mstk 0.3.3.dev11__tar.gz → 0.3.4__tar.gz

{mstk-0.3.3.dev11 → mstk-0.3.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mstk
-Version: 0.3.3.dev11
+Version: 0.3.4
 Summary: Molecular simulation toolkit
 Home-page: https://github.com/z-gong/mstk
 Author: Zheng Gong
@@ -69,3 +69,11 @@ https://mstk.readthedocs.io/en/latest/index.html
 
 - [ ] Take bond order into consideration for force field assignment
 - [ ] Re-organize algorithms scattered in topology and analyzer modules
+
+## Known issue
+`mstk` use `chemfiles` to write `XTC` trajectory. However, lastest `chemfiles` is buggy under WSL for writing binary trajectory format. 
+If you are working under WSL, please install `chemfiles 0.10.2`.
+
+```
+conda install chemfiles-lib=0.10.2 chemfiles-python=0.10.2
+```

{mstk-0.3.3.dev11 → mstk-0.3.4}/README.md

@@ -56,3 +56,11 @@ https://mstk.readthedocs.io/en/latest/index.html
 
 - [ ] Take bond order into consideration for force field assignment
 - [ ] Re-organize algorithms scattered in topology and analyzer modules
+
+## Known issue
+`mstk` use `chemfiles` to write `XTC` trajectory. However, lastest `chemfiles` is buggy under WSL for writing binary trajectory format. 
+If you are working under WSL, please install `chemfiles 0.10.2`.
+
+```
+conda install chemfiles-lib=0.10.2 chemfiles-python=0.10.2
+```

mstk-0.3.4/mstk/errors.py

@@ -0,0 +1,14 @@
+class PackmolError(Exception):
+    pass
+
+
+class GmxError(Exception):
+    pass
+
+
+class RDKitError(Exception):
+    pass
+
+
+class SchedulerError(Exception):
+    pass

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/forcefield/typer/zft.py

@@ -98,7 +98,7 @@ class ZftTyper(Typer):
 
     Notes
     -----
-    * SMARTS is parsed by using OpenBabel package. Therefore `pybel` module should be installed.
+    * SMARTS is parsed by using RDKit package. Make sure it is installed.
     * In type definition file, empty lines are ignored, and comments should start with ##.
 
     '''

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/ommhelper/reporter/statedatareporter.py

@@ -122,7 +122,7 @@ class StateDataReporter(object):
     '''
 
     def __init__(self, file, reportInterval, step=True, time=False, potentialEnergy=True,
-                 kineticEnergy=False, totalEnergy=False, temperature=True, volume=False, box=True,
+                 kineticEnergy=False, totalEnergy=False, temperature=True, volume=True, box=True,
                  density=True, progress=False, remainingTime=False, speed=True, elapsedTime=False,
                  separator='\t', systemMass=None, totalSteps=None, append=False,
                  cvs=None, pressure=True, pxx=False, pyy=False, pzz=False, extra={}):
@@ -314,8 +314,7 @@ class StateDataReporter(object):
 
         bool_press = [self._pxx, self._pyy, self._pzz]
         if any(bool_press):
-            indexes = [i for i in range(3) if bool_press[i]]
-            values.extend(self._compute_anisotropic_pressure(simulation.context, state, indexes))
+            values.extend(self._compute_anisotropic_pressure(simulation.context, state, *bool_press))
 
         if self._extra:
             values.extend(self._extra.values())
@@ -467,9 +466,20 @@ class StateDataReporter(object):
 
         return p_kinetic + p_virial
 
-    def _compute_anisotropic_pressure(self, context: mm.Context, state: mm.State, indexes: [bool]):
+    def _compute_anisotropic_pressure(self, context, state, pxx, pyy, pzz):
         '''
         Compute the anisotropic pressure of a rectangular system
+
+        Parameters
+        ----------
+        context : mm.Context
+        state : mm.State
+        pxx : bool
+            Whether or not the Pxx be calculated
+        pyy : bool
+            Whether or not the Pyy be calculated
+        pzz : bool
+            Whether or not the Pzz be calculated
         '''
         box = state.getPeriodicBoxVectors(asNumpy=True)
         positions = state.getPositions(asNumpy=True)
@@ -485,7 +495,9 @@ class StateDataReporter(object):
 
