def countWaterCNT(bgf_file, trj_file, silent=False):
# init
timestep = 0
l_timestep = []
myBGF = bgf.BgfFile(bgf_file)
myTRJ = open(trj_file)
myDAT = open(bgf_file[:-4] + ".count.dat", "w")
myDAT.write("Time\tHeight\tRadius\tNumber\n")
# extract aNos of CNT in the BGF file
aNo_CNT = []
aNo_MtOH_C = []
for atom in myBGF.a:
# Carbons in CNT
if "CNT" in atom.rName:
aNo_CNT.append(atom.aNo)
# Carbons in MeOH
if "MET" in atom.rName and "C" in atom.aName:
aNo_MtOH_C.append(atom.aNo)
N_CNT = len(aNo_CNT) # the number of CNT atoms
# for every shot in the trajectory file update BGF and manipulate
myDUMP = dump.dump(trj_file, 0) # sequential reading
while 1:
aNo_MtOH_C_in_CNT = copy.deepcopy(aNo_MtOH_C)
# initialize for the moment of inertia and the center of mass calculation
U = 0
Ut = 0
Uv = 0
Ixx = 0
Ixy = 0
Ixz = 0
Iyx = 0
Iyy = 0
Iyz = 0
Izx = 0
Izy = 0
Izz = 0
Mx = 0
My = 0
Mz = 0
time = myDUMP.next()
sys.stdout.write('\r' + "Reading timestep.. " + str(time))
sys.stdout.flush()
#print("Timestep: " + str(time))
if time == -1:
break
nu.shutup()
myDUMP.sort()
nu.say()
l_timestep = myDUMP.time() # timesteps are 'appended' incrementally
atoms = myDUMP.viz(len(l_timestep) - 1)[
2] # atom coordinate info: id,type,x,y,z for each atom as 2d array
box = myDUMP.viz(len(l_timestep) -
1)[1] # [xlo, ylo, zlo, xhi, yhi, zhi]
box = [float(i) for i in box]
boxsize = [box[3] - box[0], box[4] - box[1],
box[5] - box[1]] # simbox size
myBGF.CRYSTX = [
boxsize[0], boxsize[1], boxsize[2], myBGF.CRYSTX[3],
myBGF.CRYSTX[4], myBGF.CRYSTX[5]
]
#print("l_atoms " + str(len(atoms)))
# update the coordinates in the bgf file
for atom in atoms:
bgfatom = myBGF.getAtom(atom[0])
bgfatom.x = float(atom[2])
bgfatom.y = float(atom[3])
bgfatom.z = float(atom[4])
"""
# com of CNT
for atom in myBGF.a:
if "CNT" in atom.rName:
Mx += atom.x / N_CNT
My += atom.y / N_CNT
Mz += atom.z / N_CNT
"""
myBGF = bgftools.periodicMoleculeSort(myBGF)
#print(Mx, My, Mz)
# transpose for "all atoms in BGF"
for atom in myBGF.a:
atom.x -= Mx
atom.y -= My
atom.z -= Mz
# calculate the moment of inertia (MI) and the center of mass (COM) of CNT from "myBGF"
for aNo in aNo_CNT:
atom = myBGF.getAtom(aNo)
Ixx += (atom.y**2 + atom.z**2) / N_CNT
Iyy += (atom.x**2 + atom.z**2) / N_CNT
Izz += (atom.x**2 + atom.y**2) / N_CNT
Ixy -= (atom.x * atom.y) / N_CNT
Ixz -= (atom.x * atom.z) / N_CNT
Iyz -= (atom.y * atom.z) / N_CNT
# the moment of inertia tensor
I = numpy.array([[Ixx, Ixy, Ixz], [Ixy, Iyy, Iyz], [Ixz, Iyz, Izz]])
# eigenvalue & eigenvector calculation
eigval, eigvec = numpy.linalg.eig(
I) # eigval[0] is the minimum among the values.
# rearrange the U vector
U = numpy.matrix(eigvec)
Ut = U.T
# "myBGF" transform
for atom in myBGF.a:
v = numpy.matrix([atom.x, atom.y, atom.z]).T
Uv = Ut * v
atom.x = float(Uv[0])
atom.y = float(Uv[1])
atom.z = float(Uv[2])
# com of CNT
Mx = 0
My = 0
Mz = 0
for atom in myBGF.a:
if "CNT" in atom.rName:
Mx += atom.x / N_CNT
My += atom.y / N_CNT
Mz += atom.z / N_CNT
#print(Mx, My, Mz)
# transpose for "all atoms in BGF"
for atom in myBGF.a:
atom.x -= Mx
atom.y -= My
atom.z -= Mz
""" ### THIS PART (UPDATING LAMMPSTRJ) DOES NOT WORK ###
myDUMP.aselect.all(time) # all atoms should be selected to be changed
myDUMP.tselect.one(time)
coordx = []; coordy = []; coordz = []
for atom in myBGF.a:
coordx.append(atom.x)
coordy.append(atom.y)
coordz.append(atom.z)
myDUMP.setv("xu", coordx)
myDUMP.setv("yu", coordy)
myDUMP.setv("zu", coordz)
"""
""" ### CHECKING ORIGIN by adding an atom at (0, 0, 0) and is okay. 110914 inkim
origin = bgf.BgfAtom()
origin.aTag = 0
origin.aName = "ORIG"
origin.ffType = "Ar"
origin.chain = "A"
origin.x = 0.0
origin.y = 0.0
origin.z = 0.0
origin.rName = "ORG"
myBGF.addAtom(origin)
"""
# for CNT atoms, calculate some properties
min_x_CNT = 0.0
max_x_CNT = 0.0
radius_CNT = 0.0
height_CNT = 0.0
for aNo in aNo_CNT:
atom = myBGF.getAtom(aNo)
# minimum and maximum x coord of CNT: this will be the height of CNT
if atom.x < min_x_CNT:
min_x_CNT = atom.x
if atom.x > max_x_CNT:
max_x_CNT = atom.x
radius_CNT += math.sqrt(atom.y**2 +
atom.z**2) # average radius of CNT
radius_CNT = radius_CNT / N_CNT
height_CNT = max_x_CNT - min_x_CNT
# determine whether C in MtOH is in the CNT Cylinder
for aNo in aNo_MtOH_C:
atom = myBGF.getAtom(aNo)
dist = math.sqrt(atom.y**2 + atom.z**2)
if atom.x > min_x_CNT \
and atom.x < max_x_CNT \
and dist < radius_CNT:
pass
else:
aNo_MtOH_C_in_CNT.remove(aNo)
#print("Height of CNT: " + str(height_CNT) + " " + str(min_x_CNT) + " " + str(max_x_CNT))
#print("Average radius of CNT: " + str(radius_CNT))
#print("Average number of methanols in CNT: " + str(len(aNo_MtOH_C_in_CNT)))
myDAT.write(str(time) + "\t")
myDAT.write(str("{0:<8.3f}".format(height_CNT)))
myDAT.write(str("{0:<8.3f}".format(radius_CNT)))
myDAT.write(str(len(aNo_MtOH_C_in_CNT)) + "\n")
# write output
myBGF.saveBGF(bgf_file[:-4] + "." + str(time) + ".bgf")
#myDUMP.write(trj_file.split(".")[0] + "_mod." + trj_file.split(".")[1])
print('')
return 1
def crosslink(bgf_file, ff_file, suffix, init_script_original_file,
crosslink_script_original_file, t, ratio, out_file):
# *** variables ***
# used for calculating the number of 1", 2", and 3" amines
l_init_amines = []
l_amines = []
# used for log file output
output = ""
# used for storing number of cross_linking per each cycle
n_crosslink_per_cycle = []
# used for storing cumulative number of crosslink type
n_crosslink_pripri = 0
n_crosslink_prisec = 0
n_crosslink_secsec = 0
# initialization
f_out_file = open(out_file, 'w')
t1 = 0
t2 = 0
# clock for measuring time for a cycle
f_out_file.write("crosslink.py: version " + str(version) + "\n")
f_out_file.write("" + "\n")
f_out_file.write("Job started at " + time.asctime(time.gmtime()) + " on " +
os.environ["HOSTNAME"] + " by " + os.environ["USER"] +
"\n")
f_out_file.write("Command executed at " + os.getcwd() + "\n")
f_out_file.write("Requested options: " + str(sys.argv) + "\n")
f_out_file.write("" + "\n")
f_out_file.write("Using MPI: " + str(mpi_command) + "\n")
f_out_file.write("Using LAMMPS: " + str(lammps_command) + "\n")
f_out_file.write("Using Initial input script: " +
str(init_script_original_file) + "\n")
f_out_file.write("Using Crosslinking input script: " +
str(crosslink_script_original_file) + "\n")
curr_dir = os.path.abspath(".")
temp_dir = scratch_dir + "/" + suffix
initial_bgf_file = bgf_file
# LAMMPS executive command: defined in ~/scripts/clusterSettings.py
lammps_parallel = mpi_command + " " + lammps_command
n_crosslink = 0
f_out_file.flush()
f_out_file.write("\n* Initialization" + "\n")
# determine the initial numbers of primary, secondary, and tertiary amines
l_init_amines = bgftools.getAmineGroupInfo(initial_bgf_file)
f_out_file.write(
"\tNumber of primary, secondary, and tertiary amine groups: " +
str(l_init_amines) + "\n")
AmineGroupInfo = bgftools.setAmineGroupInfo(initial_bgf_file,
initial_bgf_file)
f_out_file.write(
"\tNumber of primary, secondary, and tertiary amine groups are investigated and updated: "
+ str(AmineGroupInfo) + "\n")
# check ratio
n_initAvailAmineSites = AmineGroupInfo[0] + AmineGroupInfo[1]
f_out_file.write("Requested ratio: " + str(ratio) + "\n")
# scratch setting and file copy
f_out_file.write("\tCreating scratch directory: " + temp_dir + "\n")
if not os.path.isdir(temp_dir):
os.makedirs(temp_dir)
# managing hostfile: "COMMENTED"
# Important Note: assume that "cat $PBS_NODEFILE > hostfile" is already executed in the .pbs file
#hostfile = open(temp_dir + "/hostfile", 'w')
#for i in range(0, 12): hostfile.write(os.environ["HOST"] + "\n")
#hostfile.close()
#shutil.copy("hostfile", temp_dir)
#shutil.copy(initial_bgf_file, temp_dir)
shutil.copy(ff_file, temp_dir)
os.chdir(temp_dir)
# LAMMPS input script files for crosslinking
#init_script_original_file = "/home/noische/research/dendrimer/simulation/600/procedures/in.noinit"
#init_script_original_file = "/home/noische/research/dendrimer/simulation/600/procedures/in.initialize"
#crosslink_script_original_file = "/home/noische/research/dendrimer/simulation/procedures/in.xlnk_nvt_1ps_k1.0_r6_0ps"
#crosslink_script_original_file = "/home/noische/research/dendrimer/simulation/600/procedures/in.xlnk_npt_1ps_k16_r28"
#crosslink_script_original_file = "/home/noische/research/dendrimer/simulation/600/procedures/in.xlnk_npt_1ps_k8_r28"
#crosslink_script_original_file = "/home/noische/research/dendrimer/simulation/600/procedures/in.xlnk_npt_1ps_k16_r35"
shutil.copy(init_script_original_file, temp_dir)
shutil.copy(crosslink_script_original_file, temp_dir)
init_script_filename = init_script_original_file.split("/")[
-1] # initial script filename
crosslink_script_filename = crosslink_script_original_file.split("/")[
-1] # crosslinking script filename
createLammpsInput = "createLammpsInput.pl -b " + bgf_file + " -f " + ff_file + " -s " + suffix + " -o 'no shake' -t " + init_script_filename + " > /dev/null"
os.system(createLammpsInput)
initial_in_file = "in." + suffix
