2

Convergence in Practice: Opportunities and Challenges

WELCOME REMARKS AND INTRODUCTION TO THE WORKSHOP

Marcia McNutt, President of the National Academy of Sciences (NAS); C. D. (Dan) Mote, Jr., President of the National Academy of Engineering (NAE); and Victor J. Dzau, President of the National Academy of Medicine (NAM), welcomed participants to the workshop. Dr. McNutt highlighted the challenges posed by adjusting the existing systems of research and training in ways that will support effective convergence and the importance of changing the culture of institutions to recognize and minimize barriers. She noted that this task is not a trivial one, and relies on developing shared understandings of a culture that will enable convergence, as well as how current incentives and disincentives affect research systems. Based on this knowledge, the strategies and practices that can alter disincentives to minimize or overcome current barriers to convergence will need to be identified. She looked forward to the workshop contributing to such efforts.

Dr. McNutt’s remarks were followed by comments from Dr. Mote, who noted that the three Academies have a long-standing interest in exploring new approaches that challenge beliefs about the right way to approach scientific research. Convergence-based approaches are common in addressing engineering systems, which draw on integrating multiple disciplines and can be undertaken by large teams of collaborators. Convergence is thus a natural and necessary culture change to enable new avenues for scientific and technical discovery, understanding, and translation. Dr. Mote illus-

trated this point with reference to the NAE Grand Challenges for Engineering that provide a vision for the planet in sustainability, security, human health, and quality of life, and he emphasized the role of addressing these challenges as humanity faces the 21st century and beyond.¹

Finally, Dr. Dzau spoke about how convergence in health and medicine is directly relevant to achieving the goals of the NAM’s Strategic Plan 2018–2023. Goal 3 of this plan includes, in part, to “build future field capacity by transcending disciplines, leveraging the convergence of emerging scientific areas, and working with new sectors in innovative ways” (NAM 2017). Integrating perspectives from multiple disciplines contributes to almost everything that the NAM does in addressing current challenges, such as achieving healthy longevity or promoting population health.² To address these and many other medical challenges, Dr. Dzau emphasized the important contributions made by the social, behavioral, and economic sciences in conjunction with natural, physical, and computational sciences disciplines. He reiterated that there are cultural and structural challenges to convergence and limited incentives for scientists to develop deep competencies beyond their primary disciplines. This meeting brings together a group of thinkers on such issues to analyze the challenges and share ideas.

Dr. Robert W. Conn, President and Chief Executive Officer (CEO) of The Kavli Foundation, completed the introductory remarks. Dr. Conn spoke about the system for U.S. research in science, engineering, and medicine, which is funded by a triad of federal government support, direct philanthropy, and indirect philanthropy through university endowments and other university-provided support. Contributions from industry form a fourth source of research funding. Dr. Conn viewed this diverse funding ecosystem as a long-term advantage for America. He noted that the integration of expertise across disciplines, as represented by convergence, is something that philanthropic funding has helped to catalyze. This approach is central to what The Kavli Foundation does, and he illustrated the point with historical examples as well as the award citations of recent winners of the Kavli Prizes.³ Dr. Conn ended with a call to action for the

___________________

¹ As expressed in the introduction and motivation for these Grand Challenges, “As the population grows and its needs and desires expand, the problem of sustaining civilization’s continuing advancement, while still improving the quality of life, looms more immediate” (NAE 2008, p. 2).

² See, for example, Dzau and Balatbat 2018, which discussed the role of convergence in population health.

³ Information on the Kavli Prizes in astrophysics, nanoscience, and neuroscience is available at http://kavliprize.org; accessed February 12, 2019. The “Announcement of the 2018 Laureates,” made on May 31, 2018, is available at http://kavliprize.org/events-andfeatures/announcement-2018-laureates; accessed February 12, 2019.

group to lead into the workshop sessions—“what can we do to ensure that our institutions support and reward those scientists, engineers, mathematicians, and physicians [and others] who today are doing convergence?”

