Merge pull request #154 from hjkgrp/code_quality

Code quality

.github/workflows/CI.yaml

@@ -79,6 +79,11 @@ jobs:
       run: |
         pytest -v --cov=molSimplify --cov-report=xml
     
+    - name: Run doctest
+      # For now still excluding several subfolders and files
+      run: |
+        pytest --doctest-modules --ignore=molSimplify/job_manager --ignore=molSimplify/Informatics/MOF --ignore=molSimplify/Scripts/in_b3lyp_usetc.py --ignore=molSimplify/Informatics/jupyter_vis.py --ignore=molSimplify/Informatics/macrocycle_synthesis.py --ignore=molSimplify/Informatics/organic_fingerprints.py molSimplify
+    
     - name: Upload coverage report to codecov
       uses: codecov/codecov-action@v3
       with:
@@ -92,7 +97,7 @@ jobs:
 
     - name: Report Status
       # Slack notifications only on the main repo
-      if: ${{ github.event_name != 'pull_request' && github.repository == 'hjkgrp/molSimplify' }}
+      if: ${{job.status == 'failure' && github.event_name != 'pull_request' && github.repository == 'hjkgrp/molSimplify' }}
       #uses: ravsamhq/notify-slack-action@v1
       uses: 8398a7/action-slack@v3
       with:

devtools/conda-envs/mols.yml

@@ -13,15 +13,20 @@ dependencies:
     - pytest
 
   # Optional
-  # GUI
+  # GUI / plotting
   - pyqt
   - qt
+  - matplotlib
   # Database
   - pymongo
+  # PDB interaction
+  - beautifulsoup4
   # Machine learning
   #- pytorch::pytorch
   #- hyperopt
   # QM interfaces
   - theochem::iodata
   - xtb
   #- psi4::psi4
+  # molscontrol
+  - scikit-image

molSimplify/Classes/README.md

@@ -1 +1 @@
-To members of the Kulik group: If you have custom data that you want to use for yourself but do not want to push to the main repo, e.g. custom ligands for molSimplify ligand construction, you can make a file `.molSimplify` and include in it a line like `CUSTOM_DATA_PATH=/Users/ralf/molSimplify_custom_data`. In that folder you specified, you can place a folder `Ligands`, and molSimplify will look there first before looking at the Ligands folder in the central repo. This functionality is implemented in `globalvars.py`.
+To members of the Kulik group: If you have custom data that you want to use for yourself but do not want to push to the main repo, e.g. custom ligands for molSimplify ligand construction, you can make a file `.molSimplify` and include in it a line like `CUSTOM_DATA_PATH=/Users/your_user/molSimplify_custom_data`. In that folder you specified, you can place a folder `Ligands`, and molSimplify will look there first before looking at the Ligands folder in the central repo. This functionality is implemented in `globalvars.py`.

molSimplify/Classes/mGUI.py

@@ -42,7 +42,10 @@
 from pkg_resources import resource_filename, Requirement
 import xml.etree.ElementTree as ET
 
-import openbabel
+try:
+    from openbabel import openbabel  # version 3 style import
+except ImportError:
+    import openbabel  # fallback to version 2
 
 # Main GUI class
 

molSimplify/Classes/mol3D.py

@@ -12,7 +12,10 @@
 import time
 import xml.etree.ElementTree as ET
 import numpy as np
-import openbabel
+try:
+    from openbabel import openbabel  # version 3 style import
+except ImportError:
+    import openbabel  # fallback to version 2
 from typing import List, Optional
 from scipy.spatial import ConvexHull
 from molSimplify.utils.decorators import deprecated
@@ -42,7 +45,7 @@ class mol3D:
     Example instantiation of an octahedral iron-ammonia complex from an XYZ file:
 
     >>> complex_mol = mol3D()
-    >>> complex_mol.readfromxyz('fe_nh3_6.xyz')
+    >>> complex_mol.readfromxyz('fe_nh3_6.xyz')  # doctest: +SKIP
 
     """
 
@@ -122,21 +125,9 @@ def __init__(self, name='ABC', loc='', use_atom_specific_cutoffs=False):
         self.use_atom_specific_cutoffs = use_atom_specific_cutoffs
 
     def __repr__(self):
-        """
-        Returns all bound methods of the mol3D class.
-
-        Returns
-        -------
-            method_string : string
-                String of methods available in mol3D class.
-        """
-
-        method_string = "\nClass mol3D has the following methods:\n"
-        for method in dir(self):
-            if callable(getattr(self, method)):
-                method_string += method + '\n'
-        return method_string
+        return f"mol3D({self.make_formula(latex=False)})"
 
