def find_RNA_anions(residue):
"""Finds all nucleic acid's residue's anions.
:param residue: nucleic acid's residue object
:type residue: openbabel.OBResidue
:return: coords of residue's anions
:rtype: list
"""
P = None
anions_RNA = []
for atom in openbabel.OBResidueAtomIter(residue):
if atom.GetAtomicNum() == config.PHOSPHORUS_NUM:
P = atom
break # Assuming there is only 1 Phosphate in residue
if P: # 5'end has no phosphate group
O = [] # List of 2 negative charged Oxygens
for oxygen in pybel.ob.OBAtomAtomIter(P):
# https://open-babel.readthedocs.io/en/latest/UseTheLibrary/migration.html
if oxygen.GetExplicitValence() == 1 or oxygen.HasDoubleBond():
O.append(oxygen)
anions_RNA.extend(O)
return anions_RNA
def find_RNA_HB_HAL_acc_don(residue):
"""Finds all hydrogen/halogen bonds acceptors with all of their neighbours and all hydrogen bonds donors together with hydrogens in nucleic acid residue.
:param residue: nucleic acid residue object
:type residue: openbabel.OBResidue
:return: coords of residue's halogen/hydrogen bonds acceptors & their neighbours
:return: coords of residue's hydrogen bonds donors together with hydrogens
:rtype: list
:rtype: list
"""
residue_atoms_list = []
for atom in openbabel.OBResidueAtomIter(residue):
residue_atoms_list.append(atom) # Create list of all residue's atoms
acceptors_RNA = [
] # List of sublists [[acceptor,acceptor',acceptor'(if the 2nd one exists)]]
donors_RNA = []
for atom in filter(lambda at: at.IsHbondAcceptor(),
residue_atoms_list): # Find all acceptors
acceptors_RNA.append([atom])
for neighbour in pybel.ob.OBAtomAtomIter(atom):
if neighbour.GetAtomicNum() != 1: # If not hydrogen
acceptors_RNA[-1].append(neighbour) # Append acceptor'
for atom in filter(
lambda at: at.IsHbondDonor(),
residue_atoms_list): # Find all donors with their hydrogens
for neighbour in pybel.ob.OBAtomAtomIter(atom):
if neighbour.IsHbondDonorH():
donors_RNA.append((atom, neighbour))
return acceptors_RNA, donors_RNA
def GetAtomByID(protein, residueNumber, name):
"""
Returns the OBAtom object given the AtomID (PDB Columns 13-16),
the protein name (ace2 or spike), and the residue number
"""
for res in ob.OBResidueIter(protein):
if res.GetNum() == residueNumber:
for atom in ob.OBResidueAtomIter(res):
if res.GetAtomID(atom) == name:
return atom
def calculate_SIMPLE(residue, ligand_name, ligand_atoms, centroid_ligand):
"""Calculates SIMPLE interaction between residue - ligand pair:
1. Check RNA residue - ligand distance
2. Compare the distance to CUTOFF:
- write down 1 if the distance <= CUTOFF
- write down 0 if the distance > CUTOFF
:param residue: residue as OpenBabel object
:param ligand_name: ligand_name^pose_number
:param ligand_atoms: coordinates of all ligand's atoms
:type residue: openbabel.OBResidue
:type ligand_name: str
:type ligand_atoms: list
:return: [ligand_name^pose_number, residue_number:residue_chain, binary info about interaction (0/1)]
:rtype: list
"""
# List of sublists of all atoms' coords of the residue
residue_atoms = []
for atom in openbabel.OBResidueAtomIter(residue):
if atom.GetAtomicNum() != 1:
residue_atoms.append(
np.array([atom.GetX(), atom.GetY(),
atom.GetZ()]))
result = [
ligand_name,
str(residue.GetNum()) + ':' + str(residue.GetChain()), 0
]
if measure_distance(
centroid(residue_atoms), centroid_ligand
) > config.RES_LIGAND_MIN_DIST: # RNA residue centroid and ligand centroid are futher than declared threshold, no chance for any contact
return result
# Flag to iterate over residue's atoms as long as we do not find an atom within CUTOFF distance from ligand
flag = True
