, and you want to be free and clear, then you can patent the work, publish your paper, get out, raise your money, license the patent. Off you go to the races. So it really depends upon what your level of training is, whether you're ready to do it, whether you're the right person to do it, or whether you're better off waiting, maturing a little bit more, finding people who have done it already, bring them in, and bring it forward.
ANDREA IPPOLITO: One very practical thing, too-- as a student, you can use the magic word. So when you're reaching out to people to get meetings, that you're looking to learn and you're a PhD student. And that's a whole lot easier to do as a student than when you are not in an academic setting. So using those magic phrases of "looking to learn" can go a long way.
DAVID ERICKSON: Steve, I wonder how do your students in your-- what are the role of the students that are the innovators and inventors of these technologies in your companies today?
STEPHEN QUAKE: I mean, it sort of depends on the timing a little bit, to your point that when kind of the invention, the publication comes as sort of a capstone of a thesis, and they're ready to finish and go do it, in many cases they've gone to be part of the company, sometimes for the full run, sometimes just to spend a year getting the experience and get the technology transferred. I've seen both of those things happen.
When it's earlier in their student career, then they pretty much have got to just make this decision of stopping out for a period to do the company thing. And I always say-- it's happened a couple of times-- come back and finish the degree, for sure. Don't forget about your PhD. And one of them took 10 years to come back, but he finally did.
DAVID ERICKSON: He did?
STEPHEN QUAKE: So yeah, that happened.
DAVID PUTNAM: Did the company work?
STEPHEN QUAKE: He had two companies by then, both very successful, very successful. He didn't need to work anymore. He mostly kite surfs around the world. He's a gentleman of leisure.
And in other cases, investors wanted the students to stop out. Student looked at him, not the right timing for me. And I want to finish my PhD, and then do something entrepreneurial. It'll be a different opportunity, probably.
DAVID ERICKSON: Yeah, feel free to step up to the microphones for other questions, by the way. What I would say is I would agree with that, as I think most of the-- I think when there's initial interest displayed in a company, or you're very enthusiastic, I think oftentimes, you don't know what you don't know. And I think a lot of the programs that we have here at Cornell can really help a person learn what those things are. So the one that David was speaking of is called Commercialization Fellows, but there's many that are like that. So Alice?
AUDIENCE: I'm Alice Lee. I manage the tech transfer here at Cornell. At first I want to congratulate Prof Quake's success, not only utilizing the technologies eventually in commercialization, but the level of impact on people's lives. It is one thing in terms of thinking of the market, and saying, what's the market penetration of my product? It's another thing to hear how many pregnant women take the test, affect their lives, and their babies, and the type of risk they can avoid. I'd really want to congratulate you for that.
My question is related to maybe with the student, you mentioned about that students involved in maybe startup, in commercialization, in various stages in their education and career choice later on, and maybe it's a question for you all. How do you see the balance between the education and the career choice in the future? And what's your advice for them?
And also, you all went into different universities. How do you see that in the different university setting these days? How does the student choose a different path in their career? Thank you.
STEPHEN QUAKE: Well, thank you. That's a very good question. So I think it, sort of as preface to answering that, I think it's important to note that my role in the companies is that I'm a consultant and board member, whatever. I'm not actually line management of the company. This is another conflict of interest thing.
Faculty aren't allowed to be line managers and faculty at the same time. So you have to decide, do you want to keep running your lab and serve an advisory role, or do you want to take a leave of absence and go take a position at a company? I've always stayed at the university, and because of the reasons I described earlier, the way I like to do research.
And the university rules, which I think are sensible, are that faculty members get one day a week to consult and do whatever they want, and that consulting. And I've taken advantage of that. And I feel it should be the same for students. Students should be allowed to consult a day a week, and get involved in some of the companies that are related to their research.
When we founded Fluidigm when I was at Caltech, I asked the university, and they were OK with that. And so my students could actually spend some time seeing their work get translated, and earn money doing it, and stock, and all that, be on the license agreement. And it was offered them as an opportunity. They could do it if they wanted, didn't have to do it.
But many of them wanted to see how industry worked. Because in academia, you sort of just see the Ivory Tower part, and industry is a bit mysterious. And to be able to see how it works without making a full commitment to a career in industry, I think, was very beneficial to them. They liked that a lot.
At Stanford, it's been a bit trickier to get that. They've got different rules and so it's been harder. But I do think it's a good experience for students to be able to participate in that sort of experience, without having to commit to it as a full-time job.
DAVID PUTNAM: And I like it when the students, when they graduate, they want to be the CSO of a company or CEO of the company. And they get into the world, of the startup world, and they realize, boy, I don't know that. I don't know how to do that.
STEPHEN QUAKE: Exactly.
DAVID PUTNAM: I know the people that do. And then they learn to step back, and say, being a senior scientist is a good thing right now. Just that person knows a lot more than I do.
