-
Notifications
You must be signed in to change notification settings - Fork 1
Refactor RMSD analyses into MDAnalysis AnaysisBase classes
#90
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Changes from all commits
2977c01
4a960a0
afa829a
df017da
d824ab6
71a11d3
5cabcf4
e16b2e3
616bf42
90ed39e
7d6e49f
4bf598f
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -4,6 +4,7 @@ channels: | |
| dependencies: | ||
| - click | ||
| - MDAnalysis | ||
| - MDAnalysisTests | ||
| - netCDF4 | ||
| - openff-units | ||
| - pip | ||
|
|
||
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -5,10 +5,9 @@ | |
| import MDAnalysis as mda | ||
| import netCDF4 as nc | ||
| import numpy as np | ||
| import tqdm | ||
| from MDAnalysis.analysis import rms | ||
| from MDAnalysis.analysis import diffusionmap, rms | ||
| from MDAnalysis.analysis.base import AnalysisBase | ||
| from MDAnalysis.transformations import unwrap | ||
| from numpy import typing as npt | ||
|
|
||
| from .reader import FEReader | ||
| from .transformations import Aligner, ClosestImageShift, NoJump | ||
|
|
@@ -100,8 +99,141 @@ def make_Universe(top: pathlib.Path, trj: nc.Dataset, state: int) -> mda.Univers | |
| return u | ||
|
|
||
|
|
||
| class Protein2DRMSD(AnalysisBase): | ||
| """ | ||
| Flattened 2D RMSD matrix | ||
|
|
||
| For all unique frame pairs ``(i, j)`` with ``i < j``, this function | ||
| computes the RMSD between atomic coordinates after optimal alignment. | ||
| """ | ||
|
|
||
| def __init__(self, atomgroup, weights=None, **kwargs): | ||
| """ | ||
| Parameters | ||
| ---------- | ||
| atomgroup: AtomGroup | ||
| Protein atoms (e.g. CA selection) | ||
| weights: np.ndarray, optional | ||
| Per-atom weights to use in the RMSD calculation. If ``None``, | ||
| all atoms are weighted equally. | ||
| """ | ||
| super().__init__(atomgroup.universe.trajectory, **kwargs) | ||
| self._weights = weights | ||
| self._ag = atomgroup | ||
|
|
||
| def _prepare(self): | ||
| self._coords = [] | ||
| self.results.rmsd2d = [] | ||
|
|
||
| def _single_frame(self): | ||
| self._coords.append(self._ag.positions.copy()) | ||
|
|
||
| def _conclude(self): | ||
| positions = np.asarray(self._coords) | ||
| nframes = positions.shape[0] | ||
|
|
||
| output = [] | ||
| for i, j in itertools.combinations(range(nframes), 2): | ||
| posi, posj = positions[i], positions[j] | ||
| pair_rmsd = rms.rmsd( | ||
| posi, | ||
| posj, | ||
| self._weights, | ||
| center=True, | ||
| superposition=True, | ||
| ) | ||
| output.append(pair_rmsd) | ||
|
|
||
| self.results.rmsd2d = np.asarray(output) | ||
|
|
||
|
|
||
| class RMSDAnalysis(AnalysisBase): | ||
| """ | ||
| 1D RMSD time series for an AtomGroup. | ||
|
|
||
| Parameters | ||
| ---------- | ||
| atomgroup : MDAnalysis.AtomGroup | ||
| Atoms to compute RMSD for. | ||
| reference: Optional[MDAnalysis.AtomGroup] | ||
| Reference AtomGroup. If ``None``, the reference positions are captured | ||
| from the mobile AtomGroup at the start of the run (i.e. whatever frame | ||
| the trajectory is on when ``.run()`` is called). | ||
| mass_weighted : bool, optional | ||
| If True, compute mass-weighted RMSD. | ||
| superposition : bool, optional | ||
| If ``True``, perform rotational superposition before computing RMSD. | ||
| """ | ||
|
|
||
| def __init__( | ||
| self, | ||
| atomgroup, | ||
| reference=None, | ||
| mass_weighted=False, | ||
| superposition=False, | ||
| **kwargs, | ||
| ): | ||
| super().__init__(atomgroup.universe.trajectory, **kwargs) | ||
|
|
