openff-toolkit 0.18.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- devtools/scripts/build_cookiecutter_json.py +22 -0
- docs/conf.py +296 -0
- docs/conftest.py +23 -0
- examples/conformer_energies/conformer_energies.py +143 -0
- examples/deprecated/SMIRNOFF_comparison/compare_molecule_energies.py +48 -0
- examples/deprecated/SMIRNOFF_comparison/compare_set_energies.py +58 -0
- examples/deprecated/liquid_box/create_mol2_pdb/create_mol2_pdb.py +193 -0
- examples/deprecated/liquid_box/openff_simulation.py +72 -0
- examples/deprecated/mixedFF_structure/generate_mixedFF_complex.py +56 -0
- examples/deprecated/testing/test_impropers/tleap_tools.py +120 -0
- examples/examples_helper.py +61 -0
- openff/toolkit/__init__.py +101 -0
- openff/toolkit/_tests/__init__.py +0 -0
- openff/toolkit/_tests/_stale_tests.py +146 -0
- openff/toolkit/_tests/conftest.py +108 -0
- openff/toolkit/_tests/create_molecules.py +494 -0
- openff/toolkit/_tests/mocking.py +145 -0
- openff/toolkit/_tests/molecule/__init__.py +1 -0
- openff/toolkit/_tests/molecule/test_state_enumeration.py +302 -0
- openff/toolkit/_tests/test_energies.py +136 -0
- openff/toolkit/_tests/test_examples.py +109 -0
- openff/toolkit/_tests/test_forcefield.py +3864 -0
- openff/toolkit/_tests/test_interchange.py +19 -0
- openff/toolkit/_tests/test_io.py +19 -0
- openff/toolkit/_tests/test_links.py +73 -0
- openff/toolkit/_tests/test_mm_molecule.py +404 -0
- openff/toolkit/_tests/test_molecule.py +4246 -0
- openff/toolkit/_tests/test_nagl.py +169 -0
- openff/toolkit/_tests/test_parameter_plugins.py +89 -0
- openff/toolkit/_tests/test_parameters.py +2898 -0
- openff/toolkit/_tests/test_toolkit_io.py +1088 -0
- openff/toolkit/_tests/test_toolkits.py +4418 -0
- openff/toolkit/_tests/test_topology.py +2485 -0
- openff/toolkit/_tests/test_utils.py +101 -0
- openff/toolkit/_tests/test_utils_callback.py +178 -0
- openff/toolkit/_tests/test_utils_collections.py +263 -0
- openff/toolkit/_tests/test_utils_serialization.py +168 -0
- openff/toolkit/_tests/utils.py +1747 -0
- openff/toolkit/data/README.md +8 -0
- openff/toolkit/data/molecules/AlkEthOH_test_filt1_ff.mol2 +2026 -0
- openff/toolkit/data/molecules/AlkEthOH_test_filt1_tripos.mol2 +2026 -0
- openff/toolkit/data/molecules/AlkEthOH_tripos.tar.gz +0 -0
- openff/toolkit/data/molecules/CID20742535_anion.sdf +50 -0
- openff/toolkit/data/molecules/CID20742535_neutral.sdf +51 -0
- openff/toolkit/data/molecules/FreeSolv.tar.gz +0 -0
- openff/toolkit/data/molecules/MiniDrugBank.sdf +33988 -0
- openff/toolkit/data/molecules/MiniDrugBank_tripos.mol2 +35462 -0
- openff/toolkit/data/molecules/PT2385.sdf +90 -0
- openff/toolkit/data/molecules/QL8.sdf +401 -0
- openff/toolkit/data/molecules/README.md +21 -0
- openff/toolkit/data/molecules/benzene.crd +8 -0
- openff/toolkit/data/molecules/benzene.mol2 +34 -0
- openff/toolkit/data/molecules/benzene.top +166 -0
- openff/toolkit/data/molecules/butane_multi.sdf +231 -0
- openff/toolkit/data/molecules/cyclohexane.mol2 +44 -0
- openff/toolkit/data/molecules/ethanol.mol2 +25 -0
- openff/toolkit/data/molecules/ethanol.sdf +23 -0
- openff/toolkit/data/molecules/ethanol_partial_charges.sdf +26 -0
- openff/toolkit/data/molecules/ethanol_reordered.mol2 +25 -0
- openff/toolkit/data/molecules/ethanol_reordered.sdf +23 -0
- openff/toolkit/data/molecules/ethene_rdkit.sdf +17 -0
- openff/toolkit/data/molecules/laromustine_ff.mol2 +71 -0
- openff/toolkit/data/molecules/laromustine_tripos.mol2 +71 -0
- openff/toolkit/data/molecules/methane_multiconformer.sdf +32 -0
- openff/toolkit/data/molecules/methane_multiconformer_properties.sdf +45 -0
- openff/toolkit/data/molecules/nh4.pdb +7 -0
- openff/toolkit/data/molecules/po4_phenylphosphate.pdb +46 -0
- openff/toolkit/data/molecules/ruxolitinib_conformers.sdf +910 -0
- openff/toolkit/data/molecules/toluene.mol2 +38 -0
- openff/toolkit/data/molecules/toluene.mol2.gz +0 -0
- openff/toolkit/data/molecules/toluene.pdb +30 -0
- openff/toolkit/data/molecules/toluene.sdf +36 -0
- openff/toolkit/data/molecules/toluene_charged.mol2 +38 -0
- openff/toolkit/data/molecules/z_3_hydroxy_propenal.sdf +23 -0
- openff/toolkit/data/proteins/5tbm_complex_solv.pdb +1928 -0
- openff/toolkit/data/proteins/5tbm_complex_solv_box.pdb +19563 -0
- openff/toolkit/data/proteins/CTerminal_ALA.pdb +20 -0
- openff/toolkit/data/proteins/CTerminal_ALA.sdf +43 -0
- openff/toolkit/data/proteins/CTerminal_ALA_ALA.pdb +30 -0
- openff/toolkit/data/proteins/CTerminal_ALA_ALA_misnamedH.pdb +30 -0
- openff/toolkit/data/proteins/CTerminal_ARG.pdb +34 -0
- openff/toolkit/data/proteins/CTerminal_TRP.pdb +34 -0
- openff/toolkit/data/proteins/MainChain_ALA.pdb +25 -0
- openff/toolkit/data/proteins/MainChain_ALA.sdf +52 -0
- openff/toolkit/data/proteins/MainChain_ALA_ALA.pdb +35 -0
- openff/toolkit/data/proteins/MainChain_ALA_ALA.sdf +69 -0
- openff/toolkit/data/proteins/MainChain_ALA_ALA_icodes.pdb +35 -0
- openff/toolkit/data/proteins/MainChain_ALA_ALA_no_hydrogens.pdb +17 -0
- openff/toolkit/data/proteins/MainChain_ARG.pdb +39 -0
- openff/toolkit/data/proteins/MainChain_ARG.sdf +81 -0
- openff/toolkit/data/proteins/MainChain_CYM.pdb +25 -0
- openff/toolkit/data/proteins/MainChain_CYS.pdb +26 -0
- openff/toolkit/data/proteins/MainChain_CYX.pdb +48 -0
- openff/toolkit/data/proteins/MainChain_CYX.sdf +96 -0
- openff/toolkit/data/proteins/MainChain_GLH.sdf +64 -0
- openff/toolkit/data/proteins/MainChain_GLU.sdf +63 -0
- openff/toolkit/data/proteins/MainChain_HID.pdb +32 -0
- openff/toolkit/data/proteins/MainChain_HIE.pdb +32 -0
- openff/toolkit/data/proteins/MainChain_HIE.sdf +67 -0
- openff/toolkit/data/proteins/MainChain_HIP.pdb +33 -0
- openff/toolkit/data/proteins/MainChain_HIP.sdf +70 -0
- openff/toolkit/data/proteins/MainChain_PRO.pdb +29 -0
