WormBase-Caltech Weekly Calls

From WormBaseWiki
Jump to navigationJump to search

Previous Years

2009 Meetings

2011 Meetings

2012 Meetings

2013 Meetings

2014 Meetings

2015 Meetings

2016 Meetings

2017 Meetings

2018 Meetings

2019 Meetings

2020 Meetings

2021 Meetings

2022 Meetings

January

February

March

April

May


August 3rd, 2023

Caltech

  • Reminder to do research integrity training
  • Not everyone at WormBase got the email - Paul S. can double-check if this was based on job titles or something else

Genetics Database Update Paper

  • Is there enough material for a WB update?
    • Explanation of transition to the Alliance
      • Start by creating a table of data types in WB vs Alliance
    • New WB data, e.g. single cell data
  • If so, who's turn is it to write it?

July 20th, 2023

Remaining debrief from IWM

  • PIs wanting to undergraduates to do more in terms of curation and training
    • check into training support with NSF community outreach grants

Motivating the community

    • Chris's trophys a huge hit - people wanted pictures and were motivated
    • $250 uPub voucher useful
    • Paul's sense of swag is off, staff have great ideas
    • WB blog supports images - can post pictures of trophy winners

Place for teaching material

want to have a public location to for people to access webinars and slides etc. ex. XenBase
Paul suggests we use an Alliance Forum to post these, will also motivate alliance members to do the same>
Chris suggests the Alliance User Forum
Todd suggests Google drive is good, but many problems with things getting muddled, e.g, by inconsistent naming = things get lost and hard to find
If we use google drive, need to have some management
Do we want a place for the community to share things with the community
Alliance User Support Group should be tapped for help here, and we can reach out to GSA to see what they do to organize community teaching docs.
Raymond - CMS system is designed for this, but we would need to set that up == Wen is interested
Todd- need to identify what type of material would we be putting up. Slide shows? - maybe use a ready built CMS that is good with side-shows, will also need to have maintenance over time.
Maybe the lowest hanging fruit is just Maintaining an Index page for these, then it doesn't matter where the material really exits
Start with generic index page at Alliance, with brief metadata - can we store it in CaltechData = Karen and Wen will talk to the library

Making Web pages for uPub papers

Immediate problem - make a splash page for paper not found at WB. something more general - ask Alliance for splash page
gene paper links is going to be more complicated

WormBook and GSA

Genetics thinking that WormBook is not that active, but possibly they are missing out on activity since community uses WormBook.org as index to find articles.
Dainel has been adding Genetics WormBook chapters to WormBook.org, so they should be available there and findable.

Connectome

Raymond fixed the link-outs to these sites, not updated on the site yet
Todd- hotfix will be run that should fix all of that

Release announcements

Not automated, Caltech needs to synchronize things, Raymond and Wen will be added to Slack release channel.




July 13th, 2023

Update/Discussion on WormBase -> Alliance paper pipeline migration

  • Alliance data model for entity/topic tagging
  • Alliance Tags for Papers Ontology
  • ECO requests
  • Bibliographic data - Alliance will be the source of truth for all references in PubMed (except legacy author data); WB will be the source of truth for all other paper, e.g. meeting abstracts
  • Entity tags
    • Species
      • Modeling evidence
        • 'Manually_connected' plus 'Curator_confirmed' - Alliance doesn't yet have people, so we will store 'WBPerson' in a created_by field
        • 'Inferred_automatically' text string (from ACKnowledge pipeline)
        • No explicit evidence in postgres, but notes on what we did, e.g. scripts and string matching
    • Genes
      • Modeling evidence
  • Topic tags
    • Will port over the WB topic and map to an ATP ontology term
    • Will not populate species (we don't know this definitively at WB)
    • Each flagging 'method' will be a unique combination of: source, source details, MOD, and creator
      • For example, author, ACKnowledge, WB, WBPerson625
      • For example, author, Fast Track Your Paper, FB, Norbert Perrimon
    • Some data types, e.g. antibody, are modeled as topics and entities; we will only populate topic unless we also have a specific curie to populate an entity value
      • New topic tags: antibody data
    • Existing validation will come from manual curator entries in curation status form and from OA values and from author ACKnowledge submissions
      • Each validation is tracked separately; no manual validation will overwrite any other manual validation
      • Manual entries will validate AI entries (more specific entries can validate less specific entries)
    • Modeling and porting over curation status pre-canned comments
      • SVM positive, curation negative
        • Validated right by curator, but not curatable for WB (will be a new workflow ATP tag)
        • WB may not have always been consistent here, so we may want to think about clean up after migration
      • C. elegans as heterologous expression system
      • GO curation (will not port over; we can get this information from a GAF, if need be)
      • BioGrid and BioGrid/GO - keep?
      • Curation negative, no strain name given in paper
      • Toxicology - create an ATP value?
      • Host-pathogen virulence - create an ATP value?

