Difference between revisions of "WBConfCall 2020.06.18-Agenda and Minutes"

From WormBaseWiki
Jump to navigationJump to search
 
(11 intermediate revisions by 5 users not shown)
Line 2: Line 2:
  
 
===WS278===
 
===WS278===
*Is it confirmed? July 10 Citaace upload to Hinxton.
+
*July 10 Citace upload to Hinxton.
  
 
===Testing data for WS278 model changes===
 
===Testing data for WS278 model changes===
 
*All disease data files passed Caltech and Hinxton testing
 
*All disease data files passed Caltech and Hinxton testing
 
*New genotype data file passed testing
 
*New genotype data file passed testing
 +
*ECO_term file and GO_annotation files also passed testing
  
 
===AFP entity extraction===
 
===AFP entity extraction===
*Strain to paper association. Does every strain mentioned in the paper need to be connected? e.g. N2, OP50..  
+
*Strain to paper association. Does every strain mentioned in the paper need to be connected? e.g. N2, OP50. What about alleles and transgenes used as reagents?
*Species association. For species we are using a threshold that is quite high (10 mentions) that gets rid of false positives (e.g. anti-mouse, anti-rabbit antibodies, etc..). We have some papers, though that mention other nematodes with  a low count (especially for phylogenetic studies) but might be worth to record in WB. Would it be worth to set up a separate pipeline that identifies those papers and populates a reference widget in WB for a given species (e.g. brenneri, japonica, pacificus, ..)?
+
**For strains coming from the AFP pipeline we discussed a potential automated workflow yesterday (Caltech/Hinxton)
 +
**Note that for gene-paper associations, we try to only associate genes that are studied in some way and not genes used as reagents, e.g. genetic mapping, tissue-specific promoters. 
 +
We can have two tags in ACeDB: genes_mentioned and genes_studied. The first tag can be automated and complete, users can see the frequency that each gene was mentioned. The second tag can be populated via referral from the curated data, via AFP author confirmation, mentioned in abstract or title, first time mentioned genes ...
 +
*Species association. For species we are using a threshold that is quite high (10 mentions) that gets rid of false positives (e.g. anti-mouse, anti-rabbit antibodies, etc..). We have some papers, though that mention other nematodes with  a low count (especially for phylogenetic studies) but might be worth to record in WB.  
 +
**Would it be worth to set up a separate pipeline that identifies those papers and populates a reference widget in WB for a given species (e.g. brenneri, japonica, pacificus, ..)?  
 +
**We would need to update the ?Species model to include references. 
 +
**We also associate non-nematode species with papers.  It would be nice to include those, as well, especially for pathogens used in C. elegans host-pathogen studies for which we're otherwise not doing much curation.
 +
The AFP team will work with Sibyl to find out the best approach to store and display the mentioned and studied gene, species, strain, allele ...
 +
 
 +
===microPublication in PMC===
 +
*Went live yesterday
 +
We need to tweet out the info and make more people aware of this. Will contact other MODs.
  
 
=== Help Desk ===
 
=== Help Desk ===
Line 16: Line 28:
 
** Email came in Monday, June 8th
 
** Email came in Monday, June 8th
 
** Do we want to share WormBase "audience" data for research project?
 
** Do we want to share WormBase "audience" data for research project?
 +
Need clarification from this user what is exactly needed for this research.
 +
Rather not to do it that if our purpose is to service the community instead of tracking user behavior.
 +
It is also very time consuming to work on this.
 
* Question about Cre-oac-31 and Cre-oac-42; seem to have disappeared?
 
* Question about Cre-oac-31 and Cre-oac-42; seem to have disappeared?
 
** Email came in on Tuesday, June 16th
 
** Email came in on Tuesday, June 16th
 
** GitHub ticket: https://github.com/WormBase/website/issues/7747
 
** GitHub ticket: https://github.com/WormBase/website/issues/7747
 +
** Paul D has responded and is looking into
 
* 2 bp intron in Cbr-hipr-1 (helpdesk email from Marie-Anne Felix)
 
* 2 bp intron in Cbr-hipr-1 (helpdesk email from Marie-Anne Felix)
 
**https://wormbase.org/species/c_briggsae/transcript/CBG15345a.1#06--10
 
**https://wormbase.org/species/c_briggsae/transcript/CBG15345a.1#06--10
 
** She retracted her question now but 2bp-intron is still weird.
 
** She retracted her question now but 2bp-intron is still weird.
 +
Change of reference genome because of poor quality. We need to notify users.

Latest revision as of 17:22, 18 June 2020

Agenda

WS278

  • July 10 Citace upload to Hinxton.

Testing data for WS278 model changes

  • All disease data files passed Caltech and Hinxton testing
  • New genotype data file passed testing
  • ECO_term file and GO_annotation files also passed testing

AFP entity extraction

  • Strain to paper association. Does every strain mentioned in the paper need to be connected? e.g. N2, OP50. What about alleles and transgenes used as reagents?
    • For strains coming from the AFP pipeline we discussed a potential automated workflow yesterday (Caltech/Hinxton)
    • Note that for gene-paper associations, we try to only associate genes that are studied in some way and not genes used as reagents, e.g. genetic mapping, tissue-specific promoters.

We can have two tags in ACeDB: genes_mentioned and genes_studied. The first tag can be automated and complete, users can see the frequency that each gene was mentioned. The second tag can be populated via referral from the curated data, via AFP author confirmation, mentioned in abstract or title, first time mentioned genes ...

  • Species association. For species we are using a threshold that is quite high (10 mentions) that gets rid of false positives (e.g. anti-mouse, anti-rabbit antibodies, etc..). We have some papers, though that mention other nematodes with a low count (especially for phylogenetic studies) but might be worth to record in WB.
    • Would it be worth to set up a separate pipeline that identifies those papers and populates a reference widget in WB for a given species (e.g. brenneri, japonica, pacificus, ..)?
    • We would need to update the ?Species model to include references.
    • We also associate non-nematode species with papers. It would be nice to include those, as well, especially for pathogens used in C. elegans host-pathogen studies for which we're otherwise not doing much curation.

The AFP team will work with Sibyl to find out the best approach to store and display the mentioned and studied gene, species, strain, allele ...

microPublication in PMC

  • Went live yesterday

We need to tweet out the info and make more people aware of this. Will contact other MODs.

Help Desk

  • Subject line: WormBase - Study on the Audience of Scientific Databases and Data Repositories
    • Email came in Monday, June 8th
    • Do we want to share WormBase "audience" data for research project?

Need clarification from this user what is exactly needed for this research. Rather not to do it that if our purpose is to service the community instead of tracking user behavior. It is also very time consuming to work on this.

Change of reference genome because of poor quality. We need to notify users.