STEPHEN QUAKE: I couldn't agree more. I always get this question from students or postdocs. I want to start a company and spin it out. Well, biotech's a little different than IT. You can't do in a dorm room and it doesn't scale the same way.
And the best thing you can do is go to a startup where there's experienced management investors, spend some years there, learn how it's done, and then be prepared to go do your own thing. Because there's a lot about how to manage that one needs to learn. And you haven't learned it in the university.
ANDREA IPPOLITO: Yeah. And most successful entrepreneurs don't get it right their first time, although we have an exception here on the stage. And so if you look at, for instance, Mark Zuckerberg, who we also have a representative from the Zuckerberg Chan Foundation, he had multiple startups before he started Facebook. And you see that time and time again. And I personally think there's no better time than when you're a student to do that first pass, whether you're the first or second or third employee in an existing startup, where you can learn and observe, or you're just going through the motions of starting a company so you learn it, and figure out where the land mines are so that you have that to grow from going forward.
STEPHEN QUAKE: I was just thinking going to a startup company as an early employee is like getting a free MBA.
ANDREA IPPOLITO: Totally.
STEPHEN QUAKE: Because you're involved in every aspect of the business growth. And it's just incredible education.
DAVID PUTNAM: One thing is students don't want to fail, but my goodness, when a VC sees that you did that, and you did that, and you did that, and you failed, and ask you why you failed, and you say, because I did this and this and this, they want to invest in you. Because you realize what you don't know now. It's very valuable. It hurts, but it's very valuable.
AUDIENCE: So this question is specific to Cornell. I [INAUDIBLE] postdoc in the vet school. And I see there's a lot of resources for the engineering students. And I'm wondering if you have a timeline to kind of splay that out across the university, so that people-- students and postdocs from the vet school were interested in kind of these same entrepreneurial students get access to, say, a six-month program [INAUDIBLE].
DAVID ERICKSON: Yeah, so the question was, we have all these programs in engineering, and how can they be spread across the university [INAUDIBLE]? So first I would say, many, colleges have very sophisticated programs. And I think probably getting connected through Entrepreneurship at Cornell is probably the best way to look at what goes on across the university.
What I would say is that-- and I would also say that many of the programs that are, quote, in engineering are probably more accessible beyond just engineering, and perhaps in vet school. And I'll let Andrea talk about their various things. One good thing is, if you have any rich alumni who can donate and endow a program, that's one really, really easy way to get it into the vet school. So--
ANDREA IPPOLITO: I think you're exactly right, though, that Engineering is its own little special being, where it is built into the fabric. And I know there's definitely room for growth even within Engineering. For one thing, getting the faculty in your department or your college to value entrepreneurship, so that looking at department chairs and how the tenure track decisions are made, that publications aren't the only factor they're looking at. And that's definitely growing and evolving across the university.
But in terms of ecosystem resources, so very selfishly, to put a plug for the We Cornell program, which is all about generating more STEM female entrepreneurs. So this is an on-ramp onto the other entrepreneurship programs. So that's one. Second, eLab is available for anyone across the university. And that's that for-credit accelerator program I mentioned, where you get $5,000.
And then last but not least, the NSF I-Corps, the National Science Foundation Innovation Corps program. Cornell has the third highest amount of NSF funding in the country, which is crazy. And we need to celebrate that more.
And we had this great program called the NSF I-Corps, where you can apply for funding to start the customer discovery and business model validation process. And that's available for anyone across the university. So my net ID is AKI2. So if anyone has any questions about that, but I do think we need to do more and expand across the university. I'm so happy you raised that.
AUDIENCE: Since you're talking about Cornell's resources, there is also this BEST program. That is for all the students, and also postdocs across campus, for future careers in commercialization and industry, particularly, and also for modeling the Commercialization Fellow program for Engineering. And the Central is trying to create more programs like that for all students in other colleges.
And also, our office, we have expanded internship program for people for PhD students who want to learn more about commercialization. I get to do this free advertisement. OK, so please, please contact us as well. Thank you.
DAVID ERICKSON: Absolutely. Please.
AUDIENCE: Hi. I'm a BME senior. So I was just wondering, whenever we have a biomedical invention, I get the sense that we want to be selective over whether such invention is big enough to support an entire company, a de novo company or versus whether or not you just want to license it to an established enterprise. So I was wondering, are there factors or considerations that you specifically consider when making such decisions?
DAVID ERICKSON: So I think that I had a little trouble hearing that. But I think the question was, Steve mentioned there's a couple of ways in which technology may exit the university. One is through licensing to an existing entity as opposed to starting your own company. And how might one decide between those two options. Is that?
DAVID ERICKSON: So sometimes, you don't get the choice. But maybe imagine a scenario where you were trying to decide to do it yourself or you already had a big customer ready to roll.
STEPHEN QUAKE: I don't think there is any single answer. It really is situational. And not every invention is big enough to be the-- or impactful enough to be the foundation of a standalone venture. So in those cases, it makes sense as a piece of something bigger.
