Build and Apply QSAR/QSPR Model — Apply Task

In this task, you can inspect QSAR models and apply them.

To display the options for this task, open the Build and Apply QSAR/QSPR Model Panel and select View/Apply.

Apply Tab Features

Plot menu

For regression models, plot the predicted values of the Y variable against the observed values in a scatter plot. Opens a plot window. The perfect fit (45° angle) is shown as a red line. Click Save Image to save the plot.

All Structures option

Include all structures in the scatter plot. The training set data is shown as blue dots. The test set data is shown as red dots.

Training Set option

Include only the training set structures in the scatter plot.

Test Set option

Include only the test set structures in the scatter plot.

Export menu

Export the model or test results as files. After selecting an option from this menu, a file browser window opens, where you can specify the filename and location of the exported file.

Model option

Select Model to export the model as .mlzip file.

Training/Test Results option

Select this option to export the training and test results as .csv file. As columns, the .csv file has the structure title, the assignment to training or test set, the observed value of the Y variable, and the predicted value of the Y variable.

Model summary section

Shows a summary of the model with the following properties:

  • Method—the method used to build the model.
  • Property to predict—the response (Y) variable
  • Structures—the number of structures in the data set, includes training and test sets
  • Descriptors—the X variables used to build the model
  • Training set—the percentage of structures in the training set
Model settings pane

Shows model-specific settings that were chosen when building the model in the Model settings options.

Visualize Factor button

Opens the 2D Viewer Panel showing atomic contributions of the factors in red and blue colors. This button is only available for the KPLS regression model.

Summary Statistics and Prediction section

Show tables containing model statistics and show the prediction for each structure.

Training tab

Show a table with statistical properties of the model for the training set data. The table contains the standard deviation and R2. Depending on the method used to build the model, more properties might be shown. For the recursive partitioning and Bayes classification models, a truth table of predicted vs. observed categories is shown.

Test tab

Show a table with statistical properties of the model for the test set data. The table contains the root-mean-square error (RMSE) and Q2. Depending on the method used to build the model, more properties might be shown.

Factor Loadings button

For partial least squares models, show a table with the contributions of each descriptor (X variable) to the factors.

Prediction table

For each structure, show the true value of the Y variable, the prediction from the model, and the absolute deviation

Apply Model section
Predict value for structures from option menu

Choose the structure source for application of the QSAR model.

  • 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.