Dr. Sherman works closely with researchers using Schrödinger software for molecular modeling and drug design projects. Here, Dr. Sherman answers some common questions about tools for GPCR homology modeling, examining component interactions, and more. All of the scripts discussed here can be downloaded from the Schrödinger Script Center.
Q: Is there a graphical interface to filter ligands based on properties and functional groups?
A: We have a graphical interface for our LigFilter program that does this. It can be found in the Filtering section under the General
tab of the Virtual Screening Workflow interface. To open the Virtual
Screening Workflow interface, select its entry under the Workflows menu
in Maestro. Additionally, a standalone interface to LigFilter called ligfilter_wrapper.py can be downloaded from our Script Center
and installed under the Scripts menu in Maestro. This program allows
for simultaneous filtering on existing structure properties,
precomputed properties, functional group patterns, and user-defined
SMARTS patterns. Finally, LigFilter can be run from the command line by
typing $SCHRODINGER/utilities/ligfilter.
Q: I’m interested in building a GPCR homology model and would like to make some large-scale manual adjustments to the structure, especially the transmembrane helices, before refinement. Currently, manual manipulations of a GPCR are quite difficult. Are there any Schrödinger tools to make this easier?
A: Yes, our GPCR Helix Manipulator script (gpcr_helix_manipulator.py)
is a graphical interface that helps facilitate some of the typical
operations one might want to apply to a GPCR or any other structure.
For example, individual helices can be controlled through tilt, sweep,
roll, and translate buttons. Additionally, backbone phi/psi angles can
be modified while interactively monitoring their position in a
Ramachandran plot. All of this can be done while showing bad contacts
and H-bonds.
Q: After running a Glide docking calculation, I would like to get interaction energies on a residue-by-residue basis. Is there a simple way to do this?
A: The Component Interactions script (component_interactions.py) uses the MacroModel
ASET functionality to compute interactions between sets of atoms on the
receptor and docked ligands. The atom sets are determined as either
receptor residues around the binding site or ligand functional groups.
It can operate on a single complex in the Workspace or on a Glide
PoseViewer file to process a series of ligands. Also, we will soon have
residue interaction capabilities directly within Glide – stay tuned for
more information on that.
Q: I am interested in aligning a set of related proteins based on residues around the binding site. However, due to different residue numbering this is quite difficult to do in an automated fashion. What should I do?
A: Our new Binding Site Alignment script (align_binding_sites.py) takes a set of proteins and aligns them based on either the automatic detection of binding site residues or on a user-defined set of residues. The script first does a global alignment of all structures using the Protein Structure Alignment functionality in Maestro. It then performs the local alignment by detecting the closest C-alpha atom from each structure to the reference structure. This method works very well for aligning kinases, for example.