Screening Settings Dialog Box
Make settings for various aspects of screening a set of ligands against Phase pharmacophore hypotheses.
To open this dialog box, click the Screening Settings button in the Phase Ligand Screening panel.
- Features
- Additional Resources
Screening Settings Dialog Box Features
Conformers tab
Set options for the conformers to use in the screening. If the structure source has conformers already generated, you can use them if you want, or regenerate them during the search.
- Use existing conformers options
-
With this option, the search is performed with existing conformers. When searching a file or Project Table entries, consecutive structures with identical titles and connectivities are treated as conformers of a single molecule.
- Score in place option
-
Score structures without doing any alignment to the hypothesis.
- Refine matches option
-
Generate extra conformers for the top-ranked match from each molecule matched. These conformers are subjected to the hit filters and are returned in the hit file. This procedure improves the Phase Screen (fitness) score, and can return matches when excluded volumes eliminate every match for some molecules. When you select this option, the Conformer generation and refinement section is activated, so you can set parameters for the extra conformers.
- Generate sites during search option
-
Generate the sites during the search, rather than use preexisting sites. You should select this option if the feature definitions for the hypothesis differ from those in the database. This option is only available if you are searching a database, where the sites are already present. For other input sources, sites are generated automatically. These sites are not saved, but only used during the search.
- Perform pre-screen using database keys option
-
Perform a pre-screening of the database using database keys, to speed up the search. Use of these 3D keys rapidly filters out the majority of molecules that cannot possibly match the hypothesis. This option is recommended except when searching a small subset of the database or when the search is split across a large number of CPUs. The option is only available when you select Generate sites during search.
- Generate conformers during search option
-
With this option, the conformers are generated as each structure is read, and site points are generated automatically along with the conformers.
Features tab
Make settings for the way in which similar features are processed.
- Feature presets option menu
-
Set options on this menu for how certain related features are treated in screening. When you set any of these options, vector scoring is turned off for all of them, so the direction of the features makes no contribution to the scoring.
- Make hydrophobic and aromatic rings equivalent—Treat aromatic rings as if they were hydrophobic features.
- Make acceptor and negative equivalent—Treat acceptors as though they were negatively charged features.
- Make donor and positive equivalent—Treat donors as though they were positively charged features.
Scoring tab
Choose the scoring function, and if a custom function is chosen, set up the function.
- Scoring function option menu
-
Choose the scoring function to use from this option menu.
- Phase Screen Score—Use the default Phase Screen scoring function.
- Custom—Set up and use a custom scoring function, by changing the weights and alignment cutoff in the Phase Screen scoring function. The tools for setting up the scoring function are displayed below the option menu.
- Consider atom types when computing volume scores option
-
Compute volume scores using only overlaps between atoms of the same MacroModel atom type. This option favors alignments that superimpose chemically similar atoms.
- Scoring function section
-
Define the formula for scoring the hits when a custom scoring function is chosen. The score is composed of several geometric scores. For each geometric score, a weight can be specified in the scoring formula. The default weights (1.0) and cutoff (1.2 Å) define the Phase Screen score.
- site score
-
This score measures how closely the features in the structure are superimposed on those of the hypothesis. The site score (the first line of the scoring formula) is calculated from the alignment score by
site score = 1.0 − (alignment score)/(alignment cutoff)
The alignment score is the RMS deviation of the position of the features in the ligand from those of the hypothesis. The default cutoff is 1.2 Å, and can be set in a text box. If the alignment score is greater than the cutoff, then the site score is negative.
Possible range is 1.0 or less. The score is 0.0 at the RMSD cutoff and 1.0 for perfect alignment.
- vector score
-
This score measures how well the vectors for acceptors, donors, and aromatic rings in the structure are aligned to those in the hypothesis. The vector score is the average cosine of the angles formed by corresponding pairs of vector features (acceptors, donors, and aromatic rings).
Possible range is from -1.0 (perfect anti-alignment) to 1.0 (perfect alignment).
- volume score
-
This score measures how much the volume of the ligand overlaps with the volume of the reference ligand of the hypothesis when aligned to the hypothesis. The volume score is the overlap of the volume of the aligned ligand with that of the reference ligand, divided by the total volume occupied by the two ligands. If there is no reference ligand, the score is zero.
Possible range is 0.0 (no volume overlap) to 1.0 (perfect volume overlap).
