Difference between revisions of "WormBase-Caltech Weekly Calls"
|Line 14:||Line 14:|
Revision as of 08:15, 5 March 2014
- 1 Previous Years
- 2 2014 Meetings
- 2.1 February 6, 2014
- 2.2 February 13, 2014
- 2.3 February 20, 2014
- 2.4 February 27, 2014
- 2.5 March 6, 2014
February 6, 2014
- Topic Curation
- WS243,244: Innate immunity and response to pathogens
- WS242: miRNA pathway
- WS241: UPR pathways
- using the caltech FTP site for storing topic page pathway images for the web team. Where is it, what information do we want to include, can we use this to track history of pathways?
- model change to put topics in hierarchical structure.
- File cabinets in hallway?
- Variations that have lost a public_name due to Million Mutation processing in WS240.
- What to do with ones that have phenotypes attached, should they have their public_names restored, so we can see them in our OA dropdown lists?
> WBVar00249046 former public-name sy5417 > WBVar00275318 former public-name y87 > WBVar00275321 former public_name y95 > WBVar00000714 former public_names cb107, cbs107 > WBVar00000718 former public_names cb111, cbs111 > WBVar00000716 former public_names cb109, cbs109 > WBVar00000720 former public_names cb113, cbs113 > e2518 is an other_name of WBVar0014462 > g320 is an other_name of WBVar00017601
- Models meeting in Toronto March 10-11
Gene Association files
- Life stage-gene associations
- Capturing relations (e.g. "involved_in") in GAF; how to do that for life stage?
- What search tool do we plan for e.g. "what genes are expressed in a particular neuron AND during dauer larval stage?"
we can use WormMine. Example query: http://www.wormbase.org/tools/wormmine/template.do?name=Anatomy_LS_Genes&scope=all
- Expression data model needs updating to capture BOTH place AND time of gene expression
Pathway images on topic pages
- Do we store/supply a flat image file for the web team?
- Store image files on the FTP site? Canopus, nightly mirroring
- Release-specific directories/images? OICR needs to know and handle
- Daniela's images are collected cumulatively; Wen's are overwritten
- May be an issue embedding iFrame interactive viewer
Variations losing public names
- Million Mutation project, variations getting renamed
- Distinguishing natural variants vs. lab-induced alleles
- Merging event (natural variants at same location merged), existing names become "Other_name"s
- Can we have a mechanism in place to decide one Public_name if there are multiple? Temporal precedence?
- First published name can be official Public_name?
- decided to leave things as they are
February 13, 2014
- Infection model
- Reminder: Alex Pico, Mary Ann Tuli, Xiaodong visiting Caltech Feb 24th
- Sequence Feature interactors
- Viewing list of topics in OA
- Infection model discussion--Advantages/disadvantages to curating to the level of strain for pathogens like Bacteria, Fungi etc) (Ranjana)
- Do we create non-nematode Strain objects here at Caltech? (Ranjana)
- (Mary Ann) As far as I can see, no. But I don't see a reason why you can't if there is a need. We will simply need to make sure you know the right curation format, which for non-nematode strains is a bit vague. I am talking to Aric shortly about the Db11 strain you sent me earlier this week.
- Things to consider:
- Creation of "?Infection" class
- Species-species interactions versus more specific strain-strain interactions
- Should we create topics that represent species-species interactions that can then be associated to relevant strain-strain interaction objects of an "?Infection" class?
- We would like to be able to create/manage new strains, nematode and non-nematode, for these annotations
- Strain OA?
- Feb 24th with Alex Pico
- Comments and requests for WikiPathways can be posted on Wiki page
- Alex Pico will present updates to viewer/editor (new JS viewer/editor)
Topic list in OA
- Xiaodong requesting Topic list viewer
- Best to just have a browser window/tab open with topics listed, either in Process Term OA or in the Curation Status form
- Some interaction objects currently refer to multiple sequence feature objects in addition to a regulator gene and a regulated gene; this results in "hairballs" in interactions, Cytoscape view
- We probably want a separate widget/viewer for gene/protein-sequence feature interactions involving a GBrowse window, displaying the interacting sequence feature in the context of the genome (to infer relevant genes, for example)
- Need to consider options for display and annotation (e.g. should we only have binary interactions?)
February 20, 2014
- Pathways in general and WikiPathways
To plan for Alex's visit, it would be good to solidify our wants and needs for process and pathway visualization. Please add specific questions for Alex to the WikiPathways_Questions_&_Comments page.
