WormBase-Caltech Weekly Calls January 2015
- 1 Janurary 2015
- 1.1 January 8, 2015
- 1.2 January 15, 2015
- 1.3 Jan 22, 2014
- 1.4 January 29, 2015
January 8, 2015
- author participation
- Cecilia: still lots of work, need to manually process data submitted by authors
- old data kept for history and for comparison
- people mess up institution names - institution field on submission form being changed to a controlled vocabulary
- campus addresses also not standardized - will organize/create address fields, e.g., street address, mail-code, etc.,
- add example address
- ~10 author submissions a week
- most curation involves extracting people information from papers (~800 week), as automated as possible right now - people and affiliation are extracted from xml, but information still needs to be parsed
- can we mine author data from other sources (institution, laboratory); important to get laboratory address for sending strains; two sets of data- (1) easy- people associated with a laboratory, (2) difficult- people not in a registered lab
- embarking on cross-product generation- ex., male tail, versus hermaphrodite tail - better to have one 'tail' and add modifiers
- wants to move to OWL now; comment is that obo relationships are not directly translatable into OWL
- Uberon has elegans anatomy
- starting to work with Chris and Karen on developing cross-products, still in exploration phase
- not looking at OWL yet
- will get relationships from Raymond
Sequence feature curation
Daniela, Gary, Xiaodong, Mary Ann are validating papers as positive/negative in curation status form. As soon as ready, Mary Ann/Gary will start curating
- 150,000 new alleles, Mary Ann done just all bar 12 of them in time for WS247
- phenotypes were only reported for a small fraction
Mary Ann received latest update, CGC very busy with strains generated through new gene engineering techniques
WB WBook chapters
Main outline here- note this document cannot be edited, people can create their own shared docs for their respective papers
- Gene Function and Interaction google docs (Chris, Gary S., Karen, Kimberly)
- Pathways and Processes (Karen) here
- Expression (Daniela)
- Hobert- micropublication for expression = small facts, which are data that will never be published (Daniela working with Hobert on these)
- Community annotation not linked to a publication
- a couple models for WB dealing with the micropubs
- WB captures all submissions blind and posts them
- WB reviews submission - curators decide which micropubs have value
- need to establish microattributions to increase the value of submissions
- involve semi-peer review through automated fact review requests to scientists who've worked on the gene, cell, entity, etc.
- need to set up a pilot, to get a feel for the amount of participation/work this tool will be
- for topic curation
- want to annotate figures with genes - involves model change
- want dynamic display with slide show, different highlighted figures - need to work with webteam
- community voting - crowdsourcing - need to work with webteam
- WB-blessed image represented through wikipathways
Continuing work on the new GO model
- management of automated descriptions with Ranjana
- phenotype requests through RNAi and allele phenotype done in december
- playing with pulling out useful data to add to description automation (James and Ranjana)
January 15, 2015
New expression objects and pictures
- Cross-species comparison of developmental gene expression
- Have life stage annotation
- Microarray experiments
- Do we have data on significant differences from C. elegans? Can we get the analysis or algorithm?
- 3 ways to access neuro-connectivity data
- ACEDB, Kris Gunsalus' server at NYU (defunct), Nikhil Bhatla's connectivity browser
- We would like to adopt Nikhil's browser (for neural connectivity AND the cell lineage browser); we need to act before it disappears
FPKM RNA-Seq data display
- Should we suggest splitting the data display by tissue, life stage, treatment, etc.?
- Gary Williams is responsible; need to talk to him
Heavy metal toxicity/response
- New topic, all papers have been checked
- Need to send list of relevant genes to Hinxton
- Curators can now flag figures relevant as topic diagrams
Phenotype picture curation
- We will begin to collect pictures for phenotypes
- Will require model changes
Phenotype OA Strain fields
- "Parental strain" autocomplete field added to TAB 2 of Phenotype OA
- Plan is to put parental strains in the Phenotype_info hash
- Natural isolate strains can be directly annotated with phenotype?