NATIONAL SCIENCE FOUNDATION SUPPORT FOR CONVERGENCE

The workshop next heard from France Córdova, director, of the National Science Foundation (NSF), which has made convergence one of its 10 “big ideas.^”4 Dr. Córdova noted that convergence will be a catalytic approach to address numerous challenges recently identified by NSF as scientific priorities, such as predicting phenotype, advancing synthetic biology, and understanding the arctic system. She also noted steps that NSF has taken to focus on convergence through its latest generation of Engineering Research Centers (ERCs), by soliciting proposals through a Dear Colleague Letter,⁵ and via a set of proposed “convergence accelerators” that will tackle selected scientific questions. In Dr. Córdova’s view, several challenges remain. These include how to interlink fundamental and translational research approaches, how to most effectively collect and mine data and use data analytics to assist in merit review and post-award assessments, how the interplay between the structures of federal agencies and universities influences which changes will be required to facilitate convergence, and the need for risk taking and prioritization. She invited feedback from participants on what NSF can continue to do and what the agency can do better to support convergence. One example offered was to take a critical approach to “adapting the merit review process to represent the broad expertise needed” to identify the most promising convergent research ideas.

THE PROMISE OF CONVERGENCE FOR SCIENTIFIC CHALLENGES: HOW SCIENCE CAN PROPEL SUSTAINABLE AGRICULTURE AND LOWER THE SECTOR’S GREENHOUSE GAS EMISSIONS

The opening panel focused on what convergence can achieve when applied to example areas of science, in this case advances with implica-

___________________

⁴ See NSF’s 10 Big Ideas, available at https://www.nsf.gov/news/special_reports/big_ideas/index.jsp; accessed February 12, 2019.

⁵ See Dear Colleague Letter: Growing Convergence Research, available at https://www.nsf.gov/pubs/2018/nsf18058/nsf18058.jsp; accessed February 12, 2019, and the program announcement Growing Convergence Research (GCR), released in February 2019 and available at https://www.nsf.gov/funding/pgm_summ.jsp?pims_id=505637; accessed February 12, 2019.

tions for sustainable agriculture and reduced carbon dioxide emissions. Because alternatives to the current system can generate additional time, cost, and effort barriers, the session was intended to help illustrate why members of the research community from funders to practitioners are motivated to push through and overcome such barriers. Steven Chu, Professor of Physics and Molecular and Cellular Physiology and the William R. Kenan, Jr., Professor of Humanities and Sciences at Stanford University, opened the session and chaired the discussion. He spoke about agriculture and forestry as a significant source of carbon dioxide emissions into the atmosphere, representing about twice the emissions from energy production. Developments in areas such as genetics and crop management have increased food production yields while development of substitute products, such as burgers derived from nonanimal sources rather than livestock, can contribute to an order of magnitude reduction in land use and carbon dioxide release.

Dr. Chu was joined by Richard Broglie, Chief Technology Officer of Pivot Bio, and Zach Serber, Chief Science Officer and Vice President of Development of Zymergen, who each shared examples of research and development goals being pursued by these companies. Pivot Bio focuses on harnessing microorganisms to improve agriculture and replace synthetic chemical fertilizers. The company relies on microbiome discovery, rational design, and synthetic biology to create microorganisms that provide improved and accessible sources of nitrogen to plant roots. The convergence of disciplines as well as engagement with stakeholders such as farmers is required to drive innovation in their focus areas. Zymergen is similarly exploring advances enabled by synthetic biology and genetic engineering, in this case to create alternative, nonpetroleum-based pathways for producing chemicals and materials. The panel speakers commented on the increasingly important role of data analysis, computational tools, and automation in conjunction with biology to drive progress forward. In both cases, the advances they are pursuing reflect a need to control cellular phenotypes.

