Ligand Docking Panel DEPRECATED

The Glide Ligand Docking panel is used to set up and run docking jobs using previously calculated receptor grids. To generate receptor grids, see the Receptor Grid Generation Panel topic.

To open this panel: click the Tasks button and browse to Receptor-Based Virtual Screening → Ligand Docking (Deprecated).

  • Using
  • Features
  • Additional Resources

Using the Ligand Docking Panel

Glide ligand docking jobs require a set of previously calculated receptor grids and one or more ligand structures. Preparation of the ligands before docking is strongly recommended. LigPrep or MacroModel can be used to prepare ligands. If a correct Lewis structure cannot be generated for a ligand, it will be skipped by the docking job.

If you want to dock a set of ligands using a progression of precision, you can use the Virtual Screening Workflow to set up and run the docking jobs. See Virtual Screening Workflow for details.

To write out the input file and a script for running the job from the command line, click the arrow next to the Settings button and choose Write. For information on command usage and options, see glide Command Help. See also Running Glide from the Command Line.

Specifying Ligands and Settings

Use the Ligands tab of the Ligand Docking panel to specify the source of ligands to be docked or scored, set size limits for skipping ligands, and optionally to change the settings for van der Waals radii scaling of nonpolar ligand atoms.

Use the Settings tab to define the basic options for docking ligands using previously calculated grids, such as selecting the type and precision of docking job you want to run and making settings for ligand sampling.

Applying Constraints

You can add constraints that the ligand poses must satisfy, to receptor features, for superposition to a reference ligand core or shape, or on ligand torsions, in the Constraints tab.

Constraints are applied during or after docking by identifying the relevant features in the ligand, and requiring specified atoms in those features to be spatially confined to the constraint region. The ligand features and the atoms in those features that must be constrained are defined in terms of SMARTS patterns. Constraints can be applied to features in a reference ligand, or in the receptor, or on ligand torsions.

Glide constraints can be used in HTVS mode, but may result in poor pose recovery relative to unconstrained HTVS docking.

Only the constraints you select will be applied in the ligand docking job you are setting up. If there are no constraints selected when you start the docking job, no constraints will be applied.

Receptor Constraints

The receptor constraints can be set in groups, up to a maximum of four groups. Each group of constraints is applied to the ligand (a Boolean AND is applied between groups). Within each group, you can require all constraints to be satisfied, or you can set a minimum number of constraints in the group that must be satisfied. In this way you can define required or optional constraints within each group. Overall, there must be 4 or fewer required constraints, across all groups. For simple constraints, you can just use Group 1.

For example, suppose the grid contains a hydrogen bond constraint hbond1 and two positional constraints pos1 and pos2. If you want to enforce the hydrogen bond constraint, and require only one of the two positional constraints, you would click the Use check box for hbond1 in Group 1 and select All under Must match; and click the Use check boxes for pos1 and pos2 in Group 2, select At least under Must match, and enter 1 in the text box. The resulting constraints can be represented as “hbond1 AND (pos1 OR pos2)”.

The definition of a constraint includes the ligand feature to which it matches. These features are defined in terms of SMARTS patterns. To edit the ligand feature definitions, select the table row for the constraint and click Edit Feature. The Edit Feature Dialog Box DEPRECATED opens, allowing you to select the feature type; import or export feature sets; and add, edit or delete patterns. Each constraint can have its own feature definition, so you can have a different definition of a given feature type for each constraint. However, the same feature definition from the same set is used for a given constraint in all groups.

To use a different combination of the available constraints, run another docking job and select the desired constraints.

Glide constraints can be used only in flexible docking jobs. (Ensure that Dock flexibly is selected in the Settings tab.) Glide constraints can be used in HTVS mode, but may result in poor pose recovery.

Note: If you include positional constraints, you must define the ligand feature that it should match. No default feature definition is provided, and the job cannot be run until the feature is defined. To define the ligand feature, select the appropriate row in the Available constraints table and click Edit Feature. The feature can then be defined in the Edit Feature Dialog Box DEPRECATED.

If your desired constraint specification cannot be put in the general form (N1 required from Group 1) AND (N2 required from Group 2) AND (N3 required from Group 3) AND (N4 required from Group 4), you might be able to achieve your goal by running more than one docking job with a separate constraint specification for each. For example, if you want to apply the constraints “(hbond1 AND pos1) OR (hbond2 AND pos2) OR (hbond3 AND pos3)”, you could run three separate docking jobs, one with (hbond1 AND pos1) set, one with (hbond2 AND pos2) set, and one with (hbond3 AND pos3) set.

Torsional Constraints

There are situations in which you want to constrain some of the torsional degrees of freedom in the ligand. For example, a ligand in the binding site might have only one conformation of a particular rotatable group, while other groups can exist in several conformations. Or the ligand might have a large number of rotatable bonds, such as in a polypeptide. Another example is when your ligand set contains atropisomers. The constraints can be applied to prevent sampling of the torsion that is responsible for the atropisomerism.

The groups that are constrained are defined in terms of SMARTS patterns, and you can constrain all of the torsions in the group or only selected torsions in the group.

