0 ×

Ligand-based Pharmacophore Creator

LigandScout Extensions for the KNIME Workbench version 1.7.9 by InteLigand GmbH

The Ligand-based Pharmacophore Creator node generates ligand-based pharmacophores from a list of given ligands.



Pharmacophore type
Two different types of pharmacophoric models can be derived due to different generation processes. The ligand-based pharmacophore generation is a pair-wise process. In each step one pharmacophore for two molecules is drained. This we can do by either taking all common features (Shared Feature Pharamacophore) or by taking all features (Merged Feature Pharmacophore). In the latter case each feature is scored and those are removed that do not match all input molecules (see number of omitted features for merged pharmacophore). The shared feature pharmacophore creates a pharmacophore that represents all common features of a ligand-set. If you want to take all features into account and assemble them into one pharmacophore, then the merged feature pharmacophore is your right choice.
(Default: Merged feature pharmcophore)
Scoring Function
During the ligand-based pharmacophore generation all intermediate alignment results are scored according to the selected scoring function. The Pharmacophore-Fit scoring function only considers pharmacophoric features and the feature RMS deviation. The Relative Pharmacophore-Fit scores the number of matching pharmacophore features and the RMSD of the pharmacophore alignment normalized to [0..1]. If you want to consider steric properties, you can choose the atom sphere overlap (fast shape) or Gaussian shape similarity scoring functions. The first implements a fast atom sphere counting algorithm to provide an overlap score. The latter, uses Gaussian functions to approximate atom spheres. By doing so the overlap can be calculated analytically.
(Default: Pharmacophore-Fit and atom overlap)
Feature Tolerance Scaling Factor
This value is multiplied to the pre-defined feature tolerances. Lowering this scale factor would end up in more restrictive pharmacophoric models.
(Default: 1.0)
Merged Omitted Fatures
This setting specifies the number of features that need not match all input molecules if Merged feature pharmacophore type is turned on.
(Default: 3)
Optional Feature Threshold (%)
If this threshold is exceeded, e.g. more than 10% of the molecules do not match a feature, the feature is declared as optional, otherwise it is handled like a normal feature. The setting is only actually used if 'Mark partially matching features optional' is enabled.
(Default: 10)
Disable Excluded Volumes
Excluded Volumes are not generated.
(Default: disabled)
Mark partially matching features optional
The 'Optional Feature Threshold (%)' setting is used to mark respective features optional.
(Default: enabled)
Treat clusters individually
Molecules are grouped by their Cluster IDs and pharmacophore models are created for each cluster containing at least 2 molecules. An appropriate column (column title: Cluster ID) specifying the Cluster IDs has to be present in the input table.
(Default: disabled)
Use matching conformation for molecule output
If enabled, the molecules from the output table are given in aligned state to the respective created pharmacophore model.
(Default: enabled)

Input Ports

Table containing training-set molecules and possibly test-set and/or ignored-set molecules. If no column is present specifying the type of set, molecules are included in the training-set per default behaviour
Table containing one pharmacophore to be used as bias. In case this table contains multiple pharmacophores, the last one is used. If separation per cluster is activated, this table also needs to contain cluster information. For each cluster, the last pharmacophore in the input table with the respective Cluster ID is then used. In case there is no pharmacophore with the according Cluster ID for a specific sufficiently large molecule cluster, no bias is set

Output Ports

Table containing the pharmacophore model(s)
Table containing the input training and test set molecules in aligned state

Best Friends (Incoming)

Best Friends (Outgoing)


To use this node in KNIME, install LigandScout Extensions for the KNIME Workbench from the following update site:


You don't know what to do with this link? Read our NodePit Product and Node Installation Guide that explains you in detail how to install nodes to your KNIME Analytics Platform.

Wait a sec! You want to explore and install nodes even faster? We highly recommend our NodePit for KNIME extension for your KNIME Analytics Platform. Browse NodePit from within KNIME, install nodes with just one click and share your workflows with NodePit Space.


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.