GenerateModel

GenerateModel is the primary user-facing module in QMzyme. GenerateModel is used to load a starting structure, define QMzyme regions, and write calculation input. To avoid unintended behavior, the initial structure must be pre-processed. I.e., ensure hydrogens have been added, and the structure is representative of the system you hope to study. If atomic charge information is not present in the input file(s), QMzyme will guess atomic charges by referring to the residue names. Any residue name corresponding to standard protein residue names, defined here, are able to be parsed for their total charge. This library can also be found in configuration under the dictionary protein_residues. If you have a non-protein residue QMzyme will assume its charge is 0. This is important if you have a ligand with a non-zero charge that you will include in your calculation. After importing QMzyme you can update the charge dictionary to add this information by adding to the residue_charges dictionary found in configuration.

The starting structure is loaded in using MDAnalysis and converted to a Universe object. There are a variety of ways to define QMzyme region(s), and once a region has been set it can be further modified, i.e., through truncation schemes. Lastly, this module interfaces with CalculateModel, Writer and qprep to automate the creation of single- or multi-scale calculation input files.

class QMzyme.GenerateModel.GenerateModel(*args, name=None, universe=None, select_atoms='all', frame=0, pickle_file=None, **kwargs)

Bases: QMzymeModel

GenerateModel can be instantiated with an MDAnalysis Universe directly, or any combination of parameters that MDAnalysis.core.universe.Universe accepts to create a Universe. See https://userguide.mdanalysis.org/stable/universe.html for details.

Parameters:

• name (str, optional) -- Name to give to the QMzymeModel. This is used for default file naming purposes throughout the QMzyme package. If not provided, it will default to the base name of the universe filename attribute.

• universe (MDAnalysis.Universe, default=None) -- MDAnalysis Universe object. If not specified, user will need to provide file(s) that MDAnalysis can use to create a Universe object.

• select_atoms (str, default='all') -- MDAnalysis selection command to specify which atoms are stored in the universe.

• frame (int, default=0) -- If trajectory was provided, specify a frame to extract coordinates from.

• pickle_file (str, default=None) -- Provide name/path+file of previously pickled QMzymeModel object to initialize

Usage:

To instantiate a model from a PDB file called "filename.pdb":

model = QMzyme.GenerateModel("filename.pdb")

If "filename.pdb" contains any components you know you do not want included in your model, you can initialize the GenerateModel instance from a selection of atoms by using the select_atoms argument:

model = QMzyme.GenerateModel("filename.pdb", select_atoms="not resname WAT")

You can also initialize the model from a topology and trajectory file(s) and specify what frame to take coordinates from:

model = QMzyme.GenerateModel("filename.pdb", "filename.dcd", frame=100)

set_catalytic_center(selection)

This method calls the set_region method and forces the region name to be 'catalytic_center'. See set_region().

set_region(selection, name=None, **kwargs)

Method to create a QMzymeRegion and add to the QMzymeModel regions list.

Parameters:

• selection (See options below, required) --

  Determines what atoms are included in the region. A variety of input options are accepted:

  • str that can be interpreted by MDAnalysis selection commands

  • an MDAnalysis AtomGroup

  • a QMzymeRegion

  • any concrete class of SelectionScheme, i.e., DistanceCutoff. Options can be found in SelectionSchemes.

• name (str, optional) -- Name of the resulting region.

• kwargs -- Keyword arguments that might be needed if a SelectionScheme is used. For example, the parameter cutoff is required to use the DistanceCutoff scheme.

Usage:

model.set_region(selection="resid 10 or resid 15", name="two_residues")

from QMzyme.SelectionSchemes import DistanceCutoff
model.set_region(selection=DistanceCutoff, cutoff=5)

Note

When using a SelectionScheme the scheme class must be imported.

truncate(scheme=<class 'QMzyme.TruncationSchemes.TerminalAlphaCarbon'>, name=None)

Method to truncate QMzymeModel. All QMzymeModel regions with assigned methods will be combined and truncated according to the specified scheme. The resulting region will be saved as the QMzymeModel truncated attribute.

Parameters:

• scheme (TruncationScheme concrete class, default=:class:~QMzyme.TruncationSchemes.TerminalAlphaCarbon) -- Specifies the truncation scheme to use. Options can be found in TruncationSchemes.

• name (str, optional) -- Name to give the truncated model. If None, the original region name will be the combination of calculation methods and the suffix '_combined_region_truncated'.

write_input(filename=None, memory='24GB', nprocs=12, reset_calculation=False)

Method to write calculation file input. The code will automatically detect what type of calculation file to prepare based on the calculation methods that have been assigned to the model region(s). Once this is called the QMzymeModel object will automatically be serialized using the pickle library and saved under the filename {self.name+'.pkl'} in the current working directory.

Parameters:

• filename (str, optional) -- Name to use for resulting file. If not specified, the file will be named according to the region(s) name. The file format ending does not need to be included.

• memory (str, optional) -- Amount of memory to specify in the input file.

• nprocs (int, optional) -- Number of processors to specify in the input file.

Note

A QM_Method must first be assigned to a region.