To set torsional constraints, first you must define a SMARTS pattern that matches the group on the ligands that you want to constrain (and does not match groups that you do not want to constrain), and then define the torsion or torsions within that SMARTS pattern whose values are to be constrained. You can define more than one group to constrain, and more than one torsion in each group. The basic procedure is:

  1. Pick atoms in the Workspace ligand that define the group.

  2. Click New.

    The New Torsion Pattern dialog box opens.

  3. Click From Workspace Selection.

    The SMARTS pattern is entered into the text box.

  4. Click OK.

    The SMARTS pattern is added to the Patterns table and marked on the Workspace ligand.

  5. Select an option for the torsions to constrain: All torsions, or Specified torsions.

    If you chose Specified torsions, continue with the steps below.

  6. Pick four atoms in the SMARTS pattern on the Workspace ligand to define a torsion for the pattern.

    The atoms must be in adjacent bonds and not all in a ring. The torsion is added to the torsions table along with the current value of the torsional angle.

  7. Choose whether to set the angle to a fixed value for all ligands or to fix the angle independently at the input value for each ligand. You set the angle to a fixed value by selecting the check box in the Set angle column of the table, and editing the Angle table cell to set the desired torsional angle.

Running the Job

When you have chosen settings for the options in the tabs, you can click the Settings button in the lower left corner of the panel to make settings for the job (or click the arrow next to the button and choose Job Settings).

This opens the Ligand Docking - Job Settings Dialog Box . Along with the standard tools, this dialog box includes tools for distributed processing: options and text boxes for specifying the number of subjobs or the maximum number of ligands per subjob, and a table of hosts, in which you can select hosts and specify the number of processors to use on each. The ligands are divided evenly between the subjobs when the job is run.

  • To separate the docking job into a specific number of subjobs, select Exactly under Separate job into and enter the number of subjobs in the text box.

  • To separate the docking job into subjobs of a maximum size (number of ligand), select Subjobs with no more than and enter the maximum number of ligands in the text box.

  • To separate the docking job into subjobs of an optimal size (the default), select Recommended number of subjobs.

  • To use multiple processors, select the hosts and enter number of processors to use on each host in the host table. You can edit the cells in the Use column to set the number of processors to use for each host..

The number of subjobs cannot be set to fewer than the number of processors: if you specify more processors, the number of subjobs is adjusted upward to match. For optimal load balancing and for restartability, the number of subjobs should be several times the number of processors—even if you are running on a single processor. It is recommended that you split any docking job into subjobs with no more than about 50,000 ligands, to minimize the loss of work in the event of a job failure (which could be due to network problems or hardware failure). Glide tries to restart failed subjobs, and if there is only one subjob, it restarts the failed job from the beginning.

You can incorporate the docking results into the project by choosing from the Incorporate option menu; however, incorporating a large number of ligands can take a long time.

Default values are used for all other command-line options. For example, the range of ligands that is divided into subjobs is the entire range of ligands to dock as specified in the Range text boxes in the Ligands tab. If you are using the default range (dock all ligands in the file), glide creates j subjobs starting with the first ligand in the file and ending with the last. If you have specified multiple files, all ligands in all files are distributed among the subjobs. To use different settings for this and other options, run glide from the command line, as described in Running Glide from the Command Line.

If the grid files are not available on the host from which you submit the job, the job will not be submitted, as a check is done to confirm the availability of the grid. If you want to submit a job with a grid that is not available on the local host, you can do so from the command line.

Alternatively, you can choose Write from the Settings button menu to write the input files without starting the job. In addition to the input file, a shell script is written that you can use to run the job. The script is written with the current settings from the Job Settings dialog box. The files are written to a subdirectory in the current directory. Both the subdirectory and the files are named using the job name.

If you want to upload the input to LiveDesign as a Glide Docking model, choose Upload to LiveDesign from the Settings button menu. There must be a Glide Docking model protocol installed on the LiveDesign server to do this. If you are not already logged on to the LiveDesign server, you will be prompted to do so. The grid and the input file (as defined by the panel settings) are then uploaded to the server. If a ligand is specified in the Core tab, it is also uploaded as the reference ligand.

To restore the default settings in all tabs, choose Reset Panel from the Settings button menu.

Post-processing

The options in the Output tab control the final output of ligand poses that pass successfully through Glide's various scoring stages.

When the docking results are incorporated into the project, you can use the Workflow Action Menu for the entry group in the Entry List (Entries) to choose an action to apply to the results.

Ligand Docking Panel Features

Receptor grid option menu

Choose a source for the receptor grid. The option menu lists the most recently used grids, as well as From file so you can enter the grid file name or browse to it. The grid file must be accessible to the host on which you are running Maestro.

If you plan to use SP-Peptide docking, you must select a grid that was generated for this docking mode (with Generate grid suitable for peptide docking selected in the Receptor tab of the Receptor Grid Generation Panel). If you plan to use any of the other three docking modes, you must not select a grid generated for SP-Peptide docking.

If you want to use grids from earlier releases in which the grid was not compressed, you can also choose the .grd file for the grid to use that grid.

Display receptor option

Display the receptor in the Workspace.

