NIH Conference on Knowledge Environments for Biomedical
Research (KEBR) - December 11-12, 2006
Conference Summary
Purpose: The KEBR conference aims to elucidate strategies
to develop, sustain, and enhance knowledge environments for biomedical research.
Such strategies will be useful to the research communities generating data,
to those developing informatics tools and resources for biomedical research,
and to those who serve as stewards of support for the research enterprise.
Background: Knowledge environments are essential to integrated
resources combining informatics tools and data sources. These knowledge environments
are defined by the explicit understandings of concepts related to the data
upon which they compute. Key data-related concepts include: What the data
represent, how one type of data relates to another, how these terms and relationships
are defined in the real-world, and how those definitions are related to each
other. Controlled vocabularies, specified ontologies, and explicitly defined
data models are components of knowledge environments that help mediate among
data-related concepts. A community-wide knowledge environment will bring to
bear the full power of sophisticated computational approaches on the full
range of research questions posed by that community.
Knowledge Environment Ideals: Knowledge environments for
biomedical research must be characterized by sustainability,
adaptability, interoperability, and "evolvability."
- Mechanisms must exist to sustain and maintain useful knowledge environments.
- Knowledge environments must be adaptable in real time to new knowledge
and new technologies.
- Knowledge environments must permit integration across multiple data forms
and levels of biological organization.
- Mechanisms must exist to ensure that biomedical research community needs
exert the prime selection pressure on the evolution of knowledge environments.
Meeting Participants: The conference will include scientists
from across the full gamut of NIH-relevant research, as well as domain-neutral
experts in computational and informatics aspects. NIH staff representing program,
review, and policy perspectives will participate, as will staff from other
federal research funding organizations, such as NSF and DOE.
Meeting Agenda: Context in each of three main areas will
be provided through plenary and discussion sessions. These sessions will highlight
attributes of successful informatics projects that address key considerations
for knowledge environments:
- how information is represented in a structured, persistent, and computable
way
- how it is instantiated and accommodated in software tools and resources,
and
- attendant socio-cultural issues related to development of knowledge environments.
Breakout sessions will provide structured input from groups assigned specific
topics. A final summary session will present breakout reports and overall conclusions
by the session chairs.
Meeting Deliverables: The strategies identified by this
conference will be described in a report and discussed in a follow-on meeting
with the leadership of cross-cutting NIH groups, particularly those groups
leading the trans-NIH Roadmap Initiatives. However, the broad strategies developed
will be useful across NIH and all NIH-support research.