||
| self._ag = atomgroup | ||
| self._reference = reference if reference is not None else self._ag | ||
| self._mass_weighted = mass_weighted | ||
| self._superposition = superposition | ||
|
|
||
| def _prepare(self): | ||
| self.results.rmsd = [] | ||
|
|
||
| self._reference_pos = self._reference.positions.copy() | ||
|
|
||
| if self._mass_weighted: | ||
| self._weights = self._ag.masses / np.mean(self._ag.masses) | ||
| else: | ||
| self._weights = None | ||
|
|
||
| def _single_frame(self): | ||
| frame_rmsd = rms.rmsd( | ||
| self._ag.positions, | ||
| self._reference_pos, | ||
| self._weights, | ||
| center=False, | ||
| superposition=self._superposition, | ||
| ) | ||
| self.results.rmsd.append(frame_rmsd) | ||
|
|
||
| def _conclude(self): | ||
| self.results.rmsd = np.asarray(self.results.rmsd) | ||
|
|
||
|
|
||
| class LigandCOMDrift(AnalysisBase): | ||
| """ | ||
| Ligand center-of-mass displacement from initial position. | ||
| """ | ||
|
|
||
| def __init__(self, atomgroup, **kwargs): | ||
| super().__init__(atomgroup.universe.trajectory, **kwargs) | ||
| self._ag = atomgroup | ||
|
|
||
| def _prepare(self): | ||
| self.results.com_drift = [] | ||
| self._initial_com = self._ag.center_of_mass() | ||
|
|
||
| def _single_frame(self): | ||
| # distance between start and current ligand position | ||
| # ignores PBC, but we've already centered the traj | ||
| drift = mda.lib.distances.calc_bonds( | ||
| self._ag.center_of_mass(), | ||
| self._initial_com, | ||
| ) | ||
| self.results.com_drift.append(drift) | ||
|
|
||
| def _conclude(self): | ||
| self.results.com_drift = np.asarray(self.results.com_drift) | ||
|
|
||
|
|
||
| def gather_rms_data( | ||
| pdb_topology: pathlib.Path, dataset: pathlib.Path, skip: Optional[int] = None | ||
| pdb_topology: pathlib.Path, | ||
| dataset: pathlib.Path, | ||
| skip: Optional[int] = None, | ||
| ) -> dict[str, list[float]]: | ||
| """ | ||
| Compute structural RMSD-based metrics for a multistate BFE simulation. | ||
|
|
@@ -161,105 +293,57 @@ def gather_rms_data( | |
| # max against 1 to avoid skip=0 case | ||
| skip = max(n_frames // 500, 1) | ||
|
|
||
| pb = tqdm.tqdm(total=int(n_frames / skip) * n_lambda) | ||
|
|
||
| u_top = mda.Universe(pdb_topology) | ||
|
|
||
| for i in range(n_lambda): | ||
| for state_idx in range(n_lambda): | ||
| # cheeky, but we can read the PDB topology once and reuse per universe | ||
| # this then only hits the PDB file once for all replicas | ||
| u = make_Universe(u_top._topology, ds, state=i) | ||
| u = make_Universe(u_top._topology, ds, state=state_idx) | ||
|
|
||
| prot = u.select_atoms("protein and name CA") | ||
| ligand = u.select_atoms("resname UNK") | ||
|
|
||
| # save coordinates for 2D RMSD matrix | ||
| # TODO: Some smart guard to avoid allocating a silly amount of memory? | ||
| prot2d = np.empty((len(u.trajectory[::skip]), len(prot), 3), dtype=np.float32) | ||
|
|
||
| prot_start = prot.positions | ||
| ligand_start = ligand.positions | ||
| ligand_initial_com = ligand.center_of_mass() | ||
| ligand_weights = ligand.masses / np.mean(ligand.masses) | ||
|
|
||
| this_protein_rmsd = [] | ||
| this_ligand_rmsd = [] | ||
| this_ligand_wander = [] | ||
|
|
||
| for ts_i, ts in enumerate(u.trajectory[::skip]): | ||
| pb.update() | ||
|
|
||
| if prot: | ||
| prot2d[ts_i, :, :] = prot.positions | ||
| this_protein_rmsd.append( | ||
| rms.rmsd( | ||
| prot.positions, | ||
| prot_start, | ||
| None, # prot_weights, | ||
| center=False, | ||
| superposition=False, | ||