- openff/toolkit/data/proteins/MainChain_TRP.pdb +39 -0
- openff/toolkit/data/proteins/NTerminal_ALA.pdb +21 -0
- openff/toolkit/data/proteins/NTerminal_ALA.sdf +45 -0
- openff/toolkit/data/proteins/NTerminal_ALA_ALA.pdb +31 -0
- openff/toolkit/data/proteins/NTerminal_ARG.pdb +35 -0
- openff/toolkit/data/proteins/NTerminal_TRP.pdb +35 -0
- openff/toolkit/data/proteins/T4-protein-unprocessed-4w51.pdb +1892 -0
- openff/toolkit/data/proteins/T4-protein.mol2 +5461 -0
- openff/toolkit/data/proteins/T4-protein.pdb +2637 -0
- openff/toolkit/data/proteins/T4-protein.sdf +5318 -0
- openff/toolkit/data/proteins/T4_protein_waters.pdb +2965 -0
- openff/toolkit/data/proteins/TwoMol_SER_CYS.pdb +50 -0
- openff/toolkit/data/proteins/aa_residues_substructures_explicit_bond_orders_with_caps.json +6435 -0
- openff/toolkit/data/proteins/aa_residues_substructures_explicit_bond_orders_with_caps_explicit_connectivity.json +6675 -0
- openff/toolkit/data/proteins/aa_residues_substructures_with_caps.json +6297 -0
- openff/toolkit/data/proteins/ace-ZZZ-gly-nme.pdb +78 -0
- openff/toolkit/data/proteins/ace-ZZZ-gly-nme.sdf +98 -0
- openff/toolkit/data/proteins/ace-a10-nme.sdf +229 -0
- openff/toolkit/data/proteins/ace-ala-nh2.pdb +21 -0
- openff/toolkit/data/proteins/fluoresceine_dyed_helix_capped.pdb +338 -0
- openff/toolkit/data/proteins/neutral_arginine.pdb +31 -0
- openff/toolkit/data/proteins/split_chain.pdb +74 -0
- openff/toolkit/data/proteins/split_chain_reordered.pdb +74 -0
- openff/toolkit/data/reference_energies/README.md +6 -0
- openff/toolkit/data/reference_energies/reference_0.8.0.json +1 -0
- openff/toolkit/data/reference_energies/reference_0.8.0.json.txt +180 -0
- openff/toolkit/data/reference_energies/reference_after_1007.json +1 -0
- openff/toolkit/data/reference_energies/reference_after_1007.json.txt +350 -0
- openff/toolkit/data/reference_energies/reference_after_802.json +1 -0
- openff/toolkit/data/reference_energies/reference_after_802.json.txt +291 -0
- openff/toolkit/data/systems/README.md +3 -0
- openff/toolkit/data/systems/amber/cyclohexane_ethanol_0.4_0.6.inpcrd +1511 -0
- openff/toolkit/data/systems/amber/cyclohexane_ethanol_0.4_0.6.prmtop +13367 -0
- openff/toolkit/data/systems/monomers/butanol.mol2 +38 -0
- openff/toolkit/data/systems/monomers/butanol.sdf +88 -0
- openff/toolkit/data/systems/monomers/cyclohexane.mol2 +45 -0
- openff/toolkit/data/systems/monomers/cyclohexane.sdf +105 -0
- openff/toolkit/data/systems/monomers/ethanol.mol2 +26 -0
- openff/toolkit/data/systems/monomers/ethanol.sdf +58 -0
- openff/toolkit/data/systems/monomers/methane.mol2 +18 -0
- openff/toolkit/data/systems/monomers/methane.sdf +38 -0
- openff/toolkit/data/systems/monomers/propane.mol2 +30 -0
- openff/toolkit/data/systems/monomers/propane.sdf +68 -0
- openff/toolkit/data/systems/monomers/tip3p_water.mol2 +14 -0
- openff/toolkit/data/systems/monomers/water.mol2 +14 -0
- openff/toolkit/data/systems/monomers/water.sdf +28 -0
- openff/toolkit/data/systems/packmol_boxes/cyclohexane_ethanol_0.4_0.6.pdb +6272 -0
- openff/toolkit/data/systems/packmol_boxes/cyclohexane_water.pdb +5524 -0
- openff/toolkit/data/systems/packmol_boxes/ethanol_water.pdb +5174 -0
- openff/toolkit/data/systems/packmol_boxes/propane_methane_butanol_0.2_0.3_0.5.pdb +7959 -0
- openff/toolkit/data/systems/test_systems/1_cyclohexane_1_ethanol.pdb +61 -0
- openff/toolkit/data/systems/test_systems/1_ethanol.pdb +24 -0
- openff/toolkit/data/systems/test_systems/1_ethanol_no_conect.pdb +15 -0
- openff/toolkit/data/systems/test_systems/1_ethanol_reordered.pdb +24 -0
- openff/toolkit/data/systems/test_systems/PE.pdb +67 -0
- openff/toolkit/data/systems/test_systems/README.md +3 -0
- openff/toolkit/data/systems/test_systems/T4_lysozyme_water_ions.pdb +29240 -0
- openff/toolkit/data/test_forcefields/Frosst_AlkEthOH.offxml +61 -0
- openff/toolkit/data/test_forcefields/Frosst_AlkEthOH_GBSA.offxml +78 -0
- openff/toolkit/data/test_forcefields/Frosst_AlkEthOH_MDL.offxml +62 -0
- openff/toolkit/data/test_forcefields/Frosst_AlkEthOH_parmAtFrosst.offxml +64 -0
- openff/toolkit/data/test_forcefields/GBSA_HCT-1.0.offxml +21 -0
- openff/toolkit/data/test_forcefields/GBSA_OBC1-1.0.offxml +18 -0
- openff/toolkit/data/test_forcefields/GBSA_OBC2-1.0.offxml +18 -0
- openff/toolkit/data/test_forcefields/README.md +23 -0
- openff/toolkit/data/test_forcefields/ammonia_minimal.offxml +27 -0
- openff/toolkit/data/test_forcefields/benzene_minimal.offxml +29 -0
- openff/toolkit/data/test_forcefields/chargeincrement-test.offxml +16 -0
- openff/toolkit/data/test_forcefields/frcmod.Frosst_AlkEthOH_withIDs +85 -0
- openff/toolkit/data/test_forcefields/ion_charges.offxml +18 -0
- openff/toolkit/data/test_forcefields/mangled.offxml +9 -0
- openff/toolkit/data/test_forcefields/old/README.md +9 -0
- openff/toolkit/data/test_forcefields/old/hbonds.offxml +11 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_2_spec_0_1.offxml +332 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_2_spec_0_2.offxml +338 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_2_spec_0_3.offxml +329 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_fixed_spec_0_1.offxml +342 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_fixed_spec_0_2.offxml +348 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_fixed_spec_0_3.offxml +337 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_spec_0_1.offxml +340 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_spec_0_2.offxml +346 -0
- openff/toolkit/data/test_forcefields/old/test_ff_0_0_4_spec_0_3.offxml +337 -0
- openff/toolkit/data/test_forcefields/smirff99Frosst_reference_0_1_spec.offxml +349 -0
- openff/toolkit/data/test_forcefields/smirnoff99Frosst_experimental.offxml +343 -0
