def get_RW(fn, inpro):
''' Select the HOH in [2.0, 3.5] of protein '''
outfile = open("RW_info.txt", "w")
pro = get_pdbinfo.pdbinfo(fn, file=inpro)
pro_atoms = pro.getPolarAtoms()
protein, waters = get_pdbinfo.pdbinfo(fn,
lines=pro_atoms).getProteinWaters()
waters_coord = get_pdbinfo.pdbinfo(fn, lines=waters).getCoords()
protein_coord = get_pdbinfo.pdbinfo(fn, lines=protein).getCoords()
### calculate distance ###
waters_coord = np.expand_dims(waters_coord, 1)
protein_coord = np.expand_dims(protein_coord, 0)
if waters_coord.shape[0] == 0:
print("No Receptor Water")
outfile.close()
else:
distance = np.linalg.norm(waters_coord - protein_coord, axis=2)
distance_min = np.min(distance, axis=1)
rw_index = []
for idx, i in enumerate(distance_min):
if i > 2.0 and i < 3.5:
rw_index.append(idx)
for i in rw_index:
rw_line = waters[i]
rw_distance = distance[i]
rw_chain = get_pdbinfo.chid(rw_line)
if rw_chain != " ":
rw_name = str(int(get_pdbinfo.resi(
rw_line))) + "." + get_pdbinfo.chid(rw_line)
else:
rw_name = str(int(get_pdbinfo.resi(rw_line)))
for idx, d in enumerate(rw_distance):
if d < 3.5 and d > 2.0:
pro_line = protein[idx]
pro_chain = get_pdbinfo.chid(pro_line)
pro_name = get_pdbinfo.resn(pro_line)
if pro_chain != " ":
pro_idx = str(int(
get_pdbinfo.resi(pro_line))) + "." + pro_chain
else:
pro_idx = str(int(get_pdbinfo.resi(pro_line)))
pro_aname = get_pdbinfo.atmn(pro_line).strip()
outline = fn + "," + pro_name + "," + pro_idx + "," + pro_aname + "," + rw_name + "," + str(
round(d, 2)) + "\n"
outfile.write(outline)
outfile.close()
def get_Ions(fn, lig, pro, infile):
outfile = open(infile, "w")
lig = get_pdbinfo.pdbinfo(fn, file=lig)
lig_atoms = lig.getPolarAtoms()
pro = get_pdbinfo.pdbinfo(fn, file=pro)
pro_ions = pro.getIons()
ion_coord = get_pdbinfo.pdbinfo(fn, lines=pro_ions).getCoords()
lig_coord = get_pdbinfo.pdbinfo(fn, lines=lig_atoms).getCoords()
ion_coord = np.expand_dims(ion_coord, 1)
lig_coord = np.expand_dims(lig_coord, 0)
if ion_coord.shape[0] == 0:
print("No Ion")
outfile.close()
else:
distance = np.linalg.norm(ion_coord - lig_coord, axis=2)
distance_min = np.min(distance, axis=1)
ion_index = []
for idx, i in enumerate(distance_min):
if i < 3.5:
ion_index.append(idx)
for i in ion_index:
ion_line = pro_ions[i]
ion_distance = distance[i]
ion_chain = get_pdbinfo.chid(ion_line)
ion_name = get_pdbinfo.atmn(ion_line).strip()
ion_chain = get_pdbinfo.chid(ion_line)
if ion_chain != " ":
ion_idx = str(int(
get_pdbinfo.resi(ion_line))) + "." + ion_chain
else:
ion_idx = str(int(get_pdbinfo.resi(ion_line)))
for idx, d in enumerate(ion_distance):
if d < 3.5:
lig_line = lig_atoms[idx]
lig_name = get_pdbinfo.atmn(lig_line).strip()
outline = fn + "," + ion_idx + "," + ion_name + "," + lig_name + "," + str(
round(d, 2)) + "\n"
outfile.write(outline)
outfile.close()
def addH(fn):
''' Add H to water molecule file (Vina need) '''
for filename in os.listdir("."):
if filename.startswith("RW") and filename.endswith(".pdb"):
atoms = get_pdbinfo.pdbinfo(file=filename).getAtoms()
if len(atoms) == 3:
continue
else:
newfilename = filename.split(".")[0] + "_addh.pdb"
os.system("obabel " + filename + " -O " + newfilename + " -h")
lines = [
line for line in open(newfilename)
if line[0:6] in ["ATOM ", "HETATM"]
]
if "ATOM " in lines[0]:
header = "ATOM "
else:
header = "HETATM"
for i in range(1, 3):
lines[i] = header + lines[i][6:]
out = open(filename.split(".")[0] + "_addh_correct.pdb", "w")
out.write("".join(lines))
