def test_add_missing_atoms():
# add missing atom at tryptophan
molfile = os.path.join(test_dir, '5dhh_missingatomTRP.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=True)
mol = Chem.RemoveHs(mol, sanitize=False)
assert mol.GetNumAtoms() == 26
mol = PreparePDBMol(mol, add_missing_atoms=True)
assert mol.GetNumAtoms() == 27
atom = mol.GetAtomWithIdx(21)
assert atom.GetAtomicNum() == 6
info = atom.GetPDBResidueInfo()
assert info.GetResidueName() == 'TRP'
assert info.GetResidueNumber() == 175
assert info.GetName().strip() == 'C9'
assert atom.IsInRing()
assert atom.GetIsAromatic()
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
# add whole ring to tyrosine
molfile = os.path.join(test_dir, '3cx9_TYR.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=True)
mol = Chem.RemoveHs(mol, sanitize=False)
assert mol.GetNumAtoms() == 23
mol = PreparePDBMol(mol, add_missing_atoms=True)
assert mol.GetNumAtoms() == 29
atom = mol.GetAtomWithIdx(17)
assert atom.GetAtomicNum() == 6
info = atom.GetPDBResidueInfo()
assert info.GetResidueName() == 'TYR'
assert info.GetResidueNumber() == 138
assert info.GetName().strip() == 'C6'
assert atom.IsInRing()
assert atom.GetIsAromatic()
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
# missing protein backbone atoms
molfile = os.path.join(test_dir, '5ar7_HIS.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False)
mol = Chem.RemoveHs(mol, sanitize=False)
assert mol.GetNumAtoms() == 21
assert mol.GetNumBonds() == 19
mol = PreparePDBMol(mol, add_missing_atoms=True)
assert mol.GetNumAtoms() == 25
assert mol.GetNumBonds() == 25
# missing nucleotide backbone atoms
molfile = os.path.join(test_dir, '1bpx_missingBase.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False)
mol = Chem.RemoveHs(mol, sanitize=False)
assert mol.GetNumAtoms() == 301
assert mol.GetNumBonds() == 333
mol = PreparePDBMol(mol, add_missing_atoms=True)
assert mol.GetNumAtoms() == 328
assert mol.GetNumBonds() == 366
def test_sulphur_bridge():
"""Test sulphur bridges retention"""
molfile = os.path.join(test_dir, '2qwe_Sbridge.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
atom1 = mol.GetAtomWithIdx(5)
atom2 = mol.GetAtomWithIdx(11)
bond = mol.GetBondBetweenAtoms(atom1.GetIdx(), atom2.GetIdx())
assert atom1.GetPDBResidueInfo().GetName().strip() == 'SG'
assert atom1.GetPDBResidueInfo().GetResidueNumber() == 92
assert atom2.GetPDBResidueInfo().GetName().strip() == 'SG'
assert atom2.GetPDBResidueInfo().GetResidueNumber() == 417
assert bond is not None
def test_intraresidue_bonding():
"""Test if fixer removes wrong connections within single residue"""
molfile = os.path.join(test_dir, '1idg_connectivity.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
# check if N and C from GLU20 are not connected
atom1 = mol.GetAtomWithIdx(11)
assert atom1.GetAtomicNum() == 7
assert atom1.GetPDBResidueInfo().GetResidueName() == 'GLU'
assert atom1.GetPDBResidueInfo().GetResidueNumber() == 20
atom2 = mol.GetAtomWithIdx(13)
assert atom2.GetAtomicNum() == 6
assert atom2.GetPDBResidueInfo().GetResidueName() == 'GLU'
assert atom2.GetPDBResidueInfo().GetResidueNumber() == 20
assert mol.GetBondBetweenAtoms(11, 13) is None
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_interresidue_bonding():
"""Test if fixer removes wrong connections between residues"""
molfile = os.path.join(test_dir, '4e6d_residues.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
# check if O from PRO
atom1 = mol.GetAtomWithIdx(11)
assert atom1.GetAtomicNum() == 8
assert atom1.GetPDBResidueInfo().GetResidueName() == 'PRO'
# ...and N from GLN
atom2 = mol.GetAtomWithIdx(22)
assert atom2.GetAtomicNum() == 7
assert atom2.GetPDBResidueInfo().GetResidueName() == 'GLN'
# ...are not connected
assert mol.GetBondBetweenAtoms(11, 22) is None
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_aromatic_ring():
"""Test aromaticity for partial matches"""
# ring is complete and should be aromatic
molfile = os.path.join(test_dir, '5ar7_HIS.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
atom = mol.GetAtomWithIdx(6)
assert atom.GetAtomicNum() == 7
info = atom.GetPDBResidueInfo()
assert info.GetResidueName() == 'HIS'
assert info.GetResidueNumber() == 246
assert info.GetName().strip() == 'ND1'
assert atom.GetIsAromatic()
atom = mol.GetAtomWithIdx(9)
assert atom.GetAtomicNum() == 7
info = atom.GetPDBResidueInfo()
assert info.GetResidueName() == 'HIS'
assert info.GetResidueNumber() == 246
assert info.GetName().strip() == 'NE2'
assert atom.GetIsAromatic()
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
# there is only one atom from the ring and it shouldn't be aromatic
molfile = os.path.join(test_dir, '3cx9_TYR.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
atom = mol.GetAtomWithIdx(14)
assert atom.GetAtomicNum() == 6
info = atom.GetPDBResidueInfo()
assert info.GetResidueName() == 'TYR'
assert info.GetResidueNumber() == 138
assert info.GetName().strip() == 'CG'
assert not atom.GetIsAromatic()
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_ring():
"""Test if fixer adds missing bond in ring"""
molfile = os.path.join(test_dir, '4yzm_ring.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
# check if there is double bond between N and C from MSE
atom1 = mol.GetAtomWithIdx(12)
assert atom1.GetAtomicNum() == 6
assert atom1.GetPDBResidueInfo().GetResidueName() == 'PHE'
atom2 = mol.GetAtomWithIdx(13)
assert atom2.GetAtomicNum() == 6
assert atom2.GetPDBResidueInfo().GetResidueName() == 'PHE'
# there is a bond and it is aromatic
bond = mol.GetBondBetweenAtoms(12, 13)
assert bond is not None
assert_almost_equal(bond.GetBondTypeAsDouble(), 1.5)
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_bondtype():
"""Test if fixer deals with non-standard residue and fixes bond types"""
molfile = os.path.join(test_dir, '3rsb_bondtype.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
# check if there is double bond between N and C from MSE
atom1 = mol.GetAtomWithIdx(13)
assert atom1.GetAtomicNum() == 6
assert atom1.GetPDBResidueInfo().GetResidueName() == 'MSE'
atom2 = mol.GetAtomWithIdx(14)
assert atom2.GetAtomicNum() == 8
assert atom2.GetPDBResidueInfo().GetResidueName() == 'MSE'
# there is a bond and it is double
bond = mol.GetBondBetweenAtoms(13, 14)
assert bond is not None
assert_almost_equal(bond.GetBondTypeAsDouble(), 2.0)
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_HOH_bonding():
"""Test if fixer unbinds HOH"""
molfile = os.path.join(test_dir, '2vnf_bindedHOH.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
# don't use templates and don't remove waters
mol = PreparePDBMol(mol, removeHOHs=False)
atom = mol.GetAtomWithIdx(5)
assert atom.GetPDBResidueInfo().GetResidueName() == 'HOH'
assert atom.GetDegree() == 0 # HOH should have no bonds
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_metal_bonding():
"""Test if fixer disconnects metals"""
molfile = os.path.join(test_dir, '1ps3_zn.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
atom = mol.GetAtomWithIdx(36)
assert atom.GetAtomicNum() == 30 # is it Zn
assert atom.GetDegree() == 0 # Zn should have no bonds
assert atom.GetFormalCharge() == 2
assert atom.GetNumExplicitHs() == 0
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_multivalent_Hs():
"""Test if fixer deals with multivalent Hs"""
# TODO: require mol without Hs in the future (rdkit v. 2018)
molfile = os.path.join(test_dir, '2c92_hypervalentH.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol, residue_whitelist=[], removeHs=False)
atom = mol.GetAtomWithIdx(84)
assert atom.GetAtomicNum() == 1 # is it H
assert atom.GetDegree() == 1 # H should have 1 bond
for n in atom.GetNeighbors(): # Check if neighbor is from the same residue
assert atom.GetPDBResidueInfo().GetResidueName() == n.GetPDBResidueInfo().GetResidueName()
# mol can be sanitized
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_many_missing():
"""Test parsing residues with **many** missing atoms and bonds"""
molfile = os.path.join(test_dir, '2wb5_GLN.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol)
assert mol.GetNumAtoms() == 5
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
assert mol.GetAtomWithIdx(4).GetDegree() == 0
# test if removal works
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
mol = PreparePDBMol(mol, remove_incomplete=True)
assert mol.GetNumAtoms() == 0
assert Chem.SanitizeMol(mol) == Chem.SanitizeFlags.SANITIZE_NONE
def test_pocket_extractor():
"""Test extracting pocket and ligand"""
molfile = os.path.join(test_dir, '5ar7.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
# there should be no pocket at 1A
pocket, ligand = ExtractPocketAndLigand(mol, cutoff=1.)