| ) | ||
| ) | ||
| if ligand: | ||
| this_ligand_rmsd.append( | ||
| rms.rmsd( | ||
| ligand.positions, | ||
| ligand_start, | ||
| ligand_weights, | ||
| center=False, | ||
| superposition=False, | ||
| ) | ||
| ) | ||
| this_ligand_wander.append( | ||
| # distance between start and current ligand position | ||
| # ignores PBC, but we've already centered the traj | ||
| mda.lib.distances.calc_bonds(ligand.center_of_mass(), ligand_initial_com) | ||
| ) | ||
|
|
||
| if prot: | ||
| # can ignore weights here as it's all Ca | ||
| rmsd2d = twoD_RMSD(prot2d, w=None) # prot_weights) | ||
| output["protein_RMSD"].append(this_protein_rmsd) | ||
| output["protein_2D_RMSD"].append(rmsd2d) | ||
| if ligand: | ||
| output["ligand_RMSD"].append(this_ligand_rmsd) | ||
| output["ligand_wander"].append(this_ligand_wander) | ||
|
|
||
| output["time(ps)"] = list(np.arange(len(u.trajectory))[::skip] * u.trajectory.dt) | ||
|
|
||
| return output | ||
|
|
||
|
|
||
| def twoD_RMSD(positions, w: Optional[npt.NDArray]) -> list[float]: | ||
| """ | ||
| Compute a flattened 2D RMSD matrix from a trajectory. | ||
|
|
||
| For all unique frame pairs ``(i, j)`` with ``i < j``, this function | ||
| computes the RMSD between atomic coordinates after optimal alignment. | ||
|
|
||
| Parameters | ||
| ---------- | ||
| positions : np.ndarray | ||
| Atomic coordinates for all frames in the trajectory. | ||
| w : np.ndarray, optional | ||
| Per-atom weights to use in the RMSD calculation. If ``None``, | ||
| all atoms are weighted equally. | ||
|
|
||
| Returns | ||
| ------- | ||
| list of float | ||
| Flattened list of RMSD values corresponding to all frame pairs | ||
| ``(i, j)`` with ``i < j``. | ||
| """ | ||
| nframes, _, _ = positions.shape | ||
|
|
||
| output = [] | ||
|
|
||
| for i, j in itertools.combinations(range(nframes), 2): | ||
| posi, posj = positions[i], positions[j] | ||
|
|
||
| rmsd = rms.rmsd(posi, posj, w, center=True, superposition=True) | ||
|
|
||
| output.append(rmsd) | ||
| prot_rmsd = RMSDAnalysis(prot).run(step=skip) | ||
| output["protein_RMSD"].append(prot_rmsd.results.rmsd) | ||
| # # Using the MDAnalysis RMSD class instead | ||
| # gs = ["protein and name CA"] | ||
| # prot_rmsd = rms.RMSD( | ||
|
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. The two RMSD classes are approximately equal in timing (on the test data) |
||
| # u, select="protein and name CA", groupselections=gs, weights="mass") | ||
| # prot_rmsd.run(step=skip) | ||
| # # The results contain: | ||
| # # - frame number | ||
| # # - time | ||
| # # - RMSD based on select (after superimposing) | ||
| # # - RMSD based on groupselections, one array per selection | ||
| # output["protein_RMSD"].append(prot_rmsd.results.rmsd.T[3]) | ||
|
|
||
| prot_rmsd2d = Protein2DRMSD(prot).run(step=skip) | ||
| output["protein_2D_RMSD"].append(prot_rmsd2d.results.rmsd2d) | ||
| # # Using the MDAnalysis DistanceMatrix class | ||
| # prot_rmsd2d = diffusionmap.DistanceMatrix(u, select="protein and name CA") | ||
|
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. This MDA code is much slower, on the test data 10s vs. 0.4s. |
||
| # prot_rmsd2d.run(step=skip) | ||
| # dist_mat = prot_rmsd2d.results.dist_matrix | ||
| # i, j = np.triu_indices_from(dist_mat, k=1) | ||
| # flattened = dist_mat[i, j] | ||
| # output["protein_2D_RMSD"].append(flattened) | ||
|
|
||
| if ligand.n_atoms > 0: | ||
| lig_rmsd = RMSDAnalysis(ligand, mass_weighted=True).run(step=skip) | ||
|
Contributor
Author
