Difference between revisions of "WormBase-Caltech Weekly Calls"

From WormBaseWiki
Jump to navigationJump to search
(37 intermediate revisions by 4 users not shown)
Line 41: Line 41:
 
[[WormBase-Caltech_Weekly_Calls_August_2019|August]]
 
[[WormBase-Caltech_Weekly_Calls_August_2019|August]]
  
 +
[[WormBase-Caltech_Weekly_Calls_September_2019|September]]
  
== September 12, 2019 ==
 
  
=== Update on SVM pipeline ===
+
== October 3, 2019 ==
* New SVM pipeline: more analysis and more parameter tuning
 
* avoiding precision (and F-value) as a measure (dependent on ratio of positives and negatives in test set)
 
* For example shown, "dumb" machine starts out with precision above 0.6
 
* G-value (Michael's invention); does not depend on distribution of sets
 
* Applied to various data types
 
* Analysis: 10-fold cross validation
 
** Randomly select 10% pos and neg (without replacement) and repeat until all papers sampled
 
* F-value changes over different p/n values; G-value does not (essentially flat)
 
* Area Under the Curve (AUC): probability that a random positive scores higher than random negative
 
* AUC values for many WB data types upper 80%'s into 90%'s
 
* Ranjana: How many papers for a good training set? Michael: we don't know yet
 
* Can't reproduce old training sets (for old SVM); provide Michael better training sets if you want improved SVM
 
* If SVM still not good enough, Michael will work on deep neural networks (Tensor Flow)
 
* Michael can provide training sets he has used recently
 
  
=== Clarifying definitions of "defective" and "deficient" for phenotypes ===
+
=== SObA comparison graphs ===
* WB phenotype ontology has many "variant/abnormal" terms and distinct subclass terms for "defective/deficient"
+
* Raymond and Juancarlos have worked on a SObA-graph based comparison tool to compare two genes for ontology-based annotations
* Have tried to create a logical definition pattern for these terms, but the vagueness of the meaning of "defective" and how it is distinct from "abnormal" has stalled the process
+
* [http://wobr2.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=Gene+Pair+to+SObA+Graph Prototype 1]
* What do we mean exactly by "defective" and how, specifically, is this distinct from "abnormal"?
+
** [http://wobr2.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=annotSummaryCytoscape&filterForLcaFlag=1&filterLongestFlag=1&showControlsFlag=0&datatype=phenotype&geneOneValue=lin-3%20(Caenorhabditis%20elegans,%20WB:WBGene00002992,%20-,%20F36H1.4)&autocompleteValue=let-23%20(Caenorhabditis%20elegans,%20WB:WBGene00002299,%20-,%20ZK1067.1 Example comparison between lin-3 and let-23]
* Definitions include meanings or words:
+
* [http://wobr1.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=Gene+Pair+to+SObA+Graph Prototype 2]
** "Variations in the ability"
+
** [http://wobr1.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=annotSummaryCytoscape&filterForLcaFlag=1&filterLongestFlag=1&showControlsFlag=0&datatype=phenotype&geneOneValue=lin-3%20(Caenorhabditis%20elegans,%20WB:WBGene00002992,%20-,%20F36H1.4)&autocompleteValue=let-23%20(Caenorhabditis%20elegans,%20WB:WBGene00002299,%20-,%20ZK1067.1) Example comparison between lin-3 and let-23]
** "aberrant"
+
* What information does a user most care about?
** "defect"
+
# What terms (nodes) are annotated to gene 1 and what terms to gene 2
** "defective"
+
# For a given term, what is the relative number of annotations between gene 1 and gene 2.
** "defects"
+
# For a given node, what is the relative number of annotations each gene has to the total annotations of that gene.
** "deficiency"
+
* # 3 is actually what we applied to size the nodes in the single-gene version of SObA. Thus, not surprisingly, I think it is important.
** "deficient"
+
* Generally people like Prototype 2 as a default view; we could possibly have a toggle to see the other view
** "disrupted"
+
* In either case users need a good legend and/or documentation
** "impaired"
+
* Jae, it would be good if a user could specifically highlight nodes specific to each gene and gray-out or de-emphasize the common nodes
** "incompetent"
 