         scale = 0.0001
         pressures = []
-        for index in indexes:
+        for index, _bool in enumerate([pxx, pyy, pzz]):
+            if not _bool:
+                continue
             scale_array = np.array([1.0, 1.0, 1.0])
             scale_array[index] = 1 + scale
 

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/ommhelper/utils.py

@@ -1,7 +1,6 @@
-import openmm
-import openmm.openmm as mm
 import numpy as np
-from openmm import app
+import openmm
+from openmm import openmm as mm, app
 from .grofile import GroFile
 from .unit import kelvin, bar, nm, ps
 
@@ -68,6 +67,12 @@ def apply_mc_barostat(system, pcoupl, P, T, nsteps=100, logger=None):
     '''
     Add a MonteCarlo barostat to the system and return the index of the added barostat
 
+    Parameters
+    ----------
+    system : mm.System
+    pcoupl : str
+        The type of barostat. Can be one of the following: iso, semi-iso, aniso, xy, z.
+
     Returns
     -------
     idx : index of the added barostat force
@@ -80,7 +85,7 @@ def apply_mc_barostat(system, pcoupl, P, T, nsteps=100, logger=None):
         force = mm.MonteCarloMembraneBarostat(P * bar, 0 * bar * nm, T * kelvin,
                                               mm.MonteCarloMembraneBarostat.XYIsotropic,
                                               mm.MonteCarloMembraneBarostat.ZFree, nsteps)
-    elif pcoupl == 'xyz':
+    elif pcoupl == 'aniso':
         msg = 'Anisotropic barostat'
         force = mm.MonteCarloAnisotropicBarostat([P * bar] * 3, T * kelvin, True, True, True, nsteps)
     elif pcoupl == 'xy':

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/scheduler/slurm.py

@@ -109,8 +109,8 @@ class Slurm(Scheduler):
         workdir = Path(workdir).absolute().as_posix()
         sh = sh or self.sh
         sh_basename = Path(sh).stem
-        out = os.path.join(workdir, sh_basename + '.out')
-        err = os.path.join(workdir, sh_basename + '.err')
+        out = sh_basename + '.out'
+        err = sh_basename + '.err'
         node_cmd = f'#SBATCH --nodes={self.n_node}\n' if self.n_node > 0 else ''
         gpu_cmd = f'#SBATCH --gres=gpu:{self.n_gpu}\n' if self.n_gpu > 0 else ''
         dep_cmd = f'#SBATCH --dependency=afterok:{id_prior}\n' if id_prior is not None else ''  # id_prior can be 0

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/simsys/ommexporter.py

@@ -143,26 +143,29 @@ class OpenMMExporter:
             elif angle_class == SDKAngleTerm:
                 logger.debug('Setting up SDK angles...')
                 aforce = mm.CustomCompoundBondForce(
-                    3, 'k*(theta-theta0)^2+step(rmin-r)*LJ96;'
-                       'LJ96=6.75*epsilon*((sigma/r)^9-(sigma/r)^6)+epsilon;'
+                    3, 'k*(theta-theta0)^2+step(rmin-r)*LJ;'
+                       'LJ=C*epsilon*((sigma/r)^n-(sigma/r)^m)+epsilon;'
                        'theta=angle(p1,p2,p3);'
                        'r=distance(p1,p3);'
-                       'rmin=1.144714*sigma')
+                       'C=n/(n-m)*(n/m)^(m/(n-m));'
+                       'rmin=(n/m)^(1/(n-m))*sigma')
                 aforce.addPerBondParameter('theta0')
                 aforce.addPerBondParameter('k')
                 aforce.addPerBondParameter('epsilon')
                 aforce.addPerBondParameter('sigma')
+                aforce.addPerBondParameter('n')
+                aforce.addPerBondParameter('m')
                 for angle in top.angles:
                     if angle in system.constrain_angles:
                         continue
                     aterm = system.angle_terms[angle]
                     if type(aterm) != SDKAngleTerm:
                         continue
-                    vdw = ff.get_vdw_term(ff.atom_types[angle.atom1.type], ff.atom_types[angle.atom2.type])
-                    if type(vdw) != MieTerm or vdw.repulsion != 9 or vdw.attraction != 6:
-                        raise Exception(f'Corresponding 9-6 MieTerm for {aterm} not found in FF')
+                    vdw = ff.get_vdw_term(ff.atom_types[angle.atom1.type], ff.atom_types[angle.atom3.type])
+                    if type(vdw) != MieTerm:
+                        raise Exception(f'Corresponding MieTerm for {aterm} not found in FF')
                     aforce.addBond([angle.atom1.id, angle.atom2.id, angle.atom3.id],
-                                   [aterm.theta, aterm.k, vdw.epsilon, vdw.sigma])
+                                   [aterm.theta, aterm.k, vdw.epsilon, vdw.sigma, vdw.repulsion, vdw.attraction])
             elif angle_class == LinearAngleTerm:
                 logger.debug('Setting up linear angles...')
                 aforce = mm.CustomCompoundBondForce(

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/topology/connectivity.py

@@ -1,5 +1,6 @@
 import numpy as np
 from .atom import Atom
+from .geometry import periodic_distance, periodic_angle, periodic_dihedral
 
 __all__ = [
     'Bond',
@@ -115,14 +116,21 @@ class Bond():
         '''
         return self.atom1.is_drude or self.atom2.is_drude
 
-    def evaluate(self):
+    def evaluate(self, cell=None):
         '''
         Evaluate the length of this bond
 
+        Parameters
+        ----------
+        box : UnitCell, Optional
+
         Returns
         -------
         value : float
         '''
+        if cell:
+            return periodic_distance(self.atom1.position, self.atom2.position, cell.size)
+
         delta = self.atom2.position - self.atom1.position
         return float(np.sqrt(delta.dot(delta)))
 
@@ -225,14 +233,21 @@ class Angle():
 
         return bond12, bond23
 
-    def evaluate(self):
+    def evaluate(self, cell=None):
         '''
         Evaluate the value of this angle in unit of radian
 
+        Parameters
+        ----------
+        cell : UnitCell, Optional
+
         Returns
         -------
         value : float
         '''
+        if cell:
+            return periodic_angle(self.atom1.position, self.atom2.position, self.atom3.position, cell)
+
         vec1 = self.atom1.position - self.atom2.position
         vec2 = self.atom3.position - self.atom2.position
         cos = vec1.dot(vec2) / np.sqrt(vec1.dot(vec1) * vec2.dot(vec2))
@@ -361,14 +376,22 @@ class Dihedral():
 
         return angle123, angle234
 
-    def evaluate(self):
+    def evaluate(self, cell=None):
         '''
         Evaluate the value of this dihedral in unit of radian
 
+        Parameters
+        ----------
+        cell : UnitCell, Optional
+
         Returns
         -------
         value : float
         '''
+        if cell:
+            return periodic_dihedral(self.atom1.position, self.atom2.position, self.atom3.position, self.atom4.position,
+                                     cell.size)
+
         vec1 = self.atom2.position - self.atom1.position
         vec2 = self.atom3.position - self.atom2.position
         vec3 = self.atom4.position - self.atom3.position
@@ -450,15 +473,23 @@ class Improper():
         '''
         return self.atom1, self.atom2, self.atom3, self.atom4
 
-    def evaluate(self):
+    def evaluate(self, cell=None):
         '''
         Evaluate the value of this improper torsion in unit of radian.
         The improper is defined as the angle between plane a1-a2-a3 and a2-a3-a4.
 
