Manage Ontology

BMS 4.0 Manual

Introduction

The Breeding Management System’s crop ontology management tool uses a structured vocabulary to allow descriptors to act as variables in database queries and statistical analysis. Core crops come preloaded with a set of variables, or ontology terms, recommended by their community of practice. Preloaded ontologies are fully customizable. The BMS also provides a generic crop database to support any breeding program. Consistent variable definitions are essential for data sharing and collaboration. As you consider ontology customization, we recommend that you coordinate with collaborators and your crop community to similarly define terms to facilitate data sharing and meta analysis. See more about Crop Ontology Curation.

  • Select Manage Ontologies from the information management menu to browse, edit, or add ontology terms.
r
Maize Ontology Browser

 

Default BMS Trait Dictionaries in Excel File Format

Variable Identification

Each variable name is identified by 3 details. Only one ontology term can be associated with a combination of unique details.

  • Property: What is measured
  • Method: How the variable is measured
  • Scale: Categorical scale or unit of measure for the variable

Edit Existing Variable

  • Select Edit to change variable details. Once a variable is already in use the database some fields cannot be edited. Editable fields include description and variable type.

Details about the variable name, ACCNO

The variable can be tagged with additional "variable types" to facilitate search & browse functions.

Default Maize Variable, ACCNO, given the alias, Accession_Number. Accession_Number is an assigned germplasm descriptor, an independent variable in statistical analysis.

Variable Name

Variable name is generally an abbreviation or acronym describing the variable. Variable names in the default BMS crop ontology will be recognizable between all members of a crop community. Variable name is the ontology term specified in BMS data collection files and the output of statistical analyses. Variable names cannot contain spaces, numbers, or special characters to ensure compatibility with Breeding View (BV) statistical engine. If variable names exceed 15 characters truncation of variable name will occur in in the graphical output of BV. Changing a variable name does not change the functional identity (property, method, and scale) of the variable.

Variable Alias

Although it is important for crop communities to similarly define traits and share a common vocabulary, the BMS provides the ability to create custom aliases for ontology terms, allowing breeders to communicate variables in familiar language within their breeding programs. Once an alias has been set, the custom alias will replace the variable name in the user interface, data collection files, and the output of statistical analyses. Giving a variable an alias does not change the functional identity (property, method, and scale) of the variable. Variables which are given aliases are automatically added to your program favorites.

Variable Definition

Variable definition is a plain language description of the variable. Editing or expanding a variable's definition does not change the functional identity (property, method, and scale) of the variable.

Add New Variable

  • Select Add New. Choose Variable.

Name & Describe Variable

  • Add a new biochemical trait to the the maize ontology.
    • Name: B_Carotene.
    • Describe: Beta-carotene content, mg/kg dry matter (mg/kg d.m.)

Add New Property

Property describes what is being measured. The default maize ontology has no property describing beta-carotene content, so a new one must be added.

  • Select Add a New Property.

  • Name the property, Beta-Carotene. Tag the classes, Quality and Physiological. Tagging the property helps the ontology search engine provide selections of relevant traits. Save.

Add New Method

Method describes how the property is measured. Beta-carotene content is measured by high performance liquid chromatography, which is not included in the default maize ontology.

  • Select Add a New Method. Name and describe the method, HPLC, high performance liquid chromatography.

Add New Scale

Beta-carotene content is a numeric variable measured in units of milligram per kilogram of dry matter.

  • Name: mg/kg d.m.
  • Description: milligram per kilogram of dry matter
  • Data Type: Numeric
  • Valid Range: Numeric variables offer the option of setting a range of valid values for quality control. Data outside of this range is flagged as a possible error during data import.

Variable Type

Variable type differentiates traits, or phenotype, from different independent variables. Variable type determines which variables are presented in different BMS tools.

  • Trait
  • Analysis
  • Environment Detail
  • Germplasm Descriptor
  • Nursery Condition
  • Selection Method
  • Study Detail
  • Treatment Factor
  • Study Detail
  • Like most terms in the crop ontology, beta-carotene, is a trait or phenotype. Select Trait from the drop down menu of different variable types. Save.

Duplicate Variable Error

The BMS will not allow you to add a new variable with an identical combination of property, method, and scale to other variables.

Search Ontology

Searching the maize database for "beta" now reveals the newly added B_Carotene variable.

Favorite Terms

Users can select terms to add to their program's favorites list.

  • Scroll the cursor over an ontology term to reveal a blue star. Select the blue star to save the associate term in the program favorites. Alternatively, giving a term an alias (see above) automatically results in that term joining the favorites list.

  •  Several places in the BMS allow users to narrow their search by program favorites, making ontology term selection much easier. Select the star from the favorites list to remove the term.

Sub-Samples & Repeated Measures

BMS version 4.0 does not allow a single ontology term to be associated with sub-samples and repeated measures. (Expect see this flexibility in future versions of the BMS.) Take acid abscisic acid concentration for example. A breeder may want to measure this trait on multiple maize ears per experimental plot or at multiple time points. The current solution is to create multiple ontology terms for each measurement.

  • New ontology terms need to be created by adding a sub-sampling or time series code to both the variable name and method. Ideally subsampling and time series nomeclature should be standardized for each crop.
Sub-Sampling Unit Code Examples
Sub-sample s s1,s2...
Quadrat qdt qdt1, qdt2...
Plant plt plt1, plt2...
Leaf lf lf1, lf2...
Stem st st1, st2...
Tiller tlr tlr1, tlr2...
Branch bch bch1, bch2...
Flower flr flr1, flr2...
Fruit frt frt1, frt2...
Spike spk spk1, spk2...
Ear ear ear1, ear2...
Pod pod

pod1, pod2...

Seed sd sd1, sd2...
Grain grn grn1, grn2...
Root rt rt1, rt2...
Tuber tbr tbr1,tbr2...

 

Time Code Type Time Code Examples
Cereals Zadoks growth/development stages GS(00-99) or DS(00-99)
  • DS65 = heading
  • DS75 = anthesis
  • DS87 = physiological maturity
Maize growth/development stages

Vegetative: VE, V1-Vn, VT

Reproductive: R1-R6

  • VE = emergence
  • V3 = third Leaf
  • VT = tasseling
  • R1 = silking
  • R6 = physiological maturity
Soybean growth/development stages VE, VC, V1-Vn, R1-R8
  • VE = emergence
  • V4 = second trifoliate
  • R2 = full flowering
  • R8 = full maturity
Generic Codes
  • hd = heading
  • flw = flowering
  • ant = anthesis
  • vg = vegetative period
  • prehd = pre-heading
  • gf = grain filling
  • mat = maturity
 
Days after emergence dae 45dae, 65dae....
Days after sowing das 45das, 65das...
Days after planting dap 45dap, 65dap...
Weeks after planting wap 1wap, 2wap...
Months after planting map 1map, 2map
Date yyyymmdd 20150315
Date+hr+min yyyymmdd 201503151135
Time t t1, t2...
  • First create a new method with the subsampling code.
  • Second create a new trait with the subsampling code.
  • Repeat. Create a new ontology term for each required sub-sample or time series measurement.

Related Tutorial

Maize: Import Trial Data