3D QSAR Model: Interpretation

Category Paths

Follow one of these paths in the Orion user interface, to find the floe.

  • Models

Description

The 3D QSAR Model: Interpretation Floe is a tool for interpreting kernel-PLS models.

The model interpretation floe requires a model dataset obtained from using the 3D QSAR Model: Builder Floe.

The default output from this floe is a dataset containing the interpretation of the kernel-PLS models inherent within the uploaded model. The interpretation is provided in the form of grids and high/low surfaces obtained from probing the binding domain with carbon and different color atoms defined based on the Implicit Mills Dean color force field. If EON-KPLS model is also selected under Selected Models, EON-kPLS model interpretation results obtained using positive and negative charge probes are available as well. The floe also outputs a ROCS query dataset based on interpretation results from ROCS-kPLS model.

Note: If the training set is relatively large (e.g., greater than 300), consider (1) manually selecting the kPLS model record(s) of interest as floe input to save cost; (2) increasing cube memory under “Cube Memory Parameters” section to avoid cube memory error. An eightfold increase is usually sufficient for a training set up to 1000 molecules.

Promoted Parameters

Title in user interface (promoted name)

Outputs

Interpretation Output Dataset (out): Output dataset containing kPLS model interpretation results.

  • Required

  • Type: dataset_out

  • Default: Output for 3D QSAR Model: Interpretation

Failed Dataset (failed): Output dataset of failed calculations.

  • Required

  • Type: dataset_out

  • Default: Failed Output for 3D QSAR Model: Interpretation

ROCS Query Output Dataset (query_out): Optional output dataset containing ROCS query.

  • Required

  • Type: dataset_out

  • Default: ROCS Query Output

Model Interpreter Parameters

Selected Models (models_selected): Selected KPLS models to interpret.

  • Type: string

  • Default: [‘ROCS-KPLS’]

  • Choices: [‘ROCS-KPLS’, ‘EON-KPLS’]

Grid Spacing (grid_size): Grid cell dimension in angstroms.

  • Type: decimal

  • Default: 1

Percentage for Surface Contour Values (surface_percentage): Grid contour values for generating high and low surfaces. 2 means using top 2% of positive values for high surfaces and bottom 2% of negative values for low surfaces.

  • Type: decimal

  • Default: 2.0

Cube Memory Parameters

Cube Memory (memory): Minimum amount of memory in MiBs (1048576 B).

  • Type: decimal

  • Default: 1800