Also important to achieving success for these companies is seeding their internal system with people possessing technical expertise but also broad-based interests in expanding their knowledge, along with empowering employees and partners to ask difficult questions and push through communication barriers across fields to develop a shared language. In the case of Zymergen, Dr. Serber outlined an informal hiring process in which he categorizes applicants as having deep expertise or competency in automation, deep expertise or competency in biology, or deep expertise or competency in computation. He estimated that 25% of his staff have translatable ability, categorized by some combination of expertise in all three areas. The nature of this process at Zymergen high-

lights an attribute of convergence—that the critical presence of individual disciplinary expertise should not be minimized. A major goal is to seed the system with a sufficient number of transdisciplinary thinkers who can promote teaming and bridge across the work of disciplinary thinkers to access what would have otherwise been lost in the intersections of bounded knowledge.

STRENGTHENING THE ECOSYSTEM FOR CONVERGENCE DISCOVERY AND TRANSLATION

Daniel Stokols, Chancellor’s Professor Emeritus at the University of California, Irvine (UCI), oriented the workshop to the components that make up an ecosystem for successful convergence, including institutional and research cultures and the policies and practices that can support such work (see Figure 2-1). Dr. Stokols reiterated the challenge for institutions of keeping costs in check while optimizing leaps forward from areas that contribute to convergence, such as transdisciplinary research and team science, and staying aligned with what is happening both nationally and at the funding agencies that support research. He presented a set of institutional leverage points for convergence, including leadership priorities and campus mission statements, practices for promotion and tenure and for awarding academic credit, seed grant programs, shared space and facilities, and education and mentorship.

Dr. Stokols has particular expertise in the evidence base and practices that support working effectively in teams. While not the only way in which a convergent approach can be applied to a problem, the integration of expertise through team collaboration is now a common way in which basic research and translational advances are made. As a result, insights in this area can contribute to convergence cultures. Dr. Stokols discussed the levers of team composition and assembly, education and training, and leadership and management as critical to team functioning. Where multiple disciplines need to be integrated, their scope and diversity influence these levers. For example, “horizontal integration” brings together disciplines that share common features such as a category of analysis or approach (for example, analyses at the molecular level), while “vertical integration” links disciplines across multiple analytic levels and types and provides additional challenges. Dr. Stokols noted that much of the data collected on team performance comes from the military, emergency response, and business worlds. In contrast, science teams tend to be more diverse, more transdisciplinary in scope, and more fluid as they form, change, and re-form to tackle particular research questions.

Dr. Stokols shared information on a new “team science accelerators” program at UCI. The program includes a consultation unit within the UCI

Institute for Clinical and Translational Science⁶ that provides strategies and tools to collaborative groups. He also pointed to the collection of resources available through the National Cancer Institute’s Team Science

___________________

⁶ Further information on the UCI Team Science Consult Service is available at https://www.icts.uci.edu/services/team.php; accessed February 12, 2019.

Toolkit,⁷ including the Team Science Field Guide and transdisciplinary orientation scale, and noted that factors associated with having a transdisciplinary orientation can be assessed and nurtured as part of education and training. Northwestern University’s “Coalesce” site provides useful information and materials as well.⁸ Dr. Stokols concluded by emphasizing the need for multiscale thinking, illustrated by connecting local technologies and policies to national and global actions to promote sustainability.

SUPPORTING THE LIFE CYCLE OF CONVERGENCE

The panel that followed convened leaders of convergence-focused programs and institutes, who discussed lessons learned from successful activities and remaining challenges. The discussions thus provided experience-based examples of concepts introduced by Dr. Stokols and explored in more depth in the themes covered in Chapter 3. The session was chaired by Gene E. Robinson, Director of the Carl R. Woese Institute for Genomic Biology and Swanlund Chair of Entomology at the University of Illinois at Urbana-Champaign (UIUC), with panelists George Poste, Del E. Webb Professor of Health Innovation and Chief Scientist of the Complex Adaptive Systems Initiative (CASI) at Arizona State University (ASU); James C. Carrington, President of the Donald Danforth Plant Science Center in St Louis; Charles A. Wilson, Chief Scientific Officer of the Gulf of Mexico Research Initiative (GoMRI); and Edison Liu, President and CEO of the Jackson Laboratory.