Show grid boxes option

Display the outer (purple) and inner (green) grid boxes in the Workspace, and the grid center. These are the boxes that are displayed when you set up the grid calculation. The box parameters are stored in the Maestro file for the receptor.

File name text box and Browse button

Enter the grid file name in this text box, or click Browse and navigate to the grid file. The name of the grid file you selected is displayed in the text box.

  • Ligands
  • Settings
  • Constraints
  • Output

Ligands Tab Features

Ligands to be docked section

Use this section to specify the source of ligands to be docked (or scored in place), the partial charges to use, and to set size thresholds for skipping ligands.

For accurate docking, the ligands you specify must satisfy these four conditions:

  1. Ligands must be three-dimensional (3D) structures.
  2. Ligands must each consist of a single molecule with no covalent bonds to a receptor and no accompanying fragments such as counter ions or solvent molecules. Glide automatically skips fragmented ligands, e.g., salts with counter ions present.
  3. Ligand files must be in Maestro, SD, PDB, or MOL2 format; or provided as SMILES strings in .smi or .csv files, which are automatically converted to 3D and prepared with LigPrep. Ligand files in other formats can be converted by importing them into Maestro and exporting them in Maestro format. If you wish, you can specify a compressed Maestro file (.mae.gz or .maegz) or SD file (.sdf.gz or .sdfgz).
  4. Ligands must have all their hydrogens (filled valences). These can be added in Maestro using the Build toolbox or Edit → Add Hydrogens.

Ligand preparation, for example using LigPrep, is strongly recommended before Glide ligand docking. See Ligand Preparation for Glide.

Glide automatically skips ligands containing unparametrized elements, e.g., tin, or atom types not supported by the OPLS force fields, such as explicit lone pair "atoms".

Use ligands from option menu

Choose the source of the ligands from this option menu, from the following options:

  • Workspace (included entries)—Dock the structures in the Workspace. If you select this option, the entries in the Workspace must all be valid ligands, as described above. You must ensure that the structures in the Workspace satisfy the requirements.

  • Project Table (selected entries)—Dock ligands that are selected entries in the Project Table. For more information about the Project Table and entry selection, see the Project Table Panel topic.

  • Files—Dock ligands from one or more files. Enter the file names in the File name text box, or click the Browse button to navigate to and select the ligand files.

    The files can be in any of the supported formats (Maestro, SD, MOL2, PDB, SMILES), but you can only have files of one format in a single job. By default, all structures in each file are docked. LigPrep is run automatically on the generated 3D structures from the SMILES input. You can specify LigPrep options by clicking the Preparation Options link and making the settings in the Ligand Preparation Options Dialog Box.

    If you select a single file, you can specify a particular range of ligands to dock, using the Range controls:

    • To start at a ligand other than the first, edit the value in the first text box.
    • To end at a value other than the last, deselect End and edit the value in the second text box.

    If a Glide job run with a single ligand file terminates abnormally, you can set the initial ligand number to pick up after the point in the input ligand file at which the problem occurred.

  • Phase Database—Extract the ligands from a Phase database. Specify the path to the database or click Browse and navigate to the database in the file selector that opens. Phase databases have the extension .phdb. Older databases (ending in _phasedb) must be converted first, with phase_database.

    You can restrict the ligand set to the subset of the database specified in the Subset text box. Enter the path to the subset or click Browse and navigate to the database subset in the file selector that opens. The subset has the extension _phase.inp.

Use input partial charges option

Select this option to use partial charges from the input structures instead of those from the force field. This option enables you to use improved partial charges, for example from a Jaguar calculation. If you select this option with SMILES input, the input partial charges are the ones generated by LigPrep.

Do not dock or score ligands with more thanaatoms

Set the maximum number of atoms a ligand structure may have if it is to be docked. Ligand structures in the input file that have more than the specified number of atoms are skipped. The default (and maximum) is 500 atoms. You can reduce the maximum number of atoms a, if the active-site region is small and enclosed, to speed up a docking calculation on a large ligand database.

Do not dock or score ligands with more thanrrotatable bonds

Set the maximum number of rotatable bonds a ligand structure may have if it is to be docked flexibly. Ligand structures in the input file that have more than this number of rotatable bonds are skipped. The default (and maximum) number is 100 rotatable bonds. If only relatively small or rigid ligand hits are wanted, you can decrease the value of r. If you use torsional constraints, the rotatable bonds that are constrained are excluded from the number of rotatable bonds. The text box is not available if you have selected Refine or Score in place in the Settings tab.

Scaling of van der Waals radii section

Glide does not generally allow for flexible receptor docking, except for reorientation of selected hydroxyl and thiol groups (see the Rotatable Groups tab in the Receptor Grid Generation Panel topic). If you want flexibility in the receptor, you can use induced fit docking (see Induced Fit Docking). However, successful docking sometimes requires that the ligand or the receptor "give" a bit in order to bind. To model this behavior, Glide can scale the van der Waals radii of nonpolar atoms (where nonpolar is defined by a partial charge threshold you can set), thereby decreasing penalties for close contacts. By default, scaling is performed for qualifying atoms in the ligand, but not those in the receptor. Ligand atom radii scaling settings can be changed using the options in this section.

