Dec 8th, 2015 Ontology Working Group Meeting: Difference between revisions

From Planteome.org
Jump to navigation Jump to search
No edit summary
 
(28 intermediate revisions by the same user not shown)
Line 3: Line 3:
Time: 8:15am PST (GMT-8)
Time: 8:15am PST (GMT-8)
* Connection details:  '''Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/600315375'''
* Connection details:  '''Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/600315375'''
* Attendees: LC, AM, PJ, DWS, BS
* [[File:Ontology WG Meeting 12-08-2015 audio.m4a|thumb|Ontology WG Meeting 12-08-2015 audio]]
* [[File:Ontology WG Meeting 12-08-2015 video.mp4|thumb|Ontology WG Meeting 12-08-2015 video]]


* Attendees: LC, AM, PJ, DWS, BS
==Updates==
==Updates==
-DWS is working on a plant anatomy glossary- images can be linked to the PO terms in the digital version.
- DWS is working on a plant anatomy glossary- images can be linked to the PO terms in the digital version.
Main interest in terms for spines, prickles, stipules,  
Main interest in terms for spines, prickles, stipules,  
''LC: should use the purl rather than the PO webpage or planteome page.''
''LC: should use the purl rather than the PO webpage or planteome page.''
Line 25: Line 29:
* From BS: Related effort: [https://www.force11.org/ Force 11]/Force 2016- Open Science Initiative (Melissa Haendel, Maryanne Martone)
* From BS: Related effort: [https://www.force11.org/ Force 11]/Force 2016- Open Science Initiative (Melissa Haendel, Maryanne Martone)


== Mapping of TO terms to trait dictionaries for lentil, soybean, cassava, rice: Overview of progress ==
== Mapping of TO terms to trait dictionaries for lentil, soybean, cassava, rice: ==
   
[[File:Visualize.png|thumbnail|lentil plant height on the TO hierarchy]]
 
* Overview of Progress: CO curators are used to using the spreadsheets
* Afola requested access so he could fill in the mappings to cassava
* Link to mapping spreadsheet: [https://docs.google.com/spreadsheets/d/1iP6l_PEc0ESLHNpimkjJWFI-aezfJwQuNYZEzCLawJA/edit?usp=sharing CO Mappings to TO]
* If you want to edit it, you have to request access
* Goal- matching to ref Trait Ontology  Terms from the more granular CO terms. Can be an exact match or a match to a higher level
** Can create a composite tree as shown in the graphic for lentil plant height.  We would like to avoid massive term inflation.
** Idea is to prefix the CO term with the species name e.g. lentil plant height (this can be added programmatically?)
* The TO terms would be linked to the PO and PATO terms through the equivalence axioms (formerly called cross products)
 
*From PJ: example of tedril types in lentil:
** See image: [http://www.prota4u.org/plantphotos/Lens%20culinaris%201.gif lentil tendrils image]- "This is a leaflet tendril." 
 
''It looks like a rachis tendril to me (PO:0025366)??''  In lentil his tendril has different "types"- simple, dichotomous, bristle type.
 
* Notes from Email discussion (see: [https://github.com/Planteome/plant-trait-ontology/issues/351 #351] and [https://github.com/Planteome/plant-trait-ontology/issues/350 #350]
 
"PJ: It is a modification of a normal plant structure (7 such types and even more to be added -->inflorescence tendril, inflorescence branch tendril, etc.)"
[http://harvardmagazine.com/2013/03/cucumber-coil-conundrum]
 
"In Karthika's case, if the species is ''lens culinaris'' it is a 'leaf rachis tendril' based on 'Leaves alternate, pinnately compound, with 5–16 leaflets; rachis (1–)2.5–3.5(–5) cm long, usually ending in a tendril or bristle' "
[http://www.prota4u.org/protav8.asp?en=1&p=Lens+culinaris]; [http://www.prota4u.org/plantphotos/Lens%20culinaris%201.gif]
* There is a known tendril-less gene from lentil  [http://www.uniprot.org/uniprot/B6DEF5]  and also in Peas: [http://www.plantcell.org/content/21/2/420.abstract]
* We need more types of tendrils???
 
== Discussion about modified variables with scale info ==
- MAL has been working on implementing the suggestion from PJ to add additional variable ids which are specific to each of the categorical scales. That way there is a unique id to assign to a data annotation.  From this you can determine the trait, method, and the scale.
 
- ID scale-  each scale has a specific ID, these IDs are not linked to TO, but to each specific CO-variables
 
- color:  can be measured using a color wheel, or subjective 1-10, or whatever breeders have used.
 
- The CO curators/breeders have their own system for quality control, somewhat subjective.  Comparison within the crop species it is fine, but would be difficult to compare across species.  At that point, we should just compare at the level of the TO trait term.
 