July 6th, 2023

June 8th, 2023

  • Update on WormBase -> Alliance paper pipeline migration
    • Newly added WB papers for retraction notices and corrections

June 1st, 2023

  • Raymond: We need to prepare some slides to introduce our tools at IWM. It would be better to drop them to the google doc folder for IWM.
  • From Tazendra to Amazon Cloud server: Valerio finished the backup setting and now restoring data on the cloud server. Now testing Wen's script and the full build will be tested after IWM
  • Textpresso will be modified and prepared for ABC in Alliance. This will be discussed again after IWM.

May 25th, 2023

  • Waterston data
    • Noticed that Bob's list had 69 dead genes and that the graphs were displayed on the dead gene page
e.g. WBGene00009022
https://wormbase.org/species/c_elegans/gene/WBGene00009022#0g1--10
    • Asked Paulo to map the dead genes to the live ones and display on the live gene page.
    • While doing this Paulo noticed that 55 genes had values for both dead and live genes, and the values are quite different.
https://wormbase.org/species/c_elegans/gene/WBGene00007084#0g1--10
    • Notify Bob, can ditch dead genes

May 11th, 2023

  • Waterston data update
    • Generated a file with the correct mappings
    • Renamed Canopus the image files on Canopus
    • Generated a -D file for Citace Minus
      • Send the .ace file to stavros
      • Contact Paulo and Todd
    • Tested on spica, all good
    • Patch the file for WS289. What needs to be done?
  • Next upload
    • Falls right after the IWM and around the 4th of July holiday
    • Do we want to do a trial upload with our new system beforehand?
    • We will upload the files on spica as we normally do but will also test the new system as much as we can
    • Will do upload on July 11th
  • Person wormbase_comments in acedb vs internal comments in postgres, but neither on WBPerson pages on website.
    • No one recalls why they're in acedb, email sent to wormbase@caltech with questions for Cecilia. There are ~14k wormbase_comments

May 4th, 2023

  • Bug with molecule OA
    • JSON parse failed error -> fixed
    • Juancarlos will look into this
  • Alliance postgres migration
    • Chris started working on this

April 27th, 2023

  • 445 image files provided by Bob Waterston contain the wrong WBGene name, e.g. egl-1_WBGene00001186_embryo_lineage.jpg (egl-1 is WBGene00001170)
  • noticed while doing spot check in discussing Eyleen J. O'Rourke data
  • Paulo programmatically went through the files, the graphs are unaffected, the images are.
  • Option 1: ask Juancarlos to rename the wrong image files, based on gene public name. Prepare a -D file for CitaceMinus and re-upload a .ace file for the wrong objects.
  • Option 2: ask Bob to send again the original files with the right name.
  • Regardless, notify Bob

March 23rd, 2023

  1. Some updates on Neural Networks development (postponed)

March 16th, 2023

  1. Getting data from acedb to load postgres. (e.g. Variation) Do we still need this ? Automated scripts seemed to break a lot, while running them manually always worked.

