def test_model_3zg0():
if not PC.MODELLER:
return
from ProtoCaller.Wrappers.modellerwrapper import modellerTransform
with Dir(PC.TESTDIR + "/shared"):
with Dir("temp", temp=True):
copyfile("../3ZG0.pdb", "3ZG0.pdb")
copyfile("../3ZG0.fasta", "3ZG0.fasta")
obj = PDB.PDB("3ZG0.pdb")
obj_mod = PDB.PDB(
modellerTransform("3ZG0.pdb",
"3ZG0.fasta",
add_missing_atoms=False))
assert len(obj.missing_atoms) == len(obj_mod.missing_atoms) == 0
assert len(obj.missing_residues) == 7
assert len(obj_mod.missing_residues) == 0
assert len(obj.disulfide_bonds) == len(
obj_mod.disulfide_bonds) == 0
assert len(obj.modified_residues) == len(
obj_mod.modified_residues) == 0
assert len(obj.site_residues) == len(obj_mod.site_residues) == 120
assert len(obj_mod[0]) == 642
assert len(obj_mod[1]) == 642
assert len(obj_mod[2]) == 9
assert len(obj_mod[3]) == 9
assert len(obj_mod[4]) == 239
assert len(obj_mod[5]) == 171
def test_prepare_1bji():
with Dir(PC.TESTDIR + "/Ensemble"):
with Dir("temp", temp=True):
protein = Protein("1bji", ligand_ref="479G")
assert isinstance(protein.ligand_ref, Ligand)
assert len(protein.ligands) == 9
assert protein._pdb_obj.numberOfAtoms == 3259
protein.filter(waters=None, include_mols=["480"], ligands=None)
assert isinstance(protein.ligand_ref, Ligand)
assert len(protein.ligands) == 0
assert protein._pdb_obj.numberOfAtoms == 3069
# this time keep only site waters and parametrise with some end state checks
protein = Protein("1bji", ligand_ref="479G")
protein.filter(ligands=None, waters="site")
protein.prepare()
protein.parametrise(Params())
if PC.BIOSIMSPACE and isinstance(protein.complex_template,
BSS._SireWrappers._system.System):
assert protein.complex_template.nAtoms() == 6006
assert pytest.approx(
protein.complex_template.charge().magnitude()) == 2
else:
# TODO: add support for ParmEd topologies when the Morph class is fully functional
pass
def test_MCS_EZ():
with Dir(PC.TESTDIR + "/shared"):
# test mapping of an asymmetric double bond onto symmetric one
ref = openFileAsRdkit("EZ_ref1.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol1.mol2", removeHs=False)
results = getMCSMap(ref, mol)
assert all([len(x) == 38 for x in results])
assert all([{(11, 12), (12, 11)}.issubset(x) for x in results])
# test mapping between two esters of different E/Z conformations
ref = openFileAsRdkit("EZ_ref2.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol2.mol2", removeHs=False)
results = getMCSMap(ref, mol)
assert all([len(x) == 28 for x in results])
results_rev = getMCSMap(mol, ref)
assert all([len(x) == 28 for x in results_rev])
# test mapping of a double bond to a single bond with unfavourable
# conformation
ref = openFileAsRdkit("EZ_ref3.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol3.mol2", removeHs=False)
results = getMCSMap(ref, mol, timeout=1, two_way_matching=True)
assert all([len(x) == 38 for x in results])
results = getMCSMap(ref, mol, timeout=1, two_way_matching=False)
assert all([len(x) == 27 for x in results])
# test mapping of a double bond to a single bond with favourable
# conformation
ref = openFileAsRdkit("EZ_ref4.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol4.mol2", removeHs=False)
results = getMCSMap(ref, mol, timeout=1, two_way_matching=True)
assert all([len(x) == 41 for x in results])
results = getMCSMap(ref, mol, timeout=1, two_way_matching=False)
assert all([len(x) == 27 for x in results])
# test the recursive algorithm for > 1 mismatching bonds
ref = openFileAsRdkit("EZ_ref5.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol5.mol2", removeHs=False)