Some global discussion points for our in-house Caltech meeting:
- what do we want our pathways to show?
- how do we want pathways displayed?
- Schedule your meeting with Mary Ann and or Alex. Either add yourself to the Calendar or let Karen know when you have time to chat.
Alex Pico's visit for WikiPathways
- WormBase calendar; add in when you want to speak to Alex
- Alex will present, have general discussion in Kerckhoff 024 at 10am on Monday
- Chris, Karen, et al meet with Alex
- Alex leaving at 2:30pm
- Web meeting 1:30pm (Google Hangout); Alex talk to web team to integrate WikiPathway images, etc.
- Integrating WormBase data to automatically inform construction of pathways?
- We can try to setup a live stream (Google Hangout?) of Alex's talk for Kimberly and Daniela
Mary Ann's visit
- Meet with Paul at 9am
- Alex Pico's talk at 10am
- 10:45am - 12pm, Mary Ann can circulate around curators
- Chris and Karen can meet after 2:30pm or Tuesday
- Tuesday morning Skype with Kimberly
- Daniela can talk with Mary Ann on Europe time
- Arrive ~2pm tomorrow (Friday 21st)
- Dinner 21st, 22nd, 23rd?
- ?Construct model intro discussion Monday afternoon?
- Tuesday morning ?Construct model call (w/Daniela)
LEGO tool prototype
- Paul Thomas made a new LEGO prototype tool
- Allows visual GUI-based LEGO annotations
- Example link: http://go-genkisugi.rhcloud.com/seed/model/gomodel:wb-GO_0043053
- Base link: http://go-genkisugi.rhcloud.com/
- See also: File:PThomaslego-Whitepaper-2010-03.pdf
Paul met with ZFIN
- Interested in human disease relevance & pathways
Human Disease relevance
- FlyBase should have data out in March
- Ranjana et al working on paper
- OMIM connnections
Odd OA behavior
- Karen pointed out (Chris confirms) strange behavior where data persists in fields that should be empty
- Browser specific? Occurs on both Firefox and Chrome
February 27, 2014
- Changes to ?Construct and ?Transgene models
- Sequence feature (?Feature) curation
- Mapping constructs to the genome, when not sequence features, i.e. artificial constructs used for expression. Gary W suggests to capture the info in a DNA_text field as it happens for RNAi. Who should pull up the sequence? Gary says:
I would expect that someone at Caltech could write a simple script to pull out this sequence for you, given the input coordinates, a GFF file containing CDS objects and a genome sequence. If not, we could knock one up for you. Example: -4770 to +3592 (relative to the gene translation start)
- Topic curation
- Innate immune response and defense response terms in GO (Kimberly)
- GO annotations for many of the genes involved in innate immune response have been annotated to both innate immune response and to a defense response term, such as 'defense response to Gram negative bacterium'. However, for a number of C. elegans genes, these two annotations are really referring to the same process.
- I'd like to propose new GO terms that have the specificity of the defense response terms wrt the specific type of pathogen, but are child terms of innate immune response. Then, where appropriate, we'd consolidate our annotations onto one new term, for example, 'innate immune response to Gram negative bacterium'.
- Also, I'd like to propose a revision to the current definition of 'innate immune response' in GO:
- Innate immune response and defense response terms in GO (Kimberly)
Current definition: Innate immune responses are defense responses mediated by germline encoded components that directly recognize components of potential pathogens.
Proposed new definition: A protective response induced in an organism by the presence of a pathogen (e.g., bacterial, viral, microsporidial). The innate immune response is mediated by molecules that directly recognize pathogen components and by signaling pathways that coordinate the subsequent defense response.
?Construct & ?Transgene
- Engineered alleles need an annotation method
- Crispr, Mos Sci, etc.
- Inserted transgenes (engineered (e.g. "Si") or integrated ("Is")), modified genes
- We need ?Variations and ?Transgenes to have a common link via ?Construct (and identity to one another)
- If there are known flanking sequences (unique insertion site in the genome) for any transgene, it will also become an allele (?Variation) object
- Mary Ann will create the ?Variation objects for ?Transgenes that meet the criteria
- Consequences for ?Expr_pattern
- ?Transgenes are used in expression pattern curation
- ?Construct can be assigned an ?Expr_pattern
- ?Feature objects (sequence features) can be assigned to a ?Construct
- ?Feature objects have been sent to Mary Ann and Gary Williams by curators (Xiaodong, Daniela, etc.) to create and approve the ?Features for a paper
- We will want a faster, more efficient pipeline for requesting ?Feature objects