The Jackson Lab, GoMRI, and Danforth Center are independent nonprofit research organizations, while the Woese Institute and ASU’s CASI are based within universities. As a result, panelists brought perspectives from different backgrounds and types of organizations. Whether freestanding or part of universities, these convergence centers and organizations were generally all established to address focused missions, which in some cases evolve over time. For example, the Danforth Center’s mandate is to “improve the human condition through plant science,”⁹ while GoMRI was established by BP as an independent research program in the wake of the 2010 Deepwater Horizon oil spill.¹⁰ The research themes of the Woese Institute of UIUC are reviewed every 5 years with decisions

___________________

⁷ See the website for the Team Science Toolkit, available at https://www.teamsciencetoolkit.cancer.gov/Public/Home.aspx; accessed February 12, 2019.

⁸ Coalesce resources and modules are available at https://www.teamscience.net; accessed February 12, 2019.

⁹ See the website of the Donald Danforth Plant Science Center at https://www.danforthcenter.org; accessed February 12, 2019.

¹⁰ See the Gulf of Mexico Research Initiative at http://gulfresearchinitiative.org; accessed February 12, 2019.

then made to continue, reinvent, or retire them. Similarly, ASU has retired several centers that remained too siloed to achieve the set objectives.

The panelists discussed their differing relationships to departments, faculty tenure systems, and engagement with the larger university and/or community ecosystem. These differences affect the constraints and incentives under which they can support convergence. For example, reimbursement of facilities and administrative (F&A) costs from research grants can be an important source of funds as well as a source of tension. At the University of Illinois, this F&A reimbursement is shared by the Woese Institute and by faculty members’ departments, helping facilitate relationships with departments and enable faculty participation in the center. The existence of a director’s fund or other mechanism can also be helpful in enabling faculty to travel and engage with colleagues from across research areas. As identified by all panelists, resources are critical to success, although institutions differ in the available mix of competitive grants, direct and indirect philanthropy, and other funding sources (such as business revenue in the case of the Jackson Lab).

Panelists also discussed how convergence collapses the linear model of research and development, suggesting that researchers, venture capital investors, start-up companies, large corporations, and others now increasingly engage with each other across the multiple stages of innovation, including in basic research. This requires the integration of a broad spectrum of skill sets and new kinds of partnerships. Dr. Carrington from the Danforth Center noted that the Center has been able to use a decentralized model that leverages the contributions and expertise of other organizations to contribute to convergent research, without having to create all components itself. The Center engages, for example, with local start-ups and has made successful use of St. Louis “Venture Cafes” to build relationships with this community. Several panelists also noted the important role of program management and business expertise as an essential component of successful convergence. This was true for the independent Jackson Lab, and was echoed by Dr. Poste from CASI, which found that combining skilled project managers and research faculty has been critical in progress and successful efforts to date.

LEADING FOR CONVERGENCE

The next panel convened academic leaders and focused particularly on campus efforts to support and incentivize convergence. It was chaired by Peter Schiffer, Vice Provost for Research and Professor of Physics at Yale University, and included panel members Taylor Eighmy, President of The University of Texas at San Antonio; Ann Arvin, Vice Provost and Dean of Research Emerita and Lucile Salter Packard Professor of Pediatrics and

Professor of Microbiology and Immunology at Stanford University; and Pramod P. Khargonekar, Vice Chancellor for Research and Distinguished Professor of Electrical Engineering and Computer Science at UCI.

