Self-Organizing Map Panel

Create and apply Kohonen self-organizing maps from fingerprints or properties of a set of structures.

To open this panel: click the Tasks button and browse to Discovery Informatics and QSAR → Self-Organizing Map.

Self-Organizing Map Panel Features

Task options

Choose the task to perform: Create map or Apply or View map. The options determine what is displayed in the rest of the panel.

Start with map option menu

Choose a map from the menu, or choose Browse to open a file selector and load a map (.som file). The menu is populated with the recently visited maps.

Create map features

Make settings to create a self-organizing map.

Structures option menu

Choose the structure source for the self-organizing map.

  • Project Table (n selected entries)—Use the entries that are currently selected in the Project Table or Entry List. The number of entries selected is shown on the menu item. An icon is displayed to the right which you can click to open the Project Table and select entries. When this option is selected, a Load button is displayed to the right.
  • File—Use the specified file. When this option is selected, the File name text box and Browse button are displayed.
Open Project Table button

Open the Project Table panel, so you can select the entries for the structure source.

File name text box and Browse button

Enter the file name in this text box, or click Browse and navigate to the file. The name of the file you selected is displayed in the text box.

Descriptor properties options

Choose the class of properties you want to use for the map.

  • Fingerprints—Use structural fingerprints for the descriptors. Click Change to change the type of fingerprint. The types are linear, radial, molprint2D, torsion, pairwise, triplet, dendritic. Click Advanced Options to set the number of informative bits in the fingerprint.

  • Entry properties—Select properties that are available in the selected entries. Click the green + button to add properties to the list; select properties in the list and click the red − sign to remove properties from the list.

Lattice section

Specify the properties of the map lattice.

Height text box

Specify the height of the lattice as a number of nodes (cells).

Width text box

Specify the width of the lattice as a number of nodes (cells).

Topology options

Choose between a rectangular and a hexagonal lattice for the map.

Use periodic boundaries option

Select this option to make the lattice topology periodic. This eliminates edge effects when updating the neighbors of the winning node.

Training section

Specify parameters for the training of the map.

Maximum # cycles text box

Specify the maximum number of training cycles.

Train until options

Train the model until the chosen condition is met.

  • All cyles are completed—continue training until the maximum number of cycles is completed.
  • RMSE increases—continue training while the RMS error is decreasing, and stop if it increases.
  • RMSE reaches—continue training until the RMS error is less than the specified value.
Decay option menu

Specify the type of decay used to decrease the gain term as the training proceeds.

Use previous randomization option

Use a seed from a previous run for random number generation. Displays the Seed text box. If not selected, a new seed is chosen.

Seed text box

Enter the desired seed for random number generation. By default, the seed from the last selected or used map is shown in the text box.

Job toolbar

Manage job submission and settings. See Job Toolbar for a description of this toolbar.

The Job Settings button opens the Self-Organizing Map - Job Settings Dialog Box, where you can make settings for running the job.

Status bar

Use the Reset button to reset the panel to its default settings and clear any data from the panel. If the panel has a Job toolbar, you can also reset the panel from the Settings button menu.

If you can submit a job from the panel, the status bar displays information about the current job settings and status for the panel. The settings include the job name, task name and task settings (if any), number of subjobs (if any) and the host name and job incorporation setting. The job status can include messages about job start, job completion and incorporation.

The status bar also contains the Help button , which opens an option menu with choices to open the help topic for the panel (Documentation), launch Maestro Assistant, or if available, choose from an option menu of Tutorials. If the panel is used by one or more tutorials, hover over the Tutorials option to display a list of tutorials. Choosing a tutorial opens the tutorial topic.

Apply or view map features

View the map or apply the map to a new set of structures.

View Map tab

This tab displays the map.

Map area

The map has tooltips that display the value in the cell. Double-clicking a cell shows the structures that contribute to the cell in the Workspace.

Display value in cells option

Display the value of the property that was used to determine the color of each cell in the cells.

Color by option menu

Choose an option for coloring the map.

  • Distance to selected cell—Color the map by the distance to the selected cell, which is outlined in red. Click on a cell in the map to select it and color the map by the distance to this cell.
  • Cell population—Color the map by the number of compounds in each cell.
  • Cell property value—Color the cell according to the property chosen from the option menu.
  • Average property value—Color the cell according to the average value of the property chosen from the option menu.
  • # Structures in category—Color the cell according to the number of structures in the cell with the selected value of the categorical property chosen from the option menu.

For the last three options, a menu is displayed below this menu for choosing the property. For the last, another menu is displayed for choosing the categorical property value.

Gradient color squares and slider

Pick colors to represent the ends of the property range used to color the cells, and adjust the range of the property represented by the colors with the text boxes or sliders. Intermediate color values are interpolated on a spectrum. Properties outside the range are colored with the appropriate color terminus value.

Distances between cells (border color)

The distance between adjacent cells is indicated by grayscale coloring of the border between the cells. This legend provides a scale of the distances.

Apply to Structures tab

Apply the map to a set of structures. The features in this tab are the same as for Create map; the Seed text box is displayed, with the seed used for the map