Difference between revisions of "WBConfCall 2023.04.20-Agenda and Minutes"

From WormBaseWiki
Jump to navigationJump to search
(Created page with "= Agenda and Minutes = *Below is mostly agenda and minutes from 2023-04-06 meeting - will update before 2023-04-20 == Help Desk == * Please double-check any tickets assigned...")
 
 
(22 intermediate revisions by 5 users not shown)
Line 1: Line 1:
 
= Agenda and Minutes =
 
= Agenda and Minutes =
  
−
*Below is mostly agenda and minutes from 2023-04-06 meeting - will update before 2023-04-20
+
== New 18 C. elegans wild isolate genomes from Cristian Riccio ==
 +
* Sarah sent an email with details on 28th of March to Paul, Todd and Scott.
 +
 
 +
<blockquote>
 +
Cristian generated 18 high quality C. elegans genomes assembled from PacBio data from wild isolates which came from Erik Andersen’s lab, and were collected from across the world. The isolates were cultured a few times to increase the homozygosity and are now frozen in Eric Miska’s lab in Cambridge. Cristian lifted over the annotations from N2, ~98% mapped, and he has been exploring structural variants (50bp - 10kb, NGMLR & Sniffles) with a subset experimentally validated, and noted some missing genes due to this i.e. natural knock-outs. Cristian has deposited the reads into ENA, and the assemblies and VCFs (SVs against N2) into Zenodo. He can share Assemblies, mapped annotations, VCFs and bams with us as necessary.
 +
 
 +
Potential options moving forward:
 +
- We could add these 18 new genomes in as tier 3 species - gene trees / synteny would probably be the most useful views, keeping in mind that the annotation is only lifted over. Timelines on this with current resourcing are not great...
 +
- We showed Cristian the SV view shared by Scott and he was interested in this for his data mapped to N2 - but commented that he only has SVs up to 10kb, and no translocations. Would this data work? Is this view based on mapped reads / bams? Can the VCF of SVs be visualised too?
 +
 
 +
I explained that we would need to discuss whether bringing in all of these genomes was the right way forward and what the timelines would be, but if not we can explore whether WormBase ParaSite or Ensembl Metazoa could take them i.e. if we only want to display the SVs as tracks on N2 rather than bringing them in as new species for WB.
 +
 
 +
What do you think?
 +
</blockquote>
 +
 
 +
* Still awaiting response - do we want these as tier 3 or as SVs wrt N2?
 +
* Should these be rendered in JBrowse 2?
 +
* Todd: What would happen to this data going forward into the Alliance? If this data won't be supported at the Alliance, it may not be the long term solution the authors are looking for.
 +
* Paul S: in the future (not this year, necessarily) we will figure out how we want to handle this data; can ask the WB SAB in May
 +
** What is an immediate stop gap measure to implement and send to Cristian? Just make data available and work on additional implementation later
 +
** Minimal implementation?
 +
*** What does it take to build JBrowse tracks? Scott: FASTA files, VCFs; nice-to-haves: BAM files, descriptions (data meta data), gene models if available
 +
*** Make files available for download: whatever they provide  (FASTA, etc.)
 +
** Scott Cain can get back in touch with Cristian and keep Sarah and Stavros in the loop
  
 
== Help Desk ==
 
== Help Desk ==
 
* Please double-check any tickets assigned to you and close, if you can
 
* Please double-check any tickets assigned to you and close, if you can
−
* WBPaper00049101 Curation question - this looks to be a question about community phenotype curation; I've assigned it to Chris for now.
+
** My script is using downloads.wormbase.org - Todd replied; will leave open for now in case there's more follow up
−
* My script is using downloads.wormbase.org - Todd replied; will leave open for now in case there's more follow up
+
**Hello, could you please reference.... - I'm leaving this open as a reminder to follow-up with Ranjana and Valerio about references for automated descriptions
−
* Hello, could you please reference.... - I'm leaving this open as a reminder to follow-up with Ranjana and Valerio about references for automated descriptions
+
** {"readyState": 4,"requestURL": "/rest/widget/variation/WBVar00543472/molecular_details","status": 500,"statusText": "Internal Server Error"} - there are three similar tickets for this issue.  Paolo replied on one, but I'm leaving them open in case there's still work to do.
−
* {"readyState": 4,"requestURL": "/rest/widget/variation/WBVar00543472/molecular_details","status": 500,"statusText": "Internal Server Error"} - there are three similar tickets for this issue.  Paolo replied on one, but I'm leaving them open in case there's still work to do.
+
** The search directory on the top right.... - request to help with search behavior.  Todd, Scott, and Paolo have commented.
−
* The search directory on the top right.... - request to help with search behavior.  Todd, Scott, and Paolo have commented.
+
*** Paulo pushed to staging for WS289 (ticket: https://github.com/WormBase/website/issues/9012) touch base with Michael Nonet
  
