|
|
LigPrep Panel |
The LigPrep panel is used to to set up and start ligand preparation calculations. The main objective of LigPrep is to take 2D or 3D structures and produce the corresponding low-energy 3D structures for use by programs such as Glide and QikProp. The input and output can be in SD or Maestro format. LigPrep can produce multiple output structures for each input structure by generating different protonation states, stereochemistry, tautomers, and ring conformations.
To open the LigPrep panel, you can:
Choose Applications → LigPrep in the main window.
Choose Tasks → Ligand Preparation in the main window.
To run a LigPrep job, choose a structure source, make settings in the panel, then click Run to run the job with the current job settings. If you want to change the job settings, click the Settings button and choose Job Settings. In the Job Settings dialog box, you can choose a job name and one or more hosts to run the job on, and click Run to run the job. If you choose to run on multiple hosts, you can select the hosts from the Hosts table, and set the number of subjobs in the Separate job into N subjobs text box. By default the number of subjobs is set to the number of processors used.
If you obtain the structures from a file, the file can be in Maestro, SD, or SMILES format. The SMILES file can be in one of two formats:
.smi file—Text file with one SMILES string and an
optional title per line.csv file—Comma-separated file with SMILES string as the
first field, title as the second, followed by optional properties.If you want to run the job later from the command line, you can write an input file by clicking Write. This file can then be used with the ligprep command to run the job.
The number of structures produced can vary a great deal, depending on the ligands being processed. The factors affecting the number of structures produced are described below:
In the Stereoisomers section, only Generate all combinations should result in many more structures. For individual ligands the increase can be as large as 2n, where n is the number of chiral centers. For large data sets the increase is typically around 1.5, but is very dependent on the classes of molecules being processed. You can reduce the number of structures produced by entering a smaller value in Generate stereoisomers (maximum), however this does increase the risk of missing important stereoisomers.
In the Ionization section, only Generate possible states at target pH should increase the number of structures. The increase is typically a factor of approximately 1.4 for a pH range of 2.0. Reducing the pH range significantly reduces this factor.
The Generate tautomers option typically increases the number of structures by less than 20%.
The number of structures is roughly proportional to, but less than, the number of low energy ring conformations per ligand.
The structures that are produced can be filtered on the basis of various predefined descriptors. These descriptors are counts of various kinds of structural components, including a range of functional groups. You can filter the structures generated by LigPrep using a file that defines conditions on these descriptors. The descriptors are:
| Num rings | Num heteroaromatic rings | |
| Num aromatic rings | Num aliphatic rings | |
| Num rotatable bonds | Num atoms | |
| Molecular weight | Num chiral centers | |
| Num neutral amines | Num amide hydrogens | |
| Num divalent oxygen atoms | Num neutral donor groups | |
| Num charged donor groups | Num neutral acceptor groups | |
| Num charged acceptor groups | Num reactive groups | |
| Num acidic hydrogens | Num donor groups | |
| Num acceptor groups |
This file can be created by clicking Create, and constructing the list of criteria in the Ligand Filtering dialog box. An example input file is given below. Comment lines start with ##.
##Remove molecules that have a molecular weight of greater than 650 Molecular weight >650.0 ## Remove molecules with too many H-bond acceptor and donor atoms Num acceptor groups >3 Num donor groups >3 ##Remove molecules with fewer than 10 atoms Num atoms <10
For more information, see the
chapter on ligfilter in the
LigPrep User Manual.
Generating large numbers of output structures (>150,000) in a single run can be very time consuming. If you want to generate large numbers of structures, you can distribute the LigPrep job across multiple processors on a multiprocessor host, by selecting the host and entering the number of processors in the Job Settings dialog box.
If structures cannot be processed for some reason, the input structures
are copied to a file named jobname-failed.ext, which
is in the same format as the input file. The structure numbers from the input
file are reported at the end of a log file, along with the stage in which the
failure occurred.
Choose the source of the structures for the current task.
If you chose File from the Use structures from option menu, enter the file name in this text box, or click Browse and navigate to the structure file. The name of the file you selected is displayed in the text box.
Enter the name of a file that contains criteria on molecular descriptors for filtering, click Browse to select an existing filter criteria file, or click Create to create a new filter criteria file in the Ligand Filtering dialog box.
Select the force field to use for minimization of structures. Three choices are available: OPLS_2005, MMFFs, and OPLS3. The default is OPLS_2005, unless you set the default in the Preferences panel.
Select a strategy for determining the ionization state of the ligand. The ionization states are generated by adding or removing protons from the ligand.
Do not change: Do not change the ionization state of the ligand.
Neutralize: Ensure that the ligand is neutral by adding or removing protons, according to the pKa values of the acidic or basic sites.
Generate possible states at target pH: Generate ionization states that are significantly populated in the pH range specified by the values in the two text boxes. There are two programs available for generating the states, listed next to Using:
Ionizer—use the Ionizer (the default)
Epik—use Epik. Epik has two options that can be used:
Add metal binding states—Generate states that are appropriate for binding to metals in addition to the normal ionization states.
Include original state—Include the original ionization state in addition to the generated states.
To use Epik, you must have an Epik license, as Epik is a separate product. Ionizer is included with LigPrep. Epik is more comprehensive in its treatment than Ionizer.
Structures from some databases can consist of multiple molecules, one of which is the ligand. In many cases, the extra molecules are counter ions or water molecules. Select this option to remove all but the molecule with the largest number of atoms.
Tautomers are thermally accessible and interchangeable isomers of molecules in which the non-hydrogen topology of the molecule remains unmodified while hydrogen atoms are relocated and bond orders change. Select this option to generate tautomers with significant populations for each input structure. This option is recommended for preparing ligands for both Glide and QikProp calculations. If you selected Epik for ionization, it will also be used for generating tautomers.
Select the treatment of stereochemical information in the input and the generation of stereoisomers in the output. There are three options under the Computation heading that control the source of stereoisomers:
Retain specified chiralities (vary other chiral
centers)— Keep the information on chiralities from the input file,
and fix these chiralities for the entire calculation. Chirality information
includes parities and bond directions from SD files, and the chirality property
from Maestro files. Maestro chiralities are only written by the
stereoizer utility. If the configuration or chirality of a chiral
center is not specified, the two possible chiralities are generated in the
output.
Determine chiralities from 3D structure—
Discard all chirality information in the input file, and determine the chirality
from the 3D geometry. These chiralities are held fixed. For centers whose
chirality is indeterminate, structures for the two possible chiralities for each
center are generated. (Sets the -g option of the ligprep command.)
Generate all combinations— Discard all
chirality information, both from input file properties and from the 3D geometry,
and generate all possible configurations that result from the combination of
chiralities on each chiral center. (Sets the -ac option of the
ligprep command.)
You can limit the number of stereoisomers to generate in the Generate at most N per ligand text box.
Select For SD V2000 input, generate enantiomers if the chiral flag is 0, if you have SD input and want both enantiomers to be generated if the chiral flag is set to zero. The default is to generate only the enantiomer represented by the specified chiralities. This setting is stored as a preference, so it is preserved across Maestro sessions.
When stereoisomers are generated for fused ring systems, the correlation between chiralities on chiral centers in the rings is taken into account, and invalid combinations are not generated.
LigPrep can generate ring conformations and evaluate an approximate conformational energy to determine which conformations are likely to be lowest in energy. Specify how many conformations are kept, counting from the lowest in energy.
Specify the file format for the output file. If you want to incorporate the
structures into the current project, you must select Maestro for the format. In Maestro format, the structures
are written to the file jobname.maegz. In SDF format, the
structures are written to the file jobname.sdf.
The Job toolbar is used to make settings for a job and to start it.
Enter a name for the job in this text box.
This button opens the Job Settings dialog box. The arrow to the right opens a menu, from which you can make settings or perform actions that are related to the job.
The menu items are described below.
This icon indicates when there are jobs running for the application that belong to the current project. It starts spinning when the first job starts, and stops when there are no more jobs running. If a job fails to start, it changes to an exclamation point.
Clicking the button shows a small job status window that lists the job name and status for all active jobs submitted from the current panel (for Jaguar and MacroModel this means all jobs from any of the application panels). You can double-click on a row to open the Monitor panel and monitor the job, or click the Monitor button to open the Monitor panel and close the job status window. The job status is updated while the window is open. If a job finishes while the window is open, the job remains displayed but with the new status. The rows are colored according to the status: yellow for submitted, green for launched, running, or finished, red for incorporated, died, or killed. Click anywhere outside the window to close it.
Run the job with the current job settings.
The status bar displays information about the current job settings for the panel. This includes the job name, task name and task settings (if any), number of subjobs (if any) and the host name and job incorporation setting. It also displays a message for a short time when you start a new job.
The status bar also contains the Help button, which opens the help topic for the panel.
|