National Academies Press: OpenBook

Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium (2023)

Chapter: 2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation

« Previous: 1 Reevaluating the Structure of Institutions and the Scientific Enterprise
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

2

The “Valleys of Death”: Addressing the Translational Gap Between Discovery and Innovation

When Vannevar Bush wrote Science, the Endless Frontier,1 he envisioned, “with incredible brilliance, the importance of universities and research institutes, and particularly their role in generating the next generation of scientists and combining that with the conduct of research,” said Shirley Tilghman, professor of microbiology at Princeton University and president emeritus of Princeton, who moderated the second panel at the symposium. He then envisioned that new knowledge generated by that research would be transferred seamlessly into the private sector, where it would ultimately serve the public good.

“We’ve discovered in the course of the last 75 years that that seamlessness is a little rockier than he imagined,” said Tilghman. The three speakers on the symposium’s second panel—one from government, one from academia, and one from the private sector—discussed these difficulties and possible ways to speed the transfer of new knowledge into societal benefits.

THE NEED FOR FLEXIBILITY AND RISK TAKING

At the Department of Energy, Under Secretary for Science and Innovation Geraldine Richmond oversees both basic research funded by the Office of Science and applied research focused on such topics as energy efficiency, the electric grid, and carbon sequestration. Reflecting the department’s mission to “take basic research and put it to good use,” Richmond also oversees 14 of the department’s 17 national laboratories, “which are our engines of innovation and have been working a long time to bridge this gap.” As such, a major aspect of her job has been to get new knowledge through the so-called valley of death.

Richmond did basic science for 35 years at the University of Oregon and said that she was never “going to leave behind my love and dedication to the basic sciences.” But current circumstances, including the COVID-19 pandemic, climate change, and the invasion of Ukraine and resulting loss of access to critical materials and other resources, call for doing more with the results of basic science. Also, the next generation of science and technology professionals have “very different goals in mind about how they want to use their science or do their science,” said Richmond.

___________________

1 National Academies of Sciences, Engineering, and Medicine, 2020, The Endless Frontier: The Next 75 Years in Science, Washington, DC: The National Academies Press, https://doi.org/10.17226/25990; V. Bush, 1945, Science, the Endless Frontier, A Report to the President by Vannevar Bush, Director of the Office of Scientific Research and Development, July 1945, Washington, DC: United States Government Printing Office, https://www.nsf.gov/about/history/nsf50/vbush1945.jsp.

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

The line between basic and applied research is often “in the eye of the beholder,” Richmond observed. That has been a problem, because funding agencies and the scientists doing research may have different ideas about what constitutes basic research, and these different conceptions may constrain scientists to the types of research that they think a funding agency would approve. “Those are the kinds of things that we have to break down,” Richmond said, “to let our scientists go from a discovery to creating some kind of an invention.”

The CHIPS and Science Act of 2022 (where CHIPS stands for Creating Helpful Incentives to Produce Semiconductors; P.L. 117-167) and other recent federal legislation have added resources along the entire spectrum of research. “The funding is there,” said Richmond. “We have to be ready to take it on and do what we need to do.” This requires being willing to take risks, which can be harder to do when funding is scarce. It also requires looking beyond publications and citations in evaluating impact. “If our departments and scientific societies and even our funding agencies are judging research impact solely by the number of publications, citations, and H-indices, that can reduce risk taking and stand in the way of those willing to step out of their norm and work on solutions to the crises that we face today.” Richmond has been working to broaden the evaluation of scientific impact. “This requires deeper thinking in our academic and research institutions to develop impact criteria for science and engineering endeavors that match the mission and values of each institution.”

Institutions as well as funders need to be more flexible in letting researchers and students think as broadly as possible in their research endeavors, she said.

Basic science will continue to be the seed corn that generates new technologies and well-trained students, but “we all have to embrace flexibility and being able to take risks because we can’t stay in the mode that we’ve worked in for so long.”

