def addsolvent(bgf_file,
solvent_bgf,
size,
margin,
out_file,
ff_file,
silent=True):
### initialize
water = False
### load the solute bgf file
if not silent: print("Initializing..")
myBGF = bgf.BgfFile(bgf_file) # str
#solventBGF = bgf.BgfFile("/home/noische/scripts/dat/WAT/f3c_box.bgf") # F3C waterbox
if not silent: print("Loading the solvent file " + solvent_bgf + " ..")
solventBGF = bgf.BgfFile(solvent_bgf)
### Generate error when the solvent box is not periodic:
if not solventBGF.PERIOD:
nu.die(
"addSolvent: The solvent file is not periodic. Use a box full of solvent."
)
### Check the type of solvent
if not silent:
print("(the solvent box seems to be full of " +
os.path.basename(solvent_file)[:-4] + " )")
if "f3c" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
if "spc" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
if "tip" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
### calculate the box size
if not silent: print("Analyzing box information..")
if len(myBGF.CRYSTX) > 2:
strsize = [myBGF.CRYSTX[0], myBGF.CRYSTX[1], myBGF.CRYSTX[2]]
else:
box = bgf.getBGFSize(myBGF, 0) # [xlo, xhi, ylo, yhi, zlo, zhi]
strsize = [box[1] - box[0], box[3] - box[2], box[5] - box[4]]
###
if size == "" and margin != "":
strsize = [
strsize[0] + 2 * margin[0], strsize[1] + 2 * margin[1],
strsize[2] + 2 * margin[2]
] # add margin on str size
elif size != "" and margin == "":
strsize = size
waterboxsize = solventBGF.CRYSTX[:3] # REMARK: This is a kind of constant.
copyNumber = [0, 0, 0]
for index, i in enumerate(copyNumber):
copyNumber[index] = math.ceil(
strsize[index] /
waterboxsize[index]) # how many times to replicate the water box
if not silent: print("Creating box information: " + str(strsize))
### replicate the solvent box
bigboxBGF = bgftools.replicateCell(solventBGF, copyNumber, True)
bigboxBGF.saveBGF("_replicate.bgf")
if not silent:
print("- Number of atoms in the created box: " + str(len(bigboxBGF.a)))
### trim the water box
if water:
if not silent:
print("Generating water box.. Calculating water molecules")
delatom = []
delwater = []
delwaterindex = []
for atom in bigboxBGF.a:
if atom.x > strsize[0] or atom.y > strsize[1] or atom.z > strsize[
2]:
delatom.append(atom.aNo)
for aNo in delatom:
water_molecule = bgftools.is_water(bigboxBGF, aNo)
if not water_molecule in delwater:
delwater.append(water_molecule)
delwater = nu.flatten(delwater)
delwater = nu.removeRepeat(delwater)
delwater.sort()
delwater.reverse()
for aNo in delwater:
delwaterindex.append(bigboxBGF.getAtomIndex(aNo))
del (delwater)
if not silent: print("Generating water box.. Trimming")
bigboxBGF.delAtoms(delwaterindex, False)
bigboxBGF.renumber()
elif not water:
if not silent:
print("Generating solvent box.. Extracting solvent molecules")
delatom = []
delsolvent = []
delsolventindex = []
for atom in bigboxBGF.a:
if atom.x > strsize[0] or atom.y > strsize[1] or atom.z > strsize[
2]:
delatom.append(atom.aNo)
for aNo in delatom:
molecule_list = []
molecule = bgftools.getmolecule(bigboxBGF, bigboxBGF.getAtom(aNo),
molecule_list)
for number in molecule_list:
if not number in delsolvent: delsolvent.append(number)
delsolvent = nu.flatten(delsolvent)
delsolvent = nu.removeRepeat(delsolvent)
delsolvent.sort()
delsolvent.reverse()
for aNo in delsolvent:
delsolventindex.append(bigboxBGF.getAtomIndex(aNo))
if not silent: print("Generating solvent box.. Trimming")
bigboxBGF.delAtoms(delsolventindex, False)
bigboxBGF.renumber()
### merge two structure
# REMARK: it is natural to have the periodic information of water box for the output BGF file.
# REMARK: HETATOM should be located on the first of the BGF file. So use dummy for merging.
if not silent: print("\nAdding trimmed solvent box to the structure..")
bigboxcenter = [strsize[0] / 2, strsize[1] / 2, strsize[2] / 2]
a, b, c = bgftools.getCom(myBGF, ff_file)
bgf.moveBGF(myBGF, bigboxcenter[0] - a, bigboxcenter[1] - b,
bigboxcenter[2] - c)
## remove bad contacts between solutes and solvents
if not silent: print("Atom distance calculation for contacts..")
