Minutes from 04/12/07 Conference Call

Attendees:

Cristian Cocos, Jim Zheng, Richard Scheuermann, Jamie Lee, Simona Carini, Jennifer Fostel, Barry Smith, Dave Parrish,Wenle Zhao

Introductions:

Richard – UTSW, Division of Clinical Sciences, Department of Pathology, OBI, ImmPort

Jamie – UTSW, Department of Pathology, GO, OBI, ImmPort

Cristian – EE, CS, philosophy background; ontology building for European consortium ACGTC (Advanced Clinical Trials Cancer or something like that); extensive experience with OWL/Protégé

Jennifer – computational biology background; building NIHIS database for clinical info; OBI; interest in analogies in different fields and structure of clinical trials and toxicology experiments

Simona – UCSF, Ida Sim’s group; background in clinical trials, nursing, and data management; development of RCT schema

Jim Zheng – Medical University South Carolina; biology/CS background; biostatistics/bioinformatics department; clinical trials data management; interest in developing computational models and ontologies using engineering approaches to to formalize process

Wenle – new to group, Medical University South Carolina; project management for clinical trials

Discussion Items:

WG organization

• Conveyed on wiki
• Agreed on Jennifer and Richard as WG chairs; this can be changed in future based on performance

Funding issues

• Anyone who wants to take initiative for grant submission is encouraged
• Barry has pending grant application to develop BOLT (Biomedical Ontologies and Logic something) center which is under review. Proposal is to develop ontologies and there may be funding available for CTO

WG Roles

• We should establish responsibilities for different roles (chairs, secretary, developers, etc)
• Jamie is volunteered for secretary role to keep wiki updated and organized
• Wiki – individuals can also add their own stuff freely without going through secretary

Managing discussion threads

• Discussions through Email makes more sense since people check often but key points or decisions made during discussions should be posted on the wiki
• Agreed that both Chairs will summarize conclusions of discussions and secretary will compile and post on wiki
• Wiki posting a good idea so people can see the progress being made
• Find a time for weekly calls to review discussions – tentatively 6pm Germany, noon Eastern, 11am Central, 9am Pacific on Thursday. Cristian will check his calendar and free call-in options. Tuesday is the other possible day

Development and WG policies

• Metadata policy
  • Briefly, when terms are being developed and structured, there should be minimum information included like term name, definition, authorship, updates/changes to terms, etc. All agreed to take a look at draft policies (“stolen” from OBI’s metadata policy) and vote on whether we should adopt these.
  • Where would we enter these metadata fields in protégé? OBI is still working this out with Protégé Stanford but in the short term, use Comments field.
• Check-out policy
  • Basic idea: whenever you want to modify ontology, make sure no one else is working on it, do your work, then let everyone know that you are done. Also, record a log of the changes. Details on wiki. People should be considerate about check-out/in.
  • We agree to adopt this policy and be considerate about check-out lengths of time.
• Membership policy – do we need this? Probably not for now but thing about it

Relationship between CTO/BFO/OBI

• We should leverage the work OBI has done and make sure our efforts are parallel to OBI.
• We should develop CTO independently using OBI branch structure as a “template” since there is high-level commonality between CTO and OBI domains. If we see a need for changes in OBI we should collaborate with them to reach an agreement. We should import the “stable” version of OBI especially at the high level.
• We can’t avoid flux within OBI but OBI coordinators are receptive to feedback from the clinical trails perspectives; WG chairs will take role to interface with OBI and make sure they understand that changes in high-level structure will impact development of parallel ontologies

Scope

• What is the CTO domain? Issue of clinical trials versus clinical research.
• The assumption is that if we adequately cover clinical trials, we would also cover 90% of the clinical research domain
• It is a good idea to define the scope using a top-down and bottom-up approach?
• We also need to define priorities within the domain
• Keep “CTO” name for time being until scope is more fully defined

Steps for early development progress

• Assemble lists of terms to combine different perspectives within CTO domain
• Merge lists of terms to remove redundancy
• Position terms into main branches
• Begin work on definitions for terms
• Please give feedback for these proposed steps and proposed goals before NIH meeting for next week’s conference call

Other

• Whenever anything is posted on wiki, please send email to the group.