The panel members spoke about the role of leadership in creating and communicating a shared vision and disrupting entrenched institutional policies and practices. One strategy universities are using to create shared interests and foster convergence is to identify and address challenge areas, focusing especially on areas that are timely and that are relevant to their surrounding community and state. Speakers noted that having key leaders on board with sufficient leadership continuity over time is needed for convergence to become part of the fabric of an institution. Several university leaders also noted the concurrent need to protect the researchers and valuable discoveries that remain rooted primarily within disciplines.

The discussions helped to further elucidate differences among universities in relation to oversight and financial systems. In most universities, faculty apply or choose to participate in convergent centers while keeping their school and departmental identities. However, numerous structural and financial differences exist. At Stanford, the university provost controls F&A reimbursement revenue from centers and institutes, while at the University of Illinois and UCI, revenue is shared and flows to the centers and faculty departments. In each case, these differences affect the incentives or disincentives offered to researchers working on convergence.

Panel members suggested that a convergence approach has increasingly become part of how universities conduct research at scale, but how to configure space for large projects and institutes remains an issue, along with the need to rethink traditional user-fee-based shared facilities. Another important dimension raised by the panel is how to identify the faculty leaders who will make convergence happen and how to facilitate and support their ability to do so. Early signs from recent waves of junior faculty indicate that these recruits increasingly seek appointments in multiple departments, while fewer want to be limited to a single department. Universities are looking at strategies to address this demand.

Finally, panel members shared examples of initiatives their universities have undertaken to stimulate convergence. These commonly include the use of seed funding programs. UCI has established a program, for example, in which faculty submit proposals for university seed funding that are peer reviewed and the investigators of successful projects are encouraged to consult with a UCI team science experts group. Seed funding can also be provided in the form of university professorships, and there were mentions of how such professorships provide productive latitude and capital to undertake more unusual, risky, or creative ideas that may not survive grant peer review. On the other hand, panelists

noted that seed funding programs are not a panacea and need to be intentional in design and management to be successful. In sharing examples of approaches to supporting convergence that had not been as effective as intended, one panelist noted that a former seed funding effort made relatively large grants with little oversight, leading to reduced impact and lower than anticipated success.

CONVERGENCE IN ACTION

The final panel of the first day featured a moderated discussion among Investigators undertaking convergent research and representatives of funding organizations supporting such activities. It was chaired by Cato Laurencin, University Professor, CEO of the Connecticut Institute for Clinical and Translational Science (CICATS), and Director of the Institute for Regenerative Engineering at the University of Connecticut, with panelists Joanna Aizenberg, Amy Smith Berylson Professor of Materials Science and Professor of Chemistry and Chemical Biology at Harvard University; Joseph DeSimone, Chancellor’s Eminent Professor of Chemistry at the University of North Carolina at Chapel Hill, the William R. Kenan, Jr., Professor of Chemical Engineering at North Carolina State University and of Chemistry at the University of North Carolina, and CEO of Carbon; Maria Pellegrini, Executive Director of Programs at the W.M. Keck Foundation; and Dragana Brzakovic, Senior Staff Associate in the Office of Integrative Activities at NSF.

Researchers play an important role in fostering a culture of convergence by leading their laboratories and bringing together members with diverse expertise and ways of thinking. Members of the panel noted that academia is seen as risk averse and they have had to create some changes for themselves, including pursuing affiliations at additional schools to bring together the expertise and resources necessary to address their research goals, creating spaces and opportunities for people of different skill sets to mingle within their laboratories, or even founding start-up companies to pursue ideas. Institutional trust and flexibility were seen as important to faculty success, with one panelist noting the advantage of starting at an institution such as Bell Laboratories, which did not follow an academic promotion and tenure process. This provided a degree of freedom from the metrics and constraints academic promotion and tenure can entail.