results = getMCSMap(ref, mol, timeout=1)
assert all([len(x) == 17 for x in results])
results_rev = getMCSMap(ref, mol, timeout=1)
assert all([len(x) == 17 for x in results_rev])
# test the recursive algorithm for > 1 mismatching bonds and matching
# a double bond onto a single bond
ref = openFileAsRdkit("EZ_ref6.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol6.mol2", removeHs=False)
results = getMCSMap(ref, mol, timeout=1, two_way_matching=True)
assert all([len(x) == 17 for x in results])
# test mapping of mismatching amide onto an ester
ref = openFileAsRdkit("EZ_ref7.mol2", removeHs=False)
mol = openFileAsRdkit("EZ_mol7.mol2", removeHs=False)
results = getMCSMap(ref, mol, timeout=1)
assert all([len(x) == 30 for x in results])
def test_Dir():
orig_dir = os.getcwd()
os.chdir(PC.TESTDIR + "/Utils")
target_dirname = os.path.join(PC.TESTDIR, "Utils", "temp")
f = Dir("temp", temp=True, purge_immediately=False)
with f:
assert os.getcwd() == target_dirname
with f:
assert os.getcwd() == target_dirname
assert os.getcwd() == target_dirname
assert os.path.exists(target_dirname)
assert os.getcwd() == os.path.join(PC.TESTDIR, "Utils")
target_dirname2 = os.path.join(PC.TESTDIR, "Utils", "temp2")
with Dir("temp2", temp=True, purge_immediately=True):
pass
assert not os.path.exists(target_dirname2)
os.chdir(orig_dir)
def test_model_3zg0():
with Dir(PC.TESTDIR + "/shared"):
with Dir("temp", temp=True):
copyfile("../3ZG0.pdb", "3ZG0.pdb")
obj = PDB.PDB("3ZG0.pdb")
obj_mod = PDB.PDB(charmmguiTransform("3ZG0.pdb"))
assert len(obj.missing_atoms) == len(obj_mod.missing_atoms) == 0
assert len(obj.missing_residues) == 7
assert len(obj_mod.missing_residues) == 0
assert len(obj.disulfide_bonds) == len(obj_mod.disulfide_bonds) == 0
assert len(obj.modified_residues) == len(obj_mod.modified_residues) == 0
assert len(obj.site_residues) == len(obj_mod.site_residues) == 120
assert len(obj_mod[0]) == 642
assert len(obj_mod[1]) == 642
assert len(obj_mod[2]) == 9
assert len(obj_mod[3]) == 9
assert len(obj_mod[4]) == 239
assert len(obj_mod[5]) == 171
def test_protonate_1bji():
with Dir(PC.TESTDIR + "/shared"):
with Dir("temp", temp=True):
copyfile("../1bji.pdb", "1bji.pdb")
file_pdb2pqr = pdb2pqrTransform("1bji.pdb")
obj_protonated = PDB.PDB(file_pdb2pqr)
assert len(obj_protonated.missing_atoms) == 0
assert len(obj_protonated.missing_residues) == 0
assert len(obj_protonated.disulfide_bonds) == 9
assert len(obj_protonated.site_residues) == 74
assert len(obj_protonated[0]) == 388
assert len(obj_protonated[1]) == 202
for i, residue in enumerate(obj_protonated[0]):
# don't deal with terminal amino acids
if i == 0 or i == len(obj_protonated[0]) - 1:
continue
if residue.resName in size_dict.keys():
assert len(residue) == size_dict[residue.resName]
def test_align():
with Dir(PC.TESTDIR + "/shared"):
with Dir("temp", temp=True):
lig_ref = Ligand("../toluene.sdf", protonated=True, minimise=False)
lig1 = Ligand("c1ccccc1CC", minimise=False)
lig2 = Ligand("c1ccccc1CCC", minimise=False)
morph = Perturbation(lig1, lig2)
mol, mcs = morph.alignAndCreateMorph(lig_ref)
assert len(mcs) == 18
positions = {
(-32.355, 7.263, 2.207),
(-33.181, 6.606, 3.301),
(-33.238, 7.160, 4.565),
(-33.992, 6.571, 5.570),
(-34.711, 5.417, 5.332),
(-34.638, 4.878, 4.060),
(-33.894, 5.447, 3.048),
(-31.882, 8.141, 2.595),
(-32.993, 7.533, 1.392),
(-31.608, 6.578, 1.864),
(-32.712, 8.025, 4.765),
(-34.016, 7.001, 6.508),
