Trial Manager - Curation

The Curation Tool enables users to explore, visualise, and curate data for quality assurance and control purposes. Data is collected in the field can be transferred to Trial Manager where the Curation tool can be used to curate the data. Once data has been curated, it can be saved to the KDXplore database. It is not a standalone plug-in, but is found within Trial Manager.


Curation Main Window

The Curation Main Window can be accessed through Trial Manager. See the instructions below for more information:


Accessing the Curation Tool
Accessing the Curation Tool (select to zoom)


Accessing the Curation Tool
Step Action
1. Open the Trials Tab of the Trial Manager Plug-in and select a trial to be curated (such as at circle1). This will display it’s details in the Trial Details Panel.
2. Select the edit_btn Edit Current Trial Button (located at circle2) which will open the Curation tool. You can also double-click a trial which will open the Curation tool straight away.

The image below details the Curation Main Window and the sections that it is made up of:


Curation Main Window
Curation Main Window (select to zoom)


Curation Main Window
Item UI Element Description
circle1 Curation Messages Panel Displays information and messages about events that are occurring in the Curation tool.
circle2 Values Panel Displays information on selected sample values and allows for the editing of values. It is comprised of the Trait Value Tab which contains information about one or more samples that have been selected in the Curation Samples Table, and the Sample Data Tab which provides information on the data set as a whole.
circle3 Plot Info & Traits Panel Lists all attributes and traits of the trial which can be used to select plot info and trait measurements to display.
circle4 Curation Samples Panel Contains the Trait Instances Table which lists all selected traits with their information, and the Curation Samples Table which displays all sample measurements from selected traits.
circle5 Toolbar Provides options for saving, exporting, and visualistion.


Plot Info & Traits Panel

The Plot Info & Traits Panel lists all of the attributes and traits of the trial which can be used to select plot info and trait measurements to display. Any selected items will appear in the Curation Samples Panel as a column.


Plot Info & Traits Panel
Plot Info & Traits Panel (select to zoom)


UI Elements of the Plot Info & Traits Panel
UI Element Description
Attribute/Trait Checkboxes Each row on the table is representative of an attribute or a trait. Select the corresponding checkbox for each one to display that attribute or trait in the Curation Samples Panel
check_all_btn Check All Traits Checkbox Checks all traits in the trial. This does not check any of the trail attributes.
uncheck_all_btn Uncheck All Traits Checkbox Unchecks all traits that are currently checked.
check_selected_btn Check Selected Trait Checkbox Checks only a trait that is currently selected (such as the DFF Plot trait in the above image).
Data Entry Mode Checkbox Allows for editing of data in Data Entry Mode.

One or more traits can also be selected by using ctrl+click and then a right+click provides visualisation options as seen in the image below.


Plot Info & Traits Panel Visualisation Options
Plot Info & Traits Panel Visualisation Options (select to zoom)


Values Panel

The Values Panel is made up of the Trait Value Tab and the Sample Data Tab which are outlined in the sections below.


Trait Value Tab

The Trait Value Tab contains information about one or more samples that have been selected in the Curation Samples Table as seen in the image above. The tab will display the soure of each sample, the date it was measured, it’s value, and its row and column numbers (X/Y cordinates).


Trait Value Tab
Trait Value Tab (select to zoom)


Changes to a value are reflected in the trait value tab by the creation of another row called Edited.

UI Elements of the Trait Value Tab
UI Element Description
Show Position Checkbox Displays/hides the plot column for sample values.
Values From A dropdown menu which allows for choosing a source of data. This is relevant if there are multiple data sources.
… Button Displays buttons which can set the min, max, median, mean, and mode.
Delete Button Deletes the values for the selected samples.
NA Button Sets the value of the selected samples to Not Applicable.
Missing Button Sets the value of the selected samples to Missing.
Set Value Button Sets the value for the selected trait samples. The value can be entered into the text field above the button.
Curated Checkbox Enables the suppressed and accepted options.
Supressed Samples Checkbox Sets values for suppressed samples.
Accepted Samples Checkbox Sets values for accepted samples.
Uncurated Checkbox Sets values for uncurated samples.
help_btn Help Button Opens a window with help instructions for this tab.