March 9th, 2023

  1. [migration to the cloud]
    1. The generic.cgi, we need to know who uses these:
      1. tempVariationObo: this brings in variations into the OA that are in WB but not in the OA. Note this is a different functionality compared to the Name server. We need this. Used by Ranjana, ??
        1. CG: Yes, I need this
      2. UpdateVariationObo--can't get to this now on Tazendra
      3. AddToVariationObo--can't get to this now on Tazendra
      4. TempStrainObo --brings in strains into the OA that are in WB, seems to be the counterpart of the tempVariationObo for strains
        1. CG: I need this
      5. AddTempObjectObo--used to add data to the database, used by AddToVariationObo and AddToStrainObo
      6. PapIdToWBPaper
        1. CG: I need this
      7. When people added things by mistake they would sometimes remove them from the flatfile, does anyone still do that ? -- Chris and Ranjana don't, Karen or Daniela might
      8. Does the public need access to this, or should we move this to a private / protected URL ? -- Nick/micropub probably needs this
    2. WSVariation and nightly geneace variation file
      1. Overlapping but WSVariation has a lot more variation-related info, e.g. phenotype and phenotype_not_observed
      2. Confirming that we still need both
    3. There's also WpaXref and WpaXrefBackwards and ListPmids and ConvertPaperIdentifiers
    4. The only thing is that it writes to flatfiles so when cronjobs update variation/strains from nameserver or WS, it also loads from the flat files. The location of the flatfiles isn't final yet, but it's dockerized. People used to manually edit that when they wanted to remove them from the list, do people still want to manually edit those flatfiles ?
    5. ContinentPIs
    6. Valerio has set up ssh access into textpressolab.com
    7. There's 3 cases for using acedb:
      1. your local use for testing stuff
      2. spica replacement for Wen
      3. extract things automatically for tazendra/postgres. Do we need this ?
  2. Some updates on Neural Networks development

Feb 23rd, 2023

New paper folder organization may affect downstream pipelines

Feb 9th, 2023

IWM 2023 booth cost

  • From Mary Anne this morning: "I had the cost wrong, it would be $750 for you at cost. Sorry, I had been looking at the wrong line item."

January 26th, 2023

Alliance issues

  • monitoring API usage?
    • We have Google Analytics but since Patrick Kalita left, no one has been monitoring it
    • Paul: we need for now a simple way of tracking number of users. Chris: seems reinventing the wheel to write something from scratch. tracking API calls would be useful for our funders.
    • Olin said you can track UI events (e.g. page clicks) and API access (including programmatic access) but we are not currently. Chris will see how much work is involved in this. We need someone with Google Analytics experience.
    • Would be good to put a call out to all people at Alliance asking who has experience with Google Analytics.
  • Simplemine: reverse proxy hosted at alliancegenome.org. If it is dockerized we can do it, otherwise not. Eventually dockerize Simplemine and move it to Alliance servers.
    • Olin concerned about security over CGI technology.
    • We will talk about this (implementation and security concerns) at the developer call
  • Embedding alliance tools or processes into MODs.
    • The API is there but is not in all MOD interest to write code to pull the data out.
    • iFrame is not very flexible on how you display data (e.g. narrow, little flexibility in configuration).
    • Could set up widgets as we did for the GO ribbon. Still requires MOD implementation of code base.
    • Paul: eventually MOD sites will be retired. Chris: One concern from Christian (ZFIN): clicking on a zebrafish gene name will open up an alliance page, could it link to ZFIN gene page or provide users with an option?
    • Maybe opening up an Alliance widget at the MOD preferably happens in a separate window (e.g. pop-up window, new browser window/tab; as opposed to in the same page/tab).

Alliance All Hands

  • Not all the curators are going to be there. Focus on PI discussion and development teams?

January 5th, 2023

  • Chris will help to create a reverse proxy for the Alliance SimpleMine. It will function like WormBase SimpleMine. People will need to visit the Alliance before going to tazendra. It will prevent users from seeing the tazendra URL and create more traffic for the Alliance
  • Alliance SimpleMine with XenBase data is ready but needs XenBase feedback before it goes public. It is based on Alliance 5.4.0 release, so it can be released when Alliance 5.4.0 goes public in late Jan or mid-Feb.
  • Karen will visit Pasadena 1/13-15

December 8th, 2022

Data Type Flagging Pipelines and Alliance Infrastructure

  • We currently run the neural network algorithms at Caltech
  • What will we need to do to transition this to run on Alliance infrastructure?
  • Can we start to run the neural networks on a more frequent basis, e.g. monthly? Could that be automated?

