CTSA Ontology Workshop

Venue: Hilton Garden Inn Orlando Airport, 7300 Augusta National Drive, Orlando, Florida 32822

Date: February 11-12, 2013.

This is the second meeting of the new IKFC Clinical and Translational Science Ontology Affinity Group which has been established to leverage the use of common ontologies to support different aspects of information-driven clinical and translational research, The focus of this meeting is to explore new and existing uses of the common ontologies to support sharing of data, particularly in the evaluation of scientific research, and even more particularly in the evaluation of CTSA activities, resource sharing, researcher networking, and biomedical informatics.

Schedule Monday, February 11, 2013

9:00-9:30 Participant introductions: What are your interests? What do you hope to gain out of participating in the ontology interest group?

9:30 Barry Smith: Limning the CTSA Ontology Landscape

10:00 Existing initiatives to support consortium-wide data standardization and integration

CTSAconnect/eagle-i/VIVO/ISF Integrated Semantic Framework (Carlo Torniai)

Impact assessment (Kristi Holmes)

Permissions and consent (jihad’s group)

Ontology-based resource and information tracking to support evaluation of translational research (Dagobert Soergel)

11:15 Coffee

11:30 Practical applications of ontology to support translational research Brief updates from:

Warren Kibbe: Using Disease Ontology for mining medical records

Yongqun He: OGMS, IDO and VO

Rob Wynden: The HOM Health Ontology Mapper / Federated i2b2

(TBC) Shahim Essaid: Mining Pathology Reports for Better Specimen Query

12:15 Lunch

13:00 Breakouts to discuss ontology-based support for some or all of the following (for more details see here

Research Networking

Evaluation (including evaluation metrics)

Community engagement

Comparative effectiveness research

Regulatory knowledge

Researcher training

Research profiling

Each group discusses:

1. What has been done thus far?

2. What are the different roles that a CTSA can or should play in these contexts?

3. How can ontologies help?

2:45 Coffee

3:00 Report back and review in large group

3:30 Continued small group work in the areas identified in the breakouts. to address

5:00 Close

Schedule Tuesday, February 12, 2013

9:00- Regrouping of breakouts to initiate creation of working teams in areas such as:

Use of ontologies in research tracking and evaluation

Use of ontologies in management of patient data (including regulatory and consent data)

9:15 Breakout sessions

10:45 Coffee

11:00 Report back and review in large group

12:00 Lunch

13:00 Identifying goals and key partnerships for the CTSOAG

14:30 Coffee

14:45 Outreach plan, documentation, summary, action items

Possible deliverables:

A list of ontology-related efforts across the CTSA Consortium compiled from the knowledge of participants

A draft list of requirements for an ontology-based system to support CTSA research and evaluation

A draft list of the major families of data which will be collected through the activities of a CTSA institution

16:00 Close

Participants

Sivaram Arabandi (Smart Content Team, Elsevier)

Mathias Brochhausen (Arkansas CTSA)

Jennifer L. Bufford (Arkansas CTSA)

Michael Conlon (University of Florida CTSA / VIVO)

James Demery (University of Florida CTSA)

Davera Gabriel (University of California, Davis CTSC)

Solomon T. Garner (RCMI Translational Research Network, Jackson State University)

Yongqun He (Ann Arbor, Michigan CTSA)

William Hogan (Arkansas CTSA)

Warren Kibbe (Northwestern CTSA)

Layne Johnson (University of Minnesota)

Fabian Neuhaus (National Institute of Standards and Technology)

Clara M. Pelfrey (Evaluation Director, Case Western Reserve CTSC)

Katherine G. Reilly (Medical University of South Carolina)

Nicholas Rejack (University of Florida)

Barry Smith (Buffalo)

Dagobert Soergel (Buffalo)

Carlo Torniai (Oregon / CTSAconnect)

Patricia Whetzel (NCBO / Stanford)

Rob Wynden (HOM / UCSF)

Organizers: William Hogan and Barry Smith

Sponsor: Translational Research Institute of the University of Arkansas for Medical Sciences

Further information: Please write to Barry Smith

Background

• The IKFC Clinical and Translational Science Ontology Affinity Group has been established to leverage the use of common ontologies to support different aspects of information-driven clinical and translational research,? The focus of this meeting is to explore new and existing uses of the common ontologies to support which are being developed in any case to support researcher networking and sharing of data, particularly in the evaluation of scientific research, and even more particularly in the evaluation of CTSA activities, resource sharing, .researcher networking, and biomedical informatics.