From the funding agency and foundation perspective, there was recognition that the more convergent a proposal, the more challenging it can be to review. It is hard to capture the review criteria that should be used, and convergent, complex projects can seem to be higher risk than other types of proposals. Such projects represent an investment in people and

in their ability to work effectively within and across disciplines and as an integrated team. Panelists recognized that such activities take resources, especially people’s time, and often require sustained funding. From a funder perspective, such projects require extra contact with researchers and additional oversight to ensure that projects remain deeply rooted in expertise, both disciplinary and transdisciplinary, and that they progress successfully. The panelists also noted that convergent science can be risky in pushing the boundaries of knowledge, and that some failure is to be expected and would need to be tolerated. Even in the case of failures, however, there can be collateral benefits to having attempted such convergent activities and lessons can be captured to refine and improve future activities.

ACHIEVING INSTITUTIONAL CULTURE CHANGE

The second day of the workshop began with a presentation from Holden Thorp, Provost and Rita Levi-Montalcini Distinguished University Professor at Washington University in St. Louis, who was introduced by Tamar Gendler, Dean of the Faulty of Arts and Sciences at Yale University.

Dr. Thorp spoke on how to move from discussion to ideas and from ideas to action. He noted that common impediments to convergence include a lack of information about convergence and what it entails; lack of practical resources (including physical spaces, capital, and materials); and lack of structural or institutional resources. He emphasized the role of incentives and cultural norms in nudging people’s behavior in desired directions, such as toward convergence, as well as the human dimensions involved in convergence efforts. Organizational structures alone cannot dictate these human elements.

Dr. Thorp illustrated a number of these challenges across multiple parts of a university. Turnover among university presidents and provosts can be high, with leaders frequently staying only several years. Faculty often do not meet their colleagues from other disciplines unless they are in administrative roles. There is a need to keep senior faculty engaged, especially at their productivity peak when they may become fatigued by their many years in the proposal-grant cycle. For a university, keeping its highly productive and highly cited faculty engaged is necessary to accomplish goals, including culture change toward convergence. Meanwhile, junior faculty are under intense pressure to obtain grants and make progress along the tenure pathway. In some fields, these pressures involve low funding success rates, increased time before obtaining major grant funding, and long tenure timelines. In addition, graduate students continue to be presented with the idea that the optimal career path is

to become a faculty member; alternative career paths thus need to be respected and seen as valuable.

The discussions also helped to elucidate the diversity of situations across different fields. While there appear to be numerous ongoing structural, financial, and workforce challenges affecting academic biomedical research, other disciplines face different challenges. For example, in areas of computation, it was noted by some participants that graduates are in such high demand that it is hard to keep trainees and faculty in academia. The human dimensions, especially the critical roles of institutional culture and of how to nudge and influence human behaviors, along with distinct differences in the current challenges faced by different disciplines, will all affect the success of efforts to integrate expertise across and within knowledge boundaries and toward convergence.

STRATEGIC PERSPECTIVES

The final panel of the workshop brought together perspectives on the tension points and challenges in fostering effective cultures of convergence and possible steps to move forward. The session was chaired by Peter Lee, Corporate Vice President of Artificial Intelligence and Research at Microsoft. Panel members were Robert W. Conn, President and CEO of The Kavli Foundation; Christopher Martin, Interim Vice President of Science Programs at The Kavli Foundation; Megan Smith, CEO of Shift7 and third U.S. Chief Technology Officer; Carrie Wolinetz, Associate Director for Science Policy and Director of the Office of Science Policy at the National Institutes of Health (NIH); Peter Highnam, Deputy Director of the Defense Advanced Research Projects Agency (DARPA); and Fleming Crim, Chief Operating Officer of NSF.

Panelists commented on the tensions associated with pushing convergence through “top down” actions by leadership and federal programs versus those arising “bottom up” by demand from researchers. Although agencies such as DARPA, NIH, and NSF differ in their missions and operations, they have a number of levers with which to drive culture change among their research communities. DARPA focuses on taking informed risks through programs that aim to track and drive the leading edge of science and technology, and success requires program managers within the agency to innovate, think creatively, and bridge across multiple areas. NIH plays a central role in funding biomedical research through its 27 autonomous institutes and centers. Convergence represents one means to solve mission-relevant scientific questions at NIH. For example, the agency had not previously funded many of the principal investigators

on its awards for the BRAIN Initiative,¹¹ but new awardees include many from fields such as chemistry and physics. NSF supports a significant amount of fundamental, investigator-driven research, although it has also become more intentional in incorporating a focus area on convergence as part of its portfolio. But both funding bodies and research-conducting institutions and companies face the difficult challenge of balancing enthusiasm for convergence with the many competing priorities and existing demands on their budgets.