** "ineffective"
 
** "perturbation that disrupts"
 
** Failure to execute the characteristic response = abnormal?
 
** abnormal
 
** abnormality leading to specific outcomes
 
** fail to exhibit the same taxis behavior = abnormal?
 
** failure
 
** failure OR delayed
 
** failure, slower OR late
 
** failure/abnormal
 
** reduced
 
** slower
 
  
=== Citace upload ===
+
=== Germ line discussion ===
** Tuesday, Sep 24th
+
* Currently, the anatomy ontology has "germ line" as a type of "Cell" and a type of "Tissue", and "germ cell" as a type of "germ line"
 +
* Chris would like to (1) remove "germ line" from under "Cell" and leave it under "Tissue" and (2) move "germ cell" out from under "germ line" and place directly under "Cell"
 +
** [https://github.com/obophenotype/c-elegans-gross-anatomy-ontology/pull/23 Made pull request]
 +
* Chris will update pull request to include a change to move "germline precursor cell" out from under "germ line" and place it under "Cell" (done)
  
=== Strain to ID mapping ===
+
=== Script to remove blank entries from Postgres ===
* Waiting on Hinxton to send strain ID mapping file?
+
* Chris stumbled across several entries in the OA that were blank (empty strings) or consisted of only whitespace, some of which were causing errors upon upload to ACEDB
* Hopefully we can all get that well before the upload deadline
+
* Juancarlos has written a script to look for all such entries; 66 tables have them on sandbox (likely same on live OA)
* Will do global replacement at time of citace upload (at least for now)
+
* Does anyone object to removing these entries throughout Postgres?
 +
* Juancarlos will remove all the empty fields identified by his script
  
=== New name server ===
 
* When will this officially go live?
 
* Will we now be able to request strain IDs through the server? Yes
 
  
=== SObA Graphs ===
+
== October 10, 2019 ==
* New graphs now live on site (Expression, Gene Ontology, Human Diseases, Phenotypes)
 
* A lot of whitespace padding above and below graph; maybe trim? trimming vertically would ultimately limit the view pane when user wants to zoom in, so we should leave as is for now
 
* Diff tool: Raymond and Juancarlos created a prototype diff tool (for comparing two genes, for example)
 
** Paul: compared two genes that should be very similar, but there are a lot of differences; may reflect annotation coverage rather than biology
 
  
 +
=== Biocuration 2020 ===
 +
* Held in Bar Harbor, Maine (organized by JAX, including MGI's Sue Bello and Cindy Smith)
 +
* Dates: Sunday May 17th to Wednesday May 20th, 2020
 +
* Will have 3rd POTATO workshop
 +
* [https://www.jax.org/education-and-learning/education-calendar/2020/05-may/biocuration-2020-conference Meeting website]
 +
* Key Dates
 +
** October 31, 2019 - Paper Submission Deadline
 +
** January 24, 2020 - Abstract  and Workshop Submission Deadline
 +
** March 6, 2020 - Notification of Acceptance
 +
** April 6, 2020 - Early Bird Registration Ends
 +
** May 8, 2020 - Registration Deadline
 +
* Academic ISB Member, early bird registration fee is $250
 +
* Author First Pass form paper, submitting to Database, biocuration issue (managed by biocuration group); authors have an opportunity to present at Biocuration conference
  
== September 19, 2019 ==
+
=== ICBO 2020 ===
 +
* International Conference on Biomedical Ontologies
 +
* [https://icbo2020.inf.unibz.it/ Meeting website]
 +
* Held in Bozen-Bolzano, Italy
 +
* 16 - 19 September 2020
  
=== Strains ===
+
=== SObA comparison tool ===
* Need to wait for new strain IDs from Hinxton before running dumping scripts
+
* [http://wobr2.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=Gene+Pair+to+SObA+Graph Prototype #1] updated
* Don't edit multi-ontology strain fields in OA for now!
 