delatom = []
delsolvent = []
delsolventindex = []
for atom1 in myBGF.a:
for atom2 in bigboxBGF.a:
# if the distance between atom1 and atom2 is less than 2.8, add to a delete list
dist_sq = (atom1.x - atom2.x)**2 + (atom1.y - atom2.y)**2 + (
atom1.z - atom2.z)**2
if dist_sq < 7.84:
delatom.append(atom2.aNo)
# delete bad atoms!
if not silent: print("Removing bad contacts..")
for aNo in delatom:
molecule_list = []
molecule = bgftools.getmolecule(bigboxBGF, bigboxBGF.getAtom(aNo),
molecule_list)
for number in molecule_list:
if not number in delsolvent: delsolvent.append(number)
delsolvent = nu.flatten(delsolvent)
delsolvent = nu.removeRepeat(delsolvent)
delsolvent.sort()
delsolvent.reverse()
for aNo in delsolvent:
delsolventindex.append(bigboxBGF.getAtomIndex(aNo))
bigboxBGF.delAtoms(delsolventindex, False)
bigboxBGF.renumber()
## compute stats for adding solvents
if not silent: print("\nComputing stats..")
mol_list = bgftools.getMoleculeList(bigboxBGF)
n_mol = len(mol_list)
n_atom = len(nu.flatten(mol_list))
if not silent:
print(
str(n_mol) + " molecules (" + str(n_atom) +
" atoms) will be added.")
## merge
bigboxBGF = myBGF.merge(bigboxBGF, True)
if not silent: print("Total atoms in the file: " + str(len(bigboxBGF.a)))
## some paperworking for periodic box
bigboxBGF.OTHER = solventBGF.OTHER
bigboxBGF.PERIOD = solventBGF.PERIOD
bigboxBGF.AXES = solventBGF.AXES
bigboxBGF.SGNAME = solventBGF.SGNAME
bigboxBGF.CRYSTX = solventBGF.CRYSTX
bigboxBGF.CELLS = solventBGF.CELLS
## adjust the size of the box
bigboxBGF.CRYSTX = [
strsize[0], strsize[1], strsize[2], solventBGF.CRYSTX[3],
solventBGF.CRYSTX[4], solventBGF.CRYSTX[5]
]
"""
### remove bad contacts and save
if water:
if not silent: print("Removing bad contacts: distance criteria is 2.8 A")
bigboxBGF = removebadcontacts(bigboxBGF, bigboxBGF, 2.8)
### renumber residue numbers
# RULE: all rNos for water molecules will be renumbered from 2. (for createLammpsInput.pl)
if not silent: print("Renumbering water molecules..")
max_rNo_in_hetatm = 0;
oxygen_list = bgftools.listOxygenAtoms(bigboxBGF)
### find the biggest rNo among HETATM
for atom in bigboxBGF.a:
if atom.aTag == 1:
if max_rNo_in_hetatm < atom.rNo: max_rNo_in_hetatm = atom.rNo
### update rNo in water molecules
rNo_for_water = max_rNo_in_hetatm + 500;
for aNo in oxygen_list:
water_aNo = bgftools.is_water(bigboxBGF, aNo) # get aNo in a water molecule by checking the oxygen atom
if water_aNo != []:
for atom_aNo in water_aNo:
bigboxBGF.getAtom(atom_aNo).rNo = rNo_for_water
rNo_for_water += 1
"""
### record BGF remarks
bigboxBGF.REMARK.insert(
0, "Solvents added by " + os.path.basename(sys.argv[0]) + " by " +
os.environ["USER"] + " on " + time.asctime(time.gmtime()))
bigboxBGF.REMARK.insert(0, "Solvents: " + str(solvent_file))
### save BGF
if not silent: print("Saving the file.. see " + str(out_file))
bigboxBGF.saveBGF(out_file)
return 1
def anneal(bgf_dir, ff_file, suffix, out_file):
# *** variables ***
# used for log file output
output = ""
# initialization
#t1 = 0; t2 = 0; # clock for measuring time for a cycle
curr_dir = os.path.abspath(bgf_dir)
temp_dir = curr_dir + "/anneal/"
# LAMMPS executive command: defined in ~/scripts/clusterSettings.py
#lammps_parallel = mpi_command + " -np 8 " + lammps_command
lammps_parallel = "/home/tpascal/codes/openmpi/1.4.3/bin/mpirun -np 8 " + lammps_command
# Annealing procedure
anneal_proc = "/home/noische/script/dat/in.lammps.anneal"
# get BGF file list in the directory
pei_files = glob.glob(bgf_dir + "/*.bgf")
pei_files.sort()
os.chdir(bgf_dir)
#print("pei_files: ", pei_files)
# for file in the directory
for file in pei_files:
### working filename
print("*** Working on %s" % file)
filename = file.split(".bgf")[0]
print("filename: " + filename)
# centerBGF
print("Centering %s" % file)