+    @deprecated("Preliminary testing showed this function is unreliable at best")
     def ACM(self, idx1, idx2, idx3, angle):
         """
         Performs angular movement on mol3D class. A submolecule is
@@ -152,8 +143,6 @@ def ACM(self, idx1, idx2, idx3, angle):
                 Index of anchor atom 2.
             angle : float
                 New bond angle in degrees.
-
-        >>> complex_mol.ACM(2, 1, 0, 180) # Make atom 2 form a 180 degree angle with atoms 1 and 0.
         """
 
         atidxs_to_move = self.findsubMol(idx1, idx2)
@@ -231,7 +220,8 @@ def addAtom(self, atom: atom3D, index: Optional[int] = None, auto_populate_BO_di
             auto_populate_BO_dict : bool, optional
                 Populate bond order dictionary with newly added atom. Default is True.
 
-        >>> C_atom = atom3D('C',[1, 1, 1])
+        >>> complex_mol = mol3D()
+        >>> C_atom = atom3D('C', [1, 1, 1])
         >>> complex_mol.addAtom(C_atom) # Add carbon atom at cartesian position 1, 1, 1 to mol3D object.
         """
 
@@ -484,7 +474,13 @@ def BCM(self, idx1, idx2, d):
             d : float
                 Bond distance in angstroms.
 
-        >>> complex_mol.BCM(1, 0, 1.5) # Set distance between atoms 0 and 1 to be 1.5 angstroms. Move atom 1.
+        >>> complex_mol = mol3D()
+        >>> complex_mol.addAtom(atom3D('H', [0, 0, 0]))
+        >>> complex_mol.addAtom(atom3D('H', [0, 0, 1]))
+        >>> complex_mol.BCM(1, 0, 0.7) # Set distance between atoms 0 and 1 to be 1.5 angstroms. Move atom 1.
+        >>> complex_mol.coordsvect()
+        array([[0. , 0. , 0. ],
+               [0. , 0. , 0.7]])
         """
 
         bondv = self.getAtom(idx1).distancev(self.getAtom(idx2))  # 1 - 2
@@ -1663,7 +1659,7 @@ def getBondedAtoms(self, idx):
 
         """
 
-        if len(self.graph): # The graph exists.
+        if len(self.graph):  # The graph exists.
             nats = list(np.nonzero(np.ravel(self.graph[idx]))[0])
         else:
             ratom = self.getAtom(idx)
@@ -2577,25 +2573,33 @@ def RCAngle(self, idx1, idx2, idx3, anglei, anglef, angleint=1.0, writegeo=False
                 The directory to which generated reaction coordinate
                 geoemtries are written, if writegeo=True.
 
-        >>> complex_mol.RCAngle(2, 1, 0, 90, 180, 0.5, True, 'rc_geometries') # Generate reaction coordinate
-        >>>             # geometries using the given structure by changing the angle between atoms 2, 1,
-        >>>             # and 0 from 90 degrees to 180 degrees in intervals of 0.5 degrees, and write the
-        >>>             # generated geometries to 'rc_geometries' directory.
-        >>> complex_mol.RCAngle(2, 1, 0, 180, 90, -0.5) # Generate reaction coordinates
-        >>>             # with the given geometry by changing the angle between atoms 2, 1, and 0 from
-        >>>             # 180 degrees to 90 degrees in intervals of 0.5 degrees, and the generated
-        >>>             # geometries will not be written to a directory.
+        >>> complex_mol = mol3D()
+        >>> complex_mol.addAtom(atom3D('O', [0, 0, 0]))
+        >>> complex_mol.addAtom(atom3D('H', [0, 0, 1]))
+        >>> complex_mol.addAtom(atom3D('H', [0, 1, 0]))
+
+        Generate reaction coordinate geometries using the given structure by changing the angle between atoms 2, 1,
+        and 0 from 90 degrees to 160 degrees in intervals of 10 degrees
+        >>> complex_mol.RCAngle(2, 1, 0, 90, 160, 10)
+        [mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2)]
+
+        Generate reaction coordinates with the given geometry by changing the angle between atoms 2, 1, and 0 from
+        160 degrees to 90 degrees in intervals of 10 degrees, and the generated geometries will not be written to
+        a directory.
+        >>> complex_mol.RCAngle(2, 1, 0, 160, 90, -10)
+        [mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2), mol3D(O1H2)]
         """
         if writegeo:
             os.mkdir(dir_name)
-            temp_list = []
-            for ang_val in np.arange(anglei, anglef+angleint, angleint):
-                temp_angle = mol3D()
-                temp_angle.copymol3D(self)
-                temp_angle.ACM(idx1, idx2, idx3, ang_val)
-                temp_list.append(temp_angle)
+        temp_list = []
+        for ang_val in np.arange(anglei, anglef+angleint, angleint):
+            temp_angle = mol3D()
+            temp_angle.copymol3D(self)
+            temp_angle.ACM(idx1, idx2, idx3, ang_val)
+            temp_list.append(temp_angle)
+            if writegeo:
                 temp_angle.writexyz(str(dir_name)+"/rc_"+str(str("{:.4f}".format(ang_val)))+'.xyz')
-            return temp_list
+        return temp_list
 