# apply a patch for initialization
patch_crosslink.pair_coeff_patch_wo_xlinker(initial_in_file, False)
# first run
t1 = time.time()
runLammpsParallel = lammps_parallel + " -screen none -in " + initial_in_file + " -log " + os.path.join(
temp_dir, suffix + ".log")
f_out_file.write("\tRunning Lammps: see " +
str(os.path.join(temp_dir, suffix + ".log")) + "\n")
os.system(runLammpsParallel)
t2 = time.time()
f_out_file.write("\tElapsed time for the initialization: " + str(t2 - t1) +
" sec\n")
f_out_file.flush()
# update coordinates
initial_trj_file = suffix + "_init.lammpstrj"
f_out_file.write("\tUpdating the BGF file " + initial_bgf_file +
" with LAMMPS trajectory file to _trj_updated.bgf" + "\n")
#updatebgf(initial_bgf_file, initial_trj_file, "_trj_updated.bgf", True, silent=True);
LAMMPS_trj2bgf.getLAMMPSTrajectory(initial_bgf_file, initial_trj_file,
"_trj_updated.bgf", -1, False, True)
f_out_file.write("\tReassigning the origin of " + "_trj_updated.bgf" +
" and saving " + "_centered.bgf" + "\n")
#nu.shutup(); centerbgf("_trj_updated.bgf", "_centered.bgf", "", "com_center"); nu.say()
nu.shutup()
os.system("~tpascal/scripts/centerBGF.pl -b _trj_updated.bgf -f " +
ff_file + " -s _centered.bgf -c com_center")
nu.say()
bgf_file = suffix + "1.bgf"
shutil.copy("_centered.bgf", bgf_file)
f_out_file.flush()
# repeat
if t == 0: t = 10
n_process = 0
endFlag = False
# crosslinking start!
for i in range(1, t + 1):
print("====== " + str(i) + " ======")
n_process = i
t1 = time.time()
N_prisecter = [0, 0, 0]
# N counter init.
f_out_file.write("\n\n====== Cycle " + str(i) + " ======" + "\n")
sname = suffix + str(i)
bgf_file = sname + ".bgf"
input_file = "in." + sname
f_out_file.flush()
# create input
f_out_file.write(
"\n* Minimization, heating, and NVT dynamics for 1ps" + "\n")
f_out_file.write("\tCreating Lammps input for " + bgf_file + "\n")
################## IMPORTANT: min heat nvt -> min nvt for crosslinking ##################### 2011.4.20.
# crosslinking with spring
createLammpsInput = "createLammpsInput.pl -b " + bgf_file + " -f " + ff_file + " -s " + sname + " -o 'no shake' -t 'min " + crosslink_script_filename + "' > /dev/null"
nu.shutup()
os.system(createLammpsInput)
nu.say()
##### apply a patch after the first step
if n_crosslink != 0:
patch_crosslink.pair_coeff_patch_w_xlinker("in." + sname, False)
else:
patch_crosslink.pair_coeff_patch_wo_xlinker("in." + sname, False)
f_out_file.flush()
# run lammps
f_out_file.write("\tRunning LAMMPS simulation on " + sname + "\n")
f_out_file.write("\tRunning Lammps: see " +
str(os.path.join(temp_dir, sname + ".log")) + "\n")
runLammpsParallel = lammps_parallel + " -screen none -in " + input_file + " -log " + sname + ".log"
os.system(runLammpsParallel)
# load trajectory
f_out_file.write("\tLoading Trajectory." + "\n")
initial_trj_file = sname + "_nvt.lammpstrj"
# update bgf
f_out_file.write("\tUpdating BGF file from trajectory." + "\n")
######updatebgf(bgf_file, temp_trj_file, "_trj_updated.bgf", True, silent=True)
#updatebgf(bgf_file, initial_trj_file, "_trj_updated.bgf", True, silent=True)
LAMMPS_trj2bgf.getLAMMPSTrajectory(bgf_file, initial_trj_file,
"_trj_updated.bgf", -1, False, True)
## align bgf periodic box ### added at 110719
#bgftools.periodicMoleculeSort("_trj_updated.bgf", "_aligned.bgf", silent=True)
#bgftools.periodicMoleculeSort("_trj_updated.bgf", "_trj_updated.bgf", [], silent=True)
# center bgf
#nu.shutup(); centerbgf("_trj_updated.bgf", "_centered.bgf", "", "com_center"); nu.say()
nu.shutup()
os.system("~tpascal/scripts/centerBGF.pl -b _trj_updated.bgf -f " +
ff_file + " -s _centered.bgf -c com_center")
nu.say()
shutil.copy("_centered.bgf", bgf_file)
f_out_file.flush()
# check the criteria from BGF file
### 1. open the BGF file
f_out_file.write("\n* Crosslinking module execution" + "\n")
clusterBGF = BgfFile(bgf_file)
clusterBGF.renumber()
f_out_file.write("\tChecking the connectivities in " + bgf_file + "\n")
### 2. list up all N atoms
N_index = []
# REMARK: actually, this is aNo
NN_dist_ok = []
for atom in clusterBGF.a:
if string.strip(atom.ffType) == "N_3":
N_index.append(atom.aNo)
"""
if "PRI" in atom.rName:
N_prisecter[0] += 1
elif "SEC" in atom.rName:
N_prisecter[1] += 1
elif "TER" in atom.rName:
N_prisecter[2] += 1
"""
N_prisecter = bgftools.getAmineGroupInfo(clusterBGF, silent=True)
f_out_file.write("\tThe number of 1', 2', and 3' amine: " +
str(N_prisecter) + "\n")
N_number = len(N_index)
criteria = 5.0