OVERCOMING THE BARRIERS TO TRANSLATION

Mark Fishman, professor in the Harvard Department of Stem Cell and Regenerative Biology, pointed to two fundamental reasons why basic discoveries can fail to produce improvements in human health or other societal benefits. The first is that translational science, which in biomedicine includes such steps as assessing toxicity and first clinical trials in humans, is difficult. “It’s hard to discover a medicine and take it all the way through development,” he said. “A lot of these studies are performed in cell culture or in animals, which of course are surrogates for humans. [But] there aren’t good animal models for most diseases. What is a depressed mouse?”

The second issue is that institutional structures can impede the movement of new knowledge to new products. The biomedical enterprise is an ecosystem dispersed among academia, biotechnology and pharmaceutical companies, and medical centers, and “there’s no good means for efficient transfer between them.” Each of these entities wants to profit from its activities, so it has set up bureaucracies like tech transfer offices in academia or lawyers in the private realm directed toward protecting and patenting intellectual property. Thus, many of the relationships among parts of the ecosystem are formalized and carried out among bureaucracies rather than among scientists and clinicians.

Also, each kind of institution has unwritten but real biases, Fishman noted. Academic institutions, for example, tend not to reward discoveries involving biotechnologies or pharmaceuticals. For their part, pharmaceutical companies can feel that basic scientists, whether in academia or in their own companies, overvalue the potential of drug candidates from their discoveries and are not interested in taking the steps needed to commercialize a drug. Finally, medical centers, which play a critical role in disease phenotyping, have become businesses focused in many cases on patient management rather than inquiry.

Fishman pointed to several solutions that can help resolve these problems. First, people and institutions need to build up the translational sciences, recognizing that activities such as preclinical and early clinical research are

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

worthy forms of science and then instructing undergraduate and graduate students in these sciences. There remains much to learn, for example, about toxicity, metabolism, animal models of disease, genomics, and proteomics. “This will give us a new textbook of medicine,” Fishman said. “In addition to the ways we standardly subdivide patients by pathology, we could have textbooks based on therapeutic implications, as is done now, for example, for breast cancer,” thereby bringing a higher level of analytic rigor to potential treatments.

The barriers between academia and private ventures also need to be broken down, Fishman observed. Increasing numbers of scientists and students are interested in translation and innovation, and support for these activities is available from philanthropic and venture biotech organizations. Encouraging academic laboratories to take on biotechnological issues, and giving rewards based on success in this arena, would help to break down barriers. So would sharing space and staff between academic and private enterprises, including the appointment of investigators who have one foot in academia and the other in the private sector. Students should be supported with funds from private enterprise, he said, so that they can get a living wage early in their careers and be able to expand academic research into private enterprise. Tech transfer offices and lawyers need to be tasked with facilitating the science rather than maximizing profits. Finally, new not-for-profit drug discovery entities independent of both academia and private bureaucracies and with long time horizons could host work that lasts a decade or more in going from a basic discovery to a therapeutic application.

The current set of institutions have “relatively short time horizons based on financial imperatives,” said Fishman. For example, given the current opioid crisis, how can fundamental neuroscience, with support from philanthropies or from the new Advanced Research Projects Agency for Health, generate new modalities to cure pain?

A HYBRID COMPANY TO BRIDGE THE VALLEY OF DEATH

When the company DeepMind Technologies was founded in 2010, co-founder and chief executive officer Demis Hassabis sought to combine the cultures of several different kinds of research institutions. One inspiration was the golden era culture of Bell Labs and government programs like the Apollo program. Another was that of highly innovative and creative companies such as Apple and Pixar.

By a variety of measures, this approach has been a success. DeepMind has published more than 1,000 research papers, including more than 25 in Nature and Science. Even more importantly, said Hassabis, DeepMind has delivered major breakthroughs, such as the programs AlphaGo, which beat the world champion at the game Go a decade before experts had predicted it would happen, and AlphaFold, which has revolutionized the ability to predict a protein’s three-dimensional structure from its amino acid sequence.

Games have been a very efficient way to develop ideas and algorithms, Hassabis remarked. Millions of instantiations can be run efficiently in the cloud. Games have clear metrics against which to optimize code. That code and the experience derived from games can then be used to develop general-purpose algorithms that have real-world impact. Many products at Google, which bought DeepMind in 2014, use some part of DeepMind’s technology.