BioCuration meeting in April
- Xiaodong is writing up an abstract, will send around before submission
- Xiaodong will run a curation tutorial for Chinese speakers
Jan 22, 2014
ad hoc curation of tier III species (Mary Ann)
WBPaper00046176 describes Ctr- genes and alleles (fig 3). Although we do not curate this (and other tier III) species, should we place this information somewhere, on an ad hoc basis? http://tazendra.caltech.edu/~acedb/daniel/00046176_Wei14_temp.pdf
Mins: Curate as much as existing models allow and hopefully expand at later date. The volume of this sort of data will increase as we encourage Community Curation. Don't seek out papers, but add data as they come along. Talk to Hinxton about adding gene models.
Complex Gene Loci (Mary Ann)
We should use existing Gene_cluster model and add new Type tag to differentiate between e.g. rrn Gene cluster and e.g. lin-15A lin-15B complex loci.
nb. Maybe hard to discuss w/o Paul Davies.
Mins: The above was agreed. Need to agree on Types and make model proposal. Detailed curation (e.g. of alleles) will not be done within the Gene_cluster but within the allele as appropriate.
- Daniela starting to work on author submission forms for expression patterns where people can submit images
- Requirements? Paul says needs sequence - minimal important stuff, sequence, expression pattern, construct, reporter gene
- define minimal standards (minimal amount to reproduce experiment)
- genome coordinates are ok for elegans, but not for other species whose genomes
- making a micropublication - identifier, research object identifier(?)
- put them in paper class? no
- look research gate, academia U, mendeley etc - social media/science community person catalog.
- ultimate goal to convince NIH to capture and catalog these bits of information
- review function - how to build this in? how many reviewers do you need, how and who to get to participate
- continue work on submission form
January 29, 2015
WormBase WormBook chapter
- Screenshots of WormBase.org website?
- Yes, the WormBook chapter should be updatable
- We can focus the discussion on general principles and underlying logic, more so than specific user tutorials
- First draft deadline: February 28, 2015
- Not going to be maintained?
- We don't know if/when it is going to die
- ModSeek is a potential replacement: http://seek.princeton.edu/modSeek/
- Data is stable, interface is not
Workshops at international meeting
- Should we have a ~30 minute open discussion at the end of the 90-minute workshop?
- Maybe, but we should take a survey before to assess what people are interested
- Survey could be part of the registration process for the workshop
- What kind of survey do we want? What have we done in the past? Does Todd know?
- Should be simple, maybe ranking importance
- What tools do people use that they would like to see incorporated into WormBase?
- GO-enrichment analysis tools, etc.
- Plan is still to have 4-6 talks
- When/how do we want to handle/discuss user concerns and criticisms?
- Do we have a plenary session? We should, maybe Kevin was going to apply for one?
- A plenary talk would likely be short (~3 min?) and would be mostly to advertise our presence at the meeting/workshop
- We could discuss which resources we provide/link to that are third party and possibly out of our control
- Review FTP site and what data is there
- Community Annotation
- Automated concise description corrections/updates
- Data submission pipelines, author first pass forms, RNAi data submission table
- Need to be specific, not vaguely discussing community contribution
- Discuss making data submission part of the publication process
- Mention existing data submission pipelines used by publishers
- Want to make things easy for users
- Emphasize expedited data incorporation
- Competition to curate alleles?
- WBPerson T-shirts for winning lab
- We can discuss micropublications
- Soliciting developer efforts?
- Incorporation of community developed tools
WormBase IWM souvenirs
- We could have prizes for community annotators
- We could make bags again
Optogenetics as new data type
- Raymond adding as new data type
- Will add to curation status form, author first pass form, etc.
- Potential subcategories exist
- Curators can let Raymond know about important subtypes to recognize/include
Linux machines updated
- Vulnerability has been plugged for all non-Textpresso Linux machines (Raymond)
- Machines have been set up to automatically install security updates
- Eventually would be good to handle with a server administration tool