Sample Data Tab

The Sample Data Tab provides information on the data sets that are available for the trial. This means that when data has been collected by one or more scorers, each data set will be unique. The image below shows just one dataset from a user called ‘tester’ and a device called ‘test device’. Other columns show the date that the data was loaded into Trial Manager and the number of samples and traits in the trial.


Sample Data Tab
Sample Data Tab (select to zoom)


The following options are available in this tab:

UI Elements of the Sample Data Tab
UI Element Description
Set Values Button Sets the collected values for the trial (of the particular trait that is selected in the Values for Trait Dropdown Menu).
Values for Trait A dropdown menu which allows for choosing of a trait that to set the final values for.
Curated Checkbox Enables the suppressed and accepted options.
Supressed Samples Checkbox Sets values for suppressed samples.
Accepted Samples Checkbox Sets values for accepted samples.
Uncurated Checkbox Sets values for uncurated samples.
help_btn Help Button Opens a window with help instructions for this tab.


Curation Samples Panel

Once traits have been chosen in the Plot Info & Traits Panel, they appear in the Trait Instances Table and Curation Samples Table which are both within the Curation Samples Panel. The following sections outline each of these tables.


Trait Instances Table

The Trait Instances Table displays the traits and trait instances that have been selected in the Plot Info & Trait Instances Panel. Each trait and trait instance is listed as a row and is accompanied by other information including the data type, number of samples, and statistics such as the mean of sample values for the trait.

Next to the name of each trait/instance will also contain the notation (s) or (p) to indicate whether the trait is on the plot or sub-plot level. Each column can also be selected to sort the traits such as ordering the traits by the number of scored samples. See the image below and the accompanying table for more information on the UI elements of the table.


Trait Instances Table
Trait Instances Table (select to zoom)


UI Elements of the Trait Instances Table
UI Element Description
resize_btn Resize Button Resizes columns according to the dropdown option that has been chosen.
Column Sizing Menu A dropdown menu which contains options for column sizing.
columns_btn Configure Table Columns Button Opens the Configure Table Columns Window (as seen in the image below) which provides options for configuring the columns that appear in the Trait Instances Table.


Configure Table Columns Window
Configure Table Columns Window (select to zoom)



Curation Samples Table

The Curation Samples Table displays collected data according to the traits and instances that have been selected as per the Plot Info & Trait Instances Panel.

The following image and table provide information on the Curation Samples Table and UI elements of the table.


Curation Samples Table
Curation Samples Table (select to zoom)


UI Elements of the Curation Samples Table
UI Element Description
Hide Inactive Plots Checkbox Hides all plots that are inactive.
Un-Curated Checkbox Displays un-curated data.
Only Scored Checkbox Displays only plots and sub-plots that have been scored.
tag_filter_btn Filter Tags Button Displays a list of tags to filter curation samples by. The button will appear as tag_filter_active_btn if tag filtering is active.
filter_plot_sub-plot_btn Filter Plots/Sub-Plots Provides options to filter by plot or sub-plot. The button will appear as filter_plot_active_btn if plot filtering is active and as filter_sub-plot_active_btn is sub-plot filtering is active.
help_btn Help Button Opens a window with help instructions for this table.
resize_btn Resize Button Resizes columns according to the dropdown option that has been chosen.
Column Sizing Menu A dropdown menu which contains options for column sizing.

