Spatial Molding for Approachable Rigid Targets (SMART)

An open-source Python package for generation of SMART probe ensembles and calculation of SMART molecular descriptors.

Dependencies

• pandas
• scipy
• mathutils
• argparse
• RDKit
• morfeus (optional)
• dbstep (optional)
• pyvista (optional)
• chimera (optional)

See documentation for version-specific syntax (https://smart-molecular-descriptors.readthedocs.io/en/latest/)

Instructions

Step 1: Set up the System for Analysis

Reading a structure

import SMART as smart

# read from a file (.pdb, .sdf, .mol2, or .xyz)
structure = smart.ReadFile(path)

# or pass MOL directly from RDKit
structure = smart.ReadMol(mol)

Define a reference binding vector

# multiple reference vector computations are supported,
# select one from below

# define a linear reference vector from atom ids
binding_atom = 0
ref_atom = 1
structure.reference_vector(tip_id=binding_atom, ref_id=ref_atom, dist=2.0)

# define a linear reference vector from atom positions
binding_atom = np.array([0,0,0])
ref_atom = np.array([0,0,1])
structure.reference_vector(tip_pos=binding_atom, ref_pos=ref_atom, dist=2.0)

# define a cross-product reference vector from atom positions
# (follows the right-hand-rule)
structure.reference_angle(tip_pos=binding_atom, ref_pos=ref_atom, dist=2.0)

# define a geometry and find empty coordination site
binding_atom = np.array([0,0,0])
ref_atom1, ref_atom2 = 1, 2
structure.reference_geometry(tip_pos=binding_atom, refs_ids=[ref_atom1, ref_atom2], geom='trigonal', dist=2.0)

# (Optional: for debugging)
# Export .pdb file with a dummy atom at the computed binding site
structure.export_alignment(out=file.split('.')[0]+'_align', dummy='X')

Step 2: Run Molecular Probe Conformational Sampling

Read probe file and find template

from SMART import conf_search as search

# set name of probe .mol2 file (default probes in folder /Probes/)
probe_name = 'S_SiH2_12_cyclic'
if search.TEMPLATE.is_template(probe_name):
      # perform free conformer search to make template
      print('get template')
      search.TEMPLATE.GetTemplate(probe_name)
else:
      # load pre-existing template
      print('gen template')
      search.TEMPLATE.GenerateTemplate(probe_name)

Set conformational search parameters and perform template search

# performs 50 steps of fitting
search.PARAMS.read_parameters({'NSTEP':50, 'VERBOSE':True})

# !! start the search !!
ensemble = search.TEMPLATE_SEARCH(structure) # output conformers

# (Optional: export ensemble to single .pdb file)
smart.ExportStructure(ensemble, 'SMART_out')

Step 3: Compute SMART Descriptors

from SMART import descriptors as desc

# compute SMART descriptors using triangulation
properties_tri = desc.get_Cloud_Properties(structure.MOL, ensemble, id=binding_atom, prox_radius=4.0, alpha=0)

# compute SMART descriptors using Buried Volume
properties_bv = desc.get_BuriedVolume_Properties(structure.MOL, ensemble, id=binding_atom, prox_radius=4.0, sterimol=True, sasa=False, vol=True)

# compute octants and quadrants
ref_atom1, ref_atom2 = 1, 2
properties_bv_oc = desc.get_Octant_Properties(structure.MOL, ensemble, id=binding_atom, z_axis=[binding_atom], xz_plane=[ref_atom1, ref_atom2],  prox_radius=4.0, octant=True, quadrant=False)

Alternatively, compute SMART descriptors using UCSF Chimera through the script chimera_descriptors.py

See documentation for more information.

Citations

This package:

• Miller, B. R.; Cammarota, R. C; Sigman, M. A Python Package for the Generation of Cavity-Specific Steric Molecular Descriptors and Applications to Diverse Systems. ChemRxiv. 2024. 10.26434/chemrxiv-2024-gk1bp

Literature Using SMART:

• Cammarota, R. C., Liu, W., Bacsa, J., Davies, H. M. L., & Sigman, M. S. Mechanistically Guided Workflow for Relating Complex Reactive Site Topologies to Catalyst Performance in C-H Functionalization Reactions. Journal of the American Chemical Society, 2022, 144(4), 1881–1898. https://doi.org/10.1021/jacs.1c12198

• Lucas W. Souza, Beck R. Miller, Ryan C. Cammarota, Anna Lo, Ixchel Lopez, Yuan-Shin Shiue, Benjamin D. Bergstrom, Sarah N. Dishman, James C. Fettinger, Matthew S. Sigman, and Jared T. Shaw, ACS Catalysis, 2024, 14 (1), 104-115, 4256https://doi.org/10.1021/acscatal.3c04256