+        Parameters
+        ----------
+        cell : UnitCell, Optional
+
         Returns
         -------
         value : float
         '''
+        if cell:
+            return periodic_dihedral(self.atom1.position, self.atom2.position, self.atom3.position, self.atom4.position,
+                                     cell.size)
+
         vec1 = self.atom2.position - self.atom1.position
         vec2 = self.atom3.position - self.atom2.position
         vec3 = self.atom4.position - self.atom3.position

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/topology/geometry.py

@@ -62,6 +62,22 @@ def grow_particle(pos1, pos2, bond, angle):
     return pos2 + w * bond
 
 
+def relocate_hydrogen(hydrogen):
+    '''
+    Update the position of a hydrogen atom according to the parent atom and other neighbors of parent
+
+    Parameters
+    ----------
+    hydrogen : Atom
+    '''
+    vector = np.array([0., 0., 0.])
+    parent = hydrogen.bond_partners[0]
+    for atom in parent.bond_partners:
+        if atom is not hydrogen:
+            vector += atom.position - parent.position
+    hydrogen.position = parent.position - vector / np.sqrt(np.dot(vector, vector)) * 0.1  # bXH = 0.1 nm
+
+
 def periodic_distance(pos1, pos2, box, distance_max=None):
     '''
     Calculate the distance between two points under periodic boundary condition
@@ -155,7 +171,7 @@ def periodic_dihedral(pos1, pos2, pos3, pos4, box):
     return sign * value
 
 
-def find_clusters(elements, func):
+def find_clusters(elements, func, show_progress=False):
     '''
     Group elements into clusters
 
@@ -175,7 +191,12 @@ def find_clusters(elements, func):
     flag = [0] * n_element
     n_cluster = 0
     clusters = []
-    for i in range(n_element):
+    if show_progress:
+        from tqdm import tqdm
+        _iterator = tqdm(range(n_element))
+    else:
+        _iterator = range(n_element)
+    for i in _iterator:
         if flag[i]:
             continue
         path = {i}
@@ -203,7 +224,7 @@ def find_clusters(elements, func):
     return clusters
 
 
-def find_clusters_consecutive(elements, func):
+def find_clusters_consecutive(elements, func, show_progress=False):
     '''
     Group elements into clusters. If element i and j are in the same group, all elements between i and j will also be put in the same group.
 
@@ -223,7 +244,12 @@ def find_clusters_consecutive(elements, func):
     flag = [0] * n_element
     n_cluster = 0
     clusters = []
-    for i in range(n_element):
+    if show_progress:
+        from tqdm import tqdm
+        _iterator = tqdm(range(n_element))
+    else:
+        _iterator = range(n_element)
+    for i in _iterator:
         if not flag[i]:
             n_cluster += 1
             flag[i] = n_cluster

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/topology/io/pdb.py

@@ -176,7 +176,7 @@ class Pdb:
             atom_name = atom.type if atom_type else atom.name
             resname = atom.residue.name
             resid = atom.residue.id + 1
-            line = 'HETATM%5d %4s %4s %4d    %8.3f%8.3f%8.3f                      %2s\n' % (
+            line = 'HETATM%5d %4s %-4s %4d    %8.3f%8.3f%8.3f                      %2s\n' % (
                 (atom.id + 1) % 100000, atom_name[:4], resname[:4], resid % 10000,
                 pos[0], pos[1], pos[2], atom.symbol[:2])
             string += line

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/topology/molecule.py

@@ -214,27 +214,22 @@ class Molecule():
         The `rdkit.Chem.Mol` object associated with this molecule.
 
         It is required by ZftTyper typing engine, which performs SMARTS matching on the molecule.
-        The obmol attribute will be assigned if the molecule is initialized from SMILES or Pybel Molecule.
-        If this information is not available, an Exception will be raised.
+        The `rdmol` attribute will be assigned if the molecule is initialized from SMILES or RDKit Molecule.
+        If it is not available, a RDKit molecule will be constructed from atoms and bonds.
+        The positions will not be preserved.
 