Right+clicking a sample provides further options for the data. The following image and table provide more information on the options:


Sample Options
Sample Options (select to zoom)


Sample Options of the Curation Samples Table
UI Element Description
Accept Values Marks the values as correct. This is important when there are multiple datasets for the trial.
Suppress Values Marks the values incorrect. This is important when there are multiple datasets for the trial.
Activate Plots Activates any de-activated plots.
De-Activate Plots De-activates a plot.
Heat Map Opens the Heat Map Window to create a heat map with the selected plots.
Scatter Plot Opens the Scatter Plot Window to create a heat map with the selected plots.
Box Plot Opens the Box Plot Window to create a box plot with the selected plots.
View Displays all attachment files associated with the plot or sub-plot.
Inspect Plots Opens the Plot Inspection Window which provides information on the plots including measurement values.

Note

Please note that the visualisation options available in the above options are explained in more detail later on in this user guide page.


Toolbar

The toolbar provides some options for saving, curating, and visualising the data. The toolbar itself can also be detached from the Main Curation Window by clicking+dragging the bar that the Toolbar is attached to. The image and and table below provide some information on the options available in the Toolbar.


Toolbar
Toolbar (select to zoom)


UI Elements of the Curation Samples Table
UI Element Description
save_btn Save Button Saves any curated changes made to the data. Once data has been saved it will not be possible to use the undo or redo functions.
export_btn Export Button Opens the Export Curated Data Window which provides options for exporting data and configuring the output of exported data.
undo_btn Undo Button Undoes any curation changes made to the dataset. This will be disabled if not applicable.
redo_btn Redo Button Redoes any undo action made for curation changes. This will be disabled if not applicable.
heat_map_btn Heat Map Button Opens the Heat Map Window which provided options for creating a heat map visualisation.
scatter_plot_btn Scatter Plot Button Opens the Scatter Plot Window which provided options for creating a scatter plot visualisation.
box_plot_btn Box Plot Button Opens the Box Plot Window which provided options for creating a box plot visualisation.
field_btn Field View Button Opens the Field View Window which provides a top-down view of the trial field and other information.


Curating Measurements

One of the primary functions of the Trial Management Plug-in and the Curation tool is to curate data that is collected from the field before it is uploaded to KDDart. Values can be curated in a number of different ways as will be outlined below.

Tip

Values can only be curated to values permitted by the validation rules of the trait that is being measured.


Single-Selection Curation

The most simple way to curate trial data is to curate a single value. The example below demonstrates how to curate a single measurement:


Single-Selection Curation
Single-Selection Curation (select to zoom)


Single-Selection Curation
Step Action
1. Select a single sample from the Curation Samples Table such as the one at circle1 in the above image. This will display the sample’s information in the Values Panel.
2. Enter a new value into the text field at circle2. You will be able to see that as you type this value, the corresponding value in the Curation Samples Table also changes in real time.
3. It is permanently set when you select the Set Value Button at circle3. Each change to a sample will be reflected in the Trait Value Tab with a history of all value changes.


Multi-Selection Curation

The Curation Samples Table also allows users to select multiple samples at once. This allows for the curation of multiple samples at the time which makes curation easier and more efficient. Multi-select curation is particularly useful when large numbers of samples need to be modified to have the same value.


Multi-Selection Curation
Multi-Selection Curation (select to zoom)


Multi-Selection Curation
Step Action
1. Select a multiple samples from the Curation Samples Table such as the one at circle1 in the above image. This will display the sample information in the Values Panel.
2. Enter a new value into the text field at circle2. You will be able to see that as you type this value, the corresponding values in the Curation Samples Table also changes in real time. For multi-selection curation, this will be for all select trials at the same time.
3. The values are permanently set when you select the Set Value Button at circle3. Each change to a sample will be reflected in the Trait Value Tab with a history of all value changes.


Trial-Wide Curation

The values of specific trait instances can be set across an entire trial. This should only be done if you are sure that the correct values are in place for that trait instance. The below image example provides an example of how to use this function:


Trial-Wide Curation
Trial-Wide Curation (select to zoom)


Showing Outliers
Step Action
1. Navigate to the Sample Data Tab in the Value Panel.
2. Select one of the datasets listed in the tab. The example above shows the datasets called test device and Edited.
3. Choose any of the checkboxes to include setting the values that are to be set.
4. Use the Values for Trait Dropdown Menu to select a trait instance such as Rust in the example above.
5. Select the Set Value Button to confirm setting the values of that trait instance.