== Trait Ontology Revisions  ==
===Proposed Revisions to Biochemical Traits===
* Issue exists that we have a number of traits that are duplicated for the content and concentration, and many others that are just on concentration
- link to google doc: [https://docs.google.com/document/d/1oPJ-U86so9H621oRMpSO9ztOXXbH-v8cX9dKuZuIn3c/edit?usp=sharing Biochemical Traits]
** concentration = amount/ VOLUME
** content = amount/MASS
* Reference about content vs concentration: http://www.ncbi.nlm.nih.gov/pubmed/23894859
* 'copper content'
** 'copper concentration'
 
* 'iron content'
** 'iron concentration'
 
* 'manganese content'
** 'manganese concentration'


* 'potassium content'
** shoot potassium content
*** potassium concentration


* 'phosphorus content'
** 'phosphorus concentration'


From PJ: [http://www.prota4u.org/plantphotos/Lens%20culinaris%201.gif lentil tendrils image]
* 'sodium content'
** 'sodium concentration'


== Discussion about modified variables with scale info ==
* 'zinc content'
** 'zinc concentration'




* discussion of content vs. concentration
- should we merge the content and concentration terms? Or put all concentration terms as subclasses of content?


* from PATO
- content is a synonym of composition
- relying on the definitions from PATO of these terms
- content
- concentration
- method




== Trait Ontology changes ==


* upper level: biochemical traits


* BS: suggestion to include all biochem traits in the morphology branch as properties of the structure


== Planteome Curator Workshop at PAG ==
== Planteome Curator Workshop at PAG ==

Latest revision as of 20:46, 16 December 2015

Planteome Ontology WG Zoom Meeting:

Date: Tuesday Dec. 8th, 2015 Time: 8:15am PST (GMT-8)


Updates

- DWS is working on a plant anatomy glossary- images can be linked to the PO terms in the digital version. Main interest in terms for spines, prickles, stipules, LC: should use the purl rather than the PO webpage or planteome page.

- DWS recently attended a meeting with the Center for Open Science, Univ. of Virginia

  • making experiments repeatable in ecology and psychology, what kind of metadata needs to be captured- using ontologies can help
    • e.g. problem of repeatibility
  • BS: developing an ontology for biological and clinical statistics- make pipeline repeatable
  • in Evolutionary biology- all the steps are defined and documented
  • From BS: Related effort: Force 11/Force 2016- Open Science Initiative (Melissa Haendel, Maryanne Martone)

Mapping of TO terms to trait dictionaries for lentil, soybean, cassava, rice:

lentil plant height on the TO hierarchy
  • Overview of Progress: CO curators are used to using the spreadsheets
  • Afola requested access so he could fill in the mappings to cassava
  • Link to mapping spreadsheet: CO Mappings to TO
  • If you want to edit it, you have to request access
  • Goal- matching to ref Trait Ontology Terms from the more granular CO terms. Can be an exact match or a match to a higher level
    • Can create a composite tree as shown in the graphic for lentil plant height. We would like to avoid massive term inflation.
    • Idea is to prefix the CO term with the species name e.g. lentil plant height (this can be added programmatically?)
  • The TO terms would be linked to the PO and PATO terms through the equivalence axioms (formerly called cross products)
  • From PJ: example of tedril types in lentil:

It looks like a rachis tendril to me (PO:0025366)?? In lentil his tendril has different "types"- simple, dichotomous, bristle type.

  • Notes from Email discussion (see: #351 and #350

"PJ: It is a modification of a normal plant structure (7 such types and even more to be added -->inflorescence tendril, inflorescence branch tendril, etc.)" [1]

"In Karthika's case, if the species is lens culinaris it is a 'leaf rachis tendril' based on 'Leaves alternate, pinnately compound, with 5–16 leaflets; rachis (1–)2.5–3.5(–5) cm long, usually ending in a tendril or bristle' " [2]; [3]

  • There is a known tendril-less gene from lentil [4] and also in Peas: [5]
  • We need more types of tendrils???

Discussion about modified variables with scale info

- MAL has been working on implementing the suggestion from PJ to add additional variable ids which are specific to each of the categorical scales. That way there is a unique id to assign to a data annotation. From this you can determine the trait, method, and the scale.

- ID scale- each scale has a specific ID, these IDs are not linked to TO, but to each specific CO-variables

- color: can be measured using a color wheel, or subjective 1-10, or whatever breeders have used.

- The CO curators/breeders have their own system for quality control, somewhat subjective. Comparison within the crop species it is fine, but would be difficult to compare across species. At that point, we should just compare at the level of the TO trait term.

Trait Ontology Revisions

Proposed Revisions to Biochemical Traits

  • Issue exists that we have a number of traits that are duplicated for the content and concentration, and many others that are just on concentration

- link to google doc: Biochemical Traits

  • 'iron content'
    • 'iron concentration'
  • 'manganese content'
    • 'manganese concentration'
  • 'potassium content'
    • shoot potassium content
      • potassium concentration
  • 'phosphorus content'
    • 'phosphorus concentration'
  • 'sodium content'
    • 'sodium concentration'
  • 'zinc content'
    • 'zinc concentration'


  • discussion of content vs. concentration

- should we merge the content and concentration terms? Or put all concentration terms as subclasses of content?

  • from PATO

- content is a synonym of composition - relying on the definitions from PATO of these terms - content - concentration - method


  • upper level: biochemical traits
  • BS: suggestion to include all biochem traits in the morphology branch as properties of the structure

Planteome Curator Workshop at PAG

Updates and Agenda (PAG 2016 Planteome-CO workshop)


Next Meeting: