stjames 0.0.83__tar.gz → 0.0.85__tar.gz

{stjames-0.0.83/stjames.egg-info → stjames-0.0.85}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: stjames
-Version: 0.0.83
+Version: 0.0.85
 Summary: standardized JSON atom/molecule encoding scheme
 Author-email: Corin Wagen <corin@rowansci.com>
 Project-URL: Homepage, https://github.com/rowansci/stjames

{stjames-0.0.83 → stjames-0.0.85}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "stjames"
-version = "0.0.83"
+version = "0.0.85"
 description = "standardized JSON atom/molecule encoding scheme"
 readme = "README.md"
 requires-python = ">=3.11"

{stjames-0.0.83 → stjames-0.0.85}/stjames/method.py

@@ -43,6 +43,7 @@ class Method(LowercaseStrEnum):
     GFN0_XTB = "gfn0_xtb"
     GFN1_XTB = "gfn1_xtb"
     GFN2_XTB = "gfn2_xtb"
+    G_XTB = "g_xtb"
 
     # this was going to be removed, but Jonathon wrote such a nice basis set test... it's off the front end.
     BP86 = "bp86"
@@ -86,8 +87,8 @@ CORRECTABLE_NNP_METHODS = [Method.MACE_MP_0B2_L, Method.ORB_V2]
 NNPMethod = PrepackagedNNPMethod | CorrectableNNPMethod
 NNP_METHODS = [*PREPACKAGED_NNP_METHODS, *CORRECTABLE_NNP_METHODS]
 
-XTBMethod = Literal[Method.GFN_FF, Method.GFN0_XTB, Method.GFN1_XTB, Method.GFN2_XTB]
-XTB_METHODS = [Method.GFN_FF, Method.GFN0_XTB, Method.GFN1_XTB, Method.GFN2_XTB]
+XTBMethod = Literal[Method.GFN_FF, Method.GFN0_XTB, Method.GFN1_XTB, Method.GFN2_XTB, Method.G_XTB]
+XTB_METHODS = [Method.GFN_FF, Method.GFN0_XTB, Method.GFN1_XTB, Method.GFN2_XTB, Method.G_XTB]
 
 CompositeMethod = Literal[Method.HF3C, Method.B973C, Method.R2SCAN3C, Method.WB97X3C]
 COMPOSITE_METHODS = [Method.HF3C, Method.B973C, Method.R2SCAN3C, Method.WB97X3C]

{stjames-0.0.83 → stjames-0.0.85}/stjames/pdb.py

@@ -228,7 +228,7 @@ def pdb_object_to_pdb_filestring(
     keyword: bool = False,
     seqres: bool = True,
     hetnam: bool = True,
-    remark: bool = True,
+    remark: bool = False,
     crystallography: bool = False,
 ) -> str:
     pdb_lines: list[str] = []

{stjames-0.0.83 → stjames-0.0.85}/stjames/workflows/__init__.py

@@ -18,6 +18,7 @@ from .macropka import *
 from .molecular_dynamics import *
 from .multistage_opt import *
 from .pka import *
+from .pose_analysis_md import *
 from .protein_cofolding import *
 from .redox_potential import *
 from .scan import *
@@ -43,6 +44,7 @@ WORKFLOW_NAME = Literal[
     "molecular_dynamics",
     "multistage_opt",
     "pka",
+    "pose_analysis_md",
     "protein_cofolding",
     "redox_potential",
     "scan",
@@ -68,10 +70,11 @@ WORKFLOW_MAPPING: dict[WORKFLOW_NAME, Workflow] = {
     "molecular_dynamics": MolecularDynamicsWorkflow,  # type: ignore [dict-item]
     "multistage_opt": MultiStageOptWorkflow,  # type: ignore [dict-item]
     "pka": pKaWorkflow,  # type: ignore [dict-item]
+    "pose_analysis_md": PoseAnalysisMolecularDynamicsWorkflow,  # type: ignore [dict-item]
+    "protein_cofolding": ProteinCofoldingWorkflow,  # type: ignore [dict-item]
     "redox_potential": RedoxPotentialWorkflow,  # type: ignore [dict-item]
     "scan": ScanWorkflow,  # type: ignore [dict-item]
     "solubility": SolubilityWorkflow,  # type: ignore [dict-item]
     "spin_states": SpinStatesWorkflow,  # type: ignore [dict-item]
-    "protein_cofolding": ProteinCofoldingWorkflow,  # type: ignore [dict-item]
     "tautomers": TautomerWorkflow,  # type: ignore [dict-item]
 }