 
== Links to Alliance ==
 
== Links to Alliance ==
Line 22: Line 45:
  
 
== IWM Meeting ==
 
== IWM Meeting ==
−
* Planning for the WB workshop
+
* Planning for the WB workshop; anything to discuss?
−
* Need some separate meetings for this?
 
−
* TH: who’s attending? Do we have travel funds available?
 
  
−
== New items ==
+
== AWS credentials ==
 
* WB needs Genes from nameserver, which is on AWS, so need creds or AMI ?
 
* WB needs Genes from nameserver, which is on AWS, so need creds or AMI ?
 
** TH
 
** TH
Line 39: Line 60:
 
*** Here's a copy of the .jar file
 
*** Here's a copy of the .jar file
 
****  https://tazendra.caltech.edu/~azurebrd/var/work/todd/wb-names-export.jar
 
****  https://tazendra.caltech.edu/~azurebrd/var/work/todd/wb-names-export.jar
 +
*** Also
 +
**** export WB_DB_URI="datomic:ddb://us-east-1/WSNames/wormbase"
 +
*** And .jar was originally at
 +
**** aws s3 cp s3://wormbase/names/exporter/wb-names-export.jar wb-names-export.jar
 
** Thanks, JC! We can make that work from one of our persistent instances. I will look when I return from vacation.
 
** Thanks, JC! We can make that work from one of our persistent instances. I will look when I return from vacation.
 +
* AWS credentials leaked
 +
** AWS has become very strict; locked our account
 +
** Todd H has fully shut down nefarious resources across the globe; still had to petition AWS
 +
** Bottom line: We will use a more restrictive set of credentials for a WB AWS superuser; everyone else has access but with more restricted permissions
 +
** If anyone has permission problems accessing AWS, let Todd know
 +
** Todd retired some outdated credentials (10+ years old); it is good practice to regularly change credentials
 +
** Google accounts also affected; strongly encourage using 2-factor authentication
  
−
== IWM ==
+
== Tazendra migration ==
−
* Mary Ann Mahoney would ask for more details when the time is right.  It's accepted we have the workshop.  Ranjana has 4 talks listed.  90 minutes total.  60 mins of talks, 30 mins of open Q&A.  Ranjana would like to go to the meeting.  May is early registration deadline.  https://docs.google.com/document/d/1ZYjSX2UGcvLY2lRY5qGuVYO1Q7euHlXZ6kHJ0W0AEVA/edit?usp=sharing
+
* Caltech curation/Postgres server is migrating to AWS
−
* Paul S will look into funds, Todd should talk to Paul S.  If people want to go, go, Paul will figure out funds later.  Everyone use sick leave and vacation time so we have money, that accounts for +/- 2 people's salary.  Paul's lab will send just a couple of people.
+
* Everything is being dockerized and moved to AWS
−
* Raymond : use Ranjana's document to organize talks, review topics, and other things.
+
** Valerio: we are using names from a domain I setup with 'textpresso-central'; would be good to setup names under the wormbase.org domain
−
* Ranjana :
+
** Valerio will specify to Todd where they should point
−
** Talk 1: Starting with one and arriving at many: Using WormBase and the Alliance of Genome Resources for gene data. (10 mins + 5 mins for questions)** Talk 2: What tool do I use? Matching tools to data searching (10 mins +  5 mins for questions)
+
* Will eventually need to make changes once the switch is official
−
** Talk 3: JBrowse and recent updates to genome browsing (10 mins + 5 mins for questions)
 
−
** Talk 4: Be ACKnowledged and help us curate your paper!  (10 mins + 5 mins for questions)
 
−
** 30 mins Q and A session after Talk 4:: All questions related to WormBase or the Alliance will be answered.
 
−
* Chris could do first talk, Ranjana would like to highlight disease under gene umbrella.
 
−
* Todd could do the tools talk, otherwise Raymond could.
 
−
* Scott not sure if he's going, to talk about JBrowse, but could record a talk in advance and be on zoom.
 
−
* Daniela / Kimberly will probably be there for ACKnowledge, Valerio won't be there.
 
