JAN 13 2011

From DictyWiki

Jump to: navigation, search

Contents

Work/Project schedule as discussed in last meeting (and commented/updated today)

  1. Curators finish first pass end of January.
  • I don't think we can finish completely end of January. We are racing to the end but are hampered by RNA shifted alignment for Chr2 (supposed to get fixed immediately after this release), the pseudogene bug (fixed in this release), and not seeing when a gene has been skipped by another curator (also updated in this release). If the release and subsequent fix is not before the last week of January, we can impossibly finish. We need to decide if we'll continue to focus and push until done, or if we start doing other things and dedicate the equivalent of one or two days to curating the remaining genes. (PF)
    • Curated models: 10422
    • Pseudogenes: 357
    • Skipped: 136
    • Deleted: 167 (at least)
    • Annotated RTE/TE: 516
    • Total (taken care of): 11598
    • Inspected/Curated genes (minus deleted): 11431
  1. Fix Chr2 RNAseq (about 2 days)
  2. Work on Gene Page to include expression section and start with Harry's data, plus fix caching.

Release 2.19

Issue still to address
Couple of patches fills memory only in testdb(nothing in developmental machine). We need to test all the latest batch scripts.
Bugs
Leftovers/Discussion items
Fast track curation does not attach 'dictyBase curated' paper to new gene models which went unnoticed for all this time. Should there be a better place to put that data or existing published paper to link?
PF: The reference attached to all curated features is this [1]. We wrote an abstract with the explanation what a curated model is years ago, but it's still good. I think it's fine to attach it to all manually curated genes, including from the new tool.
Action item : add reference to fast track curated models
Protein domains would not be assigned to all proteins at once, it takes around 6 hours to analyze and load 5000 proteins. Analysis can be incremental (running one part a time).
Action item : notify users
Curation stats should they be available from dicty/data folder on server or downloadable from curator interface
Action item: send email, make grahs visible for users
Skipped genes are genes inspected by curator, but since there is no curated model, "Gene Product Information" section states "Not curator reviewed". May be this should be reworded ("Automated" or "computationally inferred") [2]
Action item: remove note for 'not curated genes'
Expected final release
~1 and half weeks/2 weeks from today(10th Jan)

More post release

  • After new OBO file upgrade, curators (with help of Sidd) need to check where terms don't map and map manually. [Reported by GOA QC script]
  • Chr2 coordinate shift for gbrowse2(~2 days)


Multigenome release 2.19

ID resolver
Bug in BLAST reported by Petra
Above project requires upgrade of our web app library(after 1 year) and three pieces of web applications. (~Approx 3 weeks). This need to be done at some point to get in sync with current dictybase release(GO and reference migrations).
Need to decide when we could do it ? - Right after release

Expression Data from Harry

  • WE NEED TO DECIDE IF WE CAN ADD HIS DATA AND IN WHICH FORM AND WHEN.
  • Harry sent expression, spatial and temporal, plus cell cycle regulation data, in an Excel sheet with Macros. User request
  • We should have a look at the data table and make a plan when and how we can use it.
  • I can visualize the graphs (not so elegantly because I can't have Macros on the Mac) and it would be great to have those linked to genes. Here is a screenshot of what the graphs look like Image:Harry_graphs.png
We could either show it on a new tab or section in a gene page.
Specs
Will not be searchable.
Rough ballpark 2 and half weeks.
Have to go through a release at least.

Conclusion: Discuss again when would be right time to do this. Sidd and Warren said we also need to update the gene page caching and these two things could well be done together. This would then take ~3 weeks. Pascale mentioned that adding Harry's data might set a precedent and 'inspire' other people (Gadi) to submit much more (in not such a neat format).


Next software project plans [Feb 2010]

  • dictyMart overhaul. [Part of Next release]
  • GenBank release preparation.


New Genomes (from Joan Strassmann)

  • We need Rex to give update where we stand.



GO Annotations

  • Once GO is in Chado (after the next release to the public - early January??)), we cannot use the old tool anymore, and thus, cannot edit GO. A major thing we do right now is deleting GO from newly annotated pseudogenes. we need to discuss if we can find an interim solution how to do this until we move to the EBI tool, or if we let everything slack and curators have to make notes/lists of what to delete and change once we have a tool again.
  • Emily also send me 6 annotations we have that don't pass the new taxon restriction script, which I'll have to check into these days. There are sometimes other user requests that involve GO annotations, that I won't be able to access for a while.
  • I told Emily that we have now decided to use their tool. We agreed on talking about things (tentatively 2nd week of January), and Sidd/Yulia should be part and if possible Pascale (and everyone else who wants to). Scheduled Jan. 11, 10 AM CST. (Petra)
Issues to solve
Syncing new DDB_G ids between genbank updates.
Syncing our literature references.
Personal tools