• The proposal is that, when data are harvested from institutional sources, these data can be aggregated using common ontologies such as are maintained by the eagle-i and VIVO initiatives. As these data accumulate they can be compared at regular intervals for purposes of tracking and evaluation of research activities and generation of reports on research activities that can be filtered by type, location, or a temporally defined range.

• Interestingly, the use of common ontologies will mean that the data that is harvested could also be exploited also for other purposes. First, they will make all research-relevant activities easily searchable in something like the way that publications databases such as PubMed and Google Scholar already make publications easily searchable. Second, they will make results of different sorts of research activities combinable, since the same ontologies will be used to annotate, for example, clinical studies, as are used to describe mentorship opportunities or patient outreach initiatives. Third, because common ontologies are being used, all of the information collected will discoverable not only be those working within the collecting institution, but also by individuals, institutions and software agents, on the outside.

Some subgoals of this meeting

• Subgoals of the meeting include some or all of:

0. To devise a plan for creating an evolving inventory of projects across the CTSA Consortium that develop and use ontologies. The inventory needs to include information about the ontologies, their uses, the groups involved.

1. To identify existing efforts in tracking and evaluation of research activities that can be facilitated by the use of ontologies, especially within the framework of the CTSA consortium, and to share the lessons learned from such experiments. One central example under this heading is the Integrated Semantic Framework (ISF), which is being developed in the context of the CTSAconnect project (http://www.ctsaconnect.org), which has as one central component an ontology for Agents, Resources, and Grants (ARG) and which draws on the eagle-i and VIVO ontologies and on the tagging taxonomy from the CTSA ShareCenter.

2. To explore strategies to create an evaluation and tracking system for an organization like a CTSA based on, or working with, a university-wide system for faculty reporting and to identify potential collaborators working on or interested in the implementation of different aspects of such a strategy. The sorts of data which might be imported into an ontology-based tracking and evaluation system would include:

university-wide reporting data

national data for example pertaining to grants, patents

social networking data for example pertaining to existing collaborations, including mentoring

clinical study data for example available through IRBnet

Electronic Health Record data pertaining to clinicians and clinician expertise

3. This meeting will serve as the second meeting of the new CTSOAG, and will provide opportunities for participants to provide updates on recent activities promoting the use of ontologies for the sharing of data across the CTSA consortium.

The proposal is that, when data are harvested from institutional sources, these data can be aggregated using common ontologies such as are maintained by the eagle-i and VIVO initiatives. As these data accumulate they can be compared at regular intervals for purposes of tracking and evaluation of research activities and generation of reports on research activities that can be filtered by type, location, or a temporally defined range.

Interestingly, the use of common ontologies will mean that the data that is harvested could also be exploited also for other purposes. First, they will make all research-relevant activities easily searchable in something like the way that publications databases such as PubMed and Google Scholar already make publications easily searchable. Second, they will make results of different sorts of research activities combinable, since the same ontologies will be used to annotate, for example, clinical studies, as are used to describe mentorship opportunities or patient outreach initiatives. Third, because common ontologies are being used, all of the information collected will discoverable not only be those working within the collecting institution, but also by individuals, institutions and software agents, on the outside.

Subgoals

The specific subgoals of the meeting are as follows (comments and suggestions welcome):

0. To devise a plan for creating an evolving inventory of projects across the CTSA Consortium that develop and use ontologies. The inventory needs to include information about the ontologies, their uses, the groups involved

Deliverables: Outline of a plan and ways to realize the plan, including a list of participants

1. To identify existing efforts in tracking and evaluation of research activities that can be facilitated by the use of ontologies, especially within the framework of the CTSA consortium, and to share the lessons learned from such experiments. One central example under this heading is the Integrated Semantic Framework (ISF), which is being developed in the context of the CTSAconnect project (http://www.ctsaconnect.org), which has as one central component an ontology for Agents, Resources, and Grants (ARG) and which draws on the eagle-i and VIVO ontologies and on the tagging taxonomy from the CTSA ShareCenter.

2. To explore strategies to create an evaluation and tracking system for an organization like a CTSA based on, or working with, a university-wide system for faculty reporting and to identify potential collaborators working on or interested in the implementation of different aspects of such a strategy. The sorts of data which might be imported into an ontology-based tracking and evaluation system would include:

university-wide reporting data

national data for example pertaining to grants, patents

social networking data for example pertaining to existing collaborations, including mentoring

clinical study data for example available through IRBnet

Electronic Health Record data pertaining to clinicians and clinician expertise

Deliverables: A list of such efforts compiled from the knowledge of participants. A draft list of requirements for such a system. This includes objectives 2 and 4 (which may be the same), a draft list of such data resulting from a brainstorming at the meeting.