- openff/toolkit/data/test_forcefields/smirnoff99Frosst_reference_0_1_spec.offxml +349 -0
- openff/toolkit/data/test_forcefields/smirnoff99Frosst_reference_0_2_spec.offxml +356 -0
- openff/toolkit/data/test_forcefields/test_forcefield.offxml +346 -0
- openff/toolkit/data/test_forcefields/tip3p.offxml +22 -0
- openff/toolkit/data/test_forcefields/unsupported_smirnoff_version.offxml +9 -0
- openff/toolkit/topology/__init__.py +43 -0
- openff/toolkit/topology/_mm_molecule.py +663 -0
- openff/toolkit/topology/molecule.py +6102 -0
- openff/toolkit/topology/topology.py +2555 -0
- openff/toolkit/typing/__init__.py +1 -0
- openff/toolkit/typing/engines/__init__.py +1 -0
- openff/toolkit/typing/engines/smirnoff/__init__.py +31 -0
- openff/toolkit/typing/engines/smirnoff/forcefield.py +1440 -0
- openff/toolkit/typing/engines/smirnoff/io.py +234 -0
- openff/toolkit/typing/engines/smirnoff/parameters.py +3839 -0
- openff/toolkit/typing/engines/smirnoff/plugins.py +91 -0
- openff/toolkit/utils/__init__.py +57 -0
- openff/toolkit/utils/_viz.py +104 -0
- openff/toolkit/utils/ambertools_wrapper.py +507 -0
- openff/toolkit/utils/base_wrapper.py +268 -0
- openff/toolkit/utils/builtin_wrapper.py +136 -0
- openff/toolkit/utils/callback.py +187 -0
- openff/toolkit/utils/collections.py +247 -0
- openff/toolkit/utils/constants.py +18 -0
- openff/toolkit/utils/exceptions.py +766 -0
- openff/toolkit/utils/nagl_wrapper.py +164 -0
- openff/toolkit/utils/openeye_wrapper.py +2750 -0
- openff/toolkit/utils/rdkit_wrapper.py +3302 -0
- openff/toolkit/utils/serialization.py +466 -0
- openff/toolkit/utils/toolkit_registry.py +509 -0
- openff/toolkit/utils/toolkits.py +217 -0
- openff/toolkit/utils/utils.py +805 -0
- openff_toolkit-0.18.0.dist-info/METADATA +109 -0
- openff_toolkit-0.18.0.dist-info/RECORD +231 -0
- openff_toolkit-0.18.0.dist-info/WHEEL +5 -0
- openff_toolkit-0.18.0.dist-info/licenses/LICENSE +21 -0
- openff_toolkit-0.18.0.dist-info/top_level.txt +5 -0
- utilities/deprecated/convert_frosst/convert_frcmod_0.1.py +251 -0
- utilities/deprecated/convert_frosst/convert_frcmod_0.2.py +249 -0
- utilities/deprecated/convert_frosst/convert_frcmod_0.3.py +249 -0
- utilities/make_substructure_dict/_cif_to_substructure_dict.py +955 -0
- utilities/make_substructure_dict/_make_chemical_substructures.py +46 -0
- utilities/make_substructure_dict/_make_metadata_assignment_substructures.py +35 -0
- utilities/test_plugins/custom_plugins/handler_plugins.py +48 -0
- utilities/test_plugins/setup.py +19 -0
|
@@ -0,0 +1,805 @@
|
|
|
1
|
+
"""
|
|
2
|
+
Utility subroutines
|
|
3
|
+
|
|
4
|
+
"""
|
|
5
|
+
|
|
6
|
+
__all__ = [
|
|
7
|
+
"all_subclasses",
|
|
8
|
+
"convert_0_1_smirnoff_to_0_2",
|
|
9
|
+
"convert_0_2_smirnoff_to_0_3",
|
|
10
|
+
"convert_all_quantities_to_string",
|
|
11
|
+
"convert_all_strings_to_quantity",
|
|
12
|
+
"deserialize_numpy",
|
|
13
|
+
"get_data_file_path",
|
|
14
|
+
"get_molecule_parameterIDs",
|
|
15
|
+
"inherit_docstrings",
|
|
16
|
+
"object_to_quantity",
|
|
17
|
+
"quantity_to_string",
|
|
18
|
+
"requires_package",
|
|
19
|
+
"serialize_numpy",
|
|
20
|
+
"string_to_quantity",
|
|
21
|
+
"string_to_unit",
|
|
22
|
+
"temporary_cd",
|
|
23
|
+
"unit_to_string",
|
|
24
|
+
]
|
|
25
|
+
|
|
26
|
+
import contextlib
|
|
27
|
+
import functools
|
|
28
|
+
import logging
|
|
29
|
+
from collections.abc import Iterable
|
|
30
|
+
from typing import TYPE_CHECKING, Any, TypeVar, overload
|
|
31
|
+
|
|
32
|
+
import numpy as np
|
|
33
|
+
import pint
|
|
34
|
+
from numpy.typing import NDArray
|
|
35
|
+
from openff.units import Quantity, Unit
|
|
36
|
+
from openff.utilities import requires_package
|
|
37
|
+
|
|
38
|
+
if TYPE_CHECKING:
|
|
39
|
+
from openff.toolkit import ForceField, Molecule
|
|
40
|
+
|
|
41
|
+
logger = logging.getLogger(__name__)
|
|
42
|
+
|
|
43
|
+
|
|
44
|
+
# Pre-create an instance of the dimensionless unit to speed up comparisons later
|
|
45
|
+
_DIMENSIONLESS = Unit("dimensionless")
|
|
46
|
+
|
|
47
|
+
|
|
48
|
+
def inherit_docstrings(cls):
|
|
49
|
+
"""Inherit docstrings from parent class"""
|
|
50
|
+
from inspect import getmembers, isfunction
|
|
51
|
+
|
|
52
|
+
for name, func in getmembers(cls, isfunction):
|
|
53
|
+
if func.__doc__:
|
|
54
|
+
continue
|
|
55
|
+
for parent in cls.__mro__[1:]:
|
|
56
|
+
if hasattr(parent, name):
|
|
57
|
+
func.__doc__ = getattr(parent, name).__doc__
|
|
58
|
+
return cls
|
|
59
|
+
|
|
60
|
+
|
|
61
|
+
T = TypeVar("T")
|
|
62
|
+
|
|
63
|
+
|
|
64
|
+
def all_subclasses(cls: type[T]) -> list[type[T]]:
|
|
65
|
+
"""Recursively retrieve all subclasses of the specified class"""
|
|
66
|
+
return cls.__subclasses__() + [g for s in cls.__subclasses__() for g in all_subclasses(s)]
|
|
67
|
+
|
|
68
|
+
|
|
69
|
+
@contextlib.contextmanager
|
|
70
|
+
def temporary_cd(dir_path):
|
|
71
|
+
"""Context to temporary change the working directory.
|
|
72
|
+
|
|
73
|
+
Parameters
|
|
74
|
+
----------
|
|
75
|
+
|
|
76
|
+
dir_path
|
|
77
|
+
The directory path to enter within the context
|
|
78
|
+
|
|
79
|
+
Examples
|
|
80
|
+
--------
|
|
81
|
+
|
|
82
|
+
>>> dir_path = '/tmp'
|
|
83
|
+
>>> with temporary_cd(dir_path):
|
|
84
|
+
... pass # do something in dir_path
|
|
85
|
+
|
|
86
|
+
"""
|
|
87
|
+
import os
|
|
88
|
+
|
|
89
|
+
prev_dir = os.getcwd()
|
|
90
|
+
os.chdir(os.path.abspath(dir_path))
|
|
91
|
+
try:
|
|
92
|
+
yield
|
|
93
|
+
finally:
|
|
94
|
+
os.chdir(prev_dir)
|
|
95
|
+
|
|
96
|
+
|
|
97
|
+
def get_data_file_path(relative_path: str) -> str:
|
|
98
|
+
"""Get the full path to one of the reference files in testsystems.
|
|
99
|
+
In the source distribution, these files are in ``openff/toolkit/data/``,
|
|
100
|
+
but on installation, they're moved to somewhere in the user's python
|
|
101
|
+
site-packages directory.