assert pocket.GetNumAtoms() == 0
assert ligand.GetNumAtoms() == 26
# small pocket of 5A
pocket, ligand = ExtractPocketAndLigand(mol, cutoff=12.)
assert pocket.GetNumAtoms() == 928
assert ligand.GetNumAtoms() == 26
# check if HOH is in pocket
atom = pocket.GetAtomWithIdx(910)
assert atom.GetAtomicNum() == 8
assert atom.GetPDBResidueInfo().GetResidueName() == 'HOH'
# Prepare and sanitize pocket and ligand
pocket = PreparePDBMol(pocket)
ligand = PreparePDBMol(ligand)
assert Chem.SanitizeMol(pocket) == Chem.SanitizeFlags.SANITIZE_NONE
assert Chem.SanitizeMol(ligand) == Chem.SanitizeFlags.SANITIZE_NONE
# Check atom/bond properies for both molecules
bond = pocket.GetBondWithIdx(39)
assert bond.GetIsAromatic()
assert bond.GetBeginAtom().GetPDBResidueInfo().GetResidueName() == 'TYR'
atom = ligand.GetAtomWithIdx(22)
assert atom.GetAtomicNum() == 7
assert atom.GetIsAromatic()
assert atom.GetPDBResidueInfo().GetResidueName() == 'SR8'
# test if metal is in pocket
molfile = os.path.join(test_dir, '4p6p_lig_zn.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
assert mol.GetNumAtoms() == 176
pocket, ligand = ExtractPocketAndLigand(mol, cutoff=5.)
assert pocket.GetNumAtoms() == 162
assert ligand.GetNumAtoms() == 14
atom = pocket.GetAtomWithIdx(153)
assert atom.GetPDBResidueInfo().GetResidueName().strip() == 'ZN'
atom = pocket.GetAtomWithIdx(160)
assert atom.GetPDBResidueInfo().GetResidueName() == 'HOH'
pocket, ligand = ExtractPocketAndLigand(mol,
cutoff=5.,
expandResidues=False)
assert pocket.GetNumAtoms() == 74
assert ligand.GetNumAtoms() == 14
atom = pocket.GetAtomWithIdx(65)
assert atom.GetPDBResidueInfo().GetResidueName().strip() == 'ZN'
atom = pocket.GetAtomWithIdx(73)
assert atom.GetPDBResidueInfo().GetResidueName() == 'HOH'
# ligand and protein white/blacklist
molfile = os.path.join(test_dir, '1dy3_2LIG.pdb')
mol = Chem.MolFromPDBFile(molfile, sanitize=False, removeHs=False)
# by default the largest ligand - ATP
pocket, ligand = ExtractPocketAndLigand(mol, cutoff=20.)
assert pocket.GetNumAtoms() == 304
assert ligand.GetNumAtoms() == 31
atom = ligand.GetAtomWithIdx(0)
assert atom.GetPDBResidueInfo().GetResidueName() == 'ATP'
# blacklist APT to get other largest ligand - 87Y
pocket, ligand = ExtractPocketAndLigand(mol,
cutoff=20.,
ligand_residue_blacklist=['ATP'])
assert pocket.GetNumAtoms() == 304
assert ligand.GetNumAtoms() == 23
atom = ligand.GetAtomWithIdx(0)
assert atom.GetPDBResidueInfo().GetResidueName() == '87Y'
# point to 87Y explicitly
pocket, ligand = ExtractPocketAndLigand(mol,
cutoff=20.,
ligand_residue='87Y')
assert pocket.GetNumAtoms() == 304
assert ligand.GetNumAtoms() == 23
atom = ligand.GetAtomWithIdx(0)
assert atom.GetPDBResidueInfo().GetResidueName() == '87Y'
# include APT in pocket to get other largest ligand - 87Y
pocket, ligand = ExtractPocketAndLigand(mol,
cutoff=20.,
append_residues=['ATP'])
assert pocket.GetNumAtoms() == 304 + 31
assert ligand.GetNumAtoms() == 23
atom = ligand.GetAtomWithIdx(0)
assert atom.GetPDBResidueInfo().GetResidueName() == '87Y'
atom = pocket.GetAtomWithIdx(310)
assert atom.GetPDBResidueInfo().GetResidueName() == 'ATP'
