def test_atomtyper_AlkEthOH(self):
# Test to make sure that AlkEthOH gets 100% at T=0 starting from correct
# types
basetypes_filename = get_data_filename('atomtypes/basetypes.smarts')
initialtypes_filename = get_data_filename(
'atomtypes/initial_AlkEthOH.smarts')
decorators_filename = get_data_filename(
'atomtypes/new-decorators.smarts')
replacements_filename = get_data_filename(
'atomtypes/replacements.smarts')
molecules = smarty.utils.read_molecules(
get_data_filename('molecules/AlkEthOH_test_filt1_tripos.mol2'),
verbose=False)
reference_typed_molecules = smarty.utils.read_molecules(
get_data_filename('molecules/AlkEthOH_test_filt1_ff.mol2'),
verbose=False)
# Construct atom type sampler.
atomtype_sampler = smarty.AtomTypeSampler(
molecules,
basetypes_filename,
initialtypes_filename,
decorators_filename,
replacements_filename=replacements_filename,
reference_typed_molecules=reference_typed_molecules,
verbose=False,
temperature=0)
# Start sampling atom types.
fracfound = atomtype_sampler.run(2)
# Ensure fraction found is 1.0
if fracfound < 1.0:
raise Exception("Not finding 100% of AlkEthOH when starting from"
" correct SMARTS.")
def test_atomtyper(self):
basetypes_filename = get_data_filename('atomtypes/basetypes.smarts')
initialtypes_filename = get_data_filename('atomtypes/basetypes.smarts')
decorators_filename = get_data_filename('atomtypes/decorators.smarts')
replacements_filename = get_data_filename(
'atomtypes/replacements.smarts')
molecules = smarty.utils.read_molecules(
get_data_filename('molecules/zinc-subset-tripos.mol2.gz'),
verbose=False)
reference_typed_molecules = smarty.utils.read_molecules(
get_data_filename('molecules/[email protected]'),
verbose=False)
# Construct atom type sampler.
atomtype_sampler = smarty.AtomTypeSampler(
molecules,
basetypes_filename,
initialtypes_filename,
decorators_filename,
replacements_filename=replacements_filename,
reference_typed_molecules=reference_typed_molecules,
verbose=False,
decorator_behavior='simple-decorators')
# Start sampling atom types.
atomtype_sampler.run(2)
def test_atomtyper_elemental(self):
"""
Test elemental atomtype sampler for hydrogen
"""
atomtype_sampler = smarty.AtomTypeSampler(
self.mols_alkethoh,
self.basetypes,
self.basetypes,
self.combine_decs,
replacements_filename=self.replacements,
reference_typed_molecules=self.mols_alkethoh_ref,
temperature=0.1,
verbose=False,
decorator_behavior='combinatorial-decorators',
element=1)
# run sampler with optional outputs
traj = 'test_smarty.csv'
plot = 'test_smarty.pdf'
atomtype_sampler.run(5, traj)
# test trajectory analysis functions on smarty output
timeseries = score_utils.load_trajectory(traj)
scores_vs_time = score_utils.scores_vs_time(timeseries)
score_utils.create_plot_file(traj, plot, True, False)
# check if score is 100% at first iteration
if scores_vs_time['all'][0] == 1.0:
raise Exception(
"Scoring problem, 100% at first iteration for total")
def test_atomtyper(self):
"""
Test atomtype sampler with simple-decorators
"""
atomtype_sampler = smarty.AtomTypeSampler(
self.mols_zinc,
self.basetypes,
self.basetypes,
self.simple_decs,
replacements_filename=self.replacements,
reference_typed_molecules=self.mols_zinc_ref,
temperature=0.1,
verbose=False,
decorator_behavior='simple-decorators',
element=0)
atomtype_sampler.run(2)
def test_atomtyper_AlkEthOH(self):
"""
Test atomtype sampler with correct "answers"
"""
atomtype_sampler = smarty.AtomTypeSampler(
self.mols_alkethoh,
self.basetypes,
self.alkethoh_answers,
self.combine_decs,
replacements_filename=self.replacements,
reference_typed_molecules=self.mols_alkethoh_ref,
temperature=0,
verbose=False)
# Start sampling atom types.
fracfound = atomtype_sampler.run(2)
# Ensure fraction found is 1.0
if fracfound < 1.0:
raise Exception("Not finding 100% of AlkEthOH when starting from"
" correct SMARTS.")
def main():
# Create command-line argument options.
usage_string = """\
Sample over atom types, optionally attempting to match atom types in a reference typed set of molecules.
