AUG 23 2012

From DictyWiki

Jump to: navigation, search

Dicty Meeting Items

Community Annotation form
  • Tried something in google forms to not use developer time [1]. But this seems not sufficient, as this form, as is, only allows for ONE gene, with literature topics, ONE strain, ONE phenotype, and ONE GO annotation - so nothing that would suffice in the real world. Without TABLES, it's not feasible to annotate strains, phenotypes, or GO.
  • Need to develop form in RAILS. Prototype here Image:Community Curation.pdf
D. lacteum Genome
  • Genome available
  • Can be used as 'test' if loading of new genome goes 'easy' as planned
Large Scale Papers
  • Robert and Chris are annoyed by large scale papers showing up on so many gene papers, and would like to 'filter them out'. But this is not a current option, although these papers are annotated with the literature topic 'genome-wide analysis', we have no way of making them disappear. Bob and I decided to not annotate new large scale papers as they offer minimal 'good' data and are not worth the huge effort of attaching 100eds of genes. I agree with Chris who said this problem will only get worse. However, the two current papers they are annoyed by are [2], [3]. Should we even delete these annotations?
  • PMID:22120990 (Investigating the macropinocytic proteome of Dictyostelium amoebae by high-resolution mass spectrometry) all genes have the GO process term "macropinocytosis" with RCA as ev code, done recently.
  • PMID:16926386 (Proteomics fingerprinting of phagosome maturation and evidence for the role of a Galpha during uptake) all genes have the GO component term "phagocytic vesicle" /IDA, an older annotation that should be RCA. Because 'centrosome' annotations from Ralph's 2006 paper were all reportedly wrong (just deleted all), maybe these annotations are also rather mis-leading?
Kimchi - Deleted Genes
  • Kimchi Strasser has requested that we resurrect 25 genes deleted during our gene model curation effort. Allmost all genes were small, with a flawed gene model and little or no support. We do not consider these to be protein-coding genes, although some may be small pseudogenes, others look like RTE fragments.
  • Kimchi has provided RNAseq evidence for these genes-but not in a format which we can add to the database. She has not expressed an opinion as to what type of RNA genes these might be.
  • It is possible that at least some of these are RNA genes, but we as curators cannot add RNA genes to the datase. Also, Kimchi just gave gene prediction coordinates, not very helpful as those are definitely wrong.
  • Update/solutions??
    • Devise a way to add these as RNA genes? But we are also not sure they all are RNA genes / petra
    • Resurrect as flawed protein-coding genes? From the notes, these are very bad / petra
    • Other?
  • Entire email exchange of this issue in issue tracker [4]

Development Updates

Multigenome use

Some stats (August 5-23):

  • Visits: 401
  • Unique Visitors: 232
Stock Center Upload
  • Integration with dictyBase UI
  • Setup for automatic deployment
  • Mac-dependent easy upload??/petra
BLAST server for dev
  • Currently trying to get BLAST server to work with OSX Lion (compatibility issue).
Integrate D.discoideum genome in GBrowse2
Currently waiting .................. for a machine to run tblastn.
* Fisher still has library problem.
* Applied for an quest test and developer accounts (last friday), haven't got any response yet.

Volunteer

  • Anything useful for us in outcome?
  • Finished project with a modular script and unit test.
  • Wrote a script to display domain statistics.
  • Code merge request to integrate with our codebase
Personal tools