Data Fusion Panel

Run three separate virtual screening methods (Glide, Shape Screening, and Fingerprint Screening) and combine the results using a data fusion approach based on standard scores (Z-scores), to select the top compounds.

To open this panel: click the Tasks button and browse to Receptor-Based Virtual Screening → Data Fusion.

  • Features
  • Additional Resources

Data Fusion Panel Features

Tabs
  • Input
  • Glide Docking
  • Phase Shape
  • Canvas 2D Fingerprints
  • Results

Input tab

Query structure options

For shape screening and fingerprint screening, a query structure is required. Select an option to identify the query structure.

  • Workspace—Use the structure in the Workspace. Only one structure must be included in the Workspace.
  • Project Table (selected entry)—Use the entry that is currently selected in the Project Table. Only one entry must be selected.
  • File—Use the first structure from the specified file. When this option is selected, the File name text box and Browse button are activated. Click Browse and navigate to the file you want to use. The file name is displayed in the text box when you click Open in the file selector. You can also enter the file name in the text box.
Screen ligands text box and Browse button

Specify the file that contains the ligands to be screened. You can enter the file name in the text box, or click Browse and navigate to the file in the file selector that opens.

Glide Docking tab

In this tab you make settings for Glide docking.

Glide grid file text box and Browse button

Specify the Glide grid file (.zip) to use for docking. You can enter the file name in the text box, or click Browse and navigate to the file in the file selector that opens.

Use constraints from the grid file option

If the grid file contains receptor-based constraints, use all of these constraints for docking.

Precision options

Select the docking precision, from HTVS, SP, and XP. The docking precision is explained in the Settings Tab section of the Ligand Docking Panel DEPRECATED topic.

Amide bonds option menu

Choose an option for treatment of amide bonds during docking.

  • Penalize non-planar conformation—penalize amide bonds that are not cis or trans (default)
  • Vary amide bond conformation— allow the conformation around amide bonds to vary according to the force field, without additional constraints
  • Retain original amide bond conformation— freeze amide bonds in their input conformation throughout docking
  • Allow trans conformers only—enforce trans conformation within a small angle range (20°)
Perform post-docking minimization option

Perform a minimization of the poses following the docking. The minimization optimizes bond lengths and angles as well as torsional angles, and rescores the poses using the scaled Coulomb-van der Waals term and the GlideScore.

Apply strain correction terms option

Apply strain correction terms when doing the final scoring. These terms are evaluated by optimizing each ligand pose as a free ligand, first with constraints on all torsions, then without these constraints. The difference is used to compute a penalty term for unreasonably high strain. This option is only available if Perform post-docking minimization is selected.

Phase Shape tab

In this tab you make settings for screening the ligand set against the query by shape and atom or pharmacophore type.

Volume scoring options

Choose the type of object that is used for volume scoring:

  • Pharmacophore types—Treat each structure as a collection of pharmacophore sites, whose locations are assigned by applying Phase pharmacophore feature definitions. Overlapping volumes are computed only between sites of the same feature type. Each site is represented by a sphere of radius 2 Å.
  • Atom types—Treat each structure as a collection of atoms, whose volumes are defined by their van der Waals radii. Overlapping volumes are computed only between atoms of the same type, which can be selected from the Atom types for volume scoring option menu.
Atom types for volume scoring option

Choose the types of atoms between which volume overlaps will be calculated. Volume overlaps are computed between atoms that have the same atom type as defined by the choice of atom typing scheme below. This option menu is only available if you choose Atom types for volume scoring.

  • MacroModel—Calculate volume overlaps only between atoms that have the same MacroModel atom type.
  • Elements—Calculate volume overlaps only between atoms of the same element.
  • QSAR—Calculate volume overlaps between atoms that have the same pharmacophore type (Acceptor, Donor, etc.) as defined for Phase QSAR models.
  • None—Don't distinguish different types of atoms when calculating volume overlaps: all atoms are treated the same.
Generate conformers option

Run a conformational search for the molecules being screened, using ConfGen. This option is not needed if the input molecules already consist of contiguous conformer sets.

Existing conformers options

Select an option for handling existing conformers:

  • Discard—Replace all existing conformers and use only the generated conformers. This includes the input structure, if there is only one.
  • Keep—Append the generated conformers to the existing set of conformers (which might be only one structure). Select this option to guarantee that the input structure is included in the search.
Maximum number of conformers text box

Enter the maximum number of conformers per input molecule to be generated.

Retain up to N conformers per rotatable bond text box

Enter the number of conformers to keep per rotatable bond for each input molecule.

Amide bonds text box

Choose the treatment of amide bonds to be used in the conformational search, from the following:

  • Vary conformation—allow the conformation around the amide bond to vary freely.
  • Retain original conformation—do not vary the conformation around the amide bond, but keep the original conformation.
  • Set to trans—Set the conformation around the amide bond to trans (180°).

Canvas 2D Fingerprints tab

In this tab you make settings for the generation and comparison of structure-based hashed fingerprints, which are used for screening the ligand set against the query.

Fingerprint type option menu

Select the type of fingerprint you want to generate. The available types are:

  • Linear—Linear fragments + ring closures.
  • Radial (ECFP)—Fragments that grow radially from each atom. Also known as extended connectivity fingerprints.
  • Dendritic—Linear and branched fragments.
  • MOLPRINT2D—A radial-like fingerprint that encodes atom environments using lists of atom types located at different toplogical distances.
  • Atom pairs—Pairs of atoms, differentiated by type, and the distance separating them.
  • Atom triplets—Triplets of atoms differentiated by type, and the three distances separating them.
  • quartet—Quartets of atoms differentiated by type, and the six distances separating them.
  • Topological torsions—Linear paths of length three.
Atom typing scheme option menu

Select the atom and bond typing scheme to be used to classify atoms and bonds in the structures when generating fingerprints. Twelve schemes are available, and are described on the option menu.

Similarity metric option menu

Choose the similarity metric used for scoring the matches to the query. Twenty-four metrics are available, and are described in Table 5.7 in the Canvas User Manual.

Results tab

In this tab you choose how to combine the results of the three methods into a score that is used to select the top compounds.

Compute average Z-score using options

Choose how to determine the score for each ligand on which the selection of the best ligands is done. The scores for each method are converted to Z-scores (i.e. normalized) so that they are on the same scale.

  • Best Z-score only—For each ligand, use the best Z-score obtained for any of the three screening methods as the score.
  • Best 2 Z-scores—For each ligand, use the average of the best two Z-scores obtained for the three screening methods.
  • All Z-scores—For each ligand, use the average of the Z-scores obtained for all three screening methods.
Return the top options

Return the top number or percentage of compounds. A compound is detected by its title, and may consist of conformers, tautomers or ionization states of a structure. The compounds are compared by their averaged Z-score.

Job toolbar

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

The Job Settings button opens the Data Fusion - Job Settings Dialog Box, where you can make settings for running the job.

Status bar

The status bar displays information about the current job settings and status for the panel. The settings includes 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.

Use the Reset button to reset the panel to its default settings and clear any data from the panel. You can also reset the panel from the Job toolbar.

The status bar also contains the Help button , which opens the help topic for the panel in your browser. If the panel is used by one or more tutorials, hovering over the Help button displays a button, which you can click to display a list of tutorials (or you can right-click the Help button instead). Choosing a tutorial opens the tutorial topic.

Related Topics