for rna_atom in residue_atoms:
if flag:
for ligand_atom in ligand_atoms:
if check_distance(ligand_atom, rna_atom,
config.CUT_OFF_SIMPLE):
result[-1] = 1 # Condition met; write down 1
flag = False
break
else:
break
return result
def GetResidueMolecule(protein, residueNumber):
"""
Gets the OBMol object corresponding to a residue number on ACE2 protein or Spike protein
protein: 'ace2' or 'spike'
residueNumber: The residue Number
returns: The OBMol object representing the molecule
"""
resMol = ob.OBMol()
for res in ob.OBResidueIter(protein):
if res.GetNum() == residueNumber:
for atom in ob.OBResidueAtomIter(res):
resMol.AddAtom(atom)
break
return resMol
def rna_coords_atom_index_dict(structure):
"""For the debug/detail mode only; creates a dictionary or nucleic acid's atoms' coords with their atom ids as values
:param structure: nucleic acid structure object
:type structure: pybel.Molecule
:return: dictionary indexed by nucleic acid's atoms' coords, with their atom ids as values
:rtype: dict
"""
dictionary = {}
for residue in openbabel.OBResidueIter(structure.OBMol):
if residue.GetNumAtoms() > 3:
for atom in openbabel.OBResidueAtomIter(residue):
dictionary[(
atom.GetX(), atom.GetY(),
atom.GetZ())] = atom.GetResidue().GetAtomID(atom).strip()
return dictionary
def atoms(self):
"""List of Atoms in the Residue"""
return [Atom(atom) for atom in ob.OBResidueAtomIter(self.OBResidue)]
def printAtomNames(mol):
for res in openbabel.OBResidueIter(mol):
for atom in openbabel.OBResidueAtomIter(res):
name = res.GetAtomID(atom)
print(name)
def __init__(self, protein, res_name, res_num, custom_settings):
self.residue_number = int(res_num) - 1
self.residue = protein.GetResidue(self.residue_number)
self.atoms = [atom for atom in ob.OBResidueAtomIter(self.residue)]
# Making atom index consistent by separating hydrogen and heavy atom
self.heavyatoms = []
self.hydrogens = []
for atom in self.atoms:
if atom.GetAtomicNum() == 1:
self.hydrogens.append(atom)
else:
self.heavyatoms.append(atom)
self.output_mode = custom_settings['output_mode']
self.simplified = False
self.full = False
self.full_nobb = False
if self.output_mode['full']:
self.full = True
if self.output_mode['full_nobb']:
self.full_nobb = True
if self.output_mode['simplified']:
self.simplified = True
self.full_bitstring = bitarray('0000000')
self.full_nobb_bitstring = bitarray('0000000')
self.simp_bitstring = self.bs_template[res_name[:3]]
self.res_name = res_name
self.AA_name = res_name[:3]
self.res_num = res_num
self.interactions = self.AAinteractionMatrix[self.AA_name]
# Classifying atoms into atom groups
self.atomGroup = {}
for AA_int, int_atom, int_name in \
zip(self.interactions, self.interactionList, self.interactionNames):
if AA_int == 1:
atomIDs = int_atom[self.AA_name]['id']
atomList = [self.heavyatoms[atomID] for atomID in atomIDs]
self.atomGroup[int_name] = atomList
if int_name == 'h_donor':
h_donorh_list = self.h_donorh[self.AA_name]['id']
h_donorh_atoms = []
for h_list in h_donorh_list:
hydrogen = [self.hydrogens[atomID] for atomID in h_list]
h_donorh_atoms.append(hydrogen)
self.atomGroup['h_donorh'] = h_donorh_atoms
if self.AA_name in self.aromatic.keys():
self.path3 = self.aromatic[self.AA_name]['id'][:3]
self.path3atoms = []
for i in self.path3:
self.path3atoms.append(self.heavyatoms[i])
self.cross, self.modulus = getCrossModulus(self.path3atoms)
self.res_weight1 = custom_settings['res_weight1']
self.res_weight2 = custom_settings['res_weight2']
self.res_weight3 = custom_settings['res_weight3']
