Difference between revisions of "WormBase-Caltech Weekly Calls"

From WormBaseWiki
Jump to navigationJump to search
Line 120: Line 120:
 
* Not clear where the problem originated; maybe some script to populate the OA
 
* Not clear where the problem originated; maybe some script to populate the OA
 
* Karen: would be good to have an immediate report of such instances after they occur
 
* Karen: would be good to have an immediate report of such instances after they occur
 +
* Would be good to sort out if the OA has a problem or if there is a script/cronjob that is creating the issues (or some of the issues)
 +
* Looking at timestamps can give a clue
 +
** for app_anatomy, the issues only occurred in 2008 and a few in 2010
 +
** Also, if timestamps are very close to each other for many entries, likely a script is doing it

Revision as of 18:31, 27 August 2020

Previous Years

2009 Meetings

2011 Meetings

2012 Meetings

2013 Meetings

2014 Meetings

2015 Meetings

2016 Meetings

2017 Meetings

2018 Meetings

2019 Meetings



2020 Meetings

January

February

March

April

May

June

July


August 6th, 2020

Experimental conditions data flow into Alliance

  • Experimental conditions in disease annotations: WB has inducers (used to recapitulate the disease condition) and modifiers (a modifier can ameliorate, exacerbate, or have no effect, on the disease condition)
  • We use the WB Molecule CV for Inducers and Modifiers in disease annotation
  • Experimental conditions in phenotype annotations: are free text (captured in remarks); will probably need to formalize later on
  • So for data flow into Alliance:
    • In the short term we will load the Molecule CV into the Alliance (Ranjana and Michael P. will work on this)
    • Groups will switch to using common data model that works for all and common ontology/ontologies in the near future.
  • How do we handle genetic sex? Part of condition?
    • Condition has been intended for external/environmental conditions, whereas genetic sex is inherent to the organism of study
    • Expression pattern curation needs genetic sex; needs a model at the Alliance for capturing sex


August 13, 2020

Species in Postgres and ACEDB/Datomic

  • Want to dump "Affected By Pathogen" fields in Phenotype OA and RNAi OA
  • Want to be sure that what gets dumped aligns with species loaded into ACEDB
  • Currently one species annotated not in WS277: Streptococcus gallolyticus subsp. gallolyticus
  • We currently have multiple Postgres tables for storing species lists:
    • pap_species_index (used by "Affected By Pathogen" fields, AFP); Kimberly uses to assign species to papers and occasionally adds new ones
    • obo_name_ncbitaxonid
    • obo_name_taxon (original, smaller list)
    • h_pap_species_index (history for pap_species_index)
  • How do species get loaded into ACEDB? Dumps from Postgres? Which table(s)?
  • WS277 has 7,906 species (1,936 have no NCBI Taxon ID)
  • Kimberly has occasionally uploaded a species.ace file in the context of GO curation; but Hinxton otherwise handles it; should ask them
  • New species are associated with paper objects, but otherwise no additional data for those species come from Caltech
  • It might be useful to have species pages in WB that at least list papers for which we have species associations, maybe include other information?

WS279 Citace upload

  • When is it happening? Not sure; not on release schedule right now

SOLR server security (IMSS)

  • IMSS network security blocked network on our server due to its open SOLR web access.
  • Part of AMIGO stack, very old version, drives our ontology browser directly, SObA, Enrichment tools indirectly.
  • Added some firewall/URL filter and IMSS opens up the network (for now). IMSS still gripes about its service is open to the world.

Alzheimer's disease portal

  • Supplement grant awarded to Alliance for an Alzheimer's disease portal
  • Could involve automated/concise descriptions, interactions, etc.
  • Could establish useful pipelines that could be reused in other contexts


August 20th, 2020

Handling sex (male/hermaphrodite) in Expression

  • In Alliance Expression group, talking about how to handle sex in expression annotations
  • In WB, will start collecting sex in a separate field (rather than integrated into ontology terms)
  • Daniela will capture sexually dimorphic expression in remarks/comments
  • Will eventually figure out how to handle/display at the Alliance
  • Daniela will go back to existing annotations to fix data retroactively

Alliance webinars

  • We won't present at Caltech BBE retreat
  • Maybe we can present later in the academic year

WormBase build

  • WS277 took about 3 months
  • If this is a general change (we need 3 months for each build), then we should reconsider our release schedule
  • This was largely due to technical issues at EBI out of our control
  • Really helpful for curators to know when we can expect to have to upload; maybe sliding schedule doesn't work for CIT
  • Proposal: Take 3 months for the next two releases and readjust in the new year; Wen will email Hinxton


August 27th, 2020

Textpresso knockout allele counter

  • Ann Rougvie asked if Textpresso could pull out references to knockout alleles (from deletion project) in C. elegans literature
  • Valerio has a tool setup for the analysis

OA bug; joinkeys with multiple entries

  • Stumbled on while querying the app_anatomy
  • Turns out there are several PG joinkeys associated with multiple entries (should only ever be one)
  • Only latest entry shows up in OA; often there is one entry (single term or list) plus additional blank entries
  • Blank entries could be latest and so show up in OA and in Postgres queries
  • Not clear where the problem originated; maybe some script to populate the OA
  • Karen: would be good to have an immediate report of such instances after they occur
  • Would be good to sort out if the OA has a problem or if there is a script/cronjob that is creating the issues (or some of the issues)
  • Looking at timestamps can give a clue
    • for app_anatomy, the issues only occurred in 2008 and a few in 2010
    • Also, if timestamps are very close to each other for many entries, likely a script is doing it