JAN 26 2010 curators

From DictyWiki

Jump to: navigation, search

Contents

How do we handle different gene sequences from different strains?

Example: Talin

  • Issue from Anette:

After Rob's talk at the Dicty meeting revealing that AX4 does have that premature stop codon that we bridged with a gap, I talked to Annette quite a bit, and she said she would use different strains and do Westerns and use her talin Ab. Today she sent me what she found:

talin protein + : AX2, AX3 Chisholm, AX3 Insall
talin protein - : AX3 Kessin (from 1982), AX4 (Kuspa)

She said AX4 has at best a very faint truncated talin. She said from just briefly looking at the cells but not really testing and quantifying, she thinks the talin neg cells are more round and look less adhesive.


Pascale suggestion: We could create a new gene 'AX2-talA' (see on test) [BTW I wasn't able to change the gene model - something's not working with Apollo on test]

http://testdb.dictybase.org/db/cgi-bin/search/results.pl?class=dicty::UI::Search::Gene&query=talA


Petra: I think we could do that like you did on test - call it talA-AX2 (I renamed it to that) with synonym talA - so when you search you find both http://testdb.dictybase.org/db/cgi-bin/search/results.pl?class=dicty::UI::Search::Gene&query=talA

I like that - of course on the talA-AX2 site we need to add the note that the correct sequence is on the wiki, and on the talA site a note that there is a talA-AX2. Questions:

  • what do we do with the papers? Look at each and decide, or attach all to both genes?
  • What about annotations?? Maybe we need to separate?

Update after another discussion P&P: Maybe we should not increase gene number by doing it like that. We have the AX4 genome and represent what's there, the truncated gene, with very obvious notes and the AX2 sequence on the wiki. Questions remain:

  • SwissProt has the correct sequence and should change their taxon to AX2 - will that affect our linkout?
  • What do we do with GO annotations? All functional annotations are based on AX2, how do we handle that?

Conclusion for now: We do the simple solution and keep one gene with clear notes and AX2 sequence on the wiki. Take care of annotations later. 27-JAN-2010

Bad link outs

  • The Baylor site has been down for several months. Sent an email to Ricky and Adam again today, 27-JAN 2010 Petra;
    • Fixed
  • The Sanger (GeneDB) has not been updated in ages.
    • [action item] : email Matt Berriman and Rob Kay

What should we do?

GO IEAs by BLAST

Follow up on Jan 20th meeting:

  • list of GO IEAs generated by BLAST as of Jan 21, 2009
  • There are only 87 annotations left :)
  • I suggest to delete those like 'T cell homeostasis' and leave the more general 'acetate fermentation'.

[agreed; script has been turned off Jan 26-2010]

- Pascale

Automated Gene products

  • in paranoid(Pascale) I downloaded the human <-> Dicty orthologs as predicted by in paranoid. There are 3,960 Dicty genes in 'clusters' (3,351 with boot strap values of 100%)
  • about 460 genes currently do not have a gene product and would probably be assigned some gene product
  • PPOD doesn't have files for pairwise groups (all species are in the same file)
  • OMA:
    • 1,800 1:1 groups
    • no bootstrap values available
  • OrthoMCL (/orthomcl.org): also uses ensembl IDs.
  • Action item: When done write announcement on front page and ListServ.

Curator Priorities

  1. User Support [ ? % ]
  2. Literature Curation, including [ ? % ]
    • strains, plasmids
    • phenotype
    • Gene ontology
  3. Manual Review of Gene Models
    • focus on genes with errors
  4. Naming genes [ ? % ]
  5. GO Consortium tasks [ ? % ]
    • reference genome: PAINT tool development; annotation of trees
    • working groups
    • ontology development
  6. Maintenance tasks [ ? % ]
    • dictyNews
    • curation statistics
    • Database testing
    • Standard operating procedures updates
    • HTML pages; techniques, educational resources, etc
    • annotation consistency exercises

Pascale: 20% + 50% = 70% GO (ref genome etc); 20% Literature (2-3 papers); 10% Other
Petra: 10% GO; 40% gene models (10 models); 40% Literature (4-6 papers); 10% Other
Bob: 90% Gene models (18-25 models); 10% Literature (gene associations)

Meetings in 2010

  • GO Consortium Meeting March 30-31, Stanford: Rex (+SAB, April 1), Pascale (+SAB), and if Pascale gets money from GOC, Petra
  • GO annotation camp, Geneva, Switzerland, June 14-18: Pascale
  • Dicty 2010, Cardiff, England August 1-6: Rex, Pascale, Bob, Petra
  • International Biocurator Meeting, Chiba, Japan, October 6-9 - wait for funding situation

Back to Curator meetings

Personal tools