self.res_weight4 = custom_settings['res_weight4']
self.res_weight5 = custom_settings['res_weight5']
def calculate_PBS(residue, ligand_name, ligand_atoms, centroid_ligand):
"""Calculates PBS interaction between residue - ligand pair:
1. Divide RNA residue into 3 groups Phosphate/Base/Sugar (P/B/S)
2. Check each group - ligand distance
3. Compare the distance to CUTOFF:
- write down 1 if the distance <= CUTOFF
- write down 0 if the distance > CUTOFF
:param residue: residue as OpenBabel object
:param ligand_name: ligand_name^pose_number
:param ligand_atoms: coordinates of all ligand's atoms
:type residue: openbabel.OBResidue
:type ligand_name: str
:type ligand_atoms: list
:return: [ligand_name^pose_number, residue_number:residue_chain, binary info about interaction in P group (0/1), binary info about interaction in B group (0/1), binary info about interaction in S group (0/1)]
:rtype: list
"""
# List of residue's atoms as OBAtoms objects
residue_atoms = []
residue_atoms_coords = []
for atom in openbabel.OBResidueAtomIter(residue):
if atom.GetAtomicNum() != 1:
residue_atoms.append(atom)
residue_atoms_coords.append(
np.array([atom.GetX(), atom.GetY(),
atom.GetZ()]))
result = [
ligand_name,
str(residue.GetNum()) + ':' + str(residue.GetChain()), 0, 0, 0
]
if measure_distance(
centroid(residue_atoms_coords), centroid_ligand
) > config.RES_LIGAND_MIN_DIST: # RNA residue centroid and ligand centroid are futher than declared threshold, no chance for any contact
return result
# Flag to iterate over residue's atoms as long as we do not find an atom of the defined P/B/S group within CUTOFF distance from ligand
flags = [True, True, True]
for rna_atom in residue_atoms:
# Flag to check if we deal with atom that does not belong to any of the defined P/B/S groups
atom_group_type = False
rna_atom_name = residue.GetAtomID(rna_atom).strip()
rna_atom_coords = np.array(
[rna_atom.GetX(),
rna_atom.GetY(),
rna_atom.GetZ()])
for g in range(len(config.GROUPS)):
# If atom present in the group
if rna_atom_name in config.GROUPS[g]:
atom_group_type = True # If atom belongs to one of the 3 groups
if flags[g]:
for ligand_atom in ligand_atoms:
if check_distance(ligand_atom, rna_atom_coords,
config.CUT_OFF_SIMPLE):
result[g - 3] = 1 # Condition met; write down 1
flags[g] = False
break
if not atom_group_type:
raise Exception('Unknown atom type %s' %
rna_atom_name) # MAYBE NEEDS TO BE CHANGED
return result
# Fill the RESULTS dictionary of keys - ligand ids and values - lists of 0
for ligand_name in ligands_hba_hbd.keys():
RESULTS[ligand_name] = [0] * RNA_LENGTH * FUNCTIONS[fingerprint]
for residue in openbabel.OBResidueIter(
structure.OBMol
): # Loop over all RNA residue to calculate hydrogen bondings, halogen bondings & cation-anion interactions
RNA_residues.append(
str(residue.GetNum()) + ':' + str(residue.GetChain()))
RNA_nucleotides.append(str(residue.GetName()))
acceptors_RNA, donors_RNA = find_RNA_HB_HAL_acc_don(residue)
anions_RNA = find_RNA_anions(residue)
residue_atoms_coords = []
for atom in openbabel.OBResidueAtomIter(residue):
if atom.GetAtomicNum() != 1:
residue_atoms_coords.append(
np.array([atom.GetX(),
atom.GetY(),
atom.GetZ()]))
for ligand_name_HB, ligand_values_HB in ligands_hba_hbd.items():
if measure_distance(
centroid(residue_atoms_coords),
centroid(ligands_all_atoms[ligand_name_HB])
) > config.RES_LIGAND_MIN_DIST: # RNA residue centroid and ligand centroid are futher than declared threshold, no chance for any contact
continue
if consider_dha:
result = calculate_HB(residue, acceptors_RNA, donors_RNA,