Other times, when it is big enough to be its own thing, often big companies have a hard time changing direction and committing resources. And so sometimes, it would be best served by being in its own entity. And so it does all factor into it.
DAVID PUTNAM: There are two cases in which a company will snatch up your technology. One is that they have the fear of missing out. And it's in their sector, and then they're worried someone else is going to get it.
Or two, it directly will take away their market share. If you directly hit their market share, they're going to license it, either to squash it, which Alice will keep that from happening, or to develop it on their own, or as an add-on to their existing products. If it's something that's still a new [INAUDIBLE] material for Johnson and Johnson, well, they've got 50 to choose from already. I have a new one. They don't really care. So in that case, it's better to develop it as a product, build value in that product, and then they'll come and take it.
ANDREA IPPOLITO: Yeah, and the other thing to think about as well is what kind of impact you want to create. If you see that companies are interested in licensing your technology for one market need, and that's not aligned with your value system, or not where you see the most impact, then that's another opportunity to start a company so that you can help be part of the process to create impact with that technology or patent.
DAVID ERICKSON: So Steve, now you are, in a sense, halftime administrator of the Chan Zuckerberg alliance, which I imagine just has enormous amounts of entrepreneurs and people just attacking it from all angles, in the sense of trying to get in on it, on the things that are happening there. And it was interesting licensing the sort of novel directions that you've had, in terms of putting together unique licensing agreements, and standards, and so forth. But I wonder, it strikes me that you would almost have the opposite problem of us, which is almost parsing who gets to play as opposed to trying to attract people to the game. And I wonder if you have any thoughts on how, from your perspective, that plays out.
STEPHEN QUAKE: Yeah, well, our philosophy is very much to defer to the inventors. I mean, we ask them what they want to do, how they think the invention is best served. Should it be in a company, in a company they've founded. Should it go somewhere else? And we're trying to help them out. And we really feel it goes back to there's no conflict, no interest thing, that the inventors have, of anybody, the best sense of where the applications are and how their invention will be served.
DAVID ERICKSON: Great. I'm going to ask one last question here. And it is you had a very unique example there, Steve, how developing a sequencer, then using somebody else's sequencer, and kind of how that opened up a new area of research and ultimately discovery, and ultimately continued entrepreneurship. And so I wonder if you can, maybe all of you, can say a little bit about how you think the entrepreneurship has fed back into the basic science elements of your research, and what you've learned from engagement. And then Andrea, maybe what you've seen with all the entrepreneurs you've interacted with, and what you've seen there. So why don't you go ahead?
ANDREA IPPOLITO: Well, I'm an Instructor in the NSF I-Corps program that I mentioned earlier. And one thing we've heard consistently from faculty and students and postdocs that have engaged in that program is that after they go through the process of interviewing potential customers, whether that's physicians, or clinicians, or patients, or nurses, or payers, especially in the health care realm, you see different unmet needs. And you see different opportunities surrounding value propositions, or perhaps a totally different customer segment than you think you were going to serve.
And so what we've heard from faculty and students that have engaged in that program is that they take those learnings and then fed it back into how they're tackling research in their lab, and looking at the portfolio of activities that they do in their lab. So it's still focused on the basic science, but on who they think potentially in the future that work can help create impact for.
DAVID PUTNAM: So the question is, how has entrepreneurship fed back to my work as an individual, I guess. So what floats my boat is basic science-- fundamentals, understanding the first principles of what's happening and why it's happening, and then making something to meet that need. But the entrepreneurship has allowed me to understand those fundamentals with the mindset of how it's going to be useful or how it's going to be commercialized or if it could be.
And so it feeds back into directing not the fundamentals, but which direction those fundamentals lead to. And all could be important. All could be completely hogwash. But at least I have the knowledge to at least choose which one would be most interesting commercially.
STEPHEN QUAKE: Yeah, I have a similar sort of sense that's helped me understand what are the best projects to do in the university environment versus in the commercial environment. And certain things are best done in the university, things that require some sort of inspiration that are bottle-necked because you haven't had the discovery yet, and need to be worked out, whereas things that require large teams of people working together across disciplines tend to be better done in companies, where nobody is worried about who's going to be the last author on the paper and all that kind of thing.
DAVID ERICKSON: Well, on that note, we'll put the last author on this paper. And again, I'll ask the audience to thank our panelists, and in particular, our newest A. D. White Professor.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
A.D. White Professor-at-Large Stephen Quake, a leading scientist in the fields of biomedical engineering and genomic medicine, led a seminar, “Academic Entrepreneurship: Creating Impact Beyond Publishing Papers,” Oct. 7 in Statler Hall’s Alice Statler Auditorium. A panel discussion followed, with David Putnam, professor of biomedical engineering; Andrea Katherine Ippolito, lecturer in civil and environmental engineering; and moderator David Erickson, professor of mechanical engineering and director of the Sibley School of Mechanical and Aerospace Engineering.