* Juancarlos will map free text and ontology-name strain entries to strain IDs once we have the complete mapping file
 
* "Requested strain" field in Disease OA; not dumped, so don't need to worry about right now
 
  
=== Alliance literature curation ===
+
=== Textpresso derived paper connections ===
* Working group will be formed soon
+
* For example for strains and constructs, maybe anatomy terms?
* Will work out general common pipelines for literature curation
+
* May want to flag Textpresso predictions (as opposed to manually connected)
 +
* Couple of options:
 +
** 1) At time of build, populate the papers (in ACEDB/Datomic) into a 'Putative_reference' tag and display in a distinct 'Putative references' widget
 +
** 2) Not part of database build, but make associations live (using RESTful API to link out to Textpresso and submit search with URL) using Textpresso with links to Textpresso and Textpresso results, giving users chance to see context of matches in sentences at the Textpresso site
 +
*** A link to Textpresso could be done regardless of other approaches; low-hanging fruit?
 +
*** Do a diff so that Textpresso pulls up only additional papers (not already associated)?
 +
** 3) Could populate WB page with connections made through a Textpresso API call (could cache results? maybe, but might as well choose 1st option?)
 +
* Transgene pipeline:
 +
** Arun wrote script, matching transgene names (using regex; Is and Si transgenes) to papers, automatically populate OA
 +
** Another script, captures Ex transgenes as well, automatically connects to construct objects
 +
** WB only displays verified papers; unverified (predicted) associations are not dumped
 +
* Could integrate author verification as part of AFP pipeline, even for older papers? Would we want to re-request AFP results for authors that have already replied in the past? Probably not
 +
* Could embed AFP predictions in WB display with link to AFP form for authors (and others?) to verify, via logged-in users? Or via a validation token sent via email?
 +
* Chris will make GitHub ticket to ask WB web team to add a link to Textpresso search from References widget on respective page; will require a Textpresso URL constructor
 +
* Can apply to: genes, transgenes, constructs, strains, alleles, AFP-vetted entities
  
=== SObA Graph relations ===
 
* Currently only integrating over "is a", "part of" and "regulates"
 
* Maybe we could provide users an option to specify which relations to include, or maybe just exclude "regulates"
 
  
=== Author First Pass ===
+
== October 17, 2019 ==
* Putting together paper for AFP
 
* Reviewing all user input for paper
 
* Asking individual curators to check input
 
  
== September 26, 2019 ==
+
=== Alliance All Hands Face-to-Face ===
 +
* Flights: has everyone already booked? No, not yet
 +
* Any coordination of flights from Pasadena/LA?
 +
** Ranjana and Valerio got a direct flight from Burbank to Boston on Sunday for premeetings
  
=== Strains ===
+
=== SObA Comparison Tool ===
* Paul D generated WBStrains for the missing TransgeneOme objects
+
* http://wobr2.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=Gene+Pair+to+SObA+Graph
* Working on a pipeline to identify new TransgeneOme strains at each upload
+
* Prototype discussed last week, updated with feedback from prior discussions
* One TransgeneOme object had 2 strains. Possible solutions: dump 2 expression objects that differ only in the Strain or remove the UNIQUE tag in the data model
+
* Would this be a stand-alone tool discoverable under the Tools menu?
 +
** Possibly; could be a gene page widget, but may be out of place
 +
** Stand-alone tool probably makes more sense
 +
* Life stage graph doesn't specify expression pattern vs. expression cluster; pretty much only expression patterns (no clusters)
 +
 
 +
=== SObA ===
 +
* Raymond intending to share progress on SObA at December Alliance All-Hands Face-to-Face
 +
* For example, share GO SObA graph for other species
 +
* Will need to be dependent on a SOLR server with all species data
 +
** Raymond has run into problems trying to setup his own SOLR server
 +
** Raymond asked Seth Carbon if we could us GO server, but he prefers not
 +
** Appear to be software versioning issues, possible memory issues
 +
 
 +
=== GO meeting ===
 +
* Kimberly can give update on recent updates to GO from the recent GO meeting
 +
* Slides are shared online
 +
 