         Returns
         -------
         rdmol : rdkit.Chem.Mol
         '''
-        if not self._is_rdmol_valid:
-            self._construct_rdmol()
-
-        return self._rdmol
-
-    def _construct_rdmol(self):
-        '''
-        Construct a RDKit molecule from atoms and bonds. The positions will not be preserved.
-        '''
         try:
             from rdkit import Chem
         except ImportError:
             raise ImportError('RDKit not found')
 
+        if self._is_rdmol_valid:
+            return self._rdmol
+
         if any(b.order == Bond.Order.UNSPECIFIED for b in self.bonds):
             logger.warning(f'Not all bond orders are specified in {self}')
 
@@ -256,6 +251,38 @@ class Molecule():
         self._rdmol = rwmol.GetMol()
         self._is_rdmol_valid = True
 
+        return self._rdmol
+
+    def generate_conformers(self, n_conformer=1):
+        '''
+        Generate several conformers with RDKit.
+
+        The positions will be generated from only elements and bonds. The chiral center will not be respected.
+
+        Parameters
+        ----------
+        n_conformer : int
+            How many conformers to generate
+
+        Returns
+        -------
+        molecules : list of Molecule
+            Each conformer will be a independent molecule object
+        '''
+        try:
+            from rdkit.Chem import AllChem as Chem
+        except ImportError:
+            raise ImportError('RDKit not found')
+
+        molecules = []
+        rdmol = Chem.Mol(self.rdmol)
+        Chem.EmbedMultipleConfs(rdmol, numConfs=n_conformer, clearConfs=True)
+        for i in range(rdmol.GetNumConformers()):
+            molecules.append(Molecule.from_rdmol(rdmol, name=self.name))
+            rdmol.RemoveConformer(i)
+
+        return molecules
+
     @property
     def topology(self):
         '''
@@ -317,6 +344,8 @@ class Molecule():
         The angle, dihedral and improper involving this atom are untouched.
         Therefore, you may call `generate_angle_dihedral_improper` to refresh the connectivity.
 
+        # TODO This operation is extremely slow
+
         Parameters
         ----------
         atom : Atom
@@ -351,13 +380,14 @@ class Molecule():
             If update_topology is True, the topology this molecule belongs to will update its atom list and assign id for all atoms and residues.
             Otherwise, you have to re-init the topology manually so that the topological information is correct.
 
+        # TODO This operation is extremely slow
+
         Returns
         -------
         ids_removed : list of int
             The number of atoms removed
         '''
         hydrogens = []
-        ids_removed = []
         for atom in self.atoms[:]:
             if atom.symbol != 'H' or len(atom.bonds) != 1:
                 continue
@@ -366,11 +396,13 @@ class Molecule():
                 continue
             neigh.mass += atom.mass
             neigh.charge += atom.charge
-            for conn in self.bonds[:] + self.angles[:] + self.dihedrals[:] + self.impropers[:]:
-                if atom in conn.atoms:
-                    self.remove_connectivity(conn)
-            ids_removed.append(atom.id_in_mol)
             hydrogens.append(atom)
+        ids_hydrogens = [atom.id_in_mol for atom in hydrogens]
+
+        for conn in self.bonds[:] + self.angles[:] + self.dihedrals[:] + self.impropers[:]:
+            ids_set = {atom.id_in_mol for atom in conn.atoms}
+            if ids_set.intersection(ids_hydrogens):
+                self.remove_connectivity(conn)
 
         for atom in hydrogens:
             self.remove_atom(atom, update_topology=False)
@@ -378,7 +410,7 @@ class Molecule():
         if self._topology is not None and update_topology:
             self._topology.update_molecules(self._topology.molecules, deepcopy=False)
 
-        return ids_removed
+        return ids_hydrogens
 
     def add_residue(self, name, atoms, update_topology=True):
         '''
@@ -563,6 +595,8 @@ class Molecule():
         Note that when a bond get removed, the relevant angles, dihedrals and impropers are still there.
         You may call `generate_angle_dihedral_improper` to refresh connectivity.
 