WB PDF Migration

  • Daniel, Juancarlos, Kimberly continue some clean up of redundant files before moving things over to the Alliance

November 17th, 2022

  • TEC_RED meeting. Meeting with Nikita Jahveri, Wouter Van de Berg (Gupta's lab). Participants, Valerio, Daniela, Kimberly, Magda, Stavros.

identified a large number of operons in C Briggsae. Baghwathi Gupta's lab. 4200 Briggsae genes identified with SL influencers. Want to contribute the information in WormBase.

  • Used WS279 for the annotations
  • they have excel documents with the location in the supplementary files
  • Magda: are the tag sequences unique? Nikita: yes. If they bind to 2 ore more genes that have classified as paralogs
  • Also identified isoforms (multiple transcripts tab in the spreadsheet)
    • Steps forward: need to update the gene models and the operon models on WormBase.
  • Can bring in data in different tracks.
  • Need to update gene models based on this new evidence
  • Nikita will send the data to Magda and Stavros (3 spreadsheets). Need to write out the categories.
  • they did not produce the GFF file, all in excel sheets.
  • Material used: mixed stage worm population

they have excel documents with the location

November 3rd, 2022

  • Booth at IWM
    • ACKnowledge will be showcased at the WB booth
    • Will also advertise Alliance

October 27, 2022

October 6th, 2022

Mangolassi

  • Mangolassi is down (network only?); Michael will be in the office in an hour
    • Start a cronjob to check for network connections? Already set up (Raymond)
    • Michael can log into his machine; Tazendra is OK (so, not network issue?)
    • Michael fixed it by unplugging and reconnecting ethernet cable (network card issue)

GO meeting next week

Alliance Literature working group

Persons

  • Are there more use cases to collect?
  • Some existing use cases:
    • ACKnowledge attribution
    • Community phenotype curators
    • When do people need person IDs? When accessing/using some forms
  • Content of earlier email
    • When someone asks for a CGC Lab code, they must have a WBPerson ID. To fill out some tazendra forms, I think they need a WBPerson ID. To connect Persons to Laboratories for their WB lab page. To track who is a PI for which Lab. To connect to other Persons for Intellectual Lineage. Connect authors to Persons so Papers have a list of authors in publication order, each author linked to the Person (ZFIN has two lists, sorted for authors, separate for persons). Our users tell us they use the WB Person page for their cv.
    • Use of Person Contact information: I think our use case for person information is mostly to have current emails for community outreach (ACKnowledge, community phenotype forms). It's been used in the past to send letters to PIs, so we could prioritize maintaining more information just for PIs. Some analysis is sometimes done by region / country, but this is rare. If anyone can think of other use cases for person data, please let us know. While we mostly only use the email addresses, Cecilia uses address / webpage / ORCID to connect authors to Persons, so having that data is important to make more connections.
  • Knowing address and affiliation can help disambiguate similarly named persons/authors
  • Questions about ethics of tracking information and person lineage?
    • Not really a new issue; people present CVs on grants and other contexts
    • Can be problematic to display people's email address and physical address, from a security perspective
    • Email addresses: if someone has a strain/reagent not in a repository, can be helpful to connect to access materials
    • We and others provide email addresses in publications
    • We don't currently ask for permission to display the information (email, address, etc.); maybe we should?
    • Should we have a survey to ask users' opinions about displaying this information? Is it worthwhile? Concerning?
    • Should we provide a waiver that people agree their information will be in the public domain?
    • Person form disclaimer: "Note: Phone and fax information won't show in your Person report for privacy reasons, but we are collecting the information in case we need to contact you. We keep old email, old institution, and old registered_lab data for history, please don't remove it when updating your information."
    • Person form is available to public and displays personal information, e.g.: http://tazendra.caltech.edu/~azurebrd/cgi-bin/forms/person.cgi?action=Display&number=WBPerson51134
      • Could someone scrape this information? Yes, person information is available from WB database dumps
    • Could institute a login; we had wanted to keep the barrier to participate low
    • Have 3 forms of privacy, where people can opt out of emails, hide their email address, or hide more information. If hidden it stays in Caltech postgres, and doesn't go to acedb.
  • Going forward, the Alliance can implement better security for person information

Aug 25th, 2022

  • Phosphoproteomic dataset
  • WB grant - the new and radically different format for a “Biomedical Knowledgebase (U24 - Clinical Trials Not Allowed)” application



Aug 18th, 2022

  • Raymond reporting on meeting with Scott Emmons and neural connectivity data

Aug 11th, 2022

Secondary species information content.