(-35.279, 4.975, 6.072),
(-35.163, 4.013, 3.860),
(-33.869, 5.013, 2.112),
}
vec1 = mol._sire_object.property("coordinates0").toVector()
positions_A = {(x[0], x[1], x[2]) for x in vec1}
vec2 = mol._sire_object.property("coordinates1").toVector()
positions_B = {(x[0], x[1], x[2]) for x in vec2}
assert positions.issubset(positions_A)
assert positions.issubset(positions_B)
assert len(positions_A) == len(positions_B) == 21
assert positions_A == positions_B
def test_align():
from ProtoCaller.Wrappers.rdkitwrapper import _nonMCSDihedrals
with Dir(PC.TESTDIR + "/shared"):
# test conservation of dihedrals
ref = openFileAsRdkit("Align_ref1.mol2", removeHs=False)
mol = openFileAsRdkit("Align_mol1.mol2", removeHs=False)
mcs = getMCSMap(ref, mol)[0]
dihedrals = _nonMCSDihedrals(mol, mcs)
assert len(dihedrals) == 3
mol_new, mcs_new = alignTwoMolecules(ref, mol, minimise_score=False)
assert mcs == mcs_new
dihedrals_new = _nonMCSDihedrals(mol_new, mcs_new)
assert dihedrals.keys() == dihedrals_new.keys()
assert approx(dihedrals.values(), dihedrals.values())
def test_deepcopy_1bji():
with Dir(PC.TESTDIR + "/shared"):
obj = PDB.PDB("1bji.pdb")
obj2 = copy.deepcopy(obj)
assert obj2.numberOfAtoms == 3398
assert len(obj2.missing_atoms) == 0
assert len(obj2.missing_residues) == 0
assert len(obj2.modified_residues) == 3
assert len(obj2.disulfide_bonds) == 9
assert len(obj2.site_residues) == 74
assert len(obj2[0]) == 388
assert len(obj2[1]) == 202
obj2.purgeResidues(obj2.filter("type=='amino_acid'"), mode="keep")
assert 3067 == obj2.numberOfAtoms != obj.numberOfAtoms
def test_read_write_1bji():
with Dir(PC.TESTDIR + "/shared"):
obj = PDB.PDB("1bji.pdb")
filestr = open("1bji.pdb").readlines()
filestr = [line.strip() for line in filestr
if line[:3] in ["TER", "END"] or line[:4] == "ATOM" or line[:6] == "HETATM"]
assert obj.numberOfAtoms == 3398
assert len(obj.missing_atoms) == 0
assert len(obj.missing_residues) == 0
assert len(obj.modified_residues) == 3
assert len(obj.disulfide_bonds) == 9
assert len(obj.site_residues) == 74
assert len(obj[0]) == 388
assert len(obj[1]) == 202
new_file = tempfile.NamedTemporaryFile()
obj.writePDB(new_file.name)
filestr_new = open(new_file.name).readlines()
# filter out only the relevant sections
filestr_new = [line for line in filestr_new
if line[:3] in ["TER", "END"] or line[:4] == "ATOM" or line[:6] == "HETATM"]
# fix a known formatting difference with the original PDB
filestr_new = [line[:12] + " " + line[12:16] + line[17:] for line in filestr_new]
filestr_new = [line.strip() for line in filestr_new]
obj_new = PDB.PDB(new_file.name)
assert obj_new.numberOfAtoms == 3398
assert len(obj_new.missing_atoms) == 0
assert len(obj_new.missing_residues) == 0
assert len(obj.modified_residues) == 3
assert len(obj_new.disulfide_bonds) == 9
assert len(obj_new.site_residues) == 74
assert len(obj_new[0]) == 388
assert len(obj_new[1]) == 202
# the ultimate test: rewriting the PDB should result in the same PDB
assert filestr_new == filestr
import logging
logging.basicConfig(level=logging.INFO)
import numpy as np
import ProtoCaller.Simulation as Simulation
from ProtoCaller.Utils.fileio import Dir
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("-m", "--mode", type=str,
help="Whether to run bound or solvated leg")
parser.add_argument("-w", "--workdir", type=str, help="Path to where "
"complex_final.* and/or morph.* are located")
args = parser.parse_args()
workdir = Dir(args.workdir) if args.workdir else Dir(".")