Showing Outliers

The Curation Tool makes it possible to see any outliers that have occurred when scoring. Outliers are any data measurement that show extreme deviation from a data set and are an indication that data has been incorrectly entered. A part of the overall curation process is to remove these outliers. Outliers can be found in the Curation Sample Panel. See the below image and instructions on finding and removing outliers from a dataset.


Showing Outliers
Showing Outliers (select to zoom)


Showing Outliers
Step Action
1. View the samples of the Curation Samples Table. The above image shows two samples where the values have been accepted but are far outside the normal range for that trait (at circle1).
2. See the Outlers Column at circle2 to view outliers for each trait. The two outlying sample values for the DFM Plot trait are visible in this column.


Accepting/Suppressing Measurements

A part of the curation process is to either accept or suppress measurements for each sample. This is particularly important if there are multiple datasets for the same trial that do not match. An example of this would be if the data from two different users (who have scored the same trial) was imported into Trial Manager and there was a sample that was scored differently by each user. One of the samples would need to be suppressed and the other would need to be accepted. This could be ascertained by going back into the field and checking the measurement if possible. The image and instructions below provide some information on how to accept and suppress samples.


Accepting/Suppressing Measurements
Accepting/Suppressing Measurements (select to zoom)


Accepting/Suppressing Measurements
Step Action
1. Right+click a sample and either select the Accept Value(s) or Suppress Value(s) option to accept or suppress the value (as seen in circle1).
2. Either option will open a small window which will ask for confirmation.
3. Accepted values will be in bold and not have the warning icon next to it - such as the values for the Rust trait at circle2. Suppressed values will have a strikethrough (or marked as MISSING or NA if the value is not present) - such as the values for the EPH trait at circle3.


Saving & Exporting Curated Data

After curating data, it will need to be saved so that any changes are kept. The data can then also be exported if need be. See the following instructions on how to save and export curated data:


Saving & Exporting Curated Data
Saving & Exporting Curated Data (select to zoom)


Saving & Exporting Curated Data
Step Action
1. Select the save_btn Save Button from the toolbar to save any curated changes. Please note that once data is saved, any changes cannot be undone so the undo_btn Undo Button and redo_btn Redo Button will become disabled.
2. Once data is saved, it can be exported. Select the export_btn to open the Export Curated Data Window, as seen in the above image.
3. Configure the export as per the options provided. This included choosing a location to export the data too, whether to export all or just selected plots, and whether to export the operator name.
4. Select the Save Button at the bottom of the window to finalise the export.


Overview

The Overview Window provides information on the status of each plot such as whether the data has been curated or if the plot is activated. The information is displayed on a per-trait basis.

Note

Please not that the Overview Window is currently disabled and is undergoing a revision.


Field View

The Field View Window provides a top-down view of the trial field with information on plots and allows for the selection of samples. It can be opened by selecting the field_btn Field View Button from the Toolbar in the Main Curation Window.

The image below demonstrates how the selection of plots in Field View will also select the associated samples from the Curation Samples Table in the Main Curation Window. The trait chosen from the dropdown menu in the Field View Window will also determine what samples are selected e.g. if the trait PH_Soil is chosen in the Field View Window and plots are selected, then the PH_Soil samples for those plots are the ones that will be selected in the Curation Samples Table.

The image and table below provide an example of the Field View Window and it’s UI elements.