Panelists also compared and contrasted features of academia and industry, and suggested there may be opportunities to learn further from these comparisons. Industry is driven by the need to understand and exploit emerging technologies to maintain competitiveness and meet shareholder expectations. Industry in general is also commonly organized around applied projects that have specific, time-limited missions and objectives, and these project teams form and re-form as needed. Although this system provides less stability than the disciplines that underpin academia, the industry system can provide opportunities for people to develop new skills and leadership.

On the other hand, a primary driver for researchers in university settings is academic influence through peer recognition, publication, and tenure. There may also be generational differences in interests and expectations. Panelists noted that the culture of convergence is already infused in younger generations of researchers, and thus there is a need to “enable young ideas and get out of the way.”

As panelists looked ahead, they noted several enduring challenges: how to build bridges to people who may not currently support convergence, how to influence patterns of behavior and create the structures, policies, and practices to make convergence work at scale, and how to measure the value of convergence approaches. Ideas that emerged from participants on how to accomplish these goals across diverse settings and research problems are captured in Chapter 3 (see, in particular, Table 3-2). For example, it was suggested that one necessary path forward will be to find ways of collating and sharing examples faster and of building communities of practice or interest.

___________________

¹¹ Brain Research through Advancing Innovative Neurotechnologies (BRAIN) is a federal and private-sector initiative announced in 2013 that “seeks to deepen understanding of the inner workings of the human mind and to improve how we treat, prevent, and cure disorders of the brain.” See https://www.braininitiative.org/alliance; accessed February 28, 2019. By bringing together expertise from multiple fields, it represents an example of convergence.

PARALLEL BREAKOUT SESSIONS AND INNOVATIVE IDEAS FROM PARTICIPANTS

At the end of Day 1, three parallel breakout sessions provided participants with time to engage in interactive and in-depth exploration of the workshop topics. The three sessions were

1. Institutional Incentives That Enable Convergence. Chair: Michael Witherell; Deputy Chair and Rapporteur: Nadine Kabbani
2. Funding Structures That Support Convergence. Chair: Sue Merrilees; Deputy Chair and Rapporteur: Norbert Tavares
3. Changing Norms and Culture of Academic Research Toward Convergence. Chair: Joseph DeSimone; Deputy Chair and Rapporteur: Gifford Wong

Participants were asked to consider the following set of questions as starting points:

• Reflecting on what you have heard so far, how do the challenges encountered and strategies to overcome them resonate with or differ from your own experiences?
• What specific actions can be taken to most effectively enable convergence? What critical limitations or pitfalls remain?
• What needs to happen next: What are the two to four concrete things that would make the most difference in fostering cultures of convergence?
• What will be needed to achieve these ideas (for example, who needs to be engaged, where could the discussions happen, what resources [not just financial] will be needed, etc.)?

Session rapporteurs shared the main messages from these breakout discussions with the full workshop on Day 2. The concluding sessions at the workshop then provided further time for discussion among all participants. A diversity of topics emerged, including a call for a deeper evaluation of the role of prizes and other incentives in supporting convergence, as well as a call for greater opportunities for coordination between philanthropic and federal agency funders, which each have particular priorities and visions related to supporting convergence. Numerous participants also reinforced the critical role that national laboratories can play in integrative research, as well as the need to strengthen collaboration and support access to large instruments that drive certain convergence projects. Key thematic points that emerged from the workshop discussions are presented in Chapter 3.