Virtual Screening

The virtual screening module enables you to perform a virtual screening campaigns using different alignment (including 3D molecular field alignment) and similarity methods. More advanced options are available to customize the scoring function.

Platform compatibility: On Windos and Mac, it is necessary to have Pharmacelera-API running. On Linux, it can work with or without API running (different options will appear if API is running or not).

Options

Input Settings

Path to library

Select the column from previous node that contains the path to the file.

Path to reference

Select the column from previous node that contains the path to the file.

URI

IP address where the Pharmacelera API is running. Default is http://localhost (local execution).

PORT

Port to access Pharmacelera API (Default 8080).

License path (only for local execution)

Use this option to indicate the path to the license file. To run the Virtual Screening Node is necessary to have a valid license file to allow you to run simulations. To obtain a valid license file, please, contact Pharmacelera (contact@pharmacelera.com).

Basic Settings

Alignment Type

skip: This option disables molecular alignment. It can be selected to use user-provided alignments (e.g. using alignments from a docking).

structure: A substructure alignment is performed maximizing field similarity.

field: Molecule alignment is performed maximizing field similarity using the molecular moments of the selected fields of interaction.

montecarlo: This option performs the molecular alignment of the field based option and then refines it using a Montecarlo optimization to maximize field similarity among the reference and the aligned structure.

hq-montecarlo: This option performs a Montecarlo optimization like in the Field based + Montecarlo option but with a more accurate similarity assessment. This option is not intended for large libraries since it is significantly slower.

Similarity Index

The Similarity index section enables to select which similarity metric should be used to compare molecules. By default, the Tanimoto coefficient is used but also the Tversky index can be selected.

Select tanimoto or tversky

Working Directory

Working directory where screening files are stored. You can enter Windows, Linux or macOS paths.

When using API, a folder name output will be created. All API executions will be saved there with the output name given.

  • Windows: C:/PATH/TO/WORKDIR

  • Linux/MacOS: /PATH/TO/WORKDIR

Output Directory

Name of the folder that contains the outputs of the node.

When using API, a folder name output will be created. All API executions will be saved there with the output name given.

Advanced Settings

First field

This option is used to select which field of interaction is going to be used to align and compare molecules. Up to three fields can be used in parallel to build the final similarity score. By default, PharmScreen is configured to use the Electrostatic (hydroele) and Cavitational (hydrocav) contributions to hydrophobicity and the H-Bond Acceptors/Donors (hbond) fields.

Default: Hydroele

Weight for first field

This option is used to select the importance that each of the three available fields have in the overall similarity score. The sum of the three weights must be 1. By default, PharmScreen is configured to use the Electrostatic (0.25) and Cavitational (0.65) contributions to hydrophobicity and the H-Bond Acceptors/Donors (0.1) fields.

Default value: 0.25

Second field

Default: Hydrocav

Weight for second field

Default value: 0.65

Third field

Default: Hbond

Weight for third field

Default value: 0.10

Enter additional options (only local execution)

Enter any option using command line format. For example:

--sanitize --threads 8

Only if you are running Knime on Linux and you want to use the executable of PharmScreen installed locally in your drive.

Input Ports

Icon

Connect Library Preparation node containing library file

Icon

Connect Library Preparation node containing reference file

Output Ports

Icon

Table with SDF compounds

Icon

CSV with ranking and properties

Popular Predecessors

  • No recommendations found

Popular Successors

  • No recommendations found

Views

This node has no views

Workflows

  • No workflows found

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.