WormBase-Caltech Weekly Calls

From WormBaseWiki
Jump to navigationJump to search

Previous Years

2009 Meetings

2011 Meetings

2012 Meetings

2013 Meetings










October 3, 2013

Development environment for Raymond

  • Todd is working on it; will do today

NAR Manuscript

  • Todd will send around for comments today

WormBase Ontology Browser (WOBr)

  • We have a stable server for the information the browser queries: gene ontology associations, gene association files
  • Raymond was trying to keep up with the AMIGO 2 development upgrades
  • Raymond will soon be getting other ontology info into the browser
  • Takes about an hour to load data into the browser

Movie objects stored as URLs

  • There are 209 RNAi objects referencing 513 movies via URLs
  • The URLs are for RNAi.org pages that have the movies; we don't have the movie files themselves
  • Daniela will add a Database tag to the ?Movie class
  • URL suffixes/endings will act as the RNAi.org Accession number that will go into the respective Movie objects

Movie curation

  • Movie files are stored on Canopus
  • Anyone importing new movies will need to get their movies into the appropriate directories on Canopus
  • Currently movie files are all from three large scale movie datasets, so names are currently unique
  • Daniela will now sort movies according to WBPaper ID or WBPerson ID (depending on the source) and place them in directories accordingly
  • Movie files will then be organized according to these directories as ?Picture objects are already

Changing "Biological Process" to "Biological Topic"

  • Changing the class name in the web code is laborious and not good practice
  • We will keep the ?WBProcess class name, but only refer to them as "Biological Topics" on the website
  • A comment will be added to the models.wrm file explaining the discrepancy

Database OA

  • Karen and Juancarlos will construct a ?Database class OA
  • Curators can enter new databases or edit existing databases there
  • There is a static file for the web code that provides URL constructors for links out to those databases and their objects
  • In the past, the file has been manually updated to accommodate new databases or updates to URLs, for example
  • There needs to be a mechanism in place to edit or replace the file to correspond to the Database OA data

Unfolded Protein Response curation - Pilot topic curation

  • Already an ER UPR Stress WikiPathway and a Mitochondrial UPR WIkiPathway
  • Do we create one parent pathway/topic, or do we keep them separate?
  • As we work on a topic, we will assess how much material there is to cover and whether or not we need to be more specific (or more broad) in topic

October 10, 2013

?Expr_pattern and ?Movie model changes

  • Daniela was proposing changes to the models
  • Daniela can/should coordinate with Wen for microarray/tiling array data/quantitative expression data
  • For now Daniela will request only Tiling Array and Microarray tags. Wen will put in the additional model changes later on.
  • Does the text entry need to be indexed (?Text vs Text) for DB_info?

Cataloging controlled vocabularies used in OA and models

  • Karen generated Wiki page: http://wiki.wormbase.org/index.php/OA_forms,_tables,_scripts,_etc#OA_form_dropdown_lists
  • Tables list Postgres tables and OA forms and the controlled vocabularies (drop down values that dump to Text fields in the .ACE file)
  • Paul Davis wanted a list of all of the controlled vocabularies we use during curation
  • We will want to create a Wiki page that clearly outlines a model-change protocol that people should follow when changing models and refer to this page


  • WormBase approval process needs to be flushed out
  • We can annotate edges in a pathway with text: references/citations, text from article, WormBase object referenes, etc.
  • We want to link all relevant WormBase objects to the pathway where possible
  • We want to establish a graded spectrum of confidence for each pathway component/annotation/edge
  • Confidence or rigor of annotation can span across evidence types: Review reference only (low score), review figure/table, primary research evidence article, primary research evidence figure/table within article, WormBase data object (e.g. interaction, etc.) (high score)

Topic curation: Unfolded Protein Response

  • Mitochondrial UPR pathway an official WormBase topic
  • WikiPathway curation would be best while curating a paper; update on WikiPathways site for other curators to see
  • We can add annotations (papers, figures, tables, WB data objects) to edges in pathways
  • While curating a data type, there are many more aspects of the biology from the paper that inform the curator of the pathway
  • Regular editing and updating of WikiPathways will keep other curators updated on what still needs adding to the pathway
  • We should continue to focus on how this pipeline can be improved for future curation

October 17, 2013

Variation Nameserver & GeneACE

  • Caltech will get JSON dump of the entire database/nameserver from Hinxton?
  • There are issues with the synchronization of the Variation Nameserver and GeneACE. Karen is asking MaryAnn about it, and we'll decide whether our daily script should look at both or just GeneAce
  • Frequency of dump: daily if possible (it should be)