     def RCDistance(self, idx1, idx2, disti, distf, distint=0.05, writegeo=False, dir_name='rc_distance_geometries'):
         """
@@ -2622,26 +2626,34 @@ def RCDistance(self, idx1, idx2, disti, distf, distint=0.05, writegeo=False, dir
                 The directory to which generated reaction coordinate
                 geoemtries are written if writegeo=True.
 
-        >>> complex_mol.RCDistance(1, 0, 1.0, 3.0, 0.01, True, 'rc_geometries') # Generate reaction coordinate
-        >>>             # geometries using the given structure by changing the distance between atoms 1 and 0
-        >>>             # from 1.0 to 3.0 angstrom (atom 1 is moved) in intervals of 0.01 angstrom, and write
-        >>>             # the generated geometries to 'rc_geometries' directory.
-        >>> complex_mol.RCDistance(1, 0, 3.0, 1.0, -0.02) # Generate reaction coordinates
-        >>>             # geometries using the given structure by changing the distance between atoms 1 and 0
-        >>>             # from 3.0 to 1.0 angstrom (atom 1 is moved) in intervals of 0.02 angstrom, and
-        >>>             # the generated geometries will not be written to a directory.
+        >>> complex_mol = mol3D()
+        >>> complex_mol.addAtom(atom3D('H', [0, 0, 0]))
+        >>> complex_mol.addAtom(atom3D('H', [0, 0, 1]))
+
+        Generate reaction coordinate geometries using the given structure by changing the distance between atoms 1 and 0
+        from 1.0 to 3.0 angstrom (atom 1 is moved) in intervals of 0.5 angstrom
+        >>> complex_mol.RCDistance(1, 0, 1.0, 3.0, 0.5)
+        [mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2)]
+
+        Generate reaction coordinates
+        geometries using the given structure by changing the distance between atoms 1 and 0
+        from 3.0 to 1.0 angstrom (atom 1 is moved) in intervals of 0.2 angstrom, and
+        the generated geometries will not be written to a directory.
+        >>> complex_mol.RCDistance(1, 0, 3.0, 1.0, -0.25)
+        [mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2), mol3D(H2)]
         """
 
         if writegeo:
             os.mkdir(dir_name)
-            temp_list = []
-            for dist_val in np.arange(disti, distf+distint, distint):
-                temp_dist = mol3D()
-                temp_dist.copymol3D(self)
-                temp_dist.BCM(idx1, idx2, dist_val)
-                temp_list.append(temp_dist)
+        temp_list = []
+        for dist_val in np.arange(disti, distf+distint, distint):
+            temp_dist = mol3D()
+            temp_dist.copymol3D(self)
+            temp_dist.BCM(idx1, idx2, dist_val)
+            temp_list.append(temp_dist)
+            if writegeo:
                 temp_dist.writexyz(str(dir_name)+"/rc_"+str(str("{:.4f}".format(dist_val)))+'.xyz')
-            return temp_list
+        return temp_list
 
     def returnxyz(self):
         """
@@ -4840,7 +4852,7 @@ def Structure_inspection(self, init_mol=None, catoms_arr=None, num_coord=5, dict
         return flag_oct, flag_list, dict_oct_info, flag_oct_loose, flag_list_loose
 
     def get_fcs(self, strict_cutoff=False, catom_list=None):
-        """ 
+        """
         Get first coordination shell of a transition metal complex.
 