To scale receptor atom radii, you must choose the appropriate options in the Receptor tab of the Receptor Grid Generation Panel prior to grid generation.

Scaling factor text box

The Scaling factor text box specifies the scaling factor. The default is 0.80. To turn van der Waals radii scaling off, set the scaling factor to 1.0. Full penalties for close contacts of nonpolar ligand atoms will then be used.

Partial charge cutoff text box

Scaling of van der Waals radii is performed only on nonpolar atoms, defined as those for which the absolute value of the partial atomic charge is less than or equal to the number in the text box. Since this is an absolute value, the number entered must be positive. The default for ligand atoms is 0.15.

Settings Tab Features

Precision option menu

Choose a docking precision option:

HTVS (high throughput virtual screening)

High-throughput virtual screening (HTVS) docking is intended for the rapid screening of very large numbers of ligands. HTVS has much more restricted conformational sampling than SP docking, and cannot be used with score-in-place. Advanced settings are not available for HTVS, but are fixed at predetermined values.

SP (standard precision)

Standard-precision (SP) docking is appropriate for screening ligands of unknown quality in large numbers.

SP-Peptide

Standard-precision docking for peptide ligands uses the same general settings as for regular standard precision but changes some of the settings to enhance the retention of poses. Specifically, it keeps 100000 poses in the initial docking stage, and uses 1000 poses per ligand for energy minimization, and sets the number of poses for post-docking minimization to 100. Although this option was designed for peptides, it is not restricted to peptides: you can dock any kind of ligand.

If you select this option, the grid must be one that was generated for this mode, i.e. with Generate grid suitable for peptide docking selected in the Receptor tab of the Receptor Grid Generation panel.

Ligand sampling option menu and options

The Ligand sampling option menu allows you to choose whether ligands are docked flexibly, rigidly, or not at all (score in place). Extra options are available for flexible docking.

Flexible

This is the default choice and directs Glide to generate conformations internally during the docking process; this procedure is known as "flexible docking". Conformation generation is limited to variation around acyclic torsion bonds, sampling of low-energy ring conformations, and generation of pyramidalizations at certain trigonal nitrogen centers, e.g. in sulfonamides. For a set of predefined functional groups, such as amides and esters, you can bias sampling of the torsion around the bond that normally adopts a particular conformation so that it adopts the desired conformation.

There are several options for conformation generation with this choice:

Sample nitrogen inversions option

Sample inversions at pyramidal nitrogen atoms (not amides). This option is selected by default.

Sample ring conformations option

Sample the conformations of rings, using the same technology as in LigPrep—see Ring Conformations: ring_conf for details. These conformations are not sampled in the main conformation generation, which focuses on sampling of rotatable bonds, leaving the core fixed. Deselect this option if you want rings to remain in their input conformations throughout docking.

This option is selected by default. The two following options are made available if this option is selected.

Sample macrocycles using Prime option

Sample the conformations of macrocycles using the Prime macrocycle sampling code (see Prime Macrocycle Sampling Panel), with special options for Glide docking. These conformations are not sampled in the main conformation generation, which focuses on sampling of rotatable bonds, leaving the core fixed. This option changes settings for the number of poses in various stages of the docking process.

Sampling of macrocycles can take from a few minutes to an hour per ligand. The job is split into subjobs for each individual ligand., which can be distributed over multiple processors. Docking results are only generated for macrocyclic ligands: non-macrocycles are skipped.

If you want to split your input file into macrocycles and non-macrocycles before docking, you can use the following command in a terminal window (Linux or Mac, prepended with $SCHRODINGER/) or a Schrodinger shell (Windows):

run -FROM glide split_macrocycles.py

Run the command with the -h option for the full syntax, including description of input and output files.

This option is deselected by default, and is only available if Sample ring conformations is selected. This option requires a Prime license and a Macrocycle license in addition to a Glide license.

Include input ring conformation option

Select this option if you want to include the input conformations of rings in addition to other conformations. The conformational sampling uses the input conformation as a seed, and does not necessarily include it in the set of conformations returned.

This option is deselected by default, and is only available if Sample ring conformations is selected.

Bias sampling of torsions for options

Choose an option for sampling of torsions that should normally be restricted to a particular conformation. The biasing can include retention of the input conformation, setting the torsion to a particular value, applying a penalty for deviating from the desired conformations, or allowing only a particular conformation to within a small angle range. The options cover different selections of functional groups:

Predefined functional groups

Bias the sampling of torsions for a set of functional groups that is defined in a resource file (default). The choice of biasing method, as outlined above, is set in the resource file. The resource file can be customized—see Customizing Torsional Controls for Docking Planar and Other Groups.

Amides bonds only

This option applies constraints or penalties to rotation around amide C–N bonds. The option menu provides a choice of the constraint type:

  • Penalize non-planar conformation—penalize amide bonds that are not cis or trans (default)
  • Retain original conformation— freeze amide bonds in their input conformation throughout docking
  • Allow trans conformation only—enforce trans conformation within a small angle range (20°). Ligands that do not dock in this range are rejected.
None

Do not penalize or constrain rotations about certain bond types, but allow them to adopt a conformation according to the force field.

Rigid

Rigid docking allows the existing ligand structure to be translated and rigidly rotated relative to the receptor, but skips the conformation generation step.

None (refine only)

With this option, the input ligand structure does not pass through the Glide docking procedure, but the input coordinates are used to perform an optimization of the ligand structure in the field of the receptor, and then the ligand is scored. The goal of this docking method is to find the best-scoring pose that is geometrically similar to the input pose. For HTVS and SP, a minimization is performed; for XP, the ligand is regrown in place. With XP mode, this option is not a substitute for a full XP docking calculation: XP mode requires accurate initial poses.

None (score in place only)

When this option is selected, Glide does no docking, but rather uses the input ligand coordinates to position the ligands for scoring. It therefore requires accurate initial placement of the ligand with respect to the receptor.

This option is useful to score the reference ligand in its cocrystallized or modeled position, or as a post-processing step on Glide-generated poses to obtain individual components of the GlideScore prediction of the binding affinity. It can also be used to check whether the scores of the known binders in their native proteins are similar enough to their scores when cross-docked to the chosen receptor protein. If this is the case, it is reasonable to expect that similar structures would also score well.

This option should not be used with Glide XP, as full XP sampling is normally needed to avoid strong XP penalties for ligands that should be able to dock correctly.

Note: you cannot use score in place for the ligand that is defined as the reference ligand for calculation of the RMSD in conformational comparisons.

Add Epik state penalties to docking score option

If the ligands have been prepared using Epik for ionization and tautomerization, the Epik penalties for adopting higher-energy states (including those where metals are present) are added to the docking score when this option is selected. Ligands that do not have this information are not penalized and will therefore have better scores, so you should ensure that the ligand set is consistent.

If the ligand interacts with a metal (distance less than 3.0 Å), the metal penalties that are computed when Epik is run with the metal binding option are used. If multiple ligand atom-metal interactions are found, the smallest value of the metal-specific penalty is used.

Use RNA receptor scoring option

Select to enable the Glide scoring function that is tuned for RNA receptors with reweighted Coulomb, vDW, and internal energy terms.

Reward intramolecular hydrogen bonds option

Add a reward for each intramolecular hydrogen bond to the GlideScore. A contribution is also added to Emodel for each intramolecular H-bond, to favor selection of poses with intramolecular H-bonds. Ligands with intramolecular hydrogen bonds pay a smaller entropic penalty upon binding, so forming intramolecular H-bonds can be important for binding.

Enhance planarity of conjugated pi groups option

Increase the torsional potential around bonds between atoms whose geometry should be planar (i.e. sp2 atoms). This option should make aromatic rings, amides, esters, and so on, less likely to adopt a nonplanar geometry. Nonplanarity of these groups is a physically reasonable effect, because the torsional potential has a finite barrier which can be overcome to some extent by other interactions. However, in Glide docking, nonplanarity can also be a result of the approximations made to reduce clashes with the receptor.

To some extent, planarity can be enforced in flexible docking by choosing one of the biasing options for sampling torsions. However, if you want to improve planarity in post-docking minimization or for non-flexible docking, you should select this option.

Apply excluded volumes penalties option and menu

If the grid has excluded volumes associated with it, select this option to apply all the excluded volumes, and choose the penalty level from the option menu. The penalty is applied if any ligand has atoms within the excluded volume. The value of the penalty ramps up from zero at the boundary of the volume to the maximum penalty at 90% of the sphere radius . The penalty is applied in both the rough scoring stage and the final docking

Show excluded volumes option

Display in the Workspace the excluded volume spheres that are associated with the grid. Only available if the grid has excluded volumes and Apply excluded volumes penalties is selected.

Advanced Settings button

Click this button to open the Settings - Advanced Settings Dialog Box DEPRECATED. This dialog box provides options for ligand conformer generation, selection of initial poses, energy minimization, and inclusion of aromatic H-bonds and halogen bonds.

Constraints Tab Features

The Constraints tab is divided into subtabs, for different kinds of constraints.

  • Receptor
  • Core
  • Shape
  • Torsional

Receptor Tab Features

The Receptor tab contains four subtabs for defining constraint groups. Each subtab has the same structure.

Show markers option

This option is selected by default. When the receptor is displayed, atoms for which H-bond/metal constraints are defined are marked with a red asterisk and padlock. Positional and NOE constraints are marked with gray translucent spheres. When a constraint is selected, the padlock and asterisk turns blue, and the cells or spheres are colored red. To undisplay the markers, deselect this option.

Group tabs

The constraints are presented in four tabs, labeled Group 1 through Group 4. The number in parentheses after the group name is the number of constraints that must be satisfied in this group, and reflects the minimum number of constraints used. This value depends on the selection made under Must match.

Each group has the full range of constraints available. You can select the same constraint in two groups.

Available constraints table

This table lists the constraints that are available for use in the docking job, and provides the means to select constraints for use. You can select a single row in the table for various actions on the constraint. The number of constraints you selected is displayed next to the table title. The table columns are described below. Any given constraint can only be used in one constraint group. If selected in one group, a given constraint will be disabled in other groups.

Use

Check box to select the constraint for use in docking. Click to select or deselect.

Name

Constraint name. This is the name defined in the Constraints tab of the Receptor Grid Generation Panel.

Receptor Constraint Type

Type of receptor constraint. Hydrogen bond constraints are classified into H-bond for hydrogen-bond acceptors and Polar Hydrogen for hydrogen-bond donors (i.e. hydrogen atoms). Note that the minimum donor and acceptor angles (90 and 60 degrees) are smaller than those used by Maestro (120 and 90 degrees).

Ligand Feature

Name of the feature type in the ligand that must match the constraint. The available feature types are: Acceptor, Charged Acceptor, Neutral Acceptor, Acceptor Including Halogens, Donor, Donor Including Aromatic H, Donor Including Halogens, Donor Including Aromatic H + Halogens, and Custom.

By default, the feature type that matches a receptor polar hydrogen or a metal is set to Acceptor; for a receptor H-bond type it is Donor, and for a positional constraint it is Custom. Custom is undefined by default, so you must edit this feature to define the patterns that match the desired ligand atoms. For donors and acceptors, the feature type is coordinated with the choice made for treatment of halogens and aromatic hydrogens in the Settings - Advanced Settings Dialog Box DEPRECATED. A change made in either place affects the setting in the other.

Edit Feature button

This button opens the Edit Feature Dialog Box DEPRECATED, in which you can select the feature type; import or export feature sets; and add, edit or delete patterns. You must use this dialog box to provide feature definitions for positional constraints.

Must match options

These options allow you to set the number of constraints in the group that must be satisfied when docking ligands.

All option

All constraints that are selected (Use column checked) must be satisfied.

At least option and text box

The number of constraints given in the text box must be satisfied. This number must be less than the number selected (Use column checked). For example, if you chose three hydrogen-bond acceptors, and you want any two out of the three to be satisfied, you would enter 2 in the text box.

The total number of constraints that are required to match must be no more than 4. This number is the sum of required constraints within each group. For example, suppose Group 1 had two constraints selected for use, and both constraints were required to match; Group 2 had three constraints selected for use, but only one was required to match; and Group 3 and 4 had no constraints selected for use. The total number of required constraints is three: two from Group 1 and one from Group 2.

The limit on the number of required constraints is enforced when you select constraints.

Test Constraint Satisfaction Only After Docking Option

Select this option to test whether poses satisfy constraints only after the ligands have been docked. With this option, constraints are used as a post-docking filter, rather than having any influence on the docking or scoring. You can therefore compare results directly with an unconstrained docking run.

Note: This option applies only to the constraints specified in the Constraints tab (receptor constraints). It does not apply to core constraints or to torsional constraints, which are ligand constraints.

Core Tab Features

In this tab you can specify the core of a reference ligand, and use it to perform core RMSD measurements of docked ligand poses or to restrict docked poses to those that lie within some RMSD tolerance of the reference core.

The constraint that you apply here is a ligand-based constraint, which means that the ligands that are subject to the constraint are those that have the same core moiety as the reference ligand. Ligands that do not match the core can be screened out in the first docking stage. In later stages, ligands that match the core pattern but do not meet the rms tolerance for the position of the core can be screened out.

Core Pattern Comparison controls

The Core pattern comparison controls allow you to choose which task to use the core definition for.

Use core pattern comparison option

Select this option to use a core definition for comparison or constraint. This option is off by default. When you select this option, the rest of the options in the tab are enabled.

Use for RMSD calculations only option

Use the core definition for computing the RMSD from the reference ligand, but not for docking.

Restrict docking to reference position option

Restrict the docking of ligands so that the ligand "core" lies within a given RMSD of the core in the reference ligand. The core is defined in terms of a set of atoms or a SMARTS pattern; if the ligand does not contain these atoms or pattern, it can be skipped.

Tolerance text box

Enter a tolerance for the RMSD in angstroms for restricting the docking to the reference position in this text box.

Retry with less tight core constraints (1.0 Å) if poses are rejected option

If the poses are rejected with tight core constraints and the core-snapping algorithm, try again with the filtering algorithm and a core constraint tolerance of 1.0 Å. This option is turned off and not available if the tolerance is set to more than 0.75 Å, as the filtering algorithm becomes the default at that tolerance.

Define core section

In this section you can define the core of the reference ligand, that is, the atoms that are used for comparison or constraint.

Pick core-containing molecule option

This option allows you to pick the molecule to use for the core. The molecule is marked in purple in the Workspace if Show markers is selected. It is automatically selected when the controls in this section become available. You must pick a core-containing molecule before you can proceed to define the core atoms. If you also define shape constraints, you must pick the same ligand for both types of constraint.

This option is automatically deselected if you choose SMARTS pattern for the Core atoms option, as the core is not defined by a molecule.

Show markers option

This option displays markers for the core-containing molecule and for the SMARTS pattern and RMSD subset atoms, as appropriate. It is selected by default, as there is no other indication of the picked atoms or molecules.

Core atoms options

These options are used to define the core atoms in the core-containing molecule:

  • Maximum common substructure—the core atoms are determined for each ligand independently from the maximum common substructure of the ligand and the core-containing molecule.

  • All heavy atoms—the core atoms are all non-hydrogen atoms in the picked core-containing molecule.

  • All atoms—the core atoms are all atoms in the picked core-containing molecule.

  • SMARTS pattern—the core atoms are defined in terms of a SMARTS pattern, not a molecule. You can pick atoms in the Workspace and click Get From Selection to define the SMARTS pattern, or you can type a SMARTS pattern into the text box. Note that it is advisable to not have the receptor displayed when you click Get From Selection, because it can slow the generation of the SMARTS pattern.

    The atoms in the Workspace structure that match the pattern are marked with green markers, if Show markers is selected. You can also define a subset of these atoms with which to evaluate the RMSD by selecting Pick RMSD subset atoms, then picking the atoms. A lock symbol appears next to the atoms you pick. Each pick adds to the set; picking an atom again removes it from the set.The RMSD atoms are used both when calculating only the RMSD and when restraining the core: it is the RMSD of these atoms that must fall below the prescribed tolerance in the latter case.

    This option disables the Pick core-containing molecule option.

Skip ligands that do not match core pattern option

When this option is selected, if a ligand does not match the core pattern it will not be docked (i.e. it is discarded in the first stage of the funnel). This option is selected by default. If it is deselected, ligands that do not match are docked, but the RMSD with the core is not calculated.

Shape Tab Features

Specify constraints to the shape of the reference ligand (or part of it).

Apply shape constraints option

Apply the constraints on the ligand shape that are set in this tab. When selected, the other options are activated.

Define reference shape section

Define a reference ligand to which the docked ligands are constrained by shape. If you also define core constraints, you must use the same reference ligand.

Workspace ligand option and text box.

Select this option to define the shape to match as that of a ligand in the Workspace, or part of it. This text box shows the ASL expression for the picked ligand. If you also define core constraints, you must pick the same ligand for both types of constraint.

Pick option

Select this option to pick the ligand in the Workspace. The ASL expression is filled in when you pick.

Show markers option

Show markers on the picked ligand, so you can identify it if there are multiple ligands in the Workspace.

File name text box and Browse button

Enter the reference ligand file name in this text box, or click Browse and navigate to the reference ligand file. The name of the reference ligand file you selected is displayed in the text box. The allowed file types are: Maestro. The ligand in this file is used as is to define the shape to match.

Use selected atoms only option and text box

Use only the atoms that are selected in the Workspace to define the shape. The noneditable text box shows the selected atoms, and is populated when you click Load Workspace Selection.

Load Workspace Selection button

Select the atoms you want to use for the shape in the Workspace, and click this button to use these atoms. The atom numbers are displayed in the text box.

Torsional Tab Features

In this tab you can set up torsional constraints for all ligands. This is done by defining SMARTS patterns that are matched in the ligands, and sets of torsions that are constrained relative to those SMARTS patterns.

This tab is only available when either Dock flexibly or Refine is selected in the Settings tab. All constraints defined in this tab are applied during docking.

Patterns table

This table lists the SMARTS patterns used to define the torsional constraints. The SMARTS pattern cannot be edited. If you want to modify a pattern, you will have to create a new one, and delete the old one.

SMARTS pattern SMARTS pattern that is to be used for a constraint.
Atoms Number of atoms in the SMARTS pattern
Torsions Number of torsions that are defined (and constrained) for the SMARTS pattern. Set to All if all torsions are constrained using the Constrain options below.
Pattern action buttons

These buttons perform actions on the rows of the Patterns table.

New button

Create a new SMARTS pattern in the table. Opens the New Torsion Pattern dialog box, in which you can type in a SMARTS pattern or get the SMARTS pattern from the current Workspace atom selection.

Delete button

Delete the rows that are selected in the table.

Delete All button

Delete all rows in the table.

Constrain options

These options allow you to choose which torsions to constrain.

  • All torsions—Constrain all torsions for the pattern that is selected in the table, subject to the conditions for a torsion to be valid for use as a constraint.
  • Selected torsions—Constrain the torsions that are listed in the torsions table, below. When you select this option, the torsions table and related tools become available.
Torsions table

This table lists the torsions that have been selected for use as constraints for the pattern that is selected in the Patterns table.

Atoms List of atom indices in the SMARTS pattern that define the torsion (noneditable).
Set Angle Check boxes for setting the corresponding angles to the value specified in the Angle column for each ligand.
Angle Angle to which the torsion should be constrained in all ligands. By default, the torsion is constrained for each ligand to the value it had when it was read by Glide. You can edit this table cell to change the angle.
Pick atoms to add a torsion option

Select this option to pick torsions in the Workspace. Four atoms must be picked to define the torsion. The atom indices (as defined by the SMARTS pattern) are listed in the Atoms column of the torsions table when the fourth atom is picked, and the angle in the Workspace structure is shown in the Angle column. The atoms must form a contiguous set of three bonds when you have finished picking, and must not define a ring torsion. They must also be atoms in the SMARTS pattern for which you are defining torsions. No checking is done for the validity of the torsion prior to docking, so you must make sure that it meets the criteria given above. The torsion cannot be edited once you have defined it, so if it is in error, you must delete it.

Delete and Delete All buttons

Delete the rows that are selected in the table, or delete all rows in the table.

Output Tab Features

Structure output options

Make settings for the output of structures and other properties.

File type options

The final list of poses that pass Glide's criteria are written to a multi-structure pose file. The file is compressed by default. You can select one of two options for the file type:

Pose viewer file (includes receptor) option

This version of the pose file, which includes the receptor, is intended for viewing poses in the receptor binding site. For more information, see the Pose Viewer Panel topic.

Ligand pose file (excludes receptor), in format format option

This option includes only the ligand structures in the output pose file. This version of the pose file cannot be used with the pose viewing facility, but may be appropriate if the output poses are intended for input to a subsequent Glide job, for example. The file can be written in Maestro or SD format; the choice is made from the option menu. The file is named jobname_lib.ext, where ext is maegz or sdfgz.

This option is not available if you chose to write XP descriptors or per-residue interactions, or if you have rotatable groups in the grid. These features require a pose viewer file.

Limit the number of poses to report option and text box

Select this option to limit the total number of the predicted best-binding poses written to the sorted pose file. By default, all poses that pass the filters are reported. If you are docking a large number of ligands and want multiple poses per ligand, the pose file may take up a large amount of disk space.

Write out at most p poses per ligand text box

This text box limits the number of poses per ligand written to the sorted pose file. The default choice of 1 pose per ligand is intended for database screening applications. A larger choice may be appropriate for lead-optimization studies or whenever several "reasonable" poses are wanted, for example, to generate a variety of docked poses for study with a post-docking script or application. However, if you have a small binding pocket or ligands with few rotatable bonds, choosing a larger number might simply retain poor poses.

Perform post-docking minimization option and settings

This option allows you to perform a minimization of the poses following the final docking, and can be used for both flexible and rigid 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. The full force-field minimization performed by post-docking minimization penalizes highly strained ligand geometries and eliminates poses with eclipsing interactions, many intraligand close contacts, and so on. If the binding pocket is small, or the ligands are fairly rigid, the post-docking minimization will not improve the results for poses that do not fit or are already poorly aligned to the receptor.

The docking process relies on rapid generation of ligand conformers and use of a grid to represent the receptor. The ligand poses generated during docking are rarely exactly at a local minimum, and post-minimization can improve the geometry of the poses. The Grid Minimization step of the Glide funnel (see Figure 2 in Glide Methodology) does not perform a full force-field minimization and excludes interactions beyond 1,5 interactions. Thus, post-docking minimization can improve the poses.

The time taken for post-docking minimization is less than 1% of the total docking time for SP and XP docking, and can be around 10% for HTVS.

Number of poses per ligand to include text box

Specify the number of poses per ligand to include in the minimization. This number must be greater than the number of poses per ligand that is written to the output file. The default is 5 for SP and HTVS, 10 for XP, and 100 for SP-Peptide.

Testing has indicated that the rescoring of poses after post-docking minimization generally finds a lower GlideScore than is reported in the top few poses from docking. Thus, it is strongly recommended to apply the minimization to a number of poses.

Threshold for rejecting minimized pose text box

For XP docking, the pose is kept if the GlideScore is no higher than that of the original pose by less than the specified threshold, otherwise the original pose is kept. This text box is not available for SP and HTVS docking, for which the minimized poses replace the original poses.

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: the strain correction is only added if it is above a threshold, and the excess strain above this threshold is scaled before adding it to the GlideScore. You can set the threshold and the scaling factor in the Output - Advanced Settings Dialog Box DEPRECATED. The strain corrections are stored as Maestro properties for each pose, corresponding to the strain values reported in the log file. Another more flexible approach is available in the Strain Energy Calculation and Rescoring Panel.

Write per-residue interaction scores option and settings

Write out per-residue interaction scores for a specified set of residues. The Coulomb, van der Waals, and hydrogen bonding scores, the sum of these three (Eint), and the minimum distances are calculated between the ligand and the specified residues. These values are written as structure-level properties for each ligand to the pose file, as well as to the log file. The residues can be specified by selecting one of the options described below.

Per-residue interactions can be visualized in the Workspace when viewing poses. See Viewing Poses for more information.

For residues within N Å of grid center option and text box

Write scores for complete residues that have any atom within the specified distance of the grid center.

Pick residues to include option and Specify Residue button

Pick the residues for which the scores are written. The residues are specified by clicking Specify Residue and using the picking controls in the Residues for Per-Residue Interaction Scores Dialog Box DEPRECATED.

Generate screening database (VSDB) for top hits option

Enable the Glide driver job to automatically generate a VSDB file for up to 10,000 hits.

Compute RMSD to input ligand geometries option

Compute the RMSD between the docked poses and the corresponding input ligand geometries. The RMSD calculation is done in place with heavy atoms only. The value is recorded as a Maestro property, named Glide rmsd to input (internal name r_i_glide_rmsd_to_input).

Advanced Settings Button

This button opens the Output - Advanced Settings Dialog Box DEPRECATED, in which you can set options to screen out poses that either have too high an energy or are too similar to other poses, and change the parameters associated with strain correction.

 
Job toolbar

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

The Job Settings button opens the Ligand Docking - 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.

Tutorials

Videos

Related Topics