TR_03MAY10
Call-in/WebEx Information
- TR_WebEx - this meeting will include visuals, so WebEx Video will be utilized
Action items
- AI: Get virtual machine for development DONE
- AI: Sheldon to investigate source modification 1st then look at lookup table solution. DONE and DONE
- AI: Status reports on development of prototype need to be made weekly. SORT OF DONE
- AI: A draft of the data model to support dependent use cases will be prepared by Sheldon & Sharon for the next meeting. SOME PROGRESS
- EnsEMBL/Compara: Albert Vilella says all four selected use cases would be supported with their data model.
- AI: Former user (Deleted) to mockup an expert tree prune/selection w/ auto-complete to find taxa. Common taxa box should be incorporated. APWEB has an example of how to show users where their species/group of interested is placed on the tree. [incomplete]
Agenda
- Prototype Data Model Discussion (Sheldon)
- How will the following be supported?
- Use Case 2 - Show all gene duplications in a gene family tree.
- Use Case 3 - Find all gene trees showing a duplication event at a specific point on a species tree.
- Use Case 12 - Show sampled gene copy number for all taxa gene family identifier
- Use Case 4 - Find points on a species tree where a set of genes (e.g. all genes within a pathway or GO category) originated and/or diversified.
- How will the following be supported?
- Flexible API Discussion (Jamie)
Notes
Attendees: Jim Leebens-Mack, Todd Vision, Adam Kubach, Sheldon McKay, Cecile Ane, Jamie Estill, Nicole Hopkins, Andrew Lenards
Sheldon:
- considers the emails to the working group mailing to be progress for last week
- looked at TreeBeST source code, can't find 5 character limit that is being enforced
- a workflow for unique 5 character IDs has began. Andy has provide an approach for a lookup table
- didn't experience errors when using species names more than 5 characters.
Cecile - did you try for conflicts with species names that have the same 5 characters - no?
Andy - thinks we should investigate if same 5-characters in a species might causes the behavior Sharon was seeing, have 2 ways to generate unique IDs.
Todd - did we get response from authors? -> No, Sheldon will check into test cases where there are species that have the same first 5-characters and follow up with the authors, if needed.
Sheldon - Virtual Machine for development environment is set up, should Albert (from ENSEMBL) to be involved? (AI) will be writing some example queries for next meeting.
Todd - communication with Jamie looks like most data models would work, except for one schema that I hope you wrote down, how necessary to support ontologies outside of GO?
Sheldon - GO and maybe KEGG? [KEGG project site]
Jamie - how flexible to supporting ontologies, select children nodes outside of parent
Todd - how useful would it be to make controlled vocabulary more general unless using GO in query?, what are pros and cons of supporting wider vocabularies or nested queries?
—
Collaborate with Albert -> what will they want to support, improve things rather than build new, workaround for GO terms.
Jamie - talk about general goals, schema, scalable to larger sizes, general API beyond schema
Andy - intended home for prototype is Discovery Environment, an API for the public will be released Q3, not sure if a flexible API is in scope for prototype? Might be out of scope, could get it started, intent for DE, write services for it, etc., we'll release RESTful API in Q3.
Jim - great, to what degree relying on the tree databases and down road, not rely on databases, estimate on fly
—
API to get information from variety of sources will have to be part of product iPlant is developing -> don't want to have to rely on particular source.
Andy - DE will end up creating some kind of integration layer, we'd do grunt work to wrap external sources and provide that interface for others to go through as RESTful services, PhyloWS, etc.
Sheldon - unconstrained access to tree data, specific requirements for trees to TreeBeST need to intercept this to ensure they conform to NHX, need to think longer term, internal nodes need to be labeled.
Andy - Any other topics that would be discussed? Statement of work for Jens' group?
—
Cecile & Todd - only one copies on email from Bengt about something moving up, meet up time -> end of June (21st-24th)? 1st 2 weeks of June most promising times to meet? do it at NESCent?
Doodle Poll for this was presented after the meeting here
Next meeting May 17th, email update Friday/Monday on Sheldon
Jim - agenda pt -> discussion of what "we" want Jamie to do? Formalize interaction between Jamie and ET