2013 Meetings

March 7, 2013

Grant

Gene Concise Description form

Ranjana, Kimberly working on form layout
Easy to use form, simple drop down menus to select, e.g., Species
Can users follow references for info?
Citing references in concise descriptions has, in the past, greatly increased workload/time to write
Prepopulated, precanned Textpresso info?
Send out to targeted members of the community, get feedback; Who should we ask for other species?
Curators could each try using form to make at least one concise descriptions, give feedback
Prepopulation: Display all (key?) experimental info on gene from WormBase, some Textpresso info like "ABC-1 encodes..."
Can promote form at the International meeting

Galaxy Workshop

Genetic Interaction Ontology

Perhaps add "what is expected" info during curation?
Good to have a comparison of curation with core set of common papers
- Cross check for consistency of curation (at least terms should lay in same lineage, if not at same depth)

March 14, 2013

Cleaning up Citace Minus

Stemming from Paul Davis' GitHub ticket pointing out small-scale data problems
Bogus molecule objects (Chris will take care of)
Homol_data from RNAi (and the housing RNAi objects) can be removed
Ghost Expr_pattern objects, Wen taking care of (Xiaodong will remove from the Gene_regulation OA)
Will need to create a better (more robust) pipeline (with SOPs) for converting, for example, object names to WBID#'s
- Once a conversion has been made, we need to check Postgres and Citace (& Citace Minus) for consistency
- Need to minimize the burden on Wen to perform quality and consistency checks
- Curators need to clean up/fix their data before a big change
- Scenarios to foramlize:
  - Whole-class name changing
  - Model changes
Is there a list of classes (and objects?) that are held in Citace Minus?
- Cross-references create complexity
We will suggest shifting the phase of the model proposal cycle: Propose to freeze all models one month before upload

Queries

AQL vs WQL vs AcePerl
We may consider working out more formal pipelines for submitting queries (maybe to Wen?)
AQL and WQL convenient but not necessarily efficient
AcePerl efficient, but has a learning curve
Todd highly recommends learning AcePerl as it has many advantages:
- Learning AcePerl will provide skills to handle other generic queries
- Whereas AQL and WQL will be obsolete when we leave ACeDB, AcePerl will still be useful (even if the details need updating)
- Curators have GitHub access to all the AcePerl scripts, including those that generate FTP files

Under Testing

The "Under Testing" tag on GitHub issues is applied when something needs to be tested
Curators can filter for "Under Testing" issues and close them as they determine they are resolved

Protein browsing

KOGs vs eggNOG; eggNOG may be the most current
Would be nice to be able to (for each species or all species in WormBase) browse protein sets as orthologous groups by functional classification
Need to investigate

Microarray data

Data that were not provided (no changes observed in an experiment) leaves gaps in data
SPELL requires data for each experiment/probe (regardless of observation of change)
FTP site for downloads of data (independent from, but related to, SPELL); Wen needs regular access to
- Can be hosted at Caltech FTP site (Todd can mirror over to main FTP site)
Add explanatory page to SPELL site

Dead genes

Historical_gene tag will be added to several class models: ?Interaction, ?Transgene, ?Antibody, etc.
How to handle GO annotations? GO will have mapping from proteins to GO terms, no need for new tag
Chris will send around model proposal to add the tag to relevant models
Curators will need to coordinate with Juancarlos how their respective OA dumpers should work

Beyond PDF conference

IWM Meeting

Workshops and talks topics
2 Workshops (Kevin made Google Doc here (requires login):
Two workshop time slots: Thursday 1 hour (1:30 - 2:30pm), Saturday 1.5 hours (1:30 - 3pm)
- Overview Workshop 1 (Thursday slot)
  - Sequence perspective (Hinxton/EBI member)
- Data mining workshop (Thursday & Saturday slots)
  - Modmine with extra half-hour on Saturday
- Overview Workshop 2 (Saturday slot)
  - Biology perspective (Caltech: Chris Grove)
    - Interactions, regulation
    - WikiPathways
    - Concise description forms
    - Expression Data in WormBase
Spreading the word:
- Catch people during poster sessions (booth with ~4 posters)
  - Posters on: Wikipathways/Interactions, Species, Gene Expression & Regulation, Textpresso, Phenotypes & Diseases
  - Have projector displaying screen demos
  - We could create a template with a standard layout
  - Create posters with Illustrator or Scribus
- WormBase events flyer to hand out or have placed in meeting program
WormBase give away ideas:
- Mugs, aprons, gummy worms, LED light keychains, etc., etc.

Populating OA with variations for variation fields

Huge number of variations (>1,000,000!)
How to limit to relevant variations?
Can we split into "useful" vs. "generated" mutations/variations?
Splitting now on "Allele" tag
Add a new tag to the variation model?
Ask Hinxton to provide a file of all variations with a phenotype for FTP site
Also get full list of variations from Hinxton for our internal processing
Get abstracts (for posters, workshops, talks) ready by next conference call (next Thursday)

"Beyond the PDF" meeting

Who's driving the whole peer-review process?
World-wide open access data meeting
Proposal to have Research Data Alliance (RDA)
How to make all data freely available?
Data repositories, publishers, researchers, universities involved
Publishers will need to ensure that published data is made publicly available
Data sharing plans
Rebecca Lawrence Faculty of 1000 (RDA launch)
Carol Tenopir - reading analysis
- Successful researchers are very data-aware, read lots
Data format standards - defined by data repositories?
Action items: breakout groups (revolution group to completely reorganize data infrastructures)
Carole Goble: no need to "publish", just release data to public