cmd_centerbgf = "/home/tpascal/scripts/centerBGF.pl -b " + file + " -f " + ff_file + " -s " + filename + ".center.bgf" + " -c com_center"
nu.shutup()
os.system(cmd_centerbgf)
nu.say()
filename = filename.split("/")[-1]
print("filename: " + filename)
# createLammps: with finite options
# Note that the script refers /home/noische/script/data/in.lammps.anneal for the annealing process
cmd_createLammpsInput = "/home/tpascal/scripts/createLammpsInput.pl -b " + filename + ".center.bgf" + " -f " + ff_file + " -t /home/noische/script/data/in.lammps.anneal " + " -s " + filename + " -o finite "
nu.shutup()
os.system(cmd_createLammpsInput)
nu.say()
# patch in.lammps file
print("Patching infile")
cmd_sed = "sed -i 's/2 5.0 90/2 4.5 5.0 90/' in." + filename
nu.shutup()
os.system(cmd_sed)
nu.say()
# patch data.lammps file
print("Patching datafile")
cmd_mv = "mv data." + filename + " data." + filename + ".bak"
os.system(cmd_mv)
file1 = open("data." + filename, "w")
file2 = open("data." + filename + ".bak")
while 1:
line = file2.readline()
if not line: break
if "xlo" in line:
line = "-50.0 50.0 xlo xhi\n"
file1.write(line)
elif "ylo" in line:
line = "-50.0 50.0 ylo yhi\n"
file1.write(line)
elif "zlo" in line:
line = "-50.0 50.0 zlo zhi\n"
file1.write(line)
elif "# X N_3 C_3 X" in line:
line = line.replace("# X N_3 C_3 X", "0 # X N_3 C_3 X")
file1.write(line)
elif "# X C_3 C_3 X" in line:
line = line.replace("# X C_3 C_3 X", "0 # X C_3 C_3 X")
file1.write(line)
else:
file1.write(line)
# run
print("Running simulation")
lammps_thermo_log_file = filename + ".anneal.log"
cmd_lammps = lammps_parallel + " -log " + lammps_thermo_log_file
nu.shutup()
os.system(cmd_lammps)
nu.say()
# read potential energy from file
# in in.lammps.anneal: "poteng" is the averaged value of potential energy after cooldown
print("Treating Potential Energy")
poteng = []
poteng_files = glob.glob(filename + "*poteng")
for poteng_file in poteng_files:
f_poteng_file = open(poteng_file)
while 1:
line = f_poteng_file.readline()
if not line:
break
if not line[0] == "#":
parse = line.split()
poteng.append([parse[0], float(parse[1])])
poteng = nu.removeRepeat(poteng)
### print potential energy
for i in poteng:
print(i)
# when is the minimum energy during annealing?
min_poteng = 0
min_poteng_step = 0
# usually poteng < 0
for i in poteng:
if i[1] < min_poteng:
min_poteng_step = i[0]
min_poteng = i[1]
# convert that timestep into BGF
trj_file = filename + ".lammpstrj"
anneal_output_bgf_file = filename + ".annealed.bgf"
result = LAMMPS_trj2bgf.getLAMMPSTrajectory(pei_file, trj_file,
anneal_output_bgf_file,
min_poteng_step, False)
print("The lowest potential energy structure found")
# create LAMMPS input for annealed BGF file
# run the NVT simulation at 300 K
# average the potential energy
# average Rg
# write data: filename, the minimum structure, avr_poteng, avr_Rg
"""
def periodicMoleculeSort(bgf_file,
boxinfo,
fragments=[],
selection='',
out_file='',
ff_file="",
silent=False,
recursion_limit=10000):
"""
def periodicMoleculeSort(bgf_file, out_file, boxinfo, silent=True, recursion_limit=10000):
read a BGF format from bgf_file and write the periodic information to out_file
fragments = [ [atom no. of molecule 1], [atom no. of molecule 2], ... ]
"""
# open bgf
if isinstance(bgf_file, bgf.BgfFile):
if not silent:
print("bgftools.periodicMoleculeSort: Got a tossed BGF format ..")
myBGF = bgf_file
else:
if not silent:
print("bgftools.periodicMoleculeSort: Reading " + bgf_file + " ..")
myBGF = bgf.BgfFile(bgf_file)
# if there is a requesting box size, assign
if boxinfo:
myBGF.CRYSTX = boxinfo
# assign requsted boxinfo to CRYSTX
# get periodic information of bgf file
l_myBGFbox = []
# box size
if myBGF.CRYSTX == []:
if not silent:
nu.die(
"bgftools.periodicMoleculeSort: ERROR: This BGF file is not periodic. Exiting."