out.close()
os.system("mv " + filename.split(".")[0] +
"_addh_correct.pdb" + " " + filename)
os.system("rm " + newfilename)
def renumber(fmt, infile, outfile):
"""
Rename atoms in file based on order
"""
if fmt == "mol2":
lines = [line for line in open(infile)]
atom_index = lines.index("@<TRIPOS>ATOM\n")
bond_index = lines.index("@<TRIPOS>BOND\n")
atom_lines = lines[atom_index + 1:bond_index]
atoms = set([atom.split()[5].split(".")[0] for atom in atom_lines])
atom_dic = {key: 1 for key in atoms}
atom_lines_new = []
for line in atom_lines:
atom_key = line.split()[5].split(".")[0]
atom_old = line.split()[1]
atom_new = atom_key.upper() + str(atom_dic[atom_key])
if len(atom_old) > len(atom_new):
atom_new = atom_new + (len(atom_old) - len(atom_new)) * " "
elif len(atom_old) < len(atom_new):
atom_old = atom_old + (len(atom_new) - len(atom_old)) * " "
newline = line.replace(atom_old, atom_new, 1)
atom_lines_new.append(newline)
atom_dic[atom_key] += 1
outfile = open(outfile, "w")
outfile.write("".join(lines[0:atom_index + 1]))
outfile.write("".join(atom_lines_new))
outfile.write("".join(lines[bond_index:]))
outfile.close()
elif fmt == "pdb":
lines = [line for line in open(infile)]
atom_lines = get_pdbinfo.pdbinfo(file=infile).getAtoms()
atom_index = lines.index(atom_lines[0])
bond_index = lines.index(atom_lines[-1]) + 1
mol = Chem.MolFromPDBFile(infile, removeHs=False)
atom_list = [atom.GetSymbol() for atom in mol.GetAtoms()]
atoms = set([atom.GetSymbol() for atom in mol.GetAtoms()])
atom_dic = {key: 1 for key in atoms}
atom_lines_new = []
for idx, line in enumerate(atom_lines):
atom_key = atom_list[idx]
atom_old = get_pdbinfo.atmn(line).strip()
atom_new = atom_key.upper() + str(atom_dic[atom_key])
if len(atom_old) > len(atom_new):
atom_new = atom_new + (len(atom_old) - len(atom_new)) * " "
elif len(atom_old) < len(atom_new):
atom_old = atom_old + (len(atom_new) - len(atom_old)) * " "
newline = line.replace(atom_old, atom_new, 1)
atom_lines_new.append(newline)
atom_dic[atom_key] += 1
outfile = open(outfile, "w")
outfile.write("".join(lines[0:atom_index]))
outfile.write("".join(atom_lines_new))
outfile.write("".join(lines[bond_index:]))
outfile.close()
def get_box(fn, inlig):
if inlig.split(".")[-1] == "mol2":
inputfile = open("../" + inlig)
x = []
y = []
z = []
flag = False
for line in inputfile:
if line[0:13] == "@<TRIPOS>ATOM":
flag = True
continue
elif line[0:13] == "@<TRIPOS>BOND":
flag = False
if flag:
x.append(float(line[16:26].split()[0]))
y.append(float(line[27:37].split()[0]))
if line[37] == "0":
z.append(float(line[38:48].split()[0]))
else:
z.append(float(line[37:48].split()[0]))
elif inlig.split(".")[-1] == "pdb":
lines = get_pdbinfo.pdbinfo(name = fn, file = "../" + inlig).getAtoms()
x,y,z = [],[],[]
for line in lines:
coords = get_pdbinfo.pdbinfo(name = fn, lines = [line]).getCoords()
x.append(float(coords[0][0]))
y.append(float(coords[0][1]))
z.append(float(coords[0][2]))
x_center =(max(x)+min(x))/2
y_center =(max(y)+min(y))/2
z_center = (max(z)+min(z))/2
size_x = max(x) - min(x) + 10
size_y = max(y) - min(y) + 10
size_z = max(z) - min(z) + 10
new_file = open("box.txt","w")
new_file.write("center_x = " + str(x_center)+ "\n")
new_file.write("center_y = " + str(y_center)+ "\n")
new_file.write("center_z = " + str(z_center) + "\n")
new_file.write("size_x = " + str(size_x)+ "\n")
new_file.write("size_y = " + str(size_y)+ "\n")
new_file.write("size_z = " + str(size_z) + "\n")
new_file.close()