−
* Paul : Can we talk about sequence ?  How is WormBase handling other assemblies.
 
−
* Stavros would like to be there, but Sarah Dyer has the last word.
 
−
* Paul will try to organize an SAB to handle some comments.  Alliance should be able to handle orthologies in the next phase, but could be delayed a year or so, but not forever.  Want more people contributing to worm, collecting info is great to get more people what they need.
 
−
* Raymond : put that in talk 4 ?
 
−
* Paul : Listen to people in booth.
 
−
* Ranjana : Use Q&A section.
 
−
* Wen : ParaSite workshop invited us.
 
−
* Karen : Acknowledge people who curated with prizes or callout ?
 
−
* Valerio : ACK has list of top contributors, meant for public acknowledgement, Kimberly / Daniela should say something, new logo, probably swag.
 
−
* Raymond : mugs and shot glasses were popular, could be special swag.
 
−
* Karen : worm oscar trophies, a worm holding a worm.
 
−
* Raymond : list people who are going / not going on document.
 
−
* Chris : Meeting organizers requested 30 min Q&A.
 
−
* Ranjana : There have been questions before and various WormBase staff around.
 
−
* Raymond : Could remove questions on 4 talks to have extra time for 5th talk.
 
−
* Ranjana : Could ask to reserve questions for Q&A
 
−
* Chris : Would be good to get questions right away so people don't wait an hour.
 
−
* Paul : Could ask people to come to the booth, and then get feedback.  Encourage people to come to booth for prizes.
 
−
 
 
−
== Grant ==
 
−
* Paul : Summary statement for WormBase grant.  Love everything.  2 criticisms.  Alliance is good, but what if it fails.  Weren't as innovative as previous years.  Because we're not proposing tools.  Score is 14 which is amazing.  We can be innovative.
 
−
* Karen : They've complained before about tool making.
 
−
* Paul : Study section has bioinformaticians instead of elegans researchers, not the same as NIH.  Don't have to worry for 5 years.  Go back and say our job is to innovate.  We should aggresively figure out how to get more resources for MOD, in and out of country.  Innovation for that is good.  Do cool stuff.  It's gotta be fun.  Rejuvenate, re-enjoy.
 
−
* Wen : Design game for community curation, exchange points for something.
 
  
 +
== WormBase ParaSite 18 has been released ==
 +
* Our Release 18 is out since last week - https://wbparasite.wordpress.com/2023/04/13/wormbase-parasite-release-18-our-biggest-update-yet/.
  
−
== WS289 model change (not happening) ==
+
== single-cell data on staging ==
−
* Ranjana has a small model change to get asserted human genes in at the point of curation.  Mark and Stavros have looked it over.  Stavros has said there's time to get that in.  Stavros will try to do it next week or on the 17th after holiday.  Wen thought upload would be the 11th, if build won't start until 27th, could postpone upload until 18th.  Stavros needs a couple of days pre-build, will start on 17th.  Ranjana could push model change build to next release, need model change.  Mark can do model change on Tuesday 11th.  Caltech needs it by upload on the 11th.  Mark will do model change tonight.  Ranjana and Juancarlos will change dumpers.  Raymond : rebuild CitaceMinus with new models and notify people.  Wen : prefer procedure to have model changes 2 weeks before.  Mark : Magdalena knew procedure, but left, so EBI needs to know.  Stavros : Asked for model changes a couple of weeks ago.  Ranjana : Came up after face 2 face, let's hold off.
+
* Paulo has implemented the latest single cell expression data on staging for the WS288 release

Latest revision as of 16:19, 20 April 2023

Agenda and Minutes

New 18 C. elegans wild isolate genomes from Cristian Riccio

  • Sarah sent an email with details on 28th of March to Paul, Todd and Scott.

Cristian generated 18 high quality C. elegans genomes assembled from PacBio data from wild isolates which came from Erik Andersen’s lab, and were collected from across the world. The isolates were cultured a few times to increase the homozygosity and are now frozen in Eric Miska’s lab in Cambridge. Cristian lifted over the annotations from N2, ~98% mapped, and he has been exploring structural variants (50bp - 10kb, NGMLR & Sniffles) with a subset experimentally validated, and noted some missing genes due to this i.e. natural knock-outs. Cristian has deposited the reads into ENA, and the assemblies and VCFs (SVs against N2) into Zenodo. He can share Assemblies, mapped annotations, VCFs and bams with us as necessary.