)
else:
print(
"bgftools.periodicMoleculeSort: ERROR: This BGF file is not periodic. Exiting."
)
sys.exit(0)
else:
l_myBGFbox = copy.deepcopy(myBGF.CRYSTX)
# get molecules information
if fragments:
l_all_molecules = fragments
else:
l_all_molecules = getMoleculeList(myBGF, recursion_limit)
if selection:
selected_atoms = [atom.aNo for atom in myBGF.a if eval(selection)]
new_molecules_list = [
mol for mol in l_all_molecules for ano in selected_atoms
if ano in mol
]
new_molecules_list = nu.removeRepeat(new_molecules_list)
else:
new_molecules_list = l_all_molecules
# for every molecules..
for molecule in new_molecules_list:
# calculate COM
cx, cy, cz = getCom(myBGF,
ff_file=ff_file,
aNo_list=molecule,
silent=True)
# if COM is out of the box then take the residue
qx, _ = divmod(cx, l_myBGFbox[0])
qy, _ = divmod(cy, l_myBGFbox[1])
qz, _ = divmod(cz, l_myBGFbox[2])
for aNo in molecule:
atom = myBGF.getAtom(aNo)
atom.x = atom.x - l_myBGFbox[0] * qx
atom.y = atom.y - l_myBGFbox[1] * qy
atom.z = atom.z - l_myBGFbox[2] * qz
# save
if out_file:
if not silent: print("saving information to " + out_file + " ..")
myBGF.saveBGF(out_file)
return myBGF
def addsolvent(bgf_file,
solvent_bgf,
min,
max,
n_solvent,
out_file,
ff_file,
margin,
mark,
silent=True):
### initialize
water = False
default_margin = 1.0
x_margin = 0.0
y_margin = 0.0
z_margin = 0.0
if "x" in margin.lower():
x_margin = default_margin
if "y" in margin.lower():
y_margin = default_margin
if "z" in margin.lower():
z_margin = default_margin
### load the solute bgf file
if not silent: print("Initializing..")
myBGF = bgf.BgfFile(bgf_file) # str
myBGF.renumber()
if not silent: print("Loading the solvent file " + solvent_bgf + " ..")
solventBGF = bgf.BgfFile(solvent_bgf)
### Generate error when the solvent box is not periodic:
if not solventBGF.PERIOD:
nu.die(
"addSolvent: The solvent file is not periodic. Use a box full of solvent."
)
### Check the type of solvent
if not silent:
print("(the solvent box seems to be full of " +
os.path.basename(solvent_file)[:-4] + " )")
if "f3c" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
if "spc" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
if "tip" in solvent_bgf:
water = True # this flag is used to remove the bad contacts with the molecule.
### calculate the box size
if not silent: print("Analyzing box information..")
if not min:
min = [0.0, 0.0, 0.0]
if not max:
max = myBGF.CRYSTX[:3]
strsize = [
max[0] - min[0] - x_margin, max[1] - min[1] - y_margin,
max[2] - min[2] - z_margin
]
waterboxsize = solventBGF.CRYSTX[:3] # REMARK: This is a kind of constant.
copyNumber = [0, 0, 0]
for index, i in enumerate(copyNumber):
copyNumber[index] = math.ceil(
strsize[index] /
waterboxsize[index]) # how many times to replicate the water box
if not silent: print("Creating box information: " + str(strsize))
### replicate the solvent box
if not silent: print("Creating box.. this may take some time.")
bigboxBGF = bgftools.replicateCell(solventBGF, copyNumber, True)
bigboxBGF.saveBGF("_replicate.bgf")
if not silent:
print("- Number of atoms in the created box: " + str(len(bigboxBGF.a)))
delatom = []
delsolvent = []
delsolventindex = []
delwater = []
delwaterindex = []
bigboxBGF = bgf.BgfFile("_replicate.bgf")
bigboxBGF.renumber()
### trim the water box
if water:
if not silent:
print("Generating water box.. Calculating water molecules")
for atom in bigboxBGF.a:
if atom.x > strsize[0] or atom.y > strsize[1] or atom.z > strsize[
2]:
delatom.append(atom.aNo)
for aNo in delatom:
water_molecule = bgftools.is_water(bigboxBGF, aNo)
if not water_molecule in delwater:
delwater.append(water_molecule)
delwater = nu.flatten(delwater)
delwater = nu.removeRepeat(delwater)
for aNo in delwater:
delwaterindex.append(bigboxBGF.a2i[aNo])
if not silent: print("Generating water box.. Trimming")
bigboxBGF.delAtoms(delwaterindex, False)
bigboxBGF.renumber()
elif not water:
if not silent:
print("Generating solvent box.. Extracting solvent molecules")
for atom in tqdm.tqdm(bigboxBGF.a, desc='Iterating', ncols=120):
if atom.x > strsize[0] or atom.y > strsize[1] or atom.z > strsize[
2]:
delatom.append(atom.aNo)
molecule = bgftools.getMoleculeList(bigboxBGF)
for aNo in tqdm.tqdm(delatom,
desc='Appending atoms to remove',
ncols=120):
for i in molecule:
if aNo in i:
delsolvent += i
break
delsolvent = list(set(delsolvent))
for aNo in delsolvent:
delsolventindex.append(bigboxBGF.a2i[aNo])
if not silent: print("Generating solvent box.. Trimming")
delsolventindex.sort()
#delsolventindex.reverse()
bigboxBGF.delAtoms(delsolventindex, False)
bigboxBGF.renumber()
bigboxBGF.saveBGF("_temp.bgf") ## debug
### remain n molecules and delete residues
bigboxBGF = bgftools.renumberMolecules(bigboxBGF, 0, False) # renumber rNo
if n_solvent:
residues = set()
for atom in bigboxBGF.a:
residues.add(atom.rNo) # scan molecule rNo
residues = list(residues)
if not silent: print("Found %d water molecules." % len(residues))
if len(residues) < n_solvent:
nu.die("Too few solvent molecules to choose %d molecules." %
n_solvent)
rNos = random.sample(residues, n_solvent) # select n molecules
if not silent: print("%d water molecules are chosen." % n_solvent)
# delete molecules
delist = []
for atom in bigboxBGF.a:
if not atom.rNo in rNos:
delist.append(bigboxBGF.a2i[atom.aNo]) # if not chosen, delete
delist.sort()
bigboxBGF.delAtoms(delist, False)
bigboxBGF.renumber()
bigboxBGF = bgftools.renumberMolecules(bigboxBGF, 0, False)
### move the water box to min position
for atom in bigboxBGF.a:
atom.x += min[0] + x_margin / 2
atom.y += min[1] + y_margin / 2
atom.z += min[2] + z_margin / 2
### add mark to the solvents
if mark:
for atom in bigboxBGF.a:
atom.chain = mark
bigboxBGF.saveBGF('_temp.bgf')
# REMARK: it is natural to have the periodic information of water box for the output BGF file.
# REMARK: HETATOM should be located on the first of the BGF file. So use dummy for merging.
## compute stats for adding solvents
if not silent: print("\nComputing stats..")
mol_list = bgftools.getMoleculeList(bigboxBGF)
n_mol = len(mol_list)
n_atom = len(nu.flatten(mol_list))
if not silent:
print(
str(n_mol) + " molecules (" + str(n_atom) +
" atoms) will be added.")
## merge
#bigboxBGF = myBGF.merge(bigboxBGF, True)
myBGF = bgf.BgfFile(bgf_file)
bigboxBGF = bgf.BgfFile("_temp.bgf")
bigboxBGF2 = myBGF.merge(bigboxBGF)
if not silent: print("Total atoms in the file: " + str(len(bigboxBGF.a)))
## some paperworking for periodic box
bigboxBGF2.OTHER = myBGF.OTHER
bigboxBGF2.PERIOD = myBGF.PERIOD
bigboxBGF2.AXES = myBGF.AXES
bigboxBGF2.SGNAME = myBGF.SGNAME
bigboxBGF2.CRYSTX = myBGF.CRYSTX
bigboxBGF2.CELLS = myBGF.CELLS
### record BGF remarks
bigboxBGF2.REMARK.insert(
0, "Solvents added by " + os.path.basename(sys.argv[0]) + " by " +
os.environ["USER"] + " on " + time.asctime(time.gmtime()))
bigboxBGF2.REMARK.insert(0, "Solvents: " + str(solvent_file))
### save BGF
if not silent: print("Saving the file.. see " + str(out_file))
bigboxBGF2.saveBGF(out_file)
return 1