  • Gone through and counted number of objects in different data classes for all species
  • https://docs.google.com/spreadsheets/d/1Q1FE7Miz0IJxultBtEq-VFoAcdz1UpxM3wfER4-qQGg/edit?usp=sharing
  • Pristionchus: Talk to Ralf Sommer and see if wee can take some of the pristionchus.org data into WB -> Ranjana; Decision: after more discussion, decided to hold off on this as the plan is to focus on C. elegans for the grant. Perhaps in the future, we can seek finding and collaboration with the Pritionchus community (such as Sommer lab, etc).
  • Brainstorm on how wee can utilize other species data -> may be a separate grant
  • remanei (4 genome versions) and briggsae (3 genome versions) problem -> several versions of the genome. Same problem in parasite. Will need to address in the future. Want to have multiple sequence alignments so researchers can compare different versions.
  • Address c elegans version assemblies in the grant. LoS from authors?
  • Parasite has a new RNAseq analysis pipeline (not single cells yet)

July 28th, 2022

Global Core Biodata Resource application https://docs.google.com/document/d/1juJ4mm1evay32lNt3OOsbKL36RY4_3cW/edit?usp=sharing&ouid=106019143058337411488&rtpof=true&sd=true


July 14th, 2022

Reference and Person Evidence

  • We need to know how best to move forward with ?Reference and ?Person evidence in the context of data exchange between WB and the Alliance
  • Does A-team need to connect those WB Persons to be able to populate data in the Alliance curation db ?
    • If so, can it wait until we have proper persons ?
    • If not, when we have persons later will the data be reconnected to Person objects ?
  • Where does data come from, which we call Person_evidence ? Some forms ? Direct emails ? Other sources ?
  • For author curation/verification, what evidence do we want to use?
  • Can we create Paper objects for the actual emails or other forms of communication (figure out details of when and what requirements), but keep Person_evidence for form submissions ?
  • How much will it muddy things to create Paper objects for emails, e.g. how will it affect NLP downstream, do we need PDFs ?
  • If we create personal communication References at Alliance only do we also want them as WBPapers, or will we just keep ?Person evidence at WB?
  • Are we okay with conflating Person data and Paper data in Reference at the Alliance. That is, duplicating the same data as a Person and as a Reference.

How to look up balancers (Df / Dp) in WB (PWS/Ryan Baugh)

June 30, 2022

June 23, 2022

Helpdesk tickets

SOba in Alliance

GCBR application due 8th Aug

Prepare for grant; nice pics for metrics eg https://pharos.nih.gov

Should Alliance gene & allele model have Clone/Sequence name as an attribute?


May 26, 2022

WormBase user survey



May 12, 2022

Subcellular localization field

  • WB has 2 separate free text fields to capture anatomical and subcellular localization statements but obviously the boundaries are fuzzy, see the example below where the statement belongs to both- there are many examples.

“miR-228 is expressed in the germline and preferentially localizes to the nuclear periphery.”

  • Other MODs do not make such distinction.
  • Daniela is asking the group advice to see if there is any objection in merging the 2 fields and slot the paragraph descriptions in the ExpressionExperimentStatement in LinkML.
  • Decision: Keep just one field that will take care of both anatomical expression and subcellular localization
ExpressionExperimentStatement:

    is_a: EntityStatement

    description: >-

      Free-text describing some aspect(s) of a gene's expression, particularly

      nuanced information that is not readily captured in annotations.

      This statement's scope is limited to the associated ExpressionExperiment.

    notes: >-

      Inherits: statement_subject, statement_type, statement_text, references.

    slot_usage:

      statement_subject:

        range: ExpressionExperiment
Retrieved from "https://wiki.wormbase.org/index.php?title=WormBase-Caltech_Weekly_Calls&oldid=38049"