v21 Graphical Queries
Graphical Filtering
Graphical filtering can be used to assist selection decisions. The graphical filtering tool was designed by the CassavaBase group at Cornell University and is an example of a BMS "plug-and-play" application using open source API calls defined by BrAPI.
Graphical Filtering is versatile, the user can base the query on different parameters: Study name, Locations, GIDs, Observation levels (PLOT, MEANS, PLANT, QUADRAT, TIMESERIES, CUSTOM datasets), Observations timestamp range start, Observations timestamp, Group By Accession, and the Query type (Graphical Filtering and Study Comparison).
Queries > Graphical Queries > Graphical Filtering Tool.
Select a study/ location(s) you are interested in
Query type> Graphical Filtering
Study of reference:
The study of reference to be used as an example is as follows: Trial as the study type and Randomized Complete Block Design as the experimental design type. It has 2 locations, 20 entries, and 2 replications. Three traits (one of them has a categorical data type, two with numerical data type), and a selection variate.
Note: The default tabular view (not grouped by accession) displays all of the recorded measurements from a selected environment. When you select this option the means values for each germplasm entry in the environment selected are revealed.
Load
…
20 germplasm entries were evaluated with 2 replications at the Africa Rice Centre and INT CENTER FOR THE IMPROVEMENT OF MAIZE AND WHEAT breeding locations. The data revealed 80 plots.
Export data and use transformations
After the data is loaded, the Export data is enabled. You will be able to export the data retrieved in CSV format using one of the five transformations available or export without a transformation. If the transformation selected is not in concordance with the experimental design of your study, a validation will be displayed.
In the example used in this manual, the data retrieved comes from a study with replication numbers, but without BLOCK_NO, so when you select Entries by Study, REP_NO, and BLOCK_NO transformation the validation is displayed.
Select by Study and REP_NO transformation and Click on the OK button
Inspect your data
Graphical Interface
Select the + button and New Range to reveal traits.
Select trait to display in the graphical interface.
Right-click on the distribution graph to draw a box across the range of values to favor selection. The list of germplasm below becomes restricted to display only those germplasm that performed in the desired range.
Six germplasm evaluated at the CIMMYT location had GY_Ad_tha yield measurements between 6.3689-7.2210 tons per hectare.
Use the + button to add additional trait filters with the following qualifiers:
And
And Not
Or
Two germplasm evaluated at the CIMMYT location had the grain yields desired with low aflatoxin level content, aleurone color scale of 1 and 2 enough to meet the filtering criteria.
Study Comparison
You will be able to compare environments from different studies (or a single study) and have a visual graphic of those comparisons.
Select one or more studies with at least one trait in common.
Select the observation level
Query Type: Study Comparison
Load
Select the variable for the visualization
Click on Compare Variable button
For the example displayed, the Variable Aflatox_M_ppb was evaluated in 4 environments, they are represented with different colors. The user can easily identify each germplasm in the environment correlation graph.