### 3. find N-N pairs which meet
# 1st criteria: distance is between 5 and 6
# 2nd criteria: only PRI or SEC amine
# 3rd criteria: avoid repeating (removed)
for x in range(N_number):
for y in range(x + 1, N_number):
if clusterBGF.distance(clusterBGF.a2i[N_index[x]], clusterBGF.a2i[N_index[y]]) < 6.0 \
and clusterBGF.distance(clusterBGF.a2i[N_index[x]], clusterBGF.a2i[N_index[y]]) > 5.0 \
and clusterBGF.getAtom(N_index[x]).rName != "TER" \
and clusterBGF.getAtom(N_index[y]).rName != "TER":
NN_dist_ok.append([N_index[x], N_index[y]])
if not NN_dist_ok == []:
f_out_file.write("\tCrosslinking candidate site: " +
str(NN_dist_ok) + "\n" +
"\tNow filtering out the candidate.\n")
##### Filtering out the nearest adjacent and the second nearst adjacent cross-linking
if NN_dist_ok != []:
f_out_file.write(
"\tDeleting the nearest adjacent or second nearest adjacent from cross-link candidate site.\n"
)
f_out_file.write("\t\tConnectivity test 1:\n")
for pair in NN_dist_ok:
amine1 = clusterBGF.getAtom(pair[0])
amine2 = clusterBGF.getAtom(pair[1])
amine1_conect = dict()
amine2_conect = dict()
amine1_conect[amine1.aNo] = amine1.CONECT
amine2_conect[amine2.aNo] = amine2.CONECT
# 4-step connection for amine1
for j in range(0, 4):
d_temp = copy.deepcopy(amine1_conect)
for ano in d_temp:
for conect_ano in d_temp[ano]:
tempatom = clusterBGF.getAtom(conect_ano)
amine1_conect[tempatom.aNo] = tempatom.CONECT
# 4-step connection for amine2
for j in range(0, 4):
d_temp = copy.deepcopy(amine2_conect)
for ano in d_temp:
for conect_ano in d_temp[ano]:
tempatom = clusterBGF.getAtom(conect_ano)
amine2_conect[tempatom.aNo] = tempatom.CONECT
# now we get a small map for determining the connectivity
amine1_conect.update(amine2_conect)
# check the connectivity between two N atoms
f_out_file.write("\t\t\t* Checking " + str(pair) +
" (in residue " + str(amine1.rNo) + " and " +
str(amine2.rNo) + "): ")
amine1_to_amine2 = bgftools.findShortestPath(
amine1_conect, amine1.aNo, amine2.aNo)
if not type(amine1_to_amine2) is NoneType:
if len(amine1_to_amine2) < 10:
f_out_file.write(
str(amine1_to_amine2) +
"\tAdjacent pair.. will not be cross-linked.\n")
NN_dist_ok.remove(pair)
else:
f_out_file.write(
str(amine1_to_amine2) +
"\tNot an Adjacent pair.. can be cross-linked.\n")
else:
f_out_file.write(str(amine1_to_amine2) + ": No path.\n")
# if they have same residue then test again
f_out_file.write("\tFiltered pair list from test 1: " +
str(NN_dist_ok) + "\n")
f_out_file.write("\t\tConnectivity test 2:\n")
for pair in NN_dist_ok:
amine1 = clusterBGF.getAtom(pair[0])
amine2 = clusterBGF.getAtom(pair[1])
amine1_aNo = amine1.aNo
amine2_aNo = amine2.aNo
amine1_rNo = amine1.rNo
amine2_rNo = amine2.rNo
tempBGF = bgf.BgfFile()
for atom in clusterBGF.a:
if atom.rNo == amine1_rNo or atom.rNo == amine2_rNo:
tempBGF.addAtom(atom)
amine1_to_amine2 = bgftools.getShortestPath(
tempBGF, amine1_aNo, amine2_aNo)
f_out_file.write("\t\t\t* " + str(pair) + " ( in residue " +
str(amine1.rNo) + " and " + str(amine2.rNo) +
"): ")
if not type(amine1_to_amine2) is NoneType:
if len(amine1_to_amine2) < 10:
f_out_file.write(
str(amine1_to_amine2) +
"\tAdjacent pair.. will not be cross-linked.\n")
NN_dist_ok.remove(pair)
else:
f_out_file.write(
str(amine1_to_amine2) +
"\tNot an Adjacent pair.. can be cross-linked.\n")
else:
f_out_file.write(str(amine1_to_amine2) + "\n")
##### Filtering out the repeat
# Tested with pair_candidate = [[1, 2], [1, 3], [1, 4], [3, 4], [3, 5], [5, 6], [7, 8], [9, 10], [9, 11]]
if NN_dist_ok != []:
f_out_file.write(
"\tRemoving repeated crosslink candidate site: \n")
result = []
pair_candidate = copy.deepcopy(NN_dist_ok)
pair_candidate_flatten = cu.flatten(pair_candidate)
for pair in pair_candidate:
if pair_candidate_flatten.count(
pair[0]) == 1 and pair_candidate_flatten.count(
pair[1]) == 1:
# unique pair in the candidate list
result.append(pair)
pair_candidate.remove(
pair
) # unique pairs will be deleted from pair_candidate
if pair_candidate != []:
# grouping
x = pair_candidate[0][0]
templist = []
grouped = []
for pair in pair_candidate:
if pair[0] == x:
templist.append(pair)
else:
grouped.append(templist)
templist = []
x = pair[0]
templist.append(pair)
grouped.append(templist)
# shuffle pairs
for group in grouped:
random.shuffle(group)
random.shuffle(grouped)
# selecting a pair from group
for group in grouped:
pair = random.choice(group)
result.append(pair)
# removing repeats
temp_repeated = []
result_flatten = cu.flatten(result)
for element in result_flatten:
if result_flatten.count(element) != 1: # if repeated..