Field View
Field View (select to zoom)


UI Elements of the Field View Window
UI Element Description
serpentine_bottom_left_up_btn Collection Path Button Opens the Collection Path Window which provides options for how to collect data whilst in the field. This button will look different depending on the point-of-origin and traversal pattern chosen e.g. serpentine_bottom_left_up_btn shows that the origin is the bottom left and a serpentine traversal pattern is chosen.
help_btn Help Button Opens a window with help instructions for Field View.
Trait Menu A dropdown menu which allows the user to select traits for selection.
resize_btn Resize Button Resizes columns according to the dropdown option that has been chosen.
Column Sizing Menu A dropdown menu which contains options for column sizing.
Plots As Field View provides a top-down view of a trial field, each plot is shown in a grid. See the image and table below for more information on plots.


Field View Plot Information
Field View Plot Information (select to zoom)


Field View Plot Information
UI Element Description
Plot Number The number of the plot in relation to the trial. The example in the image above shows that this plot is number 29.
 
① Sub-Plot Number The number of sub-plots within the plot. The example shows that plot 29 contains three sub-plots.
❶ Tags Number The number of tags for this plot. Plot 29 in the example above has 14 tags applied.
Plot Type Displays the plot type which is a plot attribute. The plot type of plot 29 is Dry.

Collection Path

The collection path determines how a KDSmart user will collect data whilst in the field. The following table provides information on the options available for the collection path:


Collection Path Window
Collection Path Window (select to zoom)


Collection Path Options
Option Description
Plots per Group Determines how many plots are scored at once. This could be used if the user wants to decides whether to scored data from only the plots on their left as they walk through a field, or to score plots on both their left and right.
Origin & Direction Sets the corner of the field that the scoring will start from and the direction that the scorer will move in.
Traversal Establishes the pattern on traversal that the user will take through the field; either a serpentine or straight traversal pattern.

Once these options have been chosen, select the confirm_btn Confirm Button to confirm the collection path options.

Note

The collection path options that are set here will carry over to the KDSmart application when the trial is transferred for scoring. However, collection path options can also be changed in KDSmart if the user wishes.


Visualisation Tools

The Curation tool provides options for various data visualisations. Users can create heat maps, box plots, and scatter plots with scored data, as well as comparing samples with a plot identification visualisation.

Visualisations for can be generated from the Trait Instance Table, Curation Sample Table (by selecting samples and then right-clicking, then choosing a visualisation option), or from the Tool Bar at the top of the Curation Main Window.


Heat Map

The Curation Tool can create a heat map which is a representation of the experiment data that is displayed as varying colours. The image and instructions below provide information on how to create a heat map from scored data.


Creating a Heat Map
Creating a Heat Map (select to zoom)


Creating a Heat Map
Step Action
1. From the Toolbar in the Curation Main Window (at circle1), select the heat_map_btn Heat Map Button to open the Heat Map Configuration Window at circle2.
2. Select a row and then the Axis Buttons on the left of the window to set the X and Y axes of the heat map. Rows and columns will be automatically set as either the X or Y, but if need to change those values then select each row and then the Deselect Button so that each row can be reset.
3. Select a row for the value to be represented in the heat map, and then select the Value Button to set the value. The example above shows selection of the PH trait.
4. Select the Generate Heat Map Button to create the heat map which will open at circle3.

Note

An appropriate heat map can only be created from an X/Y trial. Plot ID trials do not have the information necessary to create the map.


UI Elements of a Heat Map
UI Elements of a Heat Map (select to zoom)


UI Elements of a Heat Map
UI Element Description
Camera Button Provides options for saving the heat map as an image.
Refresh Button Refreshes data that may have changed in the
Messages Tab Displays information on any changes to the heat map.
Curation Tab Provides options for accepting/suppressing values and activating/deactivating plots.
Accept Values Button Accepts the values for any selected plots.
Suppress Values Button Suppresses the values for any selected plots.
Deactivate Plots Button Deactivated any selected plots that are currently active.
Activate Plots Button Activates any selected plots that are currently deactivated.
Plots Each plot in the heat map is a square with a colour that is representative of the trait value. Plot information can be displayed by hovering over the plot with the mouse (as seen in the example image). Plots can be selected by clicking+dragging over one or more plots.
Sync Menu A dropdown menu that provides options for syncing changed values from the Curation Tool.
Colour Menu A dropdown menu that provides colour scheme options for the heat map. The colour scheme in the image above is set the Rainbow.
Opacity Field A text field that allows for the configuration of opacity levels of the heat map colour values. A number can be entered into the field and adjusted with the up and down arrows.


