Group Science J. Marc Overhage MD, PhD Regenstrief Institute Indiana University School of Medicine

Team science: A combination of serendipity and intention  Collaboration through happenstance  The careful building of teams

Characteristics of scientific problems that require a team approach Urgency and complexity of scientific and clinical problems today often dictate the need for a team-based approach Different disciplines/expertise required to solve a problem; impetus is shared need project-oriented product-oriented Different approaches required to solve a problem; impetus is shared biological system or set of problems Common facility/instrumentation/database required to solve different problems; impetus is shared approach Grand challenges for which a critical mass does not exist; impetus is intellectual challenge and potential high pay-off Combinations of the above

Guiding Principles Individual creativity should be preserved while taking advantage of the synergy of team approaches Leadership, management structure, and communication are essential elements of team science Integrity, trust, and respect lay the groundwork for effective team science All teams need an impetus, a motivation that brings the team together and encourages collaboration Team members should be selected based on team needs and not necessarily location, including international participants

Models or paradigms for team science Variables: –Size: number of investigators and participants –Location of participants: co-located or distributed –Goals Team building and working at large-scale production facilities for development of research resources – project-oriented, product-oriented –Structure: Director(s), advisory boards, staff, budgetary allocation –Issues vary depending on the specific set of variable –Examples: Interdisciplinary team focused on a common problem Interdisciplinary team focused on developing a product (new tools, new technology for clinical applications, etc.) Team using shared facility/instrumentation/database Team carrying out clinicaltrials

Factors that make team science paradigms succeed or fail Leadership: vision, enthusiasm, commitment, true team spirit Communication: time, effort, technology, training Management structure: integrate leadership and communication Team-friendly environment: integrity, trust, respect, sharing Institutional commitment: space, administrative support, faculty investment Common to all models

Characteristics of Group Science Scale Specialization Requires middleware

Middleware addresses application requirements by: facilitating the discovery and utilization of scientific data, computers, software, and instruments over the network in a controlled fashion integrating remote resources and collaboration capabilities into local experimental, computational, and visualization environments diagnosing (or averting) the cause of failures in these distributed systems managing, in a community setting, the authoring, publication, curation, and evolution of scientific data, programs, computations, and other products.

secure control over who does what, where the challenging demands of DOE science lead to unique requirements information integration and access, to computers, storage, networks, code, services, instruments, and people coscheduling and quality of service, coordinating resources critical to many experimental and computational scenarios effective network caching and computing, to stage large datasets and rapidly access computing services to support collaborative scientific work among distant partners and collaborators monitoring and problem diagnosis—end-to-end and top-to-bottom for science applications and services.

The complexity of modern science and technology increasingly requires true collaborative efforts from individuals with different training and expertise to realize the benefits of these advances for health and medical care. Cooperation in project planning and execution and sharing of data and resources are essential.

In the context of this symposium "team" science is not synonymous with "large" science, but includes the full gamut from small to large teams. There are many examples of effective, collaborative teams involving as few as two investigators.

Organizational model Appropriate problems Synergies Dimensions (cross dept, academia/private) Size Rewards

Large-Scale Biomedical Science: Exploring Strategies for Future Research (2003) IOM

BECON 2003 SYMPOSIUM ON CATALYZING TEAM SCIENCE June 23-24, 2003 Natcher Conference Center National Institutes of Health Bethesda, Maryland