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

Hassabis has a special interest in how to accelerate the progress of scientific research and innovation, an interest exemplified in the development and use of AlphaFold. The program has been able to predict how all 200 million proteins that are known to science are likely to fold, and this information was released as an open-source database to maximize its influence on the scientific community. A new sister company to DeepMind is using this information and developing other new artificial intelligence and computational technologies for accelerating the drug discovery process.

With its 1,000-plus scientists and engineers, DeepMind has also sought to create a hybrid of top-down guidance and bottom-up creativity. It clearly specifies its goals and the metrics that will be used to measure progress toward those goals, but it does not specify the solutions to those goals. “The best idea wins,” said Hassabis. DeepMind has also emphasized multidisciplinary research, which is “an easy thing to say but very hard to do at a fundamental level.” Research at DeepMind combines neuroscience and computer science but also has added other disciplines, including biology, chemistry, physics, philosophy, and the social sciences to develop its technology and deploy it safely and ethically. The greatest breakthroughs of the future will come at the intersection of two or more disciplines, which means that universities are going to have to think about how to nurture more multidisciplinary leaders. Today, grant-giving bodies have a bias toward specialists, making it difficult to be a multidisciplinary scientist with expertise in multiple fields, “but I believe those types of people will be extraordinarily valuable for future progress,” said Hassabis.

In general, technology and science are becoming increasingly intertwined, which creates a new dynamic that universities need to consider. Institutions of higher education need to facilitate the creation of career paths like professional engineering in the research environment. They also need to create the technical infrastructure that will enable new technologies and new kinds of analyses, which will open up many new opportunities, such as using artificial intelligence to model complex natural phenomena. “I find that really exciting about what it could mean for the rate of innovation and progress,” Hassabis concluded.

THE GOVERNMENT’S ROLE AS A SUPPORTER OF TRANSLATION

During the discussion period, moderator Tilghman asked about the government’s role as a supporter of translation between basic science and societal benefits. In other words, can government catalyze the destruction of the valley of death?

Richmond said that a change in attitude is hugely important. One reason the valley of death exists is because companies are not going to engage in a translational process if the risk of doing so is too high, which requires that the risk be reduced. She pointed out that the Department of Energy thinks a lot about how to de-risk the translational process. For example, it supports demonstration projects in which companies can learn and then move on to deployment. It also makes loans in high-risk areas so that companies can afford to deploy a new technology.

Fishman cautioned against the government using its funds in any way that would detract from its commitment to basic science. Funding for fundamental research is “relatively small compared to what’s needed for development,” he said. In the same way, academic institutions should not necessarily be spendings lots of money on making new drugs. Philanthropies, however, could set up novel types of entities that could function in ways similar to drug companies but with, for example, time frames for delivering therapeutics to specific population targets.

Hassabis pointed out that the British government has set up a new funding body called the Advanced Research and Innovation Agency that is designed to fund riskier research in a fashion similar to that of the Defense Advanced Research Projects Agency, or DARPA. The private sector also needs to take risks, he said, and not rely entirely on government to de-risk potential innovations. The realization that venture capital investments can be successful has led to an increase in such funding in the United Kingdom. But the real limiting factor is not money or ideas, he said. “It’s leaders.”

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

Young scientists are taught how to do research, but training in these other skills remains ad hoc and haphazard. “One has to be very good at those things in order to continue to get the backing and the funding and the support that one needs to stand a chance of making it across the valley of death. But those are skills that are not traditionally valued or taught within our systems.” Multidisciplinary grants to encourage multidisciplinary thinking could build such skills as entrepreneurship, management, and articulating a vision, he said.

MEASURING IMPACT

In response to a question about how best to measure scientific impact beyond publications and citations, Richmond said that “you have to know what your values are.”

Thus, publications and citations give an indication of productivity, but they are not necessarily a measure of the influence of scientists and their research on a broader scale. In evaluating individuals, appropriate questions might be: Has this person changed the field? Has this person created something many other people are adopting? Has this person changed the culture of a department by working over a long period to do so? What has this person done over the span of a career? Annual evaluations can lose sight of long-term effects. “This is something that our institutions and agencies have to think very carefully about, because I do believe that it’s restricting our ability for scientists to do things that are more risky or interdisciplinary and to take discovery research to applied areas.”