|
|
102
|
+
|
|
103
|
+
Parameters
|
|
104
|
+
----------
|
|
105
|
+
|
|
106
|
+
relative_path
|
|
107
|
+
Name of the file to load (with respect to `openff/toolkit/data/`)
|
|
108
|
+
|
|
109
|
+
"""
|
|
110
|
+
from importlib.resources import files
|
|
111
|
+
|
|
112
|
+
_DATA_ROOT = files("openff.toolkit") / "data"
|
|
113
|
+
|
|
114
|
+
file_path = _DATA_ROOT / relative_path
|
|
115
|
+
|
|
116
|
+
if not file_path.exists(): # type: ignore[attr-defined]
|
|
117
|
+
raise ValueError(f"Sorry! {file_path} does not exist. If you just added it, you'll have to re-install")
|
|
118
|
+
|
|
119
|
+
return str(file_path)
|
|
120
|
+
|
|
121
|
+
|
|
122
|
+
@pint.register_unit_format("simple")
|
|
123
|
+
def format_unit_simple(unit, registry, **options):
|
|
124
|
+
return " * ".join(f"{u} ** {p}" for u, p in unit.items())
|
|
125
|
+
|
|
126
|
+
|
|
127
|
+
def unit_to_string(input_unit: Unit) -> str:
|
|
128
|
+
return f"{input_unit:simple}"
|
|
129
|
+
|
|
130
|
+
|
|
131
|
+
def quantity_to_dict(input_quantity):
|
|
132
|
+
value = input_quantity.magnitude
|
|
133
|
+
if isinstance(value, np.ndarray):
|
|
134
|
+
value = value.tolist()
|
|
135
|
+
|
|
136
|
+
return {
|
|
137
|
+
"value": value,
|
|
138
|
+
"unit": str(input_quantity.units),
|
|
139
|
+
}
|
|
140
|
+
|
|
141
|
+
|
|
142
|
+
def dict_to_quantity(input_dict):
|
|
143
|
+
return input_dict["value"] * Unit(input_dict["unit"])
|
|
144
|
+
|
|
145
|
+
|
|
146
|
+
def quantity_to_string(input_quantity: Quantity) -> str:
|
|
147
|
+
"""
|
|
148
|
+
Serialize a ``openff.units.Quantity`` to a string representation that is backwards-compatible
|
|
149
|
+
with older versions of the OpenFF Toolkit. This includes a " * " between numerical values and
|
|
150
|
+
their units and "A" being used in place of the unicode Å ("\N{ANGSTROM SIGN}").
|
|
151
|
+
|
|
152
|
+
Parameters
|
|
153
|
+
----------
|
|
154
|
+
input_quantity
|
|
155
|
+
The quantity to serialize
|
|
156
|
+
|
|
157
|
+
Returns
|
|
158
|
+
-------
|
|
159
|
+
output_string
|
|
160
|
+
The serialized quantity
|
|
161
|
+
|
|
162
|
+
"""
|
|
163
|
+
unitless_value: float | int | NDArray | list = input_quantity.m
|
|
164
|
+
# The string representation of a numpy array doesn't have commas and breaks the
|
|
165
|
+
# parser, thus we convert any arrays to list here
|
|
166
|
+
if isinstance(unitless_value, np.ndarray):
|
|
167
|
+
unitless_value = list(unitless_value)
|
|
168
|
+
|
|
169
|
+
unit_string = unit_to_string(input_quantity.units)
|
|
170
|
+
|
|
171
|
+
return f"{unitless_value} * {unit_string}"
|
|
172
|
+
|
|
173
|
+
|
|
174
|
+
@functools.lru_cache
|
|
175
|
+
def string_to_unit(unit_string) -> Unit:
|
|
176
|
+
"""
|
|
177
|
+
Deserializes a ``openff.units.Quantity`` from a string representation, for
|
|
178
|
+
example: "kilocalories_per_mole / angstrom ** 2"
|
|
179
|
+
|
|
180
|
+
|
|
181
|
+
Parameters
|
|
182
|
+
----------
|
|
183
|
+
unit_string
|
|
184
|
+
Serialized representation of a ``openff.units.Quantity``.
|
|
185
|
+
|
|
186
|
+
Returns
|
|
187
|
+
-------
|
|
188
|
+
output_unit
|
|
189
|
+
The deserialized unit from the string
|
|
190
|
+
"""
|
|
191
|
+
return Unit(unit_string)
|
|
192
|
+
|
|
193
|
+
|
|
194
|
+
@functools.lru_cache
|
|
195
|
+
def string_to_quantity(quantity_string: str) -> int | float | Quantity:
|
|
196
|
+
"""Attempt to parse a string into a ``Quantity``.
|
|
197
|
+
|
|
198
|
+
Note that strings representing dimensionless floats or ints are returned as floats or ints, not
|
|
199
|
+
`Quantity` objects. For example, "1.0" is returned as `1.0` (a float) not
|
|
200
|
+
`Quantity(1.0, "dimensionless")`. (This quirk can't be captured by type annotations because the
|
|
201
|
+
input type remains str.)
|
|
202
|
+
|
|
203
|
+
This function is cached, keyed by the stringified quantity, to avoid re-parsing the same strings,
|
|
204
|
+
and re-creating `Quantity` objects, for identical inputs.
|
|
205
|
+
"""
|
|
206
|
+
|
|
207
|
+
from tokenize import TokenError
|
|
208
|
+
|
|
209
|
+
from pint import UndefinedUnitError
|
|
210
|
+
|
|
211
|
+
try:
|
|
212
|
+
quantity = Quantity(quantity_string)
|
|
213
|
+
except (TokenError, UndefinedUnitError):
|
|
214
|
+
return quantity_string # type: ignore[return-value]
|
|
215
|
+
|
|
216
|
+
# TODO: Should intentionally unitless array-likes be Quantity objects
|
|
217
|
+
# or their raw representation?
|
|
218
|
+
if quantity.units == _DIMENSIONLESS and isinstance(quantity.m, int | float):
|
|
219
|
+
return quantity.m
|
|
220
|
+
else:
|
|
221
|
+
return quantity
|
|
222
|
+
|
|
223
|
+
|
|
224
|
+
def convert_all_strings_to_quantity(
|
|
225
|
+
smirnoff_data: dict,
|
|
226
|
+
ignore_keys: Iterable[str] = tuple(),
|
|
227
|
+
) -> dict:
|
|
228
|
+
"""
|
|
229
|
+
Traverses a SMIRNOFF data structure, attempting to convert all
|
|
230
|
+
quantity-defining strings into ``openff.units.Quantity`` objects.
|
|
231
|
+
|
|
232
|
+
Integers and floats are ignored and not converted into a dimensionless
|
|
233
|
+
``openff.units.Quantity`` object.
|
|
234
|
+
|
|
235
|
+
Some good keys to ignore include `ignore_keys=["smirks", "name", "id", "parent_id"]`
|
|
236
|
+
since these are commonly used in SMIRNOFF force fields in ways that are meant to
|
|
237
|
+
remain strings. For more context, see `Issue #1635`_.
|
|
238
|
+
|
|
239
|
+
.. _Issue #1635: https://github.com/openforcefield/openff-toolkit/issues/1635.
|
|
240
|
+
|
|
241
|
+
Parameters
|
|
242
|
+
----------
|
|
243
|
+
smirnoff_data
|
|
244
|
+
A hierarchical dict structured in compliance with the SMIRNOFF spec
|
|
245
|
+
ignore_keys
|
|
246
|
+
A list of keys to skip when converting strings to quantities
|
|
247
|
+
|
|
248
|
+
Returns
|
|
249
|
+
-------
|
|
250
|
+
converted_smirnoff_data
|
|
251
|
+
A hierarchical dict structured in compliance with the SMIRNOFF spec,
|
|
252
|
+
with quantity-defining strings converted to ``openff.units.Quantity`` objects
|
|
253
|
+
"""
|
|
254
|
+
from pint import DefinitionSyntaxError
|
|
255
|
+
|
|
256
|
+
if isinstance(smirnoff_data, dict):
|
|
257
|
+
for key, value in smirnoff_data.items():
|
|
258
|
+
if key in ignore_keys:
|
|
259
|
+
smirnoff_data[key] = value
|
|
260
|
+
else:
|
|
261
|
+
smirnoff_data[key] = convert_all_strings_to_quantity(
|
|
262
|
+
value,
|
|
263
|
+
ignore_keys=ignore_keys,
|
|
264
|
+
)
|
|
265
|
+
obj_to_return = smirnoff_data
|
|
266
|
+
|
|
267
|
+
elif isinstance(smirnoff_data, list):
|
|
268
|
+
for index, item in enumerate(smirnoff_data):
|
|
269
|
+
smirnoff_data[index] = convert_all_strings_to_quantity(
|
|
270
|
+
item,
|
|
271
|
+
ignore_keys=ignore_keys,
|
|
272
|
+
)
|
|
273
|
+
obj_to_return = smirnoff_data
|
|
274
|
+
|
|
275
|
+
elif isinstance(smirnoff_data, int) or isinstance(smirnoff_data, float):
|
|
276
|
+
obj_to_return = smirnoff_data
|
|
277
|
+
|
|
278
|
+
else:
|
|
279
|
+
try:
|
|
280
|
+
obj_to_return = object_to_quantity(smirnoff_data)
|
|
281
|
+
except (TypeError, DefinitionSyntaxError):
|
|
282
|
+
obj_to_return = smirnoff_data
|
|
283
|
+
|
|
284
|
+
return obj_to_return
|
|
285
|
+
|
|
286
|
+
|
|
287
|
+
@overload
|
|
288
|
+
def convert_all_quantities_to_string(
|
|
289
|
+
smirnoff_data: list[Quantity],
|
|
290
|
+
) -> list[str]: ...
