WormBase-Caltech Weekly Calls April 2015

April 2015

April 2, 2015

Adding Species information to nightly geneace dumps

• • Michael Paulini is adding Species information for the WBGene IDs in the genes.ace file
  • Does anyone want Species information for any of the other files
  • For list of files, see: ftp://ftp.sanger.ac.uk/pub/consortia/wormbase/STAFF/mh6/nightly_geneace/
  • Data in postgres in table gin_species

WormBase Ontology Browser WS248 online on Dev

• Caveat: the rest of the WormBase data on the Dev site is WS247 (one release behind)
  • Would explain some discrepancies

Micropublication form

• Daniela waiting for feedback from Paul's lab
• Will send to Alison Frand and Elissa Hallem at UCLA to get input

Mary Ann's IWM abstract for community annotation

• People can send comment over next day or two
• Will make revisions and send around next week

April 9, 2015

Tazendra

• Had some glitches with secondary hard drive this week, not sure what the problem is
• /home2 directory is not intended as a primary storage; curators need to back data up
• Paper PDFs stored on /home2 as the primary storage location
• Papers are backed up on Athena (Wen's computer)
• TIF files are stored on CDs
• Main Tazendra hard drive backed up (RAIDed/mirrored), backed up nightly up to ~1 month (or 0.5 month), monthly for a year, but no older versions

Citace data upload

• April 28th, 10am

Community Annotation Forms

• Mary Ann's IWM (poster) abstract includes community annotation forms
• Community annotation form list:
  • Micropublication
  • Gene Expression form (made 15 years ago); mostly not used; micropublication form could be co-opted
  • Gene description form (primarily for the community to edit the automated concise descriptions)
  • "Submit Data" list on WormBase (http://www.wormbase.org/about/userguide/submit_data#0--10)
• Would be good to have a community annotation portal/landing page
• Mary Ann will send around a summary of discussions about community annotation, along with prioritization of forms for curators and Juancarlos
• Relevant curators should consider reviewing existing forms for updates

modSeek

• http://seek.princeton.edu/modSeek/
• Princeton 6th year grad student developed it (leaving within 6 months)
• "Seek" developed for human on microarray and RNAseq data
• 5 model organisms in "modSeek": yeast, worm, fly, mouse, zebrafish
• Pulls in data from Gene Expression Omnibus (GEO)
• Analysis computed from raw data
• Data from SPELL should be reasonably transferred to modSeek
• We need to determine if WormBase data can be updated per release
• Has text mining
• Cross-species comparisons available

April 16, 2015

Topic Images

• Daniela importing Lipid Metabolism topic images
• Daniela may also try to create gene-image connections
• Picture OA could capture relevant genes
• For now, Daniela will only import images for which we have permissions

WormBook Chapters

• Would be good to make sure relevant FTP files are mentioned and pointed to
• We can add a section on concise description and automated description
  • Not sure yet where to put it; maybe in introduction chapter

April 23, 2015

WormBook Chapters

• Importing figures or screenshots can result in lost resolution
• We may need to play with image manipulation/editing to fix
• Gene function chapter has intermine section; maybe should be an appendix or separate chapter
  • Will make separate chapter; will reference Intermine with a single screenshot in gene function chapter
• We will make sure the gene function chapter properly links to GO chapter

Wikipathway

• Karen created Lipid metabolism pathway
• Dennis Kim and Jonathan Ewbank are updating their WormBook chapter on innate immunity; working with Karen on WikiPathway

LEGO

• Kimberly participating in conference calls
• Progress being made on back-end to handle evidence properly
• Curators have been going through specific use cases
• Talking about generally how to best model different biological scenarios
• Estimate for getting back end finished on the order of weeks
• Front-end still being developed; getting close
• Karen asks: How will LEGO integrate with Process&Pathway curation at WormBase?
  • Kimberly: LEGO is very molecular-function-centric; may be more granular than WB Topics/Process&Pathway
  • We should see soon how these two will interface

Controlled vocabulary for institutions

• How to handle synonyms of institution names
• We will accept entries of synonyms, but will only actually store/track the (English) official name

April 30, 2015

modSeek

• http://seek.princeton.edu/modSeek/
• Wen has established an account
• Wen is running scripts to generate worm data for modSeek
• Data in modSeek is G.E.O. (Gene Expression Omnibus) based
• Links at modSeek go to G.E.O. data, not WormBase
• Wen is considering a WB-specific search with links back to WB data
• modSeek only has (and supports) C. elegans data, not other nematodes for which we have data
• Maybe we could have a "WormBaseSeek"/"WB-Seek"?
• We can run a mirror at WB
• Wen is modifying SPELL scripts for modSeek
• Would we be able to perform cross-species comparisons with a WB-Seek?
• Textpresso-Dev is currently running our modSeek instance
  • Textpresso-Dev probably would not be able to handle mouse, human, fly data for species comparisons etc.
• We would like to be able to perform cross-species comparisons for all nematode species as well as the other MODs/human
• WB data is paper-centric, unlike data in modSeek
• WB data is processed data (author processed), modSeek data is raw data
  • We want to be able to keep/maintain the processed data in addition to the raw data

BioCurator Meeting 2015 (China)

• ~300 people in attendance, about 1/3 from outside China
• Mouse (MGI), yeast (SGD), EBI, NCBI, and WB represented at meeting
• Xiaodong's training session
  • Q&A about literature curation
  • Many small bio-databases in China, focused mainly on genome annotation analysis; no literature curation
  • Xiaodong demonstrated the Ontology Annotator (OA) tool
• Currently there is no bio-data center in China; Biocuration society would like to establish one
• Web Apollo, JBrowse, phylogenetic G.O. annotation info presented
• Yeast database (SGD?) use IntAct (in collaboration with BioGrid) to curate protein complexes
• There was considerable discussion about the use of Uberon as a cross-species anatomy ontology
  • Would be good for WB to establish connections to Uberon with C. elegans anatomy ontology
  • This will require specification of relationships; non-trivial
• Yuling presented work on SVM analysis
  • People had questions about whether SVM could be used for hierarchical flagging (data types and then subtypes)
    • This requires training sets; could consider moving forward
    • (Karen) We already run SVM for allele data followed by Textpresso entity recognition
    • We would likely run the entire corpus through the SVM for a whole new datatype with proper training sets
  • Can SVM be applied at the paragraph or sentence level?
    • Possible, and we have some curated sentences saved, for example with cell component curation (CCC)
• UniProt developing UniRule system for large scale protein annotation
  • Manually establish (text mining?) rules for future recognition and annotation
• Reactome is beginning to use ORCID person IDs to give attribution for Reactome pathways

FTE estimates for WB staff

• Paul S. sent around an e-mail with a spreadsheet for filling out FTE estimates
• Paul asks that people fill it out today and send it back to him
• Two e-mails; first with complete form, second with simpler form
• Please fill out complete form, or simpler form if necessary