+        # TODO This operation is extremely slow
+
         Parameters
         ----------
         connectivity : [Bond, Angle, Dihedral, Improper]

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/trajectory/trajectory.py

@@ -76,6 +76,9 @@ class Trajectory():
     def __del__(self):
         self.close()
 
+    def __repr__(self):
+        return f'<Trajectory: {self.n_frame} frames {self.n_atom} atoms>'
+
     def close(self):
         '''
         Close the opened trajectory file(s).
@@ -97,7 +100,8 @@ class Trajectory():
         this method should not be used because all the frames actually point to the same frame.
         In this case, use :func:`read_frames` instead.
 
-        i_frame should be in the range of [0, n_frame), otherwise and Exception will be raised.
+        i_frame should be in the range of [-1, n_frame), otherwise and Exception will be raised.
+        -1 means the last frame.
 
         Parameters
         ----------
@@ -123,6 +127,8 @@ class Trajectory():
         # Reset the information in self.frame in case the frames read from different trajectory files pollute each other for CombinedTrajectory
         self.frame.reset()
 
+        if i_frame == -1:
+            i_frame = self.n_frame - 1
         self._handler.read_frame(i_frame, self.frame)
         return self.frame
 
@@ -134,7 +140,8 @@ class Trajectory():
         Instead, a new Frame object is constructed for each frame.
         This method should be called when you want to multiprocess several frames in parallel.
 
-        All the items in i_frames should be in the range of (0, n_frame), otherwise and Exception will be raised.
+        All the items in i_frames should be in the range of [-1, n_frame), otherwise and Exception will be raised.
+        -1 means the last frame.
 
         Parameters
         ----------
@@ -153,6 +160,8 @@ class Trajectory():
 
         frames = [Frame(self.n_atom) for _ in i_frames]
         for ii, i_frame in enumerate(i_frames):
+            if i_frame == -1:
+                i_frame = self.n_frame - 1
             self._handler.read_frame(i_frame, frames[ii])
         return frames
 
@@ -219,8 +228,6 @@ class Trajectory():
 
         '''
         trj = Trajectory(file, 'r')
-        if i_frame == -1:
-            i_frame = trj.n_frame - 1
         frame = trj.read_frame(i_frame)
         trj.close()
         return frame

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk/utils/__init__.py

@@ -7,6 +7,7 @@ import subprocess
 import random
 import string
 import numpy as np
+from typing import Iterable
 
 
 def greatest_common_divisor(numbers):
@@ -187,3 +188,11 @@ def align_mpl_axis(ax1, ax2):
 
     ax1.set(ylim=ylims1)
     ax2.set(ylim=ylims2)
+
+
+def flatten(it):
+    for x in it:
+        if (isinstance(x, Iterable) and not isinstance(x, str)):
+            yield from flatten(x)
+        else:
+            yield x

{mstk-0.3.3.dev11 → mstk-0.3.4}/mstk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mstk
-Version: 0.3.3.dev11
+Version: 0.3.4
 Summary: Molecular simulation toolkit
 Home-page: https://github.com/z-gong/mstk
 Author: Zheng Gong
@@ -69,3 +69,11 @@ https://mstk.readthedocs.io/en/latest/index.html
 
 - [ ] Take bond order into consideration for force field assignment
 - [ ] Re-organize algorithms scattered in topology and analyzer modules
+
+## Known issue
+`mstk` use `chemfiles` to write `XTC` trajectory. However, lastest `chemfiles` is buggy under WSL for writing binary trajectory format. 
+If you are working under WSL, please install `chemfiles 0.10.2`.
+
+```
+conda install chemfiles-lib=0.10.2 chemfiles-python=0.10.2
+```