Development To Do List

From DictyWiki

Jump to: navigation, search

Contents

Prioritized list

  1. Multigenome
  2. Add gbrowse for dictyostelium
  3. Updating genbank
    1. One week for sending first batch of file. Then weekly work.
  4. Set GAF pipeline with protein2goa (2 Weeks+)
  5. Re-architecture/data migration at dictyBase.
  6. Install intermine

Not prioritized

  • Display Harry's data in gene page
    • Setup rails app by divya
    • Got the COMPLETE dataset from Adrian Tsang (July 2012 Petra)
  • GenBank loader (broken): Curators will create list of what should be added 2-4 x a year (depending on amount) and developer adds those records an wipes all others. One record also needs to be deleted, see [1]
  • Speed up caching in gene page
  • Loading blast alignments for other genomes
  • Upgrade or move Issue tracker so entries can be somewhat automated
  • Semi automatic annotations (gene models, gene names, products)
    • via Fast track curation tool
  • Integrate and display SNP data
  • Integrate expression data
  • Integrate protemics data
  • Stock center projects
    • Add to K. aerogenes strains (DBS0349837, DBS0349820) to bacterial strain list; Hardcode [2]
    • Interface for adding new strain/plasmid (HIGHLY important!!)
    • Make plasmid keywords visible on the user side / List of unique strain/plasmid keywords
    • Provide a pop-up of the keywords as a help to search plasmids
    • Show in plasmid list when plasmids have Maps and/or Sequence
    • Allow plasmids to be linked to genes by curator interface
    • Add the orders done through e-mail
    • Making strains/plasmids publicly visible/invisible through curator interface ('hide' button)
    • Stats for time of receipts -> shipped/growing from issuetracker
    • Show plasmid synonyms
    • Make plasmid description searchable
    • New field for 'parental vector' for plasmids
    • Plasmid searches do not accept wildcard character (Dave Knecht)
  • Import MESH terms for every dictybase paper.
  • Codon conversion tool [Suggested my Mike myer]

Curator wish list for better after world

  • Colleague subscription: send email when colleague submits update; ideally email contains link to view and approve - pf
  • Provide some way to update specific data from test to prod. For example, it would be nice to curate strains and phenotypes for a certain gene on test, and then transfer all directly to prod. RD
  • A history table, or some kind of way to record data from previous updates. For example, it would be good to maintain some record of old curation status notes, gene product names, etc. so that we don't have to document everything in the private curation notes. RD
  • The ability to update data from the command line which currently requires both updates at dictyBase curator central and changes to the wiki. It takes about 3 to 3 1/2 minutes to simply add a gene name. Ideally, it would nice to simplify or automate some of the below steps. For example, to simply add a gene name, one must:
    • add the new gene name at dictyBase curator central
    • update the curation status note at dictyBase curator central
    • Add a private note documenting evidence and describing why a change has been made, and the previous annotation that has been replaced at dictyBase curator central
    • Edit the dictywiki from the gene page to also add the new gene name
    • Move the wiki page to the new gene name RD
  • GO: separate manual and electronic (IEA) annotations on gene page - pf
  • Reconcile StockCenter phenotypes with annotated phenotypes; need to add some synonyms to ontology terms, like 'slugger' to 'increased slug migration'; Then map Stock center phenotypes to ontoly if we can generate a reference for the annotations. Display alongside strains of these phenotypes is questionable, but better than no info of course. See e.g. [3]. This has been mapped by Kerry/Petra. Should be in Yogesh's GitHub
  • Would be good to have a search to identify all the phenotypes associated with a pubmed id. For example, if I attempt to search for "23301051" at our search box on the home page I get no results. If I search within stock center search I can get the strain from this paper, but not the additional phenotypes which have this as a reference. RD
  • Ideally, it would be useful to be able to add a pubmed id to the search site on the main page and find all curation within dictyBase (strains, phenotypes, GO terms) associated with that pubmed id. RD - Facetted Search will allow that and much more, after overhaul (Petra 10/3/14).
  • It would be useful to see all dictyBase genes which match an HMM. For example, if I select the protein domain page for DDB_G0284015, I can find that it matches PFAM:PF00644. Maybe too difficult, but even better would be to store all dicty genes which match an HMM and to be able to easily select and view. RD
  • Would like to add a new literature topic: RD
    • Under Life Cycle
    • Topic:"Social evolution"
    • Definition: "Role of the gene product in complex social behavior,including cooperation, altruism, kin recognition and selection, cheating and resistance to cheating, etc.” This is DONE in ontology, plus other new terms as discussed, Sidd will 'hard fix' soon (Petra 10/3/14).
  • Would like to have a way to keep track of all papers we have looked at and decided are not curatable. Like a literature topic: "Not curatable" or something like this.
  • Clean up strain characteristic annotations. When a specific term is annotated, such as 'blasticidin resistant' delete the unspecific parent term 'drug resistant' if present. Another example is 'uracil auxotroph' and it's parent 'auxotroph'. These are old historical strains, but a large group of those available in the DSC. Create report and fix automatically. (Petra)
  • We should add somewhere in a dicty FAQ, instructions to users as to when to send an email to dictyBase@northwestern.edu versus when to submit an email to the listserv

After Migration (Things to remember)

  • Data: Delete all PubMed associations to deleted genes, e.g. actin has hundres of 'uncurated' papers attached.
  • Reunite all dictyBase curation tools (no difference between literature and gene model and other curation tools for login)
  • Create extra curator click to check when a paper is curated. Maybe also when a paper is inspected but 'uncuratable' would be nice)

User Requests

  • Meg Titus
    • Flag frequently cited papers: Run citation index on all Dicty papers. Look at output and mark/flag/ make easily searchable.
    • Provide codon usage table for all coding sequences DONE - simple Excel file Petra 11/26/12

Reference

Current development

Completed development

Software improvement

Previous to do list

Back to Internal_documentation_and_standard_operating_procedures

Personal tools