- included volume score
-
This score measures how much the volume of the ligand overlaps the included volume of the hypothesis, if the hypothesis has an included volume defined. The individual volume score is the overlap of the volume of the ligand with that of the included volume, divided by the total volume occupied by the ligand and the included volume when the ligand is aligned to the hypothesis. If there is no included volume, the score is zero.
Possible range is 0.0 (no volume overlap) to 1.0 (perfect volume overlap).
Constraints tab
Specify distance, angle, and dihedral constraints for combinations of individual features in the hypotheses used in the screening. To add constraints to a hypothesis, the hypothesis must be displayed in the Workspace.
- Pick features option and text box
-
Pick hypothesis features in the Workspace to define a constraint. Two sites define a distance constraint, three sites define an angle constraint, and four sites define a dihedral constraint. You must pick them in the desired order for an angle or dihedral constraint. The features are listed in the text box.
- Add Constraint button
-
When you have picked the desired features, click this button to add them to the table as a constraint. A new table row is created and populated.
- Constraints table
-
Table of constraints that have been defined. You can select a single row to change the sites (by clicking Edit) or select multiple rows to delete.
Hypothesis Name of hypothesis for which the constraint is defined. If you are screening with multiple hypotheses, you can define constraints for each hypothesis. Noneditable. Features List of features that define the constraint, separated by hyphens. Two sites define a distance constraint, three sites define an angle constraint, and four sites define a dihedral constraint. Type Lists the constraint type: distance, angle, or dihedral. Noneditable. Use Column of checkboxes. Checking a box marks the constraint for use in screening, and displays markers for the constraint in the Workspace. Value Target value of the distance, angle, or dihedral that is being constrained, in angstroms or degrees. You can edit the value by double-clicking in the cell and changing the value. Tolerance Tolerance to be applied when matching the constraint, in angstroms or degrees. You can edit the tolerance by double-clicking in the cell and changing the value. - Delete and Delete All buttons
-
Delete the constraint that is selected in the table or delete all constraints.
Groups tab
Select features in the hypotheses that you want to group for screening. The groups are specified by picking features on the hypotheses that are displayed in the Workspace. The features in each group must be selected from a single hypothesis and cannot contain features that are already in a group.
- Pick features option and text box
-
Pick the features to be grouped in the Workspace. The features are listed in the text box, which is noneditable. You can only pick features that are not already in a group.
- Create Group button
-
Create the feature group. A row is added to the table for the group. This button is only enabled when more than one feature has been picked.
- Groups table
-
Table listing the feature groups. You can select a single row in the table to operate on.
Hypothesis Name of hypothesis for which the group is defined. If you are screening with multiple hypotheses, you can define groups for each hypothesis. Noneditable. Group Index of the group. The index starts from 1 for each hypothesis. Grouped Features List of features in the group. Must Match Option menu that allows you to specify the minimum number of the features in the group required to match during a screen. - Delete and Delete All buttons
-
Delete the group that is selected in the table; delete all groups from the table (clear the groups).
Criteria tab
Set criteria for rejecting ligands.
- Match rejection criteria section
-
Set criteria for rejecting ligands based on geometric matching.
- Use default option
-
Use the default criteria for rejecting ligands.
- Use custom rejection criteria option
-
Specify the criteria for rejecting ligands based on alignment and volume scores. For information on these scores, see Scoring function section.
- Reject ligands with options and text boxes
-
Select options to reject ligands based on their scores and specify the score cutoffs used to reject ligands. The options are:
- alignment score
- vector score
- volume score
- included volume score
- Intersite distance matching tolerance option and text box
-
If this option is selected, apply the tolerance for matching intersite distances specified in the text box. If the difference in distance is less than this value, the distance is considered to match.
- Limit CPU time for matching to N seconds/molecule option and text box
-
Select this option and enter a time in the text box to restrict the amount of CPU time used per molecule in the search. Limiting the CPU time is useful when doing partial matching for hypotheses with many sites, as the time spent finding matches to a given molecule may become very large due to the number of possible combinations.
Output tab
Set options for the amount and ordering of hits and the file type.
- Sort hits by decreasing Phase Screen Score option
-
Sort the hits by decreasing values of the Phase Screen score. If this option is not selected, hits are written out in the order that they are encountered in the input structure source.
- Return at most N hits per molecule option
-
Limit the number of hits returned per molecule to the specified value. These hits are passed into the sort, if sorting is selected. Some molecules can return more than one hit because different alignments or different conformers might match the hypothesis.
- Structure output section
-
Specify the format for the output structure file.