|
|
291
|
+
|
|
292
|
+
|
|
293
|
+
@overload
|
|
294
|
+
def convert_all_quantities_to_string(
|
|
295
|
+
smirnoff_data: dict,
|
|
296
|
+
) -> dict[str, Any]: ...
|
|
297
|
+
|
|
298
|
+
|
|
299
|
+
@overload
|
|
300
|
+
def convert_all_quantities_to_string(
|
|
301
|
+
smirnoff_data: "Quantity",
|
|
302
|
+
) -> str | list[str] | dict[str, Any]: ...
|
|
303
|
+
|
|
304
|
+
|
|
305
|
+
def convert_all_quantities_to_string(
|
|
306
|
+
smirnoff_data: dict | str | Quantity | list,
|
|
307
|
+
) -> str | dict[str, Any] | list[str]:
|
|
308
|
+
"""
|
|
309
|
+
Traverses a SMIRNOFF data structure, attempting to convert all
|
|
310
|
+
quantities into strings.
|
|
311
|
+
|
|
312
|
+
Parameters
|
|
313
|
+
----------
|
|
314
|
+
smirnoff_data
|
|
315
|
+
A hierarchical dict structured in compliance with the SMIRNOFF spec
|
|
316
|
+
|
|
317
|
+
Returns
|
|
318
|
+
-------
|
|
319
|
+
converted_smirnoff_data
|
|
320
|
+
A hierarchical dict structured in compliance with the SMIRNOFF spec,
|
|
321
|
+
with ``openff.units.Quantity``s converted to string
|
|
322
|
+
"""
|
|
323
|
+
|
|
324
|
+
if isinstance(smirnoff_data, dict):
|
|
325
|
+
for key, value in smirnoff_data.items():
|
|
326
|
+
smirnoff_data[key] = convert_all_quantities_to_string(value)
|
|
327
|
+
return smirnoff_data
|
|
328
|
+
elif isinstance(smirnoff_data, list):
|
|
329
|
+
for index, item in enumerate(smirnoff_data):
|
|
330
|
+
smirnoff_data[index] = convert_all_quantities_to_string(item)
|
|
331
|
+
return smirnoff_data
|
|
332
|
+
elif isinstance(smirnoff_data, Quantity):
|
|
333
|
+
return quantity_to_string(smirnoff_data)
|
|
334
|
+
else:
|
|
335
|
+
return smirnoff_data
|
|
336
|
+
|
|
337
|
+
|
|
338
|
+
@functools.singledispatch
|
|
339
|
+
def object_to_quantity(object):
|
|
340
|
+
"""
|
|
341
|
+
Attempts to turn the provided object into `openff.units.Quantity`s.
|
|
342
|
+
|
|
343
|
+
Can handle float, int, str, `Quantity`, `openmm.unit.Quantity`, or iterators over
|
|
344
|
+
the same. Raises an exception if unable to convert all inputs.
|
|
345
|
+
|
|
346
|
+
Parameters
|
|
347
|
+
----------
|
|
348
|
+
object
|
|
349
|
+
The object to convert to a ``Quantity`` object(s).
|
|
350
|
+
|
|
351
|
+
Returns
|
|
352
|
+
-------
|
|
353
|
+
converted_object
|
|
354
|
+
|
|
355
|
+
"""
|
|
356
|
+
# If we can't find a custom type, we treat this as a generic iterator.
|
|
357
|
+
return [object_to_quantity(sub_obj) for sub_obj in object]
|
|
358
|
+
|
|
359
|
+
|
|
360
|
+
@object_to_quantity.register(Quantity)
|
|
361
|
+
def _(obj):
|
|
362
|
+
return obj
|
|
363
|
+
|
|
364
|
+
|
|
365
|
+
@object_to_quantity.register(str)
|
|
366
|
+
def _(obj):
|
|
367
|
+
import pint
|
|
368
|
+
|
|
369
|
+
try:
|
|
370
|
+
return string_to_quantity(obj)
|
|
371
|
+
except pint.errors.UndefinedUnitError:
|
|
372
|
+
raise ValueError
|
|
373
|
+
|
|
374
|
+
|
|
375
|
+
@object_to_quantity.register(int)
|
|
376
|
+
@object_to_quantity.register(float)
|
|
377
|
+
def _(obj):
|
|
378
|
+
return Quantity(obj)
|
|
379
|
+
|
|
380
|
+
|
|
381
|
+
try:
|
|
382
|
+
import openmm.unit
|
|
383
|
+
from openff.units.openmm import from_openmm
|
|
384
|
+
|
|
385
|
+
@object_to_quantity.register(openmm.unit.Quantity)
|
|
386
|
+
def _(obj):
|
|
387
|
+
return from_openmm(obj)
|
|
388
|
+
|
|
389
|
+
except ImportError:
|
|
390
|
+
pass # pragma: nocover
|
|
391
|
+
|
|
392
|
+
|
|
393
|
+
def serialize_numpy(np_array) -> tuple[bytes, tuple[int]]:
|
|
394
|
+
"""
|
|
395
|
+
Serializes a numpy array into a big-endian bytestring and tuple representing its shape.
|
|
396
|
+
|
|
397
|
+
Parameters
|
|
398
|
+
----------
|
|
399
|
+
np_array
|
|
400
|
+
Input numpy array
|
|
401
|
+
|
|
402
|
+
Returns
|
|
403
|
+
-------
|
|
404
|
+
serialized
|
|
405
|
+
A big-endian bytestring of the NumPy array.
|
|
406
|
+
shape
|
|
407
|
+
The shape of the serialized array
|
|
408
|
+
"""
|
|
409
|
+
bigendian_float = np.dtype(float).newbyteorder(">")
|
|
410
|
+
bigendian_array = np_array.astype(bigendian_float)
|
|
411
|
+
serialized = bigendian_array.tobytes()
|
|
412
|
+
shape = np_array.shape
|
|
413
|
+
return serialized, shape
|
|
414
|
+
|
|
415
|
+
|
|
416
|
+
def deserialize_numpy(
|
|
417
|
+
serialized_np: bytes | list,
|
|
418
|
+
shape: tuple[int, ...],
|
|
419
|
+
) -> NDArray:
|
|
420
|
+
"""
|
|
421
|
+
Deserializes a numpy array from a bytestring or list. The input, if a bytestring, is
|
|
422
|
+
assumed to be in big-endian byte order.
|
|
423
|
+
|
|
424
|
+
Parameters
|
|
425
|
+
----------
|
|
426
|
+
serialized_np
|
|
427
|
+
A byte or list serialized representation of a numpy array
|
|
428
|
+
shape
|
|
429
|
+
The shape of the serialized array
|
|
430
|
+
|
|
431
|
+
Returns
|
|
432
|
+
-------
|
|
433
|
+
np_array
|
|
434
|
+
The deserialized numpy array
|
|
435
|
+
"""
|
|
436
|
+
if isinstance(serialized_np, list):
|
|
437
|
+
np_array = np.array(serialized_np)
|
|
438
|
+
if isinstance(serialized_np, bytes):
|
|
439
|
+
dt = np.dtype("float").newbyteorder(">")
|
|
440
|
+
np_array = np.frombuffer(serialized_np, dtype=dt)
|
|
441
|
+
np_array = np_array.reshape(shape)
|
|
442
|
+
return np_array
|
|
443
|
+
|
|
444
|
+
|
|
445
|
+
def convert_0_2_smirnoff_to_0_3(smirnoff_data_0_2):
|
|
446
|
+
"""
|
|
447
|
+
Convert an 0.2-compliant SMIRNOFF dict to an 0.3-compliant one.