if args.mode == "bound":
dirname = "Run_Bound"
gro = "complex_final.gro"
top = "complex_final.top"
elif args.mode == "solvated":
dirname = "Run_Solvated"
gro = "morph.gro"
top = "morph.top"
else:
raise ValueError("Please choose either 'bound' or 'solvated'")
# determine the lambda values
vdw_lambdas = [0] * 9 + [round(x, 2) for x in np.linspace(0.00, 0.95, num=26)] \
+ [0.97, 0.98, 0.99, 0.999, 1.00]
# import os
# add an alternative default version for GROMACS. Otherwise, use bash default
# os.environ["GROMACSHOME"] = os.path.expanduser("~/gromacs-2018.4")
import logging
logging.basicConfig(level=logging.INFO)
from BioSimSpace._SireWrappers import System
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ligand, Perturbation
from ProtoCaller.Solvate import solvate
from ProtoCaller.IO.GROMACS import saveAsGromacs
from ProtoCaller.Parametrise import Params
from ProtoCaller.Wrappers.biosimspacewrapper import resize
with Dir("Relative_Solvation", overwrite=True):
with Dir("Ligands"):
# create two ligands from SMILES strings and name them
toluene = Ligand("C1=CC=CC=C1C", name="toluene", workdir="Ligands")
benzene = Ligand("C1=CC=CC=C1", name="benzene", workdir="Ligands")
# protonate parametrise the ligands at pH 7 with GAFF2
for ligand in [toluene, benzene]:
ligand.protonate(babel_parameters={"pH": 7.0})
ligand.parametrise(Params(ligand_ff="gaff2"))
with Dir("toluene~benzene"):
# create the perturbation from the ligands
perturbation = Perturbation(toluene, benzene)
# we create the mixed topology object by using the toluene as a reference ligand
# import os
# add an alternative default version for GROMACS. Otherwise, use bash default
# os.environ["GROMACSHOME"] = os.path.expanduser("~/gromacs-2018.4")
import logging
logging.basicConfig(level=logging.INFO)
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ligand
from ProtoCaller.Solvate import solvate
from ProtoCaller.IO.GROMACS import saveAsGromacs
from ProtoCaller.Parametrise import Params
from ProtoCaller.Wrappers.parmedwrapper import openFilesAsParmed, resize
with Dir("Absolute_Solvation", overwrite=True):
with Dir("Ligands"):
# create two ligands from SMILES strings and name them
toluene = Ligand("C1=CC=CC=C1C", name="toluene", workdir="Ligands")
benzene = Ligand("C1=CC=CC=C1", name="benzene", workdir="Ligands")
# protonate parametrise the ligands at pH 7 with GAFF2
for ligand in [toluene, benzene]:
ligand.protonate(babel_parameters={"pH": 7.0})
ligand.parametrise(Params(ligand_ff="gaff2"))
for ligand in [toluene, benzene]:
with Dir(ligand.name):
# open the parametrised ligand as a ParmEd object
lig_vac = openFilesAsParmed(ligand.parametrised_files)
# give a dummy box length of e.g. 10 nm
lig_vac = resize(lig_vac, 10)
# save the vacuum leg as .gro and .top files
def test_pdbconnect():
with Dir(PC.TESTDIR + "/Utils/temp", temp=True):
downloader = PDBDownloader("1BJI")
downloader.getPDB()
downloader.getFASTA()
assert len(downloader.getLigands()) == 10
# import os
# add an alternative default version for GROMACS. Otherwise, use bash default
# os.environ["GROMACSHOME"] = os.path.expanduser("~/gromacs-2018.4")
import logging
logging.basicConfig(level=logging.INFO)
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ligand, Protein, Ensemble
with Dir("Sialyltransferase", overwrite=False):
# create a protein and ligands using custom files
lig_ref = Ligand("lig_ref.sdf",
protonated=True,
workdir="Ligands",
name="lig_ref")
lig1 = Ligand("lig1.sdf", protonated=True, workdir="Ligands", name="lig1")
lig2 = Ligand("lig2.sdf", protonated=True, workdir="Ligands", name="lig2")