Fishman emphasized the importance of solving meaningful real-world problems that people care about. “Have you shown something’s possible that the majority of the field didn’t think was possible? Have you opened up a whole new way of making progress?” He observed that while he was at Novartis he participated in a pathway analysis to figure out which of the company’s thousands of researchers were making major contributions. “We found that the nodes were never the leaders. They were people who were buried, sometimes far away. . . . They were helping the whole place float.” Identifying such people and figuring out how they make everyone in an institution better would be a valuable contribution to the organization of research, he said.

OPEN ACCESS

A question about the implications of open access to federally funded research results led to both policy recommendations and cautions. The research community has advocated forcefully for open access for many years, said Richmond. At the same time, the world is very competitive in many areas of innovation, and protecting information can be important in realizing the benefits of innovation. “I’m fully in favor of open access,” said Richmond. “I just want us to think carefully about this and study carefully any negative implications to determine whether it is achieving its purpose.”

Hassabis pointed to the possible safety and ethical aspects of open access. “Once you open source something, then anyone can use it, including bad actors. You can’t pull that back again.”

A major constraint on open access, said Fishman, is that fields like toxicology are not considered science by many journals.

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

ELITISM IN FUNDING

Finally, in response to a question about diversifying the pool of researchers who receive funding (which is discussed in the next chapter), Hassabis acknowledged that funding often goes to a limited group of people. For its part, DeepMind has worked hard to increase the diversity of the talent pool from which it draws. It funds millions of dollars of scholarships and chairs in universities to further diversity. It directs significant funding toward the master’s level, which is “a really good place to try and do this, where we take STEM [science, technology, engineering, mathematics] students from disadvantaged backgrounds or other underrepresented backgrounds and then almost put them through a conversion course in machine learning at some of the top universities in the U.K. They come out of that being able either to apply machine learning in their own domain of expertise or go further into a Ph.D.” By increasing the diversity of people in STEM, the diversity of inputs into problem-solving is increased as well. “In the end, if we want these technologies to serve and benefit everyone in the world, we need everybody’s input and voice in the design of these systems to make sure they are not biased, that they are fair, and that they are used for the good of everyone.”

Fishman similarly noted that funding tends to gravitate toward people whom funders know and trust. “The only way to break this down is to broaden the club,” he said. “The good news is that there’s now more and more crossover in this club between biotech, venture, and academia, so you can come in through several different routes.” It is also important not just to admit people into the club but to support them, Fishman said. Private companies can play this role, both within the company and by supporting researchers in academia. “There are lots of opportunities to change the culture.”

Box 2-1 is a transcript of the pre-recorded remarks provided by Indiana Senator (R), Todd Young, during the symposium. Senator Young’s remarks consider how legislature like the CHIPS and Science Act can play a pivotal role in maintaining the competitiveness of the country.

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×

This page intentionally left blank.

Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 9
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 10
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 11
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 12
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 13
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 14
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 15
Suggested Citation:"2 The "Valleys of Death": Addressing the Translational Gap Between Discovery and Innovation." National Academies of Sciences, Engineering, and Medicine. 2023. Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium. Washington, DC: The National Academies Press. doi: 10.17226/26863.
×
Page 16
Next: 3 Producing the Right Technical and Professional Science Workforce: Ensuring Inclusivity, Increasing Diversity, and Improving Training »
Transforming Research and Higher Education Institutions in the Next 75 Years: Proceedings of the 2022 Endless Frontier Symposium Get This Book
×
Buy Paperback | $22.00 Buy Ebook | $17.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

On September 22, 2022, the National Academy of Sciences held a symposium entitled Endless Frontier 2022: Research and Higher Education Institutions for the Next 75 Years. The event was a follow up to a February 2020 NAS symposium convened to celebrate the 75th anniversary of the landmark report Science, the Endless Frontier.

Building on the 2020 symposium and on lessons learned during the COVID-19 pandemic, the September 2022 symposium sought to generate tools, strategies, and actionable steps that people and institutions can implement to ensure that science and technology continue to serve the public good. The symposium was designed to progress from broad perspectives that encompass the entire science and technology enterprise to consideration of more specific issues. This proceedings summarizes the 2022 symposium.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!