Scatterplot

A scatterplot displays data points on the X and Y axes to show the relationship between two variables. By using marker color, shape, and size, you can show up to three additional data dimensions. A scatterplot is a chemically-aware viewer and is a great choice to visualize chemical space or activity cliffs.

Controls

Context menu	Right-click
Zoom	Alt+Mouse Drag
Zoom in	Mouse Wheel Up or Plus
Zoom out	Mouse Wheel Down or Minus
Double-click	Reset view
Select	Shift+Mouse Drag, Ctrl+Click, Shift+Click
Invert selected	Ctrl+Mouse Click
Scroll	Up, Down, Left, Right
Toggle lasso tool	L
Toggle regression line	R
Show in full screen	Alt+F

Adding and configuring a scatterplot

To add a scatterplot, click the Scatterplot icon on the Toolbox.

Use the viewer controls to select columns for each axis and marker color and size. You can also drag-and-drop columns from the grid into the scatter plot. For additional configurations, click the Gear icon on top of the viewer and set your preferences in the Context Panel.

You can also access key settings from the context menu by right-clicking. Element-specific context menus appear when you click on legend, axis, or label.

Data source

To specify the rows to show on the scatter plot, use the "Table" and "Row Source" properties in the Data section on the Context Panel.

• "Table" to visualize a table other than the current one
• "Row Source" to visualize a subset of data:
  • "Filtered" (the default value) - scatter plot follows view filter
  • "All", "Selected", "SelectedOrCurrent", "FilteredSelected", "MouseOverGroup", "CurrentRow", "MouseOverRow" - other options useful for providing interactivity

You can further filter visible rows by setting the "Filter" property to an expression, such as ${AGE} > 18.

X, Y, Colors, Sizes, Markers

Use the column selectors on top of the scatterplot, and the popup menu to setup basic properties:

Filtering

In addition to visualizing filtered rows, scatter plot can also be used to filter the table, which in turn affects what you see on other viewers on this view. This behavior is controlled by the "Zoom and Filter" property:

• "filter by zoom" (default): as you zoom in, global view filter changes to show only rows that are visible on the scatterplot. In this mode, "Filter Out Invalid" property defined whether rows that could not be visualized on the scatter plot (such as negative values on log scales) should be filtered out.
• "no action": zooming in does not affect view filter
• "zoom by filter": as the view filter changes, scatter plot zooms in to the minimum area containing filtered points. This is useful for analyzing clusters of data.
• "pack and zoom by filter": mostly same as "zoom by filter", but in case categorical values are shown on an axis and some categories are completely filtered out, these categories get removed (packed) from the axis. Useful when visualizing data that has a large number of categories.

Selection

Scatterplot highlights selected rows in yellow, and lets you select points as well:

• To select area: Shift + mouse-drag
• To unselect area: Ctrl + Shift + mouse-drag
• To toggle point selection: Ctrl + click

To switch between lasso and rectangular selection modes, press L or click Lasso Tool from the context menu.

Regression lines

To show a regression line, press R or check the "Show Regression Line" property on the context panel. To hide the equation, uncheck "Show Regression Line Equation".

Formula lines

A scatterplot can show reference lines that represent formulas or equations. These lines are used to emphasize specific areas on the chart or data. Common examples include a regression line, value bands, and so on.

To show a custom formula line, right-click a scatterplot, then choose Tools > Formula Lines... This action opens a Formula Lines dialog. Here, enter your formula and configure the line settings. Your formula should refer to the columns on the X and Y axes. The syntax for the formula is similar to that used to Add New Column.

developers

You can add formula lines programmatically.

Tooltip

By default, a scatterplot inherits the tooltip from the grid. However, you can customize the scatterplot's tooltip to show the data you want using the Tooltip info pane on the Context Panel. To configure a custom tooltip:

1. Enable custom tooltip: Set Show Tooltip to Show custom tooltip.

2. Choose which columns to display: In Row Tooltip, select the columns whose values you want to show in the tooltip.

3. Control axis values in the tooltip: In Data Values, specify how axis values should appear:

   • To exclude axis values from the tooltip, choose Do not add.
   • To show only axis values, choose Data values only.
   • To add axis values, choose Merge.

In addition, a scatterplot itself can be used as a group tooltip, which may be especially useful when dealing with grouped or clustered data or when the screen space is limited.

To make scatterplot appear in a tooltip when you hover over a row group, right-click on the scatterplot and select Tooltip > Use as Group Tooltip.

Labels

To show values next to the markers, set the Labels settings either from context menu or from the properties panel:

• To select columns to show, expand Label Form and check or drag-and-drop columns
• To select a subset of rows to show, use Show Labels For
  • You can drag-and-drop labels to exact positions in the Selected or Current modes
• Use Label as Marker renders centered label instead of the marker. This is particularly useful for zooming in on molecular datasets (points become molecules).
• Check Show Column Names to show column names next to the values

To quickly adjust settings for labels, right-click on the label.

Jitter

Many points in your dataset might fall into the same X and Y coordinates (often happens with integer or categorical columns). To spread them out on the plot, set Jitter Size:

Connecting lines

You can set a column that defines order in which points are connected. Below, we see the (gdp, life expectancy) trajectory of different countries over time.

Cheminformatics

Scatterplot supports custom value renderers, and in particular it's very useful for visualizing high-dimensional chemical data. If you choose to do so, molecules could be rendered on axes, as labels, or on tooltips. To learn more, check out cheminformatics.

WebGPU acceleration

WebGPU acceleration allows you to quickly render massive datasets (10 millions rows and more). To get the maximum performance, set Zoom and Filter property to "no action".

This feature is currently in beta. To enable it, check Settings > Beta > Enable Scatter Plot Web GPU Acceleration

Videos

Properties

Property	Type	Description
Data
Filter Out Invalid	boolean	Invalid are null values and not positive numbers if axis is logarithmic.
Show Filtered Out Points	boolean	When true, filtered out points are rendered using Filtered Out Rows Color.
Axes Follow Filter	boolean	When true, scatter plot will zoom to an area defined by the range filters for X and Y columns, even if Zoom And Filter property is not set to Zoom by Filter.
Zoom And Filter	string	Determines the relationship between table filter and scatter plot area: * No action: they are disconnected * Filter by zoom: scatter plot acts as a filter; as you zoom in, points get filtered out * Zoom by filter: scatter plot focuses on the filtered points as the filter changes * Pack and zoom by filter: removes filtered out categories and focuses on the filtered points as the filter changes.
Filter	string	Formula that filters out rows to show. Examples: ${AGE} > 20 or ${WEIGHT / 2)} > 100, ${SEVERITY} == ''Medium'', ${RACE}.endsWith(''sian'')
Table	string
X
X Column Name	string	A column to use on the X axis. Could be numerical or categorical.
X Map	string	Time unit map function for x column (applicable to dates only).
X Axis Type	string
Invert X Axis	boolean
X Min	number
X Max	number
Show Vertical Grid Lines	boolean
Show X Axis	boolean
Show X Selector	boolean
X Whisker Min Column Name	string	Point lower bound for x axis whiskers. Selecting it disables X Whisker Range.
X Whisker Max Column Name	string	Point upper bound for x axis whiskers. Selecting it disables X Whisker Range.
X Whisker Range Column Name	string	Point range for x axis whiskers. Applied only if X Whisker Min and X Whisker Max are not set.
X Axis Label Orientation	string
Y
Y Column Name	string	A column to use on the Y axis. Could be numerical or categorical.
Y Map	string	Time unit map function for y column (applicable to dates only).
Y Axis Type	string
Invert Y Axis	boolean
Y Min	number
Y Max	number
Show Horizontal Grid Lines	boolean
Show Y Axis	boolean
Show Y Selector	boolean
Y Whisker Min Column Name	string	Point lower bound for y axis whiskers. Selecting it disables Y Whisker Range.
Y Whisker Max Column Name	string	Point upper bound for y axis whiskers. Selecting it disables Y Whisker Range.
Y Whisker Range Column Name	string	Point range for y axis whiskers. Applied only if Y Whisker Min and Y Whisker Max are not set.
Axes
Show X Histogram	boolean	Shows a distribution histogram along the X axis (at the top)
Show Y Histogram	boolean	Shows a distribution histogram along the Y axis (on the right)
Histogram Bins	number	Number of bins for axis histograms
Color
Color Column Name	string	A column to be used for color-coding. Could be numerical or categorical. If not set, Filtered Rows Color is used for markers that pass the filter. Color palettes could defined either for columns in the column context panel, or via Linear Color Scheme and Categorical Color Scheme properties.
Color Map	string	Categorical coloring time unit map function (applicable to dates only).
Show Color Selector	boolean
Color Axis Type	string
Invert Color Scheme	boolean
Color Min	number
Color Max	number
Size
Size Column Name	string	A numerical column to use for size-coding markers. See also Marker Min Size and Marker Max Size.
Show Size Selector	boolean
Marker
Markers Column Name	string	A categorical column that determines the shape of the markers.
Markers Map	string	Marker category time unit map function (applicable to dates only).
Marker Type	string
Marker Default Size	number	By default - automatic sizing based on current dataframe
Marker Opacity	number
Jitter Size	number	Randomly shift (x, y) marker position up to the Jitter Size pixels. Useful when multiple points fall on the same exact position. If Jitter Size Y is defined, then Jitter Size shifts x only.
Jitter Size Y	number	Randomly shift y marker position up to the Jitter Size Y pixels.
Marker Draw Border	boolean
Marker Border Width	number
Marker Min Size	number
Marker Max Size	number
General
Lines Order Column Name	string	When defined, a line would be drawn for each series (defined by the categorical color column) using the order specified by Lines Order
Lines Width	number	Defines the width of the lines connecting the markers. See Lines Width.
Show Min Max Tickmarks	boolean	Shows tickmarks and labels for minimum and maximum value on each axis.
Show Drop Lines	boolean	Shows exact X and Y coordinates for the mouse cursor.
Mouse Drag	string
Lasso Tool	boolean	When true, lasso area selector is used instead of the rectangular one. Toggle this option by pressing L.
Allow Zoom	boolean
Legend Visibility	visibilitymode
Legend Position	flexautoposition
Row Source	string	Determines the rows shown on the plot.
Allow Dynamic Menus	boolean
Show Context Menu	boolean	Properties common for all viewers todo: use code generation
Title	string
Description	string	Viewer description that gets shown at the Descriptor Position. Markup is supported.
Help	string	Help to be shown when user clicks on the ''?'' icon on top. Could either be in markdown, or a URL (starting with ''/'' or ''http'').
Description Position	flexposition
Description Visibility Mode	visibilitymode
Labels
Label Column Names	list	Label columns to show next to the markers.
Show Labels For	string	Determines the rows shown on the scatter plot.
Display Labels	visibilitymode	Determines how to show marker label: * Always - show labels for all visible markers * Auto - show labels only for markers where enough space is available * Never - show no labels.
Show Label Named Columns	visibilitymode	Determines whether to show column names next to label values.
Use Label As Marker	boolean	If checked, display a label content as marker.
Label Color As Marker	boolean	To display labels separately or as markers (works for non-text labels).
Label As Marker Size	number	Marker size in which label is inscribed.
Label Content Size	number	Label inner content size.
Lines
Show Regression Line	boolean	Regression line visibility (toggle by pressing R)
Show Regression Line Equation	boolean
Show Spearman Correlation	boolean
Show Pearson Correlation	boolean
Show Mean Absolute Error	boolean
Show Root Mean Square Error	boolean
Regression Per Category	boolean
Show Dataframe Formula Lines	boolean	Control the visibility of dataframe-originated formula lines. Edit formula lines by right-clicking and selecting Tools | Formula Lines from the popup menu. Requires the PowerPack plugin.
Show Viewer Formula Lines	boolean	Control the visibility of dataframe-originated formula lines. Edit formula lines by right-clicking and selecting Tools | Formula Lines from the popup menu. Requires the PowerPack plugin.
Selection
Show Current Point	boolean	Controls the indication of the current row
Show Mouse Over Point	boolean	Controls the indication of the mouse-over row
Show Mouse Over Row Group	boolean	Highlight ''mouse-over'' rows (such as the ones that fall into a histogram bin that the mouse is currently hovering over).
Show Selected Rows	boolean	When true, selected markers are highlighted using the selected rows color. When false, selected markers use their regular color coding.
Style
Auto Layout	boolean
Back Color	number
Filtered Rows Color	number
Filtered Out Rows Color	number
Selected Rows Color	number
Missing Value Color	number
Label Color	number
Axis Line Color	number
Axis Text Color	number
Grid Line Color	number
Regression Line Color	number
Whisker Color	number
Regression Line Transparency	number
Linear Color Scheme	list
Categorical Color Scheme	list
Axes Use Column Format	boolean	Determines whether the axes should follow the non-precision-related format (such as money) set for the corresponding column.
Auto Axis Size	boolean	If true, X Axis Height and Y Axis Width are calculated automatically to fit the required precision. If false, the specified X Axis Height and Y Axis Width properties are used.
X Axis Height	number	Requires Auto Axis Size to be turned off.
Y Axis Width	number	Requires Auto Axis Size to be turned off.
Axis Font	string
Label Font	string
Formula Font	string
Annotation Font	string
Controls Font	string	Viewer controls elements font.
Annotation regions
Show Viewer Annotation Regions	boolean
Show Dataframe Annotation Regions	boolean
Tooltip
Show Tooltip	string	Controls scatter plot tooltip visibility
Show Labels	visibilitymode
Data Values	string	Controls whether columns on X and Y axes are displayed in tooltip * Do not add: they are not shown * Data values only: only they are shown * Merge: standard behavior
Row Tooltip	string	Newline-separated list of column names to be used in a tooltip. Requires showTooltip to be enabled.
Row Group Tooltip	string
Description
Show Title	boolean

See also:

• 3D Scatterplot
• Column selectors
• Viewers
• Table view
• JS API: Scatterplot
• Community:
  • Scatterplot updates
  • Visualization-related updates