temp_repeated.append(element)
for element in temp_repeated:
templist = []
for pair in result:
if element in pair:
templist.append(pair)
result.remove(pair)
if templist != []:
selected = random.choice(templist)
result.append(selected)
NN_dist_ok = result
##### end of filtering
f_out_file.write("\t" + str(len(NN_dist_ok)) +
" N-N pairs which meets the criteria ( " +
str(criteria) + "A ) are found." + "\n")
f_out_file.write("\t" +
"The following N-N pairs will be crosslinked: " +
str(NN_dist_ok) + "\n")
f_out_file.flush()
### 4. attach crosslinker
# NN_dist_ok is the list that will be get crosslinked
if len(NN_dist_ok) > 0:
# used for counting intra- and inter-molecular cross-linking during cycle
n_intra = 0
n_inter = 0
n_crosslink_in_cycle = 0
# if the required ratio reached, stop cross-linking
r = float(n_crosslink) / float(n_initAvailAmineSites)
f_out_file.write("degree of crosslinking: " + str(n_crosslink) +
" / " + str(n_initAvailAmineSites) + " = " +
str(r) + '\n')
if r > ratio:
f_out_file.write(
"************* Requested degree of cross-link reached: EXITING ************"
)
NN_dist_ok = []
endFlag = True
f_out_file.write("\n* Attaching the crosslinker to " + bgf_file +
"\n")
for pair in NN_dist_ok:
f_out_file.write("\tLinking " + str(pair) + " Nitrogens: ")
##### a) load a crosslinker
xlinkerBGF_file = "/home/noische/research/dendrimer/xlinker/20111129/xlinker_new.bgf"
#xlinkerBGF = BgfFile("/home/noische/research/dendrimer/structure/xlinker/xlnker.bgf")
xlinkerBGF = BgfFile(xlinkerBGF_file)
f_out_file.write("\tLoaded the crosslinker file " +
str(xlinkerBGF_file) + "\n")
head_xlnk = xlinkerBGF.a[0] # head C of xlnk
tail_xlnk = xlinkerBGF.a[2] # tail C of xlnk
##### b) select a hydrogen to be detached
# /- head_detach_H(M) /- head_xlnk_detach_H(H) /- tail_detach_H(N)
# head_N(O) --- head_xlnk(J) --- tail_xlnk(K) --- tail_N(P)
# \- tail_xlnk_detach_H(I)
# Detach one H atom at random connected with J which is marked as X
# Detach one H atom at random connected with K which is marked as Xmar
head_N = clusterBGF.getAtom(pair[0])
tail_N = clusterBGF.getAtom(pair[1])
head_N.chain = "O"
tail_N.chain = "P"
head_N_rNo = head_N.rNo
tail_N_rNo = tail_N.rNo
# check whether intra- or inter-molecular crosslinking
if head_N_rNo == tail_N_rNo:
#f_out_file.write("\t\tintra-molecular crosslinking between ")
f_out_file.write(
"\t\tintra-molecular crosslinking between resNo " +
str(head_N.rNo) + " " + str(head_N.rName) + " and " +
"resNo " + str(head_N.rNo) + " " + str(tail_N.rName) +
" amines\n")
f_out_file.write(
"\t\tintra-molecular crosslinking will be passed.")
n_intra += 1
continue
# intramolecular crosslinking is prohibited
else:
f_out_file.write(
"\t\tinter-molecular crosslinking between ")
f_out_file.write("resNo " + str(head_N.rNo) + " " +
str(head_N.rName) + " and " + "resNo " +
str(head_N.rNo) + " " +
str(tail_N.rName) + " amines\n")
n_inter += 1
n_crosslink_in_cycle += 1
#f_out_file.write("resNo " + str(head_N.rNo) + " " + str(head_N.rName) + " and " + "resNo " + str(head_N.rNo) + " " + str(tail_N.rName) + " amines\n") # intramolecular prohibition
# count pri-pri, pri-sec, sec-sec crosslinking number
if "PRI" in head_N.rName and "PRI" in tail_N.rName:
n_crosslink_pripri += 1
elif "PRI" in head_N.rName and "SEC" in tail_N.rName:
n_crosslink_prisec += 1
elif "SEC" in head_N.rName and "PRI" in tail_N.rName:
n_crosslink_prisec += 1
elif "SEC" in head_N.rName and "SEC" in tail_N.rName:
n_crosslink_secsec += 1
f_out_file.write(
"\t\tCumulative PRI-PRI crosslinking number: " +
str(n_crosslink_pripri) + "\n")
f_out_file.write(
"\t\tCumulative PRI-SEC crosslinking number: " +
str(n_crosslink_prisec) + "\n")
f_out_file.write(
"\t\tCumulative SEC-SEC crosslinking number: " +
str(n_crosslink_secsec) + "\n")
f_out_file.write("\n")
f_out_file.write("\t\tChoosing atoms..\n")
# change the crosslinker's rNo as same as head_N
for atom in xlinkerBGF.a:
atom.rNo = head_N_rNo
# choose which H atom will be detached from head_N
temp_list = []
for ano in head_N.CONECT:
if clusterBGF.getAtom(ano).is_hydrogen():
temp_list.append(ano)
if temp_list == []: continue
head_H_detach_candidate_id = random.choice(temp_list)
f_out_file.write("\t\t\tChoosing the hydrogen " +
str(head_H_detach_candidate_id) + " from " +
str(temp_list) + " attached to " +
str(head_N.aNo) + "\n")
# choose which H atom will be detached from tail_N
temp_list = []
for ano in tail_N.CONECT:
if clusterBGF.getAtom(ano).is_hydrogen():
temp_list.append(ano)
if temp_list == []: continue
tail_H_detach_candidate_id = random.choice(temp_list)
f_out_file.write("\t\t\tChoosing the hydrogen " +
str(tail_H_detach_candidate_id) + " from " +
str(temp_list) + " attached to " +
str(tail_N.aNo) + "\n")
head_detach_H = clusterBGF.getAtom(
head_H_detach_candidate_id) # H which will be detached
tail_detach_H = clusterBGF.getAtom(
tail_H_detach_candidate_id) # also
head_detach_H.chain = "M"
tail_detach_H.chain = "N"
head_detach_H_charge = head_detach_H.charge
tail_detach_H_charge = tail_detach_H.charge
# choose which hydrogen will be detached from J
temp_list = []
for atom in xlinkerBGF.a:
if atom.chain == "X":
temp_list.append(atom.aNo)
head_xlnk_detach_H_aNo = random.choice(temp_list)
head_xlnk_detach_H = xlinkerBGF.getAtom(head_xlnk_detach_H_aNo)
head_xlnk_detach_H.chain = "H"
head_xlnk_detach_H_charge = head_xlnk_detach_H.charge
f_out_file.write("\t\t\tH atom " +
str(head_xlnk_detach_H_aNo) +
" in the cross-linker will be deleted.\n")
##### c) move xlinker near to the place: let N-H bond is 1.0 A: move C to 1.4 * N-H bond
#delta = [ head_detach_H.x - head_xlnk.x, head_detach_H.y - head_xlnk.y, head_detach_H.z - head_xlnk.z ] # old method: moves head_xlnk to head_detach_H
#deltax = 1.4 * head_detach_H.x - 0.4 * head_N.x - head_xlnk.x
#deltay = 1.4 * head_detach_H.y - 0.4 * head_N.y - head_xlnk.y
#deltaz = 1.4 * head_detach_H.z - 0.4 * head_N.z - head_xlnk.z
#delta = [deltax, deltay, deltaz]
delta = (1.4 * head_detach_H.x - 0.4 * head_N.x - head_xlnk.x,
1.4 * head_detach_H.y - 0.4 * head_N.y - head_xlnk.y,
1.4 * head_detach_H.z - 0.4 * head_N.z - head_xlnk.z
) # new method
#moveBGF(xlinkerBGF, -deltax, -deltay, -deltaz )
##### c2) rotate xlinker toward tail_N and move
vec = (tail_N.x - head_N.x, tail_N.y - head_N.y,
tail_N.z - head_N.z)
xlinkerBGF = rotateBGF(xlinkerBGF, 1, 3, vec,
0) # 0 means return as BgfFile
moveBGF(xlinkerBGF, delta[0], delta[1], delta[2])
##### c3) find which tail_detach_H is the nearst to tail_N
temp_list = []
temp_dist = 100
for atom in xlinkerBGF.a:
if atom.chain == "Y":
temp_list.append(atom.aNo)
tail_xlnk_detach_H_aNo = 0
for ano in temp_list:
temp = clusterBGF.distance(clusterBGF.a2i[tail_N.aNo],
xlinkerBGF.a2i[ano])
if temp < temp_dist:
temp_dist = temp
tail_xlnk_detach_H_aNo = ano
tail_xlnk_detach_H = xlinkerBGF.getAtom(tail_xlnk_detach_H_aNo)
tail_xlnk_detach_H.chain = "I"
tail_xlnk_detach_H_charge = tail_xlnk_detach_H.charge
##### d) merge
clusterBGF = clusterBGF.merge(xlinkerBGF, True)
clusterBGF.saveBGF("_xlnk_temp.bgf")
clusterBGF = BgfFile("_xlnk_temp.bgf")
##### e) detach head_detach_H and connect head_N-head_xlnk and tail_N-tail_xlnk
for atom in clusterBGF.a:
if atom.chain == "O":
head_N = atom
if atom.rName == "XLK":
if atom.chain == "J":
head_xlnk = atom
if atom.chain == "P":
tail_N = atom
if atom.rName == "XLK":
if atom.chain == "K":
tail_xlnk = atom
if atom.chain == "H":
head_xlnk_detach_H = atom
if atom.chain == "I":
tail_xlnk_detach_H = atom
# disconnect Hydrogens from head_N and tail_N
f_out_file.write("\t\t\tDisconnecting Hydrogen(aNo: " +
str(head_detach_H.aNo) +
") from Nitrogen(aNo: " + str(head_N.aNo) +
") - Head" + "\n")
f_out_file.write("\t\t\tDisconnecting Hydrogen(aNo: " +
str(tail_detach_H.aNo) +
") from Nitrogen(aNo: " + str(tail_N.aNo) +
") - Tail" + "\n")
clusterBGF.disconnect(clusterBGF.a2i[head_N.aNo],
clusterBGF.a2i[head_detach_H.aNo])
clusterBGF.disconnect(clusterBGF.a2i[tail_N.aNo],
clusterBGF.a2i[tail_detach_H.aNo])
# detach Hydrogens from head_N and tail_N
if head_detach_H.aNo > tail_detach_H.aNo:
f_out_file.write("\t\t\tDetaching Hydrogen atom: " +
str(head_detach_H.aNo) + "\n")
clusterBGF.delAtom(clusterBGF.a2i[head_detach_H.aNo])
f_out_file.write("\t\t\tDetaching Hydrogen atom: " +
str(tail_detach_H.aNo) + "\n")
clusterBGF.delAtom(clusterBGF.a2i[tail_detach_H.aNo])
else:
f_out_file.write("\t\t\tDetaching Hydrogen atom: " +
str(tail_detach_H.aNo) + "\n")
clusterBGF.delAtom(clusterBGF.a2i[tail_detach_H.aNo])
f_out_file.write("\t\t\tDetaching Hydrogen atom: " +
str(head_detach_H.aNo) + "\n")
clusterBGF.delAtom(clusterBGF.a2i[head_detach_H.aNo])
# disconnect H-J and delete H
clusterBGF.disconnect(clusterBGF.a2i[head_xlnk.aNo],
clusterBGF.a2i[head_xlnk_detach_H.aNo])
clusterBGF.delAtom(clusterBGF.a2i[head_xlnk_detach_H.aNo])
# disconnect K-I and delete I
clusterBGF.disconnect(clusterBGF.a2i[tail_xlnk.aNo],
clusterBGF.a2i[tail_xlnk_detach_H.aNo])
clusterBGF.delAtom(clusterBGF.a2i[tail_xlnk_detach_H.aNo])
# connect O-J
clusterBGF.connect(
clusterBGF.a2i[head_N.aNo],
clusterBGF.a2i[head_xlnk.aNo]) # O-J connection
# connect K-P
clusterBGF.connect(
clusterBGF.a2i[tail_N.aNo],
clusterBGF.a2i[tail_xlnk.aNo]) # K-P connection
# residue update
f_out_file.write("\t\t\tUpdating residue information." + "\n")
if head_N.rName == "PRI":
head_N.rName = "SEC"
elif head_N.rName == "SEC":
head_N.rName = "TER"
if tail_N.rName == "PRI":
tail_N.rName = "SEC"
elif tail_N.rName == "SEC":
tail_N.rName = "TER"
# chain update
f_out_file.write("\t\t\tUpdating chain information." + "\n")
head_N.chain = "V"
tail_N.chain = "V"
head_xlnk.chain = "A"
tail_xlnk.chain = "A"
f_out_file.write("\n")
# charge
f_out_file.write("\t\tCalculating charges..\n")
f_out_file.write("\t\t\tCharge of the amine group 1: " +
str("{0:8.5f}".format(head_N.charge)) + "\n")
f_out_file.write("\t\t\tCharge of the amine group 2: " +
str("{0:8.5f}".format(tail_N.charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the head C in the cross-link: " +
str("{0:8.5f}".format(head_xlnk.charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the tail C in the cross-link: " +
str("{0:8.5f}".format(tail_xlnk.charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the H detached from amine group 1: " +
str("{0:8.5f}".format(head_detach_H_charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the H detached from amine group 2: " +
str("{0:8.5f}".format(tail_detach_H_charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the H detached from head C in the cross-link: "
+ str("{0:8.5f}".format(head_xlnk_detach_H_charge)) + "\n")
f_out_file.write(
"\t\t\tCharge of the H detached from tail C in the cross-link: "
+ str("{0:8.5f}".format(tail_xlnk_detach_H_charge)) + "\n")
#head_N.charge += head_detach_H_charge
#tail_N.charge += tail_detach_H_charge
#tail_xlnk.charge += tail_xlnk_detach_H_charge
#head_xlnk.charge += head_xlnk_detach_H_charge
head_xlnk.charge += head_detach_H_charge
tail_xlnk.charge += tail_detach_H_charge
head_N.charge += head_xlnk_detach_H_charge
tail_N.charge += tail_xlnk_detach_H_charge
#leftoverCharge = clusterBGF.charge() / 2.0
#head_N.charge -= leftoverCharge
#tail_N.charge -= leftoverCharge
leftoverCharge = clusterBGF.charge()
f_out_file.write("\t\t\tExcess Charge: " +
str("{0:8.5f}".format(leftoverCharge)) + "\n")
f_out_file.write(
"\t\t\tExcess Charge will be added to head N atom.\n")
head_N.charge -= leftoverCharge
f_out_file.write("\t\tModified charge of the amine group 1: " +
str("{0:8.5f}".format(head_N.charge)) + "\n")
f_out_file.write("\t\tModified charge of the amine group 2: " +
str("{0:8.5f}".format(tail_N.charge)) + "\n")
f_out_file.write("\t\tThe charge of the system is " +
str("{0:8.5f}".format(clusterBGF.charge())) +
"\n")
f_out_file.write("" + "\n")
n_crosslink += 1
f_out_file.flush()
##### The end of the crosslinking step
##### Minimization
f_out_file.flush()
clusterBGF.renumber()
f_out_file.write("\tSaving the crosslinked file to " +
"_linked.bgf" + "\n")
clusterBGF.saveBGF("_linked.bgf")
f_out_file.write(
"\n* Crosslinking step is finished. Entering minimizing step."