|
|
448
|
+
This involves removing units from header tags and adding them
|
|
449
|
+
to attributes of child elements.
|
|
450
|
+
It also requires converting ProperTorsions and ImproperTorsions
|
|
451
|
+
potentials from "charmm" to "fourier".
|
|
452
|
+
|
|
453
|
+
Parameters
|
|
454
|
+
----------
|
|
455
|
+
smirnoff_data_0_2
|
|
456
|
+
Hierarchical dict representing a SMIRNOFF data structure according the the 0.2 spec
|
|
457
|
+
|
|
458
|
+
Returns
|
|
459
|
+
-------
|
|
460
|
+
smirnoff_data_0_3
|
|
461
|
+
Hierarchical dict representing a SMIRNOFF data structure according the the 0.3 spec
|
|
462
|
+
"""
|
|
463
|
+
# Legacy force fields sometimes specify the NonbondedForce's sigma_unit value, but then provide
|
|
464
|
+
# atom size as rmin_half. Here we correct for this behavior by explicitly defining both as
|
|
465
|
+
# the same unit if either one is defined.
|
|
466
|
+
if "vdW" in smirnoff_data_0_2["SMIRNOFF"].keys():
|
|
467
|
+
rmh_unit = smirnoff_data_0_2["SMIRNOFF"]["vdW"].get("rmin_half_unit", None)
|
|
468
|
+
sig_unit = smirnoff_data_0_2["SMIRNOFF"]["vdW"].get("sigma_unit", None)
|
|
469
|
+
if (rmh_unit is not None) and (sig_unit is None):
|
|
470
|
+
smirnoff_data_0_2["SMIRNOFF"]["vdW"]["sigma_unit"] = rmh_unit
|
|
471
|
+
elif (sig_unit is not None) and (rmh_unit is None):
|
|
472
|
+
smirnoff_data_0_2["SMIRNOFF"]["vdW"]["rmin_half_unit"] = sig_unit
|
|
473
|
+
# If both are None, or both are defined, don't overwrite anything
|
|
474
|
+
else:
|
|
475
|
+
pass
|
|
476
|
+
|
|
477
|
+
# Recursively attach unit strings
|
|
478
|
+
smirnoff_data = recursive_attach_unit_strings(smirnoff_data_0_2, {})
|
|
479
|
+
|
|
480
|
+
# Change TorsionHandler potential from "charmm" to "k*(1+cos(periodicity*theta-phase))". Note that, scientifically,
|
|
481
|
+
# we should have used "k*(1+cos(periodicity*theta-phase))" all along, since "charmm" technically
|
|
482
|
+
# implies that we would support a harmonic potential for torsion terms with periodicity 0
|
|
483
|
+
# More at: https://github.com/openforcefield/openff-toolkit/issues/303#issuecomment-490156779
|
|
484
|
+
if "ProperTorsions" in smirnoff_data["SMIRNOFF"]:
|
|
485
|
+
if "potential" in smirnoff_data["SMIRNOFF"]["ProperTorsions"]:
|
|
486
|
+
if smirnoff_data["SMIRNOFF"]["ProperTorsions"]["potential"] == "charmm":
|
|
487
|
+
smirnoff_data["SMIRNOFF"]["ProperTorsions"]["potential"] = "k*(1+cos(periodicity*theta-phase))"
|
|
488
|
+
if "ImproperTorsions" in smirnoff_data["SMIRNOFF"]:
|
|
489
|
+
if "potential" in smirnoff_data["SMIRNOFF"]["ImproperTorsions"]:
|
|
490
|
+
if smirnoff_data["SMIRNOFF"]["ImproperTorsions"]["potential"] == "charmm":
|
|
491
|
+
smirnoff_data["SMIRNOFF"]["ImproperTorsions"]["potential"] = "k*(1+cos(periodicity*theta-phase))"
|
|
492
|
+
|
|
493
|
+
# Add per-section tag
|
|
494
|
+
sections_not_to_version_0_3 = ["Author", "Date", "version", "aromaticity_model"]
|
|
495
|
+
for l1_tag in smirnoff_data["SMIRNOFF"].keys():
|
|
496
|
+
if l1_tag not in sections_not_to_version_0_3:
|
|
497
|
+
if smirnoff_data["SMIRNOFF"][l1_tag] is None:
|
|
498
|
+
# Handle empty entries, such as the ToolkitAM1BCC handler.
|
|
499
|
+
smirnoff_data["SMIRNOFF"][l1_tag] = {}
|
|
500
|
+
|
|
501
|
+
smirnoff_data["SMIRNOFF"][l1_tag]["version"] = 0.3
|
|
502
|
+
|
|
503
|
+
# Update top-level tag
|
|
504
|
+
smirnoff_data["SMIRNOFF"]["version"] = 0.3
|
|
505
|
+
|
|
506
|
+
return smirnoff_data
|
|
507
|
+
|
|
508
|
+
|
|
509
|
+
def convert_0_1_smirnoff_to_0_2(smirnoff_data_0_1):
|
|
510
|
+
"""
|
|
511
|
+
Convert an 0.1-compliant SMIRNOFF dict to an 0.2-compliant one.
|
|
512
|
+
This involves renaming several tags, adding Electrostatics and ToolkitAM1BCC tags, and
|
|
513
|
+
separating improper torsions into their own section.
|
|
514
|
+
|
|
515
|
+
Parameters
|
|
516
|
+
----------
|
|
517
|
+
smirnoff_data_0_1
|
|
518
|
+
Hierarchical dict representing a SMIRNOFF data structure according the the 0.1 spec
|
|
519
|
+
|
|
520
|
+
Returns
|
|
521
|
+
-------
|
|
522
|
+
smirnoff_data_0_2
|
|
523
|
+
Hierarchical dict representing a SMIRNOFF data structure according the the 0.2 spec
|
|
524
|
+
"""
|
|
525
|
+
smirnoff_data = smirnoff_data_0_1.copy()
|
|
526
|
+
|
|
527
|
+
l0_replacement_dict = {"SMIRFF": "SMIRNOFF"}
|
|
528
|
+
l1_replacement_dict = {
|
|
529
|
+
"HarmonicBondForce": "Bonds",
|
|
530
|
+
"HarmonicAngleForce": "Angles",
|
|
531
|
+
"PeriodicTorsionForce": "ProperTorsions",
|
|
532
|
+
"NonbondedForce": "vdW",
|
|
533
|
+
}
|
|
534
|
+
for old_l0_tag, new_l0_tag in l0_replacement_dict.items():