stjames-0.0.85/stjames/workflows/pose_analysis_md.py

@@ -0,0 +1,104 @@
+from typing import Annotated, Self
+
+from pydantic import AfterValidator, PositiveFloat, PositiveInt, model_validator
+
+from ..base import Base, round_float
+from ..pdb import PDB
+from ..types import UUID, round_list
+from .workflow import MoleculeWorkflow
+
+
+class BindingPoseContact(Base):
+    """
+    A single protein–ligand contact from an MD trajectory.
+
+    :param protein_atom_index: the index of the protein atom
+    :param ligand_atom_index: the index of the ligand atom
+    :occupancy: the probability of seeing this interaction in a frame, between 0 and 1
+    """
+
+    protein_atom_index: int
+    ligand_atom_index: int
+    occupancy: Annotated[float, AfterValidator(round_float(3))]
+
+
+class BindingPoseTrajectory(Base):
+    """
+    Represents a single trajectory looking at a binding pose.
+
+    :param uuid: the UUID of the trajectory
+    :param ligand_rmsd: the RMSD of the ligand vs. starting pose (aligning the protein)
+    :param contacts: the conserved binding-pose contacts
+    """
+
+    uuid: UUID
+    ligand_rmsd: Annotated[list[float], AfterValidator(round_list(3))] = []
+    contacts: list[BindingPoseContact] = []
+
+
+class PoseAnalysisMolecularDynamicsWorkflow(MoleculeWorkflow):
+    """
+    Pose analysis molecular dynamics workflow.
+
+    Note that the protein can be supplied either by UUID or raw PDB object.
+    We anticipate that the former will dominate deployed usage, but the latter is handy for isolated testing.
+    If, for whatever reason, the workflow is initialized with both a `target_uuid` and a `target`, the UUID will be ignored.
+
+    Inherited:
+    :param initial_molecule: Molecule of interest
+    :param mode: Mode for workflow (currently unused)
+
+    New:
+    :param protein: PDB of the protein.
+    :param protein_uuid: UUID of the protein.
+    :param num_trajectories: how many trajectories to run
+    :param equilibration_time_ns: how long to equilibrate trajectories for, in nanoseconds
+    :param simulation_time_ns: how long to run trajectories for, in nanoseconds
+    :param temperature: the temperature, in K
+    :param pressure_atm: the pressure, in atm
+    :param langevin_timescale_ps: the timescale for the Langevin integrator, in inverse picoseconds
+    :param timestep_fs: the timestep, in femtoseconds
+    :param constrain_hydrogens: whether or not to use SHAKE to freeze bonds to hydrogen
+    :param nonbonded_cutoff: the nonbonded cutoff for particle-mesh Ewald, in Å
+    :param protein_prune_cutoff: the cutoff past which residues will be deleted, in Å
+    :param protein_restraint_cutoff: the cutoff past which alpha-carbons will be constrained, in Å
+    :param protein_restraint_constant: the force constant for backbone restraints, in kcal/mol/Å**2
+    :param ionic_strength_M: the ionic strength of the solution, in M (molar)
+    :param water_buffer: the amount of water to add around the protein, in Å
+
+    Results:
+    :param trajectories: for each replicate, a UUID and the corresponding analysis results
+    :param minimized_protein_uuid: UUID of final system PDB
+    """
+
+    protein: PDB | None = None
+    protein_uuid: UUID | None = None
+
+    num_trajectories: PositiveInt = 4
+    equilibration_time_ns: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 5
+    simulation_time_ns: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 10
+
+    temperature: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 300
+    pressure_atm: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 1.0
+    langevin_timescale_ps: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 1.0
+
+    timestep_fs: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 2
+    constrain_hydrogens: bool = True
+    nonbonded_cutoff: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 8.0
+
+    protein_prune_cutoff: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 9.0
+    protein_restraint_cutoff: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 7.0
+    protein_restraint_constant: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 100
+
+    ionic_strength_M: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 0.10
+    water_buffer: Annotated[PositiveFloat, AfterValidator(round_float(3))] = 6.0
+
+    minimized_protein_uuid: UUID | None = None
+    trajectories: list[BindingPoseTrajectory] = []
+
+    @model_validator(mode="after")
+    def check_cutoff_sanity(self) -> Self:
+        """Check if protein is provided."""
+        if self.protein_prune_cutoff < self.protein_restraint_cutoff:
+            raise ValueError("Pruning cutoff must be larger than restraint cutoff")
+        return self