protein = Protein("2WNB", pdb_file="protein.pdb", ligand_ref=lig_ref)
# create the morphs from the ligands
morphs = [[lig1, lig2], [lig2, lig1]]
# create a system from the protein and the morphs and set up some default
# settings regarding system preparation
system = Ensemble("GROMACS",
protein=protein,
morphs=morphs,
box_length_complex=7,
ligand_ff="gaff2",
workdir=protein.workdir.path)
# only keep the reference ligand
# import os
# add an alternative default version for GROMACS. Otherwise, use bash default
# os.environ["GROMACSHOME"] = os.path.expanduser("~/gromacs-2018.4")
import logging
logging.basicConfig(level=logging.INFO)
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ligand, Protein, Ensemble
with Dir("Sialyltransferase", overwrite=True):
# create a protein from its PDB code and the residue number of the ligand
# we are going to use for mapping
protein = Protein("2WNB", ligand_ref="1344")
# delete any atoms with altLoc B
for chain in protein.pdb_obj:
for residue in chain:
atoms_to_purge = [x for x in residue if x.altLoc == "B"]
residue.purgeAtoms(atoms_to_purge, "discard")
protein.pdb_obj.writePDB()
# create two ligands from SMILES strings
lig1 = Ligand("O([P@]([O-])(=O)CC[C@@]1(C(=O)N)C[C@@H]([C@H]([C@H]([C@@H]"
"([C@@H](CO)O)O)O1)NC(=O)C)O)C[C@H]1O[C@@H](n2c(=O)nc(N)cc2)"
"[C@H](O)[C@@H]1O", workdir="Ligands")
lig2 = Ligand("[P@]([O-])(=O)(OC[C@H]1O[C@@H](n2ccc(nc2=O)N)[C@H](O)[C@@H]"
"1O)CC[C@]1(CO)O[C@H]([C@@H]([C@H](C1)O)NC(=O)C)[C@H](O)"
"[C@H](O)CO", workdir="Ligands")
# create the morphs from the ligands
morphs = [[lig1, lig2], [lig2, lig1]]
import logging
logging.basicConfig(level=logging.INFO)
import glob
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ensemble, Ligand
from ProtoCaller.Parametrise import Params
params = Params(ligand_ff="GAFF2")
with Dir("FXA", overwrite=False):
# create a ligand dictionary and populate it from the Ligands folder.
# this block makes sure we don't reparametrise if we rerun the script.
ligands = {}
with Dir("Ligands") as ligdir:
lignames = [x.split(".")[0] for x in glob.glob("*.sdf")]
for ligname in lignames:
sdf = "{}.sdf".format(ligname)
prmtop = "{}.prmtop".format(ligname)
inpcrd = "{}.inpcrd".format(ligname)
if len(glob.glob(prmtop)) and len(glob.glob(inpcrd)):
parametrised_files = [prmtop, inpcrd]
else:
parametrised_files = None
# here we initialise the ligand with custom structures.
ligands[ligname] = Ligand(sdf,
protonated=True,
minimise=False,
# import os
# add an alternative default version for GROMACS. Otherwise, use bash default
# os.environ["GROMACSHOME"] = os.path.expanduser("~/gromacs-2018.4")
import logging
logging.basicConfig(level=logging.INFO)
from ProtoCaller.Utils.fileio import Dir
from ProtoCaller.Ensemble import Ligand, Protein, Ensemble
with Dir("T4-lysozyme", overwrite=True):
# create a protein from its PDB code and the residue number of the ligand
# we are going to use for mapping
protein = Protein("181L", ligand_ref="400")
# create two ligands from SMILES strings and name them
benzol = Ligand("C1=CC=CC=C1", name="benzol", workdir="Ligands")
o_xylene = Ligand("CC1=CC=CC=C1C", name="o-xylene", workdir="Ligands")
# create the morphs from the ligands
morphs = [[benzol, o_xylene], [o_xylene, benzol]]
# create a system from the protein and the morphs and set up some default
# settings regarding system preparation
system = Ensemble("GROMACS",
protein=protein,
morphs=morphs,
box_length_complex=7,
ligand_ff="gaff2",
workdir=protein.workdir.path)
# only keep the reference ligand and keep all crystallographic waters
system.protein.filter(ligands=None, waters="all")