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Ligand RMSD is currently calculated on the hybrid topology, which may not be what we want long term. |
||
| output["ligand_RMSD"].append(lig_rmsd.results.rmsd) | ||
| # # Using the MDAnalysis RMSD class instead | ||
| # groupselections = ["resname UNK"] | ||
| # lig_rmsd = rms.RMSD( | ||
| # u, | ||
| # select="protein and name CA", | ||
| # groupselections=groupselections, | ||
| # weights="mass", | ||
| # ) | ||
| # lig_rmsd.run(step=skip) | ||
| # output["ligand_RMSD"].append(lig_rmsd.results.rmsd.T[3]) | ||
| lig_com_drift = LigandCOMDrift(ligand).run(step=skip) | ||
| output["ligand_wander"].append(lig_com_drift.results.com_drift) | ||
|
|
||
| output["time(ps)"] = np.arange(len(u.trajectory))[::skip] * u.trajectory.dt | ||
|
|
||
| return output | ||
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,80 @@ | ||
| import MDAnalysis as mda | ||
| import pytest | ||
| from MDAnalysisTests.datafiles import DCD, PSF | ||
| from numpy.testing import assert_allclose, assert_almost_equal | ||
|
|
||
| from openfe_analysis.rmsd import RMSDAnalysis | ||
|
|
||
|
|
||
| @pytest.fixture | ||
| def mda_universe(): | ||
| return mda.Universe(PSF, DCD) | ||
|
|
||
|
|
||
| @pytest.fixture() | ||
| def correct_values(): | ||
| return [0, 4.68953] | ||
|
|
||
|
|
||
| @pytest.fixture() | ||
| def correct_values_mass(): | ||
| return [0, 4.74920] | ||
|
|
||
|
|
||
| def test_rmsd(mda_universe, correct_values): | ||
| prot = mda_universe.select_atoms("name CA") | ||
| prot_rmsd = RMSDAnalysis(prot, superposition=True).run(step=49) | ||
| assert_almost_equal( | ||
| prot_rmsd.results.rmsd, | ||
| correct_values, | ||
| 4, | ||
| err_msg="error: rmsd profile should match" + "test values", | ||
| ) | ||
|
|
||
|
|
||
| def test_rmsd_frames(mda_universe, correct_values): | ||
| prot = mda_universe.select_atoms("name CA") | ||
| prot_rmsd = RMSDAnalysis(prot, superposition=True).run(frames=[0, 49]) | ||
| assert_almost_equal( | ||
| prot_rmsd.results.rmsd, | ||
| correct_values, | ||
| 4, | ||
| err_msg="error: rmsd profile should match" + "test values", | ||
| ) | ||
|
|
||
|
|
||
| def test_rmsd_single_frame(mda_universe): | ||
| prot = mda_universe.select_atoms("name CA") | ||
| prot_rmsd = RMSDAnalysis(prot, superposition=True).run(start=5, stop=6) | ||
| single_frame = [0.91544906] | ||
| assert_almost_equal( | ||
| prot_rmsd.results.rmsd, | ||
| single_frame, | ||
| 4, | ||
| err_msg="error: rmsd profile should match" + "test values", | ||
| ) | ||
|
|
||
|
|
||
| def test_mass_weighted(mda_universe, correct_values): | ||
| # mass weighting the CA should give the same answer as weighing | ||
| # equally because all CA have the same mass | ||
| prot = mda_universe.select_atoms("name CA") | ||
| prot_rmsd = RMSDAnalysis(prot, superposition=True, mass_weighted=True).run(step=49) | ||
|
|
||
| assert_almost_equal( | ||
| prot_rmsd.results.rmsd, | ||
| correct_values, | ||
| 4, | ||
| err_msg="error: rmsd profile should matchtest values", | ||
| ) | ||
|
|
||
|
|
||
| def test_custom_weighted(mda_universe, correct_values_mass): | ||
| prot = mda_universe.select_atoms("all") | ||
| prot_rmsd = RMSDAnalysis(prot, superposition=True, mass_weighted=True).run(step=49) | ||
| assert_almost_equal( | ||
| prot_rmsd.results.rmsd, | ||
| correct_values_mass, | ||
| 4, | ||
| err_msg="error: rmsd profile should matchtest values", | ||
| ) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
_analysis_algorithm_is_parallelizable