+ "\n")
# remove bad contacts
#f_out_file.write("\tRemoving bad contacts after crosslinking." + "\n")
#nu.shutup(); removebadcontacts("_linked.bgf", "_removed.bgf", 2.0); nu.say();
# create input for minimization
#createLammpsInput = "createLammpsInput.pl -b _removed.bgf -f " + ff_file + " -s minimization -o 'no shake' -t 'min' > /dev/null"
createLammpsInput = "createLammpsInput.pl -b _linked.bgf -f " + ff_file + " -s minimization -o 'no shake' -t 'min' > /dev/null"
runLammpsParallel = lammps_parallel + " -screen none -in in.minimization -log minimization.log"
f_out_file.write("\tMaking LAMMPS input script for minimization." +
"\n")
nu.shutup()
os.system(createLammpsInput)
nu.say()
# patch for minimization ## added at 110524
if n_crosslink != 0:
patch_crosslink.pair_coeff_patch_w_xlinker(
"in.minimization", False)
else:
patch_crosslink.pair_coeff_patch_wo_xlinker(
"in.minimization", False)
f_out_file.write("\tApplied a patch on in.minimization file.\n")
# run lammps
f_out_file.write("\tMinimizing the structure." + "\n")
f_out_file.write("\tRunning Lammps: see " +
str(os.path.join(temp_dir, "minimization.log")) +
"\n")
os.system(runLammpsParallel)
# load trajectory
f_out_file.write(
"\tLoading Trajectory from the minimization result." + "\n")
update_trj_file = "minimization.min.lammpstrj"
# update bgf
f_out_file.write(
"\tUpdating BGF file from the minimization result." + "\n")
###updatebgf("_removed.bgf", "_temp.lammpstrj", "_trj_updated.bgf", silent=True)
###updatebgf("_linked.bgf", "_temp.lammpstrj", "_trj_updated.bgf", True, silent=True)
#updatebgf("_linked.bgf", "minimization_min.lammpstrj", "_trj_updated.bgf", True, silent=True)
LAMMPS_trj2bgf.getLAMMPSTrajectory("_linked.bgf",
"minimization_min.lammpstrj",
"_trj_updated.bgf", -1, False,
True)
# align bgf periodic box ### added at 110719
#bgftools.periodicMoleculeSort("_trj_updated.bgf", "_trj_updated.bgf", [], silent=True)
# center bgf
f_out_file.write("\tAligning BGF file." + "\n")
#nu.shutup(); centerbgf("_trj_updated.bgf", "_centered.bgf", "", "com_center"); nu.say()
nu.shutup()
os.system("~tpascal/scripts/centerBGF.pl -b _trj_updated.bgf -f " +
ff_file + " -s _centered.bgf -c com_center")
nu.say()
shutil.copy("_centered.bgf", suffix + str(i + 1) + ".bgf")
# store the number of cross-link
n_crosslink_per_cycle.append([i, len(NN_dist_ok)])
f_out_file.write("" + "\n")
f_out_file.write("--- " + str(len(NN_dist_ok)) +
" times of crosslinking occured at the cycle " +
str(i) + "\n")
f_out_file.write(
"--- " + str(n_intra) +
" times of intra-molecular crosslinking occured at the cycle "
+ str(i) + "\n")
f_out_file.write(
"--- " + str(n_inter) +
" times of inter-molecular crosslinking occured at the cycle "
+ str(i) + "\n")
f_out_file.write("--- End of the crosslinking process for cycle " +
str(i) + "\n")
f_out_file.write("--- See " + suffix + str(i + 1) +
".bgf for output. \n" + "\n")
f_out_file.flush()
if endFlag:
break
else:
n_crosslink_per_cycle.append([i, 0])
f_out_file.write("--- Passing the crosslinking step in cycle " +
str(i) + "\n" + "\n")
f_out_file.write(
"--- 0 times of crosslinking occured at the cycle " + str(i) +
"\n")
f_out_file.write(
"--- 0 times of intra-molecular crosslinking occured at the cycle "
+ str(i) + "\n")
f_out_file.write(
"--- 0 times of inter-molecular crosslinking occured at the cycle "
+ str(i) + "\n")
shutil.copy(bgf_file, suffix + str(i + 1) + ".bgf")
f_out_file.flush()
if endFlag:
break
t2 = time.time()
f_out_file.write("Elapsed time for the cycle: " + str(t2 - t1) +
" sec\n")
##### The end of the crosslinking cycle (1ps)
f_out_file.write("\n\nJob completed at " + time.asctime(time.gmtime()) +
"\n")
f_out_file.write("Output structure on " + suffix + "_out.bgf" + "\n")
f_out_file.write("The number of crosslinking through this job: " +
str(n_crosslink) + "\n")
f_out_file.write("\n\n*** Statistics for Crosslinking ***\n")
f_out_file.write("Cycle\tFrequency\n")
for i in n_crosslink_per_cycle:
f_out_file.write(str(i[0]) + "\t" + str(i[1]) + "\n")
f_out_file.close()
shutil.copy(suffix + str(t + 1) + ".bgf", suffix + "_out.bgf")
# minimization
nu.shutup()
createLammpsInput = "~tpascal/scripts/createLammpsInput.pl" + " -b temp.bgf " + " -f /home/noische/ff/DREIDING2.21.ff" + " -s temp " + " -o 'no shake' -t min " + " > /dev/null"
_ = os.system(createLammpsInput)
in_file = "in.temp"
data_file = "data.temp"
os.system("sed -i 's/dielectric 1/dielectric 72/' " + in_file)
os.system("sed -i 's/kspace_style pppm 0.0001/kspace_style none/' " +
in_file)
os.system("sed -i 's/boundary p p p/boundary s s s/' " + in_file)
os.system("sed -i 's/1.0e-4 1.0e-4 500 5000/1.0e-6 1.0e-6 5000 50000/' " +
in_file)
#os.system("sed -i 's/2.0 multi/5.0 multi/' " + in_file)
os.system(
"sed -i 's/lj\/charmm\/coul\/long\/opt 7.5 8.50000/lj\/cut\/coul\/debye 0.142 10/' "
+ in_file)
os.system("sed -i 's/every 2 delay 4/every 1 delay 0/' " + in_file)
#os.system("sed -i 's/0.000000 50.000000/-10.000000 10.000000/' " + data_file)
os.system("sed -i 's/0 # X/0 0 # X/' " + data_file)
os.system("sed -i 's/Impropers//' " + data_file)
runLammps = mpi_command + lammps_command + " -in in.temp -log lammps.log >> temp.log"
_ = os.system(runLammps)
LAMMPS_trj2bgf.getLAMMPSTrajectory("temp.bgf", "temp.min.lammpstrj",
"temp.bgf", -1, False, True)
mybgf = bgf.BgfFile('temp.bgf')
nu.say()
mybgf.saveBGF(out_file)
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
"""