|
|
535
|
+
# Convert first-level smirnoff_data tags.
|
|
536
|
+
# Right now this just changes the SMIRFF tag to SMIRNOFF
|
|
537
|
+
if old_l0_tag in smirnoff_data.keys():
|
|
538
|
+
smirnoff_data[new_l0_tag] = smirnoff_data[old_l0_tag]
|
|
539
|
+
del smirnoff_data[old_l0_tag]
|
|
540
|
+
|
|
541
|
+
# SMIRFF tag will have been converted to SMIRNOFF here
|
|
542
|
+
# Convert second-level tags here
|
|
543
|
+
for old_l1_tag, new_l1_tag in l1_replacement_dict.items():
|
|
544
|
+
if old_l1_tag in smirnoff_data["SMIRNOFF"].keys():
|
|
545
|
+
smirnoff_data["SMIRNOFF"][new_l1_tag] = smirnoff_data["SMIRNOFF"][old_l1_tag]
|
|
546
|
+
del smirnoff_data["SMIRNOFF"][old_l1_tag]
|
|
547
|
+
|
|
548
|
+
# Add 'potential' field to each l1 tag
|
|
549
|
+
default_potential = {
|
|
550
|
+
"Bonds": "harmonic",
|
|
551
|
+
"Angles": "harmonic",
|
|
552
|
+
"ProperTorsions": "charmm",
|
|
553
|
+
# Note that "charmm" isn't actually correct, and was later changed
|
|
554
|
+
# in the 0.3 spec. More info at
|
|
555
|
+
# https://github.com/openforcefield/openff-toolkit/pull/311#commitcomment-33494506
|
|
556
|
+
"vdW": "Lennard-Jones-12-6",
|
|
557
|
+
}
|
|
558
|
+
for l1_tag in smirnoff_data["SMIRNOFF"].keys():
|
|
559
|
+
if l1_tag in default_potential.keys():
|
|
560
|
+
# Ensure that it isn't there already (shouldn't happen, but better to be safe)
|
|
561
|
+
if "potential" in smirnoff_data["SMIRNOFF"][l1_tag].keys():
|
|
562
|
+
assert smirnoff_data[l1_tag].keys == default_potential[l1_tag]
|
|
563
|
+
continue
|
|
564
|
+
# Issue an informative warning about assumptions made during conversion.
|
|
565
|
+
logger.warning(
|
|
566
|
+
f"0.1 SMIRNOFF spec file does not contain 'potential' attribute for '{l1_tag}' tag. "
|
|
567
|
+
f"The SMIRNOFF spec converter is assuming it has a value of '{default_potential[l1_tag]}'"
|
|
568
|
+
)
|
|
569
|
+
smirnoff_data["SMIRNOFF"][l1_tag]["potential"] = default_potential[l1_tag]
|
|
570
|
+
|
|
571
|
+
# Separate improper torsions from propers
|
|
572
|
+
if "ProperTorsions" in smirnoff_data["SMIRNOFF"]:
|
|
573
|
+
if "Improper" in smirnoff_data["SMIRNOFF"]["ProperTorsions"]:
|
|
574
|
+
# First generate an ImproperTorsions header, taking the relevant values from the ProperTorsions header
|
|
575
|
+
improper_section = {
|
|
576
|
+
"k_unit": smirnoff_data["SMIRNOFF"]["ProperTorsions"]["k_unit"],
|
|
577
|
+
"phase_unit": smirnoff_data["SMIRNOFF"]["ProperTorsions"]["phase_unit"],
|
|
578
|
+
"potential": smirnoff_data["SMIRNOFF"]["ProperTorsions"]["potential"],
|
|
579
|
+
"Improper": smirnoff_data["SMIRNOFF"]["ProperTorsions"]["Improper"],
|
|
580
|
+
}
|
|
581
|
+
|
|
582
|
+
# Then, attach the newly-made ImproperTorsions section
|
|
583
|
+
smirnoff_data["SMIRNOFF"]["ImproperTorsions"] = improper_section
|
|
584
|
+
del smirnoff_data["SMIRNOFF"]["ProperTorsions"]["Improper"]
|
|
585
|
+
|
|
586
|
+
# Add Electrostatics tag, setting several values to their defaults and
|
|
587
|
+
# warning about assumptions that are being made
|
|
588
|
+
electrostatics_section = {
|
|
589
|
+
"method": "PME",
|
|
590
|
+
"scale12": 0.0,
|
|
591
|
+
"scale13": 0.0,
|
|
592
|
+
"scale15": 1.0,
|
|
593
|
+
"cutoff": 9.0,
|
|
594
|
+
"cutoff_unit": "angstrom",
|
|
595
|
+
}
|
|
596
|
+
logger.warning(
|
|
597
|
+
"0.1 SMIRNOFF spec did not allow the 'Electrostatics' tag. Adding it in 0.2 spec conversion, and "
|
|
598
|
+
"assuming the following values:"
|
|
599
|
+
)
|
|
600
|
+
for key, val in electrostatics_section.items():
|
|
601
|
+
logger.warning(f"\t{key}: {val}")
|
|
602
|
+
|
|
603
|
+
# Take electrostatics 1-4 scaling term from 0.1 spec's NonBondedForce tag
|
|
604
|
+
electrostatics_section["scale14"] = smirnoff_data["SMIRNOFF"]["vdW"]["coulomb14scale"]
|
|
605
|
+
del smirnoff_data["SMIRNOFF"]["vdW"]["coulomb14scale"]
|
|
606
|
+
smirnoff_data["SMIRNOFF"]["Electrostatics"] = electrostatics_section
|
|
607
|
+
|
|
608
|
+
# Change vdW's lj14scale to 14scale, add other scaling terms
|
|
609
|
+
vdw_section_additions = {
|
|
610
|
+
"method": "cutoff",
|
|
611
|
+
"combining_rules": "Lorentz-Berthelot",
|
|
612
|
+
"scale12": "0.0",
|
|
613
|
+
"scale13": "0.0",
|
|
614
|
+
"scale15": "1",
|
|
615
|
+
"switch_width": "1.0",
|
|
616
|
+
"switch_width_unit": "angstrom",
|
|
617
|
+
"cutoff": "9.0",
|
|
618
|
+
"cutoff_unit": "angstrom",
|
|
619
|
+
}
|
|
620
|
+
for key, val in vdw_section_additions.items():
|
|
621
|
+
if key not in smirnoff_data["SMIRNOFF"]["vdW"].keys():
|
|
622
|
+
logger.warning(
|
|
623
|
+
f"0.1 SMIRNOFF spec file does not contain '{key}' attribute for 'NonBondedMethod/vdW'' tag. "
|
|
624
|
+
f"The SMIRNOFF spec converter is assuming it has a value of '{val}'"
|
|
625
|
+
)
|
|
626
|
+
smirnoff_data["SMIRNOFF"]["vdW"][key] = val
|
|
627
|
+
|
|
628
|
+
# Rename L-J 1-4 scaling term from 0.1 spec's NonBondedForce tag to vdW's scale14
|
|
629
|
+
smirnoff_data["SMIRNOFF"]["vdW"]["scale14"] = smirnoff_data["SMIRNOFF"]["vdW"]["lj14scale"]
|
|
630
|
+
del smirnoff_data["SMIRNOFF"]["vdW"]["lj14scale"]
|
|
631
|
+
|
|
632
|
+
# Add <ToolkitAM1BCC/> tag
|
|
633
|
+
smirnoff_data["SMIRNOFF"]["ToolkitAM1BCC"] = {}
|
|
634
|
+
|
|
635
|
+
# Update top-level tag
|
|
636
|
+
smirnoff_data["SMIRNOFF"]["version"] = 0.2
|
|
637
|
+
|
|
638
|
+
return smirnoff_data
|
|
639
|
+
|
|
640
|
+
|
|
641
|
+
def recursive_attach_unit_strings(smirnoff_data, units_to_attach):
|
|
642
|
+
"""
|
|
643
|
+
Recursively traverse a SMIRNOFF data structure, appending "* {unit}" to values in key:value pairs
|
|
644
|
+
where "key_unit":"unit_string" is present at a higher level in the hierarchy.
|
|
645
|
+
This function expects all items in smirnoff_data to be formatted as strings.
|
|
646
|
+
|
|
647
|
+
Parameters
|
|
648
|
+
----------
|
|
649
|
+
smirnoff_data
|
|
650
|
+
Any level of hierarchy that is part of a SMIRNOFF dict, with all data members
|
|
651
|
+
formatted as string.