def get_input(datadir, fn):
"""
Get input files based on pdbid
return: inlig_rdkit --> inlig mol2 or sdf for RDkit using
inlig3 --> inlig pdb file
inpro1 --> inpro with only protein
inpro2 --> inpro with both proteina and water
"""
olddir = os.getcwd()
os.chdir(datadir)
### get ligand ###
### input should be provided as either of sdf file (best choice) or mol2 file ###
inlig1 = fn + "_ligand.mol2"
inlig2 = fn + "_ligand.sdf"
inlig3 = fn + "_ligand.pdb"
### check ligand input file ###
inlig_rdkit = None
if inlig1 in os.listdir("."):
inlig = inlig1
try:
mol = Chem.MolFromMol2File(inlig, removeHs=False)
if mol == None:
if inlig2 in os.listdir("."):
inlig = inlig2
mol = Chem.SDMolSupplier(inlig, removeHs=False)[0]
if mol != None:
inlig_rdkit = inlig2
else:
inlig_rdkit = inlig1
except:
pass
else:
inlig = inlig2
try:
mol = Chem.SDMolSupplier(inlig, removeHs=False)[0]
if mol != None:
inlig_rdkit = inlig2
except:
pass
### correcting atom name for sasa calculation ###
### if inlig sdf or mol2 file might have problems, you should also provide a pdb file that can be used to conduct other calculation ###
if inlig3 not in os.listdir(".") and inlig_rdkit == None:
### if sdf and mol2 files can't be processed by RDKit, we need to generate pdb file and continue other calculations ###
### this is not good, since if that molecule has large problem, the conversion process might be wrong; but this can work when the problem is caused by RDKit ###
infmt = inlig.split(".")[-1]
outlig = inlig3
cmd = "obabel -i" + infmt + " " + outlig + " -opdb -O " + outlig
os.system(cmd)
if inlig3 not in os.listdir("."):
inlig = inlig_rdkit
infmt = inlig.split(".")[-1]
if infmt == "mol2":
outlig_num = inlig.split(".")[0] + "_rename.mol2"
renumber(infmt, inlig, outlig_num)
outlig = inlig3.split(".")[0] + "_rename.pdb"
cmd = "obabel -i" + infmt + " " + outlig_num + " -opdb -O " + outlig
os.system(cmd)
elif infmt == "sdf":
outlig = inlig3
cmd = "obabel -i" + infmt + " " + inlig + " -opdb -O " + outlig
os.system(cmd)
outlig_num = outlig.split(".")[0] + "_rename.pdb"
renumber('pdb', outlig, outlig_num)
else:
infmt = inlig3.split(".")[-1]
outlig_num = inlig.split(".")[0] + "_rename.pdb"
renumber(infmt, inlig3, outlig_num)
inlig3 = fn + "_ligand_rename.pdb"
### get protein ###
### at least one protein structure should be provided with all waters ###
inpro1 = fn + "_protein.pdb"
inpro2 = fn + "_protein_all.pdb"
if inpro1 not in os.listdir("."):
inpro = inpro2
outpro = open(inpro1, "w")
protein_lines = get_pdbinfo.pdbinfo(fn,
file=inpro).getProteinWaters()[0]
outpro.write("".join(protein_lines))
outpro.close()
### check input structures ###
if inlig_rdkit != None and os.path.isfile(
inlig_rdkit) and os.stat(inlig_rdkit).st_size != 0:
print("Ligand for conformation stability:" + inlig_rdkit)
else:
print(
"Warning:input ligand should be checked, skip ligand stability calculation, use default(dE:-300, RMSD:300)"
)
if os.path.isfile(inlig3) and os.stat(inlig3).st_size != 0:
print("Ligand for Vina, SASA, BA, ION:" + inlig3)
else:
sys.exit("Error: ligand input (pdb)")
if os.path.isfile(inpro1) and os.stat(inpro1).st_size != 0:
print("Protein without water molecules:" + inpro1)
else:
sys.exit("Error: protein input without water")
if os.path.isfile(inpro2) and os.stat(inpro2).st_size != 0:
print("Protein with water molecules:" + inpro2)
else:
sys.exit("Error: protein input with water")
os.chdir(olddir)
return inlig_rdkit, inlig3, inpro1, inpro2