  • JD has built a new version for WS239 including FASTA sequences (excluding CDS sequences) and RNAi objects
  • JD has asked for curators to review the changes before pushing it to the live site
  • The Intermine grant is up for renewal; they requested testimonials from curators; we can send them to JD

Topic Curation

  • Still working on Unfolded Protein Response (UPR) for WS241
  • Chris will send summary of UPR curation status (including gene list) to Hinxton
  • Hinxton will work on touching up/curating gene models for UPR-relevant genes in C. elegans and homologs in other nematode species
  • Hinxton suggested covering ncRNAs for WS242; we probably want to focus on a particular subtopic (one type of ncRNA with biogenesis and mechanism of action)


  • We need to establish a good way to annotate indirect associations (e.g. RNAi leads to gene expression change)
  • SBML and SBGN exist but may not be entirely compatible with the WikiPathway arrows/components

WormBase Ontology Browser

  • Raymond is waiting for development environment update from Todd
  • Raymond is working on incorporating WormBase-specific ontologies into the browser
  • Taxon-specific tag in model, could be useful for species-specific life stages

Release Schedule Details Wiki Page

  • Would be good to establish a time during the cycle in which the web team is receptive to requests for changes
  • Google Docs page here

Database OA

  • Current mechanism of updating Database URLs/URIs/constructors involves manually editing a Database flat file (on GitHub) read by the website https://github.com/WormBase/website/blob/staging/root/templates/config/external_urls
  • Karen considering the development of a Database OA
  • Issue is with denormalization of ACEDB to flat file
  • If within the release cycle, the Database OA would need to be dumped and the info overwrite the flat file, but then it needs to be committed on GitHub and pushed to production
  • Also a problem is coordinating with the Hinxton Database info
  • Although laborious, may still be better to manually edit the flat file

October 24, 2013

Disease Curation

  • Collaboration with FlyBase
  • Two branches of curation: (1) disease ontology and descriptions, (2) gene/protein orthologies
  • OMIM generates human written/readable descriptions, may want data exchange with OMIM
  • Gene-association file (GAF) for Disease Ontology (for eventual browsing in WOBr)? No file generated yet, but GO GAF scripts can be modified to create

Cell Component Curation (CCC)

  • 1100 papers prioritized for GO
  • James building Textpresso for MGI based on that corpus
  • Maybe run SVMs on the corpus and perform CCC
  • Mapping gene IDs to UniProt IDs
  • Should there be a mechanism to provide IDs for uncloned genes? Which can then be used for, for example, biological process curation

WormBase Ontology Browser

  • All WormBase ontology files have been loaded into the backend of the browser
  • Expr_pattern and Phenotype associations files included
  • "is a" and "part of" relationships used to propagate annotations to higher/lower terms

Genetic Interaction Ontology

  • Chris and Rose working on a simplified manuscript, just explaining the ontology

Person curation: Database tag

  • Establishing a database tag to capture ORCID IDs, for example
  • Social media? LinkedIn, Research Gate, Academia.edu... maybe not useful

Nightly GeneACE/Nameserver dumps

  • Dumping scripts have been modified to handle dead genes
  • OAs affected: Interaction, Gene Regulation, Transgene, Antibody, Expr_pattern, Process
  • Curators ought to test their respective dumpers on live OA (Tazendra) to make sure that dump script changes are working as intended. "I think we should test this on the sandbox, tazendra doesn't have genes with suppressed tags" -- J
  • Example genes for different types (Split, Suppressed, Merged, Dead) on Wiki pages

CRISPR/Cas9 Curation

  • Curation is tricky for insertions or deletion/insertions
  • Daniela wants to curate expression patterns for the "alleles"/"transgenes" resulting from, for example, GFP insertion events
  • How to handle?

Topic Curation

  • Wiki page setup for Topic Curation
  • UPR Curation for WS241
    • Hinxton working on a table of orthologs for relevant genes in non-core species
  • ncRNA Curation for WS242

Sequence Features as Interactors

  • Proposing a "Feature_interactor" tag for the ?Interaction model
  • Will enable curation of physical protein-DNA interactions and regulatory interactions (e.g. cis-element regulates gene expression in cis)

Conservation tracks on GBrowse

  • Not on WormBase GBrowse any more
  • UCSC has conservation track
  • Do we want 6-species comparison track in GBrowse?


  • Good news: Wen in contact with SGD SPELL curators/developers; many built in functions: there is a tag that can be added to allow search for different methods and species
  • Bad news: Wen in touch with Matt: possible to make stand alone SPELL clustering analysis, but may be very difficult