Potential options moving forward: - We could add these 18 new genomes in as tier 3 species - gene trees / synteny would probably be the most useful views, keeping in mind that the annotation is only lifted over. Timelines on this with current resourcing are not great... - We showed Cristian the SV view shared by Scott and he was interested in this for his data mapped to N2 - but commented that he only has SVs up to 10kb, and no translocations. Would this data work? Is this view based on mapped reads / bams? Can the VCF of SVs be visualised too?

I explained that we would need to discuss whether bringing in all of these genomes was the right way forward and what the timelines would be, but if not we can explore whether WormBase ParaSite or Ensembl Metazoa could take them i.e. if we only want to display the SVs as tracks on N2 rather than bringing them in as new species for WB.

What do you think?

  • Still awaiting response - do we want these as tier 3 or as SVs wrt N2?
  • Should these be rendered in JBrowse 2?
  • Todd: What would happen to this data going forward into the Alliance? If this data won't be supported at the Alliance, it may not be the long term solution the authors are looking for.
  • Paul S: in the future (not this year, necessarily) we will figure out how we want to handle this data; can ask the WB SAB in May
    • What is an immediate stop gap measure to implement and send to Cristian? Just make data available and work on additional implementation later
    • Minimal implementation?
      • What does it take to build JBrowse tracks? Scott: FASTA files, VCFs; nice-to-haves: BAM files, descriptions (data meta data), gene models if available
      • Make files available for download: whatever they provide (FASTA, etc.)
    • Scott Cain can get back in touch with Cristian and keep Sarah and Stavros in the loop

Help Desk

  • Please double-check any tickets assigned to you and close, if you can
    • My script is using downloads.wormbase.org - Todd replied; will leave open for now in case there's more follow up
    • Hello, could you please reference.... - I'm leaving this open as a reminder to follow-up with Ranjana and Valerio about references for automated descriptions
    • {"readyState": 4,"requestURL": "/rest/widget/variation/WBVar00543472/molecular_details","status": 500,"statusText": "Internal Server Error"} - there are three similar tickets for this issue. Paolo replied on one, but I'm leaving them open in case there's still work to do.
    • The search directory on the top right.... - request to help with search behavior. Todd, Scott, and Paolo have commented.

Links to Alliance


IWM Meeting

  • Planning for the WB workshop; anything to discuss?

AWS credentials

  • WB needs Genes from nameserver, which is on AWS, so need creds or AMI ?
    • TH
      • I will give a post-mortem of the recent credential leak at the next conference call.
      • The preferred and recommended way to do this would be via an assumed role, not dedicated AWS CLI credentials.
      • I think you mean instance? AMIs are images of instances. IMHO a stand-alone instance is overkill, and already have instances available to run a cron.
    • Juancarlos
      • That's great. I don't know much about AWS. If you're able to set up a cronjob to daily
        • java -cp wb-names-export.jar clojure.main -m wormbase.names.export gene genes.csv
      • And put the output genes.csv somewhere web accessible, I think that's all Caltech would need.
      • I don't know much more about it, other than Matt Russell told me to run that (we went back and forth a bunch of times before it worked).
      • Here's a copy of the .jar file
      • Also
        • export WB_DB_URI="datomic:ddb://us-east-1/WSNames/wormbase"
      • And .jar was originally at
        • aws s3 cp s3://wormbase/names/exporter/wb-names-export.jar wb-names-export.jar
    • Thanks, JC! We can make that work from one of our persistent instances. I will look when I return from vacation.
  • AWS credentials leaked
    • AWS has become very strict; locked our account
    • Todd H has fully shut down nefarious resources across the globe; still had to petition AWS
    • Bottom line: We will use a more restrictive set of credentials for a WB AWS superuser; everyone else has access but with more restricted permissions
    • If anyone has permission problems accessing AWS, let Todd know
    • Todd retired some outdated credentials (10+ years old); it is good practice to regularly change credentials
    • Google accounts also affected; strongly encourage using 2-factor authentication

Tazendra migration

  • Caltech curation/Postgres server is migrating to AWS
  • Everything is being dockerized and moved to AWS
    • Valerio: we are using names from a domain I setup with 'textpresso-central'; would be good to setup names under the wormbase.org domain
    • Valerio will specify to Todd where they should point
  • Will eventually need to make changes once the switch is official

WormBase ParaSite 18 has been released

single-cell data on staging

  • Paulo has implemented the latest single cell expression data on staging for the WS288 release
Retrieved from "https://wiki.wormbase.org/index.php?title=WBConfCall_2023.04.20-Agenda_and_Minutes&oldid=37867"