|
|
652
|
+
units_to_attach
|
|
653
|
+
Dict of the form {key:unit_string}
|
|
654
|
+
|
|
655
|
+
Returns
|
|
656
|
+
-------
|
|
657
|
+
unit_appended_smirnoff_data: dict
|
|
658
|
+
"""
|
|
659
|
+
import re
|
|
660
|
+
|
|
661
|
+
# Make a copy of units_to_attach so we don't modify the original (otherwise things like k_unit could
|
|
662
|
+
# leak between sections)
|
|
663
|
+
units_to_attach = units_to_attach.copy()
|
|
664
|
+
# smirnoff_data = smirnoff_data.copy()
|
|
665
|
+
|
|
666
|
+
# If we're working with a dict, see if there are any new unit entries and store them,
|
|
667
|
+
# then operate recursively on the values in the dict.
|
|
668
|
+
if isinstance(smirnoff_data, dict):
|
|
669
|
+
# Go over all key:value pairs once to see if there are new units to attach.
|
|
670
|
+
# Note that units to be attached can be defined in the same dict as the
|
|
671
|
+
# key:value pair they will be attached to, so we need to complete this check
|
|
672
|
+
# before we are able to check other items in the dict.
|
|
673
|
+
for key, value in list(smirnoff_data.items()):
|
|
674
|
+
if key[-5:] == "_unit":
|
|
675
|
+
units_to_attach[key[:-5]] = value
|
|
676
|
+
del smirnoff_data[key]
|
|
677
|
+
|
|
678
|
+
# Go through once more to attach units as appropriate
|
|
679
|
+
for key in smirnoff_data.keys():
|
|
680
|
+
# We use regular expressions to catch possible indexed attributes
|
|
681
|
+
attach_unit = None
|
|
682
|
+
for unit_key, unit_string in units_to_attach.items():
|
|
683
|
+
if re.match(f"{unit_key}[0-9]*", key):
|
|
684
|
+
attach_unit = unit_string
|
|
685
|
+
|
|
686
|
+
if attach_unit is not None:
|
|
687
|
+
smirnoff_data[key] = str(smirnoff_data[key]) + " * " + attach_unit
|
|
688
|
+
|
|
689
|
+
# And recursively act on value, in case it's a deeper level of hierarchy
|
|
690
|
+
smirnoff_data[key] = recursive_attach_unit_strings(smirnoff_data[key], units_to_attach)
|
|
691
|
+
|
|
692
|
+
# If it's a list, operate on each member of the list
|
|
693
|
+
elif isinstance(smirnoff_data, list):
|
|
694
|
+
for index, value in enumerate(smirnoff_data):
|
|
695
|
+
smirnoff_data[index] = recursive_attach_unit_strings(value, units_to_attach)
|
|
696
|
+
|
|
697
|
+
# Otherwise, just return smirnoff_data unchanged
|
|
698
|
+
else:
|
|
699
|
+
pass
|
|
700
|
+
|
|
701
|
+
return smirnoff_data
|
|
702
|
+
|
|
703
|
+
|
|
704
|
+
def get_molecule_parameterIDs(molecules: list["Molecule"], forcefield: "ForceField") -> tuple[dict, dict]:
|
|
705
|
+
"""Process a list of molecules with a specified SMIRNOFF ffxml file and determine which parameters are used by
|
|
706
|
+
which molecules, returning collated results.
|
|
707
|
+
|
|
708
|
+
Parameters
|
|
709
|
+
----------
|
|
710
|
+
molecules
|
|
711
|
+
List of molecules (with explicit hydrogens) to parse
|
|
712
|
+
forcefield
|
|
713
|
+
The ForceField to apply
|
|
714
|
+
|
|
715
|
+
Returns
|
|
716
|
+
-------
|
|
717
|
+
parameters_by_molecule
|
|
718
|
+
Parameter IDs used in each molecule, keyed by isomeric SMILES
|
|
719
|
+
generated from provided OEMols. Each entry in the dict is a list
|
|
720
|
+
which does not necessarily have unique entries; i.e. parameter IDs
|
|
721
|
+
which are used more than once will occur multiple times.
|
|
722
|
+
|
|
723
|
+
parameters_by_ID
|
|
724
|
+
Molecules in which each parameter ID occur, keyed by parameter ID.
|
|
725
|
+
Each entry in the dict is a set of isomeric SMILES for molecules
|
|
726
|
+
in which that parameter occurs. No frequency information is stored.
|
|
727
|
+
|
|
728
|
+
"""
|
|
729
|
+
|
|
730
|
+
from openff.toolkit.topology import Topology
|
|
731
|
+
|
|
732
|
+
# Create storage
|
|
733
|
+
parameters_by_molecule: dict[str, list] = dict()
|
|
734
|
+
parameters_by_ID: dict[str, set[str]] = dict()
|
|
735
|
+
|
|
736
|
+
# Generate isomeric SMILES for each molecule, ensuring all molecules are unique
|
|
737
|
+
isosmiles = [molecule.to_smiles() for molecule in molecules]
|
|
738
|
+
already_seen = set()
|
|
739
|
+
duplicates = set(
|
|
740
|
+
smiles
|
|
741
|
+
for smiles in isosmiles
|
|
742
|
+
if smiles in already_seen or already_seen.add(smiles) # type: ignore[func-returns-value]
|
|
743
|
+
)
|
|
744
|
+
if len(duplicates) > 0:
|
|
745
|
+
raise ValueError(
|
|
746
|
+
"Error: get_molecule_parameterIDs has been provided a list of oemols which contains some duplicates: "
|
|
747
|
+
f"{duplicates}"
|
|
748
|
+
)
|
|
749
|
+
|
|
750
|
+
# Assemble molecules into a Topology
|
|
751
|
+
topology = Topology()
|
|
752
|
+
for molecule in molecules:
|
|
753
|
+
topology.add_molecule(molecule)
|
|
754
|
+
|
|
755
|
+
# Label molecules
|
|
756
|
+
labels = forcefield.label_molecules(topology)
|
|
757
|
+
|
|
758
|
+
# Organize labels into output dictionary by looping over all molecules/smiles
|
|
759
|
+
for idx in range(len(isosmiles)):
|
|
760
|
+
# Pull smiles, initialize storage
|
|
761
|
+
smi = isosmiles[idx]
|
|
762
|
+
parameters_by_molecule[smi] = []
|
|
763
|
+
|
|
764
|
+
# Organize data for this molecule
|
|
765
|
+
data = labels[idx]
|
|
766
|
+
for force_type in data.keys():
|
|
767
|
+
for parameter_type in data[force_type].values():
|
|
768
|
+
pid = parameter_type.id
|
|
769
|
+
# Store pid to molecule
|
|
770
|
+
parameters_by_molecule[smi].append(pid)
|
|
771
|
+
|
|
772
|
+
# Store which molecule this pid occurred in
|
|
773
|
+
if pid not in parameters_by_ID:
|
|
774
|
+
parameters_by_ID[pid] = set()
|
|
775
|
+
parameters_by_ID[pid].add(smi)
|
|
776
|
+
else:
|
|
777
|
+
parameters_by_ID[pid].add(smi)
|
|
778
|
+
|
|
779
|
+
return parameters_by_molecule, parameters_by_ID
|
|
780
|
+
|
|
781
|
+
|
|
782
|
+
def sort_smirnoff_dict(data):
|
|
783
|
+
"""
|
|
784
|
+
Recursively sort the keys in a dict of SMIRNOFF data.
|
|
785
|
+
|
|
786
|
+
Adapted from https://stackoverflow.com/a/47882384/4248961
|
|
787
|
+
|
|
788
|
+
TODO: Should this live elsewhere?
|
|
789
|
+
"""
|
|
790
|
+
sorted_dict = dict()
|
|
791
|
+
for key, val in sorted(data.items()):
|
|
792
|
+
if isinstance(val, dict):
|
|
793
|
+
# This should hit each ParameterHandler and dicts within them
|
|
794
|
+
sorted_dict[key] = sort_smirnoff_dict(val)
|
|
795
|
+
elif isinstance(val, list):
|
|
796
|
+
# Handle case of ParameterLists, which show up in
|
|
797
|
+
# the smirnoff dicts as lists of dicts
|
|
798
|
+
new_parameter_list = list()
|
|
799
|
+
for param in val:
|
|
800
|
+
new_parameter_list.append(sort_smirnoff_dict(param))
|
|
801
|
+
sorted_dict[key] = new_parameter_list
|
|
802
|
+
else:
|
|
803
|
+
# Handle metadata or the bottom of a recursive dict
|
|
804
|
+
sorted_dict[key] = val
|
|
805
|
+
return sorted_dict
|