 +
=== "all stages Ce" life stage ===
 +
* Currently used to annotate that RNA was collected from, or a gene was observed to be expressed during, all C. elegans life stages
 +
* "all stages Ce" is currently the root node of the C. elegans branch, but needs to change to generic "C. elegans life stage"
 +
* Should we:
 +
** 1) Create a "C. elegans life span" or "C. elegans life cycle" term to represent the entire life span and annotate to that?
 +
*** Does this mean that, for example, a gene is expressed at some point during the life cycle or consistently throughout the entire life span?
 +
** 2) Annotate instead to, for example, "embryo Ce", "larva Ce", and "adult Ce" individually
 +
* Note: authors are often vague in their descriptions simply saying "during all stages" possibly stating a beginning and end of the range
 +
* Wen: not comfortable making a decision right now; want to discuss with Gary Williams and with other MOD members about how to handle large scale expression data
 +
* Daniela: will look through existing (old and new) expression pattern annotations made to "all stages Ce" to see if it would be reasonable to annotate each case individually to "embryo Ce", "larva Ce", and "adult Ce" individually
 +
 
 +
=== Gene class missing description ===
 +
* The gene class "aatf" has no description, so in the aatf-1 gene page Overview widget, the gene has empty parentheses next to the gene name where there should be a description of what "aatf" stands for (coming from the ?Gene_class description)
 +
* Jae or Ranjana will create a ticket and assign it to someone at Hinxton

Revision as of 17:12, 17 October 2019

Previous Years

2009 Meetings

2011 Meetings

2012 Meetings

2013 Meetings

2014 Meetings

2015 Meetings

2016 Meetings

2017 Meetings

2018 Meetings


GoToMeeting link: https://www.gotomeet.me/wormbase1


2019 Meetings

January

February

March

April

May

June

July

August

September


October 3, 2019

SObA comparison graphs

  1. What terms (nodes) are annotated to gene 1 and what terms to gene 2
  2. For a given term, what is the relative number of annotations between gene 1 and gene 2.
  3. For a given node, what is the relative number of annotations each gene has to the total annotations of that gene.
  • # 3 is actually what we applied to size the nodes in the single-gene version of SObA. Thus, not surprisingly, I think it is important.
  • Generally people like Prototype 2 as a default view; we could possibly have a toggle to see the other view
  • In either case users need a good legend and/or documentation
  • Jae, it would be good if a user could specifically highlight nodes specific to each gene and gray-out or de-emphasize the common nodes

Germ line discussion

  • Currently, the anatomy ontology has "germ line" as a type of "Cell" and a type of "Tissue", and "germ cell" as a type of "germ line"
  • Chris would like to (1) remove "germ line" from under "Cell" and leave it under "Tissue" and (2) move "germ cell" out from under "germ line" and place directly under "Cell"
  • Chris will update pull request to include a change to move "germline precursor cell" out from under "germ line" and place it under "Cell" (done)

Script to remove blank entries from Postgres

  • Chris stumbled across several entries in the OA that were blank (empty strings) or consisted of only whitespace, some of which were causing errors upon upload to ACEDB
  • Juancarlos has written a script to look for all such entries; 66 tables have them on sandbox (likely same on live OA)
  • Does anyone object to removing these entries throughout Postgres?
  • Juancarlos will remove all the empty fields identified by his script


October 10, 2019

Biocuration 2020

  • Held in Bar Harbor, Maine (organized by JAX, including MGI's Sue Bello and Cindy Smith)
  • Dates: Sunday May 17th to Wednesday May 20th, 2020
  • Will have 3rd POTATO workshop
  • Meeting website
  • Key Dates
    • October 31, 2019 - Paper Submission Deadline
    • January 24, 2020 - Abstract and Workshop Submission Deadline
    • March 6, 2020 - Notification of Acceptance
    • April 6, 2020 - Early Bird Registration Ends
    • May 8, 2020 - Registration Deadline
  • Academic ISB Member, early bird registration fee is $250
  • Author First Pass form paper, submitting to Database, biocuration issue (managed by biocuration group); authors have an opportunity to present at Biocuration conference

ICBO 2020

  • International Conference on Biomedical Ontologies
  • Meeting website
  • Held in Bozen-Bolzano, Italy
  • 16 - 19 September 2020