usage: %prog --basetypes smartsfile --decorators smartsfile [--substitutions smartsfile] --molecules molfile [--reference molfile] --iterations niterations [--temperature temperature]
example:
python %prog --basetypes=atomtypes/basetypes.smarts --decorators=atomtypes/decorators.smarts --substitutions=atomtypes/substitutions.smarts \
--molecules=molecules/zinc-subset-tripos.mol2.gz --reference=molecules/[email protected] --iterations 1000 --temperature=0.1
"""
version_string = "%prog %__version__"
parser = OptionParser(usage=usage_string, version=version_string)
parser.add_option(
"-b",
"--basetypes",
metavar='BASETYPES',
action="store",
type="string",
dest='basetypes_filename',
default=None,
help="Filename defining base atom types as SMARTS atom matches.")
parser.add_option(
"-d",
"--decorators",
metavar='DECORATORS',
action="store",
type="string",
dest='decorators_filename',
default=None,
help="Filename defining decorator atom types as SMARTS atom matches.")
parser.add_option(
"-s",
"--substitutions",
metavar="SUBSTITUTIONS",
action="store",
type="string",
dest='substitutions_filename',
default=None,
help=
"Filename defining substitution definitions for SMARTS atom matches (OPTIONAL)."
)
parser.add_option(
"-r",
"--reference",
metavar="REFMOL",
action="store",
type="string",
dest='reference_molecules_filename',
default=None,
help=
"Reference typed molecules for computing likelihood (must match same molecule and atom ordering in molecules file) (OPTIONAL)."
)
parser.add_option(
"-m",
"--molecules",
metavar='MOLECULES',
action="store",
type="string",
dest='molecules_filename',
default=None,
help=
"Small molecule set (in any OpenEye compatible file format) containing 'dG(exp)' fields with experimental hydration free energies."
)
parser.add_option("-i",
"--iterations",
metavar='ITERATIONS',
action="store",
type="int",
dest='iterations',
default=150,
help="MCMC iterations.")
parser.add_option(
"-t",
"--temperature",
metavar='TEMPERATURE',
action="store",
type="float",
dest='temperature',
default=0.1,
help=
"Effective temperature for Monte Carlo acceptance, indicating fractional tolerance of mismatched atoms (default: 0.1). If 0 is specified, will behave in a greedy manner."
)
parser.add_option(
"-l",
'--trajectory',
metavar="TRAJECTORY_FILE",
action="store",
dest="traj_file",
default="trajectory.csv",
help=
"Name for trajectory file output, trajectory saves only changes to the list of 'atomtypes' for each iteration. For now, if the file name already exists, it just won't create a trajectory file"
)
verbose = True
# Parse command-line arguments.
(options, args) = parser.parse_args()
# Ensure all required options have been specified.
if (options.basetypes_filename is
None) or (options.decorators_filename is
None) or (options.molecules_filename is None):
parser.print_help()
parser.error("All input files must be specified.")
# Load and type all molecules in the specified dataset.
import smarty.utils
molecules = smarty.utils.read_molecules(options.molecules_filename,
verbose=True)
# Read reference typed molecules, if specified.
reference_typed_molecules = None
if options.reference_molecules_filename is not None:
reference_typed_molecules = smarty.utils.read_molecules(
options.reference_molecules_filename, verbose=True)
# Construct atom type sampler.
atomtype_sampler = smarty.AtomTypeSampler(
molecules,
options.basetypes_filename,
options.decorators_filename,
replacements_filename=options.substitutions_filename,
reference_typed_molecules=reference_typed_molecules,
verbose=verbose,
temperature=options.temperature)
# Start sampling atom types.
atomtype_sampler.run(options.iterations, options.traj_file)
def main():
# Create command-line argument options.
usage_string = """\
Sample over atom types, optionally attempting to match atom types in a reference typed set of molecules.
usage: %prog --basetypes smartsfile --initialtypes smartsfile --decorators smartsfile [--substitutions smartsfile] --molecules molfile [--reference molfile] --iterations niterations [--temperature temperature]
example:
python %prog --basetypes=atomtypes/basetypes.smarts --initialtypes=atomtypes/initialtypes.smarts --decorators=atomtypes/decorators.smarts --substitutions=atomtypes/substitutions.smarts \
--molecules=molecules/zinc-subset-tripos.mol2.gz --reference=molecules/[email protected] --iterations 1000 --temperature=0.1
"""
version_string = "%prog %__version__"
parser = OptionParser(usage=usage_string, version=version_string)
parser.add_option(
"-e",
"--element",
metavar='ELEMENT',
action="store",
type="int",
dest='element',
default=0,
help=
"By default the element value is 0 corresponding to sampling all atomtypes. If another atomic number is specified only atoms with that atomic number are sampled (i.e. --element=8 will only sample atomtypes for oxygen atoms)."
)
parser.add_option(
"-b",
"--basetypes",
metavar='BASETYPES',
action="store",
type="string",
dest='basetypes_filename',
default=None,
help=
"Filename defining base or generic atom types as SMARTS atom matches; these are indestructible and normally are elemental atom types."
)
parser.add_option(
"-f",
"--initialtypes",
metavar='BASETYPES',
action="store",
type="string",
dest='initialtypes_filename',
default=None,
help=
"Filename defining initial (first) atom types as SMARTS atom matches.")
parser.add_option(
"-d",
"--decorators",
metavar='DECORATORS',
action="store",
type="string",
dest='decorators_filename',
default=None,
help="Filename defining decorator atom types as SMARTS atom matches.")
parser.add_option(
"-s",
"--substitutions",
metavar="SUBSTITUTIONS",
action="store",
type="string",
dest='substitutions_filename',
default=None,
help=
"Filename defining substitution definitions for SMARTS atom matches (OPTIONAL)."
)
parser.add_option(
"-r",
"--reference",
metavar="REFMOL",
action="store",
type="string",
dest='reference_molecules_filename',
default=None,
help=
"Reference typed molecules for computing likelihood (must match same molecule and atom ordering in molecules file) (OPTIONAL)."
)
parser.add_option(
"-m",
"--molecules",
metavar='MOLECULES',
action="store",
type="string",
dest='molecules_filename',
default=None,
help=
"Small molecule set (in any OpenEye compatible file format) containing 'dG(exp)' fields with experimental hydration free energies."
)
parser.add_option("-i",
"--iterations",
metavar='ITERATIONS',
action="store",
type="int",
dest='iterations',
default=150,
help="MCMC iterations.")
parser.add_option(
"-t",
"--temperature",
metavar='TEMPERATURE',
action="store",
type="float",
dest='temperature',
default=0.1,
help=
"Effective temperature for Monte Carlo acceptance, indicating fractional tolerance of mismatched atoms (default: 0.1). If 0 is specified, will behave in a greedy manner."
)
parser.add_option(
"-l",
'--trajectory',
metavar="TRAJECTORY_FILE",
action="store",
dest="traj_file",
default="trajectory.csv",
help=
"Name for trajectory file output, trajectory saves only changes to the list of 'atomtypes' for each iteration. If the file already exists, it is overwritten."
)
parser.add_option(
"-p",
'--plot',
metavar="PLOT_FILE",
action="store",
dest="plot_file",
default=None,
help=
"Name for output file of a plot of the score versus time. If not specified, none will be written. If provided, needs to use a file extension suitable for matplotlib/pylab. Currently requires a trajectory file to be written using -l or --trajectory."
)
parser.add_option(
"-x",
"--decoratorbehavior",
metavar='DECORATOR_BEHAVIOR',
action="store",
type="string",
dest='decorator_behavior',
default='combinatorial-decorators',
help=
"Choose between simple-decorators or combinatorial-decorators (default = combinatorial-decorators)."
)
verbose = True
# Parse command-line arguments.
(options, args) = parser.parse_args()
# Ensure all required options have been specified.
if (options.basetypes_filename is
None) or (options.decorators_filename is
None) or (options.molecules_filename is None):
parser.print_help()
parser.error("All input files must be specified.")
# Ensure the Decorator Behavior option has been specified right
if not (options.decorator_behavior == 'simple-decorators'
or options.decorator_behavior == 'combinatorial-decorators'):
parser.print_help()
parser.error("Option not valid for decorator behavior.")
# Load and type all molecules in the specified dataset.
molecules = utils.read_molecules(options.molecules_filename, verbose=True)
# Read reference typed molecules, if specified.
reference_typed_molecules = None
if options.reference_molecules_filename is not None:
reference_typed_molecules = utils.read_molecules(
options.reference_molecules_filename, verbose=True)
# Construct atom type sampler.
if options.element == 0:
if verbose: print("Sampling all atomtypes")
elif options.element > 0:
if verbose:
print("Sampling atoms with atomic number %i" % options.element)
else:
parser.print_help()
parser.error(
"Element number must be 0 for all atoms or an integer greater than 0 for an atomic number"
)
atomtype_sampler = smarty.AtomTypeSampler(
molecules,
options.basetypes_filename,
options.initialtypes_filename,
options.decorators_filename,
replacements_filename=options.substitutions_filename,
reference_typed_molecules=reference_typed_molecules,
verbose=verbose,
temperature=options.temperature,
decorator_behavior=options.decorator_behavior,
element=options.element)
# Start sampling atom types.
atomtype_sampler.run(options.iterations, options.traj_file)
if options.plot_file is not None:
if options.traj_file is None:
print("Cannot create plot file without a trajectory file")
else:
smarty.score_utils.create_plot_file(options.traj_file,
options.plot_file, False,
verbose)