         Parameters
@@ -4865,7 +4877,7 @@ def get_fcs(self, strict_cutoff=False, catom_list=None):
         return fcs
 
     def get_bo_dict_from_inds(self, inds):
-        """ 
+        """
         Recreate bo_dict with correct indices
 
         Parameters
@@ -4897,7 +4909,7 @@ def get_bo_dict_from_inds(self, inds):
         return new_bo_dict
 
     def create_mol_with_inds(self, inds):
-        """ 
+        """
         Create molecule with indices.
 
         Parameters
@@ -4926,7 +4938,7 @@ def create_mol_with_inds(self, inds):
         return molnew
 
     def make_formula(self, latex=True):
-        """ 
+        """
         Get a chemical formula from the mol3D class instance.
 
         Parameters
@@ -4961,7 +4973,7 @@ def make_formula(self, latex=True):
         return retstr
 
     def read_smiles(self, smiles, ff="mmff94", steps=2500):
-        """ 
+        """
         Read a smiles string and convert it to a mol3D class instance.
 
         Parameters
@@ -5036,7 +5048,7 @@ def get_smiles(self, canonicalize=False, use_mol2=False) -> str:
         return smi
 
     def mols_symbols(self):
-        """ 
+        """
         Store symbols and their frequencies in symbols_dict attributes.
         """
         self.symbols_dict = {}
@@ -5047,7 +5059,7 @@ def mols_symbols(self):
                 self.symbols_dict[atom.symbol()] += 1
 
     def read_bonder_order(self, bofile):
-        """ 
+        """
         Get bond order information from file.
 
         Parameters
@@ -5091,7 +5103,7 @@ def read_bonder_order(self, bofile):
                 self.bodavrg_dict.update({ii: np.mean(devi)})
 
     def read_charge(self, chargefile):
-        """ 
+        """
         Get charge information from file.
 
         Parameters
@@ -5112,7 +5124,7 @@ def read_charge(self, chargefile):
             print(("chargefile does not exist.", chargefile))
 
     def get_mol_graph_det(self, oct=True, useBOMat=False):
-        """ 
+        """
         Get molecular graph determinant.
 
         Parameters
@@ -5521,7 +5533,7 @@ def get_features(self, lac=True, force_generate=False, eq_sym=False,
         return results
 
     def getMLBondLengths(self):
-        """ 
+        """
         Outputs the metal-ligand bond lengths in the complex.
 
         Returns
@@ -5548,7 +5560,7 @@ def getMLBondLengths(self):
 
     @deprecated('Using this function might lead to inconsistent behavior.')
     def setAtoms(self, atoms):
-        """ 
+        """
         Set atoms of a mol3D class to atoms.
 
         Parameters
@@ -5560,7 +5572,7 @@ def setAtoms(self, atoms):
         self.natoms = len(atoms)
 
     def setLoc(self, loc):
-        """ 
+        """
         Sets the conformation of an amino acid in the chain of a protein.
 
         Parameters

molSimplify/Classes/monomer3D.py

@@ -7,7 +7,8 @@
 
 
 class monomer3D:
-    """Holds information about a monomer, used to do manipulations.  Reads information from structure file (pdb) or is directly built from molsimplify.
+    """Holds information about a monomer, used to do manipulations.
+    Reads information from structure file (pdb) or is directly built from molsimplify.
 
     """
 
@@ -42,6 +43,9 @@ def __init__(self, three_lc='GLY', chain='undef', id=-1, occup=1.00, loc=''):
         # Temporary list for storing conformations
         self.temp_list = []
 
+    def __repr__(self):
+        return f"monomer3D({self.three_lc}, id={self.id})"
+
     def identify(self):
         """ States whether the amino acid is (positively/negatively) charged, polar, or hydrophobic.
 
@@ -228,9 +232,6 @@ def addAtom(self, atom, index=None):
                 Index of added atom. Default is None.
             auto_populate_BO_dict : bool, optional
                 Populate bond order dictionary with newly added atom. Default is True.
-
-        >>> C_atom = atom3D('C',[1, 1, 1])
-        >>> complex_mol.addAtom(C_atom) # Add carbon atom at cartesian position 1, 1, 1 to mol3D object.
         """
 
         if index is None: