#P = 0.25
atmlist = []
idir = '/home/jujuman/Research/extensibility_test_sets/COMP6v2/TripeptideS/inputs/'
cdir = '/home/jujuman/Research/extensibility_test_sets/COMP6v2/TripeptideS/confs/'
#idir = '/home/jujuman/Research/GDB-11-wB97X-6-31gd/dnnts_h2ocluster/h2o_cluster/inputs/'
#cdir = '/home/jujuman/Research/GDB-11-wB97X-6-31gd/dnnts_h2ocluster/h2o_cluster/confs/'
files = os.listdir(idir)
files.sort()
if not os.path.exists(cdir):
os.mkdir(cdir)
dc = aat.diverseconformers(cnstfile, saefile, nnfdir, aevsize, 0, False)
for fi, f in enumerate(files):
print(fi, 'of', len(files), ') Working on:', f)
data = hdn.read_rcdb_coordsandnm(idir + f)
spc = data["species"]
xyz = data["coordinates"]
nmc = data["nmdisplacements"]
frc = data["forceconstant"]
if set(['P', 'B', 'Br']).isdisjoint(set(spc)):
#Ngen = K*frc.size
#Nkep = int(Ngen*P)
Ngen = K * 8
def normal_mode_sampling(self, T, Ngen, Nkep, maxd, sig, gpuid):
of = open(self.ldtdir + self.datdir + '/info_data_nms.nfo', 'w')
aevsize = self.netdict['aevsize']
anicv = aat.anicrossvalidationconformer(self.netdict['cnstfile'],
self.netdict['saefile'],
self.netdict['nnfprefix'],
self.netdict['num_nets'],
[gpuid], False)
dc = aat.diverseconformers(self.netdict['cnstfile'],
self.netdict['saefile'],
self.netdict['nnfprefix'] + '0/networks/',
aevsize, gpuid, False)
Nkp = 0
Nkt = 0
Ntt = 0
idx = 0
for di, id in enumerate(self.idir):
of.write(
str(di) + ' of ' + str(len(self.idir)) + ') dir: ' + str(id) +
'\n')
#print(di,'of',len(self.idir),') dir:', id)
files = os.listdir(id)
files.sort()
Nk = 0
Nt = 0
for fi, f in enumerate(files):
print(f)
data = hdt.read_rcdb_coordsandnm(id + f)
#print(id+f)
spc = data["species"]
xyz = data["coordinates"]
nmc = data["nmdisplacements"]
frc = data["forceconstant"]
if "charge" in data and "multip" in data:
chg = data["charge"]
mlt = data["multip"]
else:
chg = "0"
mlt = "1"
nms = nmt.nmsgenerator(xyz,
nmc,
frc,
spc,
T,
minfc=5.0E-2,
maxd=maxd)
conformers = nms.get_Nrandom_structures(Ngen)
if conformers.shape[0] > 0:
if conformers.shape[0] > Nkep:
ids = dc.get_divconfs_ids(conformers, spc, Ngen, Nkep,
[])
conformers = conformers[ids]
sigma = anicv.compute_stddev_conformations(conformers, spc)
sid = np.where(sigma > sig)[0]
Nt += sigma.size
Nk += sid.size
if 100.0 * sid.size / float(Ngen) > 0:
Nkp += sid.size
cfn = f.split('.')[0].split('-')[0] + '_' + str(
idx).zfill(5) + '-' + f.split('.')[0].split(
'-')[1] + '_2.xyz'
cmts = [' ' + chg + ' ' + mlt for c in range(Nk)]
hdt.writexyzfilewc(self.cdir + cfn, conformers[sid],
spc, cmts)
idx += 1
Nkt += Nk
Ntt += Nt
of.write(' -Total: ' + str(Nk) + ' of ' + str(Nt) +
' percent: ' + "{:.2f}".format(100.0 * Nk / Nt) + '\n')
of.flush()
#print(' -Total:',Nk,'of',Nt,'percent:',"{:.2f}".format(100.0*Nk/Nt))
del anicv
del dc
of.write('\nGrand Total: ' + str(Nkt) + ' of ' + str(Ntt) +
' percent: ' + "{:.2f}".format(100.0 * Nkt / Ntt) + ' Kept ' +
str(Nkp) + '\n')
#print('\nGrand Total:', Nkt, 'of', Ntt,'percent:',"{:.2f}".format(100.0*Nkt/Ntt), 'Kept',Nkp)
of.close()
def structural_sampling(self, N, sig, gpuid):
of = open(self.ldtdir + self.datdir + '/info_data_strucs.nfo', 'w')
aevsize = self.netdict['aevsize']
anicv = aat.anicrossvalidationconformer(self.netdict['cnstfile'],
self.netdict['saefile'],
self.netdict['nnfprefix'],
self.netdict['num_nets'],
[gpuid], False)
dc = aat.diverseconformers(self.netdict['cnstfile'],
self.netdict['saefile'],
self.netdict['nnfprefix']+'0/networks/',
aevsize,
gpuid, False)
files = os.listdir(self.strucsfolder)
files.sort()
Nkt = 0
Ntt = 0
cnt = 0
for fi,f in enumerate(files):
print(f)
fil = open(self.strucsfolder+f,'r')
lines = fil.readlines()
fil.close()
nlines = len(lines)
# Reading all conformations
nat = int(lines[0])
nconfs=int(round(len(lines)/(nat+2)))
crds=[]
for conf in range(nconfs):
crd =[]
if (conf==0):
if (not re.search("Charge:",lines[1]) or not re.search("Mul:",lines[1])):
raise ValueError('Error: the first comment line in %s must have charge and multiplicity. Please add something like " Charