Oct 17th, 2017 Ontology Working Group Meeting

Planteome Ontology WG Zoom Meeting

• Date: Tuesday Oct 17th, 2017
• Time: 8:15am PDT (GMT-7)

• Connection details:

Topic: Planteome Ontology WG Meeting Time: Oct 17, 2017 8:15 AM Pacific Time (US and Canada)

Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/609630127

• Meeting Recordings:
  • File:Ontology WG meeting 10-17-17 audio only.m4a
  • File:Ontology WG meeting video 10-17-17.mp4

Agenda and Notes

News:

• Newly published paper by the Wheat Data Interoperability Working group in F1000Research: https://f1000research.com/articles/6-1843/v1
• Revisions completed on the NAR Paper- resubmitted 10-13-17

Visit to BTI at Cornell next week

- LC and MAL will attend

• MAL:
  • RTB Program of CGIAR crops: root, tuber and banana- also includes cassava, sweet potato, potato, and yam
  • Fieldbook, need be able to add new traits and variables
  • Afola (cassava) would prefer to update the OBO or OWL file
  • Curators for banana, sweet potato, potato, yam would prefer to edit the TD- who is it?
  • Goal is to set up the synchronization between CO and GitHub, Need to be trained on Protege to update the file
  • Show Planteome and the mappings to the TO
  • TD related questions, find out what data is available
  • Breeding API- being widely implemented
  • Will work on get data for each crop, if available

• EA:
  • These are important 2017 Deliverables for RTB project:
  • Breeding Community of Practice (CoP) includes all the RTB crops
  • Breeding Super Products, for breeding align the BTI next-gen database for those crops, with the Breeding Management System (BMS)
  • Goal: everyone using the same workflows, variables, fieldbook etc
  • Breeding API- being widely implemented- http://docs.brapi.apiary.io/#

• PJ:
  • Are we generating a species-specific ontology graphs for each crop, with the upper TO hierarchy?
  • We should be creating one for each crop with species-specific tree
  • MAL: This can be easily done, hierarchy from the TO can be created in the Protege, each class is a subclass of the TO terms
  • This allows the curators to just maintain the TD and ref ontology tree can guide then
  • When the CO is converted to OBO file to do the mapping, it is a flat list, so the hierarchy is coming from the TO
  • Planteome APIs can also serve the CO terms, so they should be using them rather than APIs from multiple other sources
  • PJ will send an email about what he would like to see

• Can we access some publicly available data on the RTB crops?

- Annotated and cleaned data, not the raw data from the breeding programs - would like to get some for each of the crops- good example for TO paper - annotations are also served through the APIs

Paper for mapping and improvements to the TO

Marie will lead with LC (co-first authors?)

Details of the methods of mapping, peer-reviewed

Short and sweet paper on the initial 8-10 crops - perhaps for Current Plant Biology (now on PubMed)

EA: Marie needs to have a paper, focus on the mapping….

PJ: include discussions at BTI, about how the planteome is fostering the community standards, encouraging the community to adopt them and utilizing the datasets and keeping the ontologies up to date

Include revisions to TO and update, cover the design patterns

CM: what is the focus of the mapping paper, plant crop folks or bioinformatics people?

PJ: Should be both, go into a plant journal, get community involvement, LC and MAL will be in Cornell next week, working with Lukas Mueller’s group, banana and tuber crops,

Designing workflow for annotation and data collection

Second paper on the Samara work

CM, AM, LC with MAL- had discussions with the consultant Jorrit- see the gdoc: Planteome Term Matcher

Mapping the trait descriptors to the TO, sort of overlaps with what Marie has been doing, so we need to sort that out

Also covers the design patterns

Redoing the genome GO annotations for next release of the Planteome:

Get new proteome files and redo GO annotations

Gramene and PlantEnsemble are updating their genome and are out of sync with Uniprot, so would be better to get from Planteome

We should be the primary annotator for GO annotations for plants. We should be sharing GO annotations into GOC, the UniProt would pick up- only GO annotations (not PO, TO etc), also Plant Ensemble, Gramene and Plant Reactome- all would be in sync

Projections from Arabidopsis to other species- need to use something like taxon restraints to ensure that annotations are not projected in error

PJ, JE, AM- rechecking work flow for rice, adding semi-automated pipeline

Add confidence scores, transmembrane domains, chloroplast proteins, not utilized right now

Will move to doing it on all 65 species, Projection from Arabidopsis

Primary resource for the public editing interface being developed with private company which has been contracted.

Building database, registered users can edit, take responsibility of certain genes etc

Running: Interproscan, THMM, TargetP etc, rice is test case

nparanoid result, for projections

Will feed back filtered set of GO annotations into the GO database

GO’s QC pipeline should be integrated into the pipeline- should be run at CyVerse

Eric and/or Seth could show on next meeting? - Oct 31st

• CM:

New tool: Jenkins 2 for QC can be run, publishes report

Jenkins 2 demonstration at next meeting- Oct 31st

Annotations at GOA from Uniprot- Protein2GO

• PJ: Need to follow up on the Maize GO annotations from MaizeGDB

- from Lisa H? - https://github.com/Planteome/planteome-annotation-data/issues/1

Planning for Planteome renewal

• PJ: Renewal plan- submit by spring
• Renewal- community engagement, data analysis for community, biological focus, machine learning, smaller scale proposal, emphasis on data reuse, standards, not so much on ontology development, biology driven rather than resource driven
• Need to focus by biological questions: eg. plant traits link to env factors and some plant genes, use the APIs,
• What can the users get from the resource, machine learning
• Submit by spring, PGP may not exist down the road
  • Link to Google Folder: https://drive.google.com/drive/folders/0B71O1Y7eIKhsanJmUk9uaGEwMnM?usp=sharing

TO DO:

• Create folders for Planteome papers: https://drive.google.com/drive/folders/0B71O1Y7eIKhsQ3VtTDhqTkxRUnM?usp=sharing
  • TO paper with mapping to CO vocabs
  • Samara papers- see: https://docs.google.com/document/d/1p5PUqXnPQWlFYKHH5yJtbPpRcOnX0AsuGRj7M_2HxAk/edit
  • One paper on GRIN annotations, one on the stress ontology….?

High Throughput phenotyping talks on future calls

EA will contact Pascal, Francois

MAL visit to Corvallis?

- AM could use help with stress ontology, etc for the release PJ agreed, could also get CM to come or go to Berkeley When? Who will pay the travel costs?

Phenoscape Hackathon in Dec:

• https://blog.phenoscape.org/
• LC and AM planning to attend, JE and CM may attend?
• What other plant people going- who?
  • Lorena Endara will be attending, collaborator with Hong Cui
  • Others?
• Maybe Hilmar could come on a future call and show us an overview?
• there was an experimental plant phenoscape- what is plant community looking for
• Can we get a list of the attendees?

Items from the NAR Review:

• Question about the filters:

- Page 6 line 21: the fact that filters must be set before browsing the tree is a limitation of the current implementation that should indeed be documented, but on the web site, not on this paper. This latter one should remain fully relevant even after correction of this kind of little limitations. Same with line 33 with the space at the end trick.

• I think the reviewer misunderstood this...

- Page 7: the swagger include examples of human genes. it would be nice to have examples with plant genes. No impact on the paper though.

• • CM opened a tracker for it: https://github.com/Planteome/biolink-api/issues/3

* PJ: Need to update the Planteome website

Upcoming Events

• Phenoscape Hackathon in Dec

• Plant and Animal Genome meeting January 2018