{stjames-0.0.83 → stjames-0.0.85/stjames.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: stjames
-Version: 0.0.83
+Version: 0.0.85
 Summary: standardized JSON atom/molecule encoding scheme
 Author-email: Corin Wagen <corin@rowansci.com>
 Project-URL: Homepage, https://github.com/rowansci/stjames

{stjames-0.0.83 → stjames-0.0.85}/stjames.egg-info/SOURCES.txt

@@ -62,6 +62,7 @@ stjames/workflows/macropka.py
 stjames/workflows/molecular_dynamics.py
 stjames/workflows/multistage_opt.py
 stjames/workflows/pka.py
+stjames/workflows/pose_analysis_md.py
 stjames/workflows/protein_cofolding.py
 stjames/workflows/redox_potential.py
 stjames/workflows/scan.py

{stjames-0.0.83 → stjames-0.0.85}/tests/test_pdb.py

@@ -37,6 +37,17 @@ def test_read_pdb_filestring() -> None:
     PDB.model_validate(json)
 
 
+def test_read_pdb_filestring_mome() -> None:
+    """Rest reading of a pdb string."""
+    with open("tests/data/cluster_1.pdb") as f:
+        data = f.read()
+    pdb = pdb_from_pdb_filestring(data)
+
+    json = pdb.model_dump()
+    print(json)
+    PDB.model_validate(json)
+
+
 def test_read_mmcif_filestring() -> None:
     """Rest reading of a mmcif string."""
     with open("tests/data/1ema.cif") as f:
@@ -85,7 +96,9 @@ def test_from_pdb_to_pdb_2qto() -> None:
     filestring = pdb_object_to_pdb_filestring(pdb, header=True, source=True, keyword=True, crystallography=True)
     pdb2 = pdb_from_pdb_filestring(filestring)
 
-    assert pdb == pdb2
+    assert pdb.description == pdb2.description
+    assert pdb.experiment == pdb2.experiment
+    assert pdb.models == pdb2.models
 
 def test_from_pdb_to_pdb_1ema() -> None:
     with open("tests/data/1ema.pdb") as f:
@@ -94,7 +107,8 @@ def test_from_pdb_to_pdb_1ema() -> None:
     filestring = pdb_object_to_pdb_filestring(pdb, header=True, source=True, keyword=True, crystallography=True)
     pdb2 = pdb_from_pdb_filestring(filestring)
 
-    assert pdb == pdb2
+    assert pdb.description == pdb2.description
+    assert pdb.models == pdb2.models
 
 def test_from_pdb_to_pdb_2hu4() -> None:
     with open("tests/data/2HU4.pdb") as f:
@@ -105,11 +119,9 @@ def test_from_pdb_to_pdb_2hu4() -> None:
     pdb2 = pdb_from_pdb_filestring(filestring)
 
     assert pdb.description == pdb2.description
-    assert pdb.experiment == pdb2.experiment
     # not true but doesn't matter
     print(pdb.geometry == pdb2.geometry)
     assert pdb.models == pdb2.models
-    assert pdb.quality == pdb2.quality
 
 def mmcif_author_format_to_pdb_format(authors: list[str]) -> list[str]:
     return [f"{last.upper()}{first.upper()}" for first, last in