Scatter Plot

The Curation Tool also has the ability to create a scatter plot to represent scored data. The image and instructions below provide more information about how to create and use the scatter plot.


Creating a Scatter Plot
Creating a Scatter Plot (select to zoom)


Creating a Scatter Plot
Step Action
1. From the Toolbar in the Curation Main Window (at circle1), select the scatter_plot_btn Scatter Plot Button to open the Scatter Plot Configuration Window at circle2.
2. Each row is a trait that is in the trial. Select a checkbox to set that trait as the X axis. Deselect that checkbox to remove that trait as an axis. The example above has the PH_cm trait set as the X axis.
3. To set traits to be represented in the scatter plot, continue selecting the relevant checkboxes. One or more traits need to be set as the value to be displayed. The EH_cm and EarsHvst_ears_plot traits are set as the values to be represented in the scatter plot of the above image.
4. Select the Continue Button to create the scatter plot which will open at circle3.

Once the scatter plot has been created, see the image and table below for more information on using the scatter plot.


UI Elements of a Scatter Plot
UI Elements of a Scatter Plot (select to zoom)


UI Elements of a Scatter Plot
UI Element Description
Camera Button Provides options for saving the scatter plot as an image.
Refresh Button Refreshes data that may have changed in the
Messages Tab Displays information on any changes to the scatter plot.
Curation Tab Provides options for accepting/suppressing values.
Apply To Options Radio buttons that allow for the selection of one or all traits represented as values in the scatter plot.
Accept Values Button Accepts the values for any selected samples.
Suppress Values Button Suppresses the values for any selected samples.
Samples Each sample in the scatter plot is a coloured square that is representative of the trait according to the key. In the example above , EH_cm is light blue and EarsHvst_ears_plot is yellow. Samples can be selected by clicking+dragging over values in the scatter plot.
Sync Menu A dropdown menu that provides options for syncing changed values from the Curation Tool.
Colour Menu A dropdown menu that provides colour scheme options for the heat map. The colour scheme in the image above is set to Rainbow.
Opacity Field A text field that allows for the configuration of opacity levels of the heat map colour values. A number can be entered into the field and adjusted with the up and down arrows.


Box Plot

A box plot can also be created with the curation tool. The image and instructions below provide more information about how to create and use the box plot.


Creating a Box Plot
Creating a Box Plot (select to zoom)


Creating a Box Plot
Step Action
1. From the Toolbar in the Curation Main Window (at circle1), select the box_plot_btn Box Plot Button to open the Box Plot Configuration Window at circle2.
2. Each trait is a row within the Box Plot Configuration Window. Select one or more checkboxes to display that trait in the box plot.
3. Select the Continue Button to create the box plot which will open at circle3.

Once the box plot has been created, see the image and table below for more information on using the box plot.


UI Elements of a Scatter Plot
UI Elements of a Scatter Plot (select to zoom)


UI Elements of a Box Plot
UI Element Description
Camera Button Provides options for saving the box plot as an image.
Refresh Button Refreshes data that may have been changed.
Messages Tab Displays information on any changes to the box plot.
Accept Values Button Accepts the values for any selected samples.
Suppress Values Button Suppresses the values for any selected samples.
Samples Samples are grouped into a box plot which demonstrates the data in quartiles, the box being Q2 and Q3. Samples can be selected by clicking+dragging over an area of the box plot.
Sync Menu A dropdown menu that provides options for syncing changed values from the Curation Tool.
Show Parameters Options Checkboxes for displaying the outliers, mean, and median of the data.