SObA comparison tool

Textpresso derived paper connections

  • For example for strains and constructs, maybe anatomy terms?
  • May want to flag Textpresso predictions (as opposed to manually connected)
  • Couple of options:
    • 1) At time of build, populate the papers (in ACEDB/Datomic) into a 'Putative_reference' tag and display in a distinct 'Putative references' widget
    • 2) Not part of database build, but make associations live (using RESTful API to link out to Textpresso and submit search with URL) using Textpresso with links to Textpresso and Textpresso results, giving users chance to see context of matches in sentences at the Textpresso site
      • A link to Textpresso could be done regardless of other approaches; low-hanging fruit?
      • Do a diff so that Textpresso pulls up only additional papers (not already associated)?
    • 3) Could populate WB page with connections made through a Textpresso API call (could cache results? maybe, but might as well choose 1st option?)
  • Transgene pipeline:
    • Arun wrote script, matching transgene names (using regex; Is and Si transgenes) to papers, automatically populate OA
    • Another script, captures Ex transgenes as well, automatically connects to construct objects
    • WB only displays verified papers; unverified (predicted) associations are not dumped
  • Could integrate author verification as part of AFP pipeline, even for older papers? Would we want to re-request AFP results for authors that have already replied in the past? Probably not
  • Could embed AFP predictions in WB display with link to AFP form for authors (and others?) to verify, via logged-in users? Or via a validation token sent via email?
  • Chris will make GitHub ticket to ask WB web team to add a link to Textpresso search from References widget on respective page; will require a Textpresso URL constructor
  • Can apply to: genes, transgenes, constructs, strains, alleles, AFP-vetted entities


October 17, 2019

Alliance All Hands Face-to-Face

  • Flights: has everyone already booked? No, not yet
  • Any coordination of flights from Pasadena/LA?
    • Ranjana and Valerio got a direct flight from Burbank to Boston on Sunday for premeetings

SObA Comparison Tool

  • http://wobr2.caltech.edu/~azurebrd/cgi-bin/soba_multi.cgi?action=Gene+Pair+to+SObA+Graph
  • Prototype discussed last week, updated with feedback from prior discussions
  • Would this be a stand-alone tool discoverable under the Tools menu?
    • Possibly; could be a gene page widget, but may be out of place
    • Stand-alone tool probably makes more sense
  • Life stage graph doesn't specify expression pattern vs. expression cluster; pretty much only expression patterns (no clusters)

SObA

  • Raymond intending to share progress on SObA at December Alliance All-Hands Face-to-Face
  • For example, share GO SObA graph for other species
  • Will need to be dependent on a SOLR server with all species data
    • Raymond has run into problems trying to setup his own SOLR server
    • Raymond asked Seth Carbon if we could us GO server, but he prefers not
    • Appear to be software versioning issues, possible memory issues

GO meeting

  • Kimberly can give update on recent updates to GO from the recent GO meeting
  • Slides are shared online

"all stages Ce" life stage

  • Currently used to annotate that RNA was collected from, or a gene was observed to be expressed during, all C. elegans life stages
  • "all stages Ce" is currently the root node of the C. elegans branch, but needs to change to generic "C. elegans life stage"
  • Should we:
    • 1) Create a "C. elegans life span" or "C. elegans life cycle" term to represent the entire life span and annotate to that?
      • Does this mean that, for example, a gene is expressed at some point during the life cycle or consistently throughout the entire life span?
    • 2) Annotate instead to, for example, "embryo Ce", "larva Ce", and "adult Ce" individually
  • Note: authors are often vague in their descriptions simply saying "during all stages" possibly stating a beginning and end of the range
  • Wen: not comfortable making a decision right now; want to discuss with Gary Williams and with other MOD members about how to handle large scale expression data
  • Daniela: will look through existing (old and new) expression pattern annotations made to "all stages Ce" to see if it would be reasonable to annotate each case individually to "embryo Ce", "larva Ce", and "adult Ce" individually

Gene class missing description

  • The gene class "aatf" has no description, so in the aatf-1 gene page Overview widget, the gene has empty parentheses next to the gene name where there should be a description of what "aatf" stands for (coming from the ?Gene_class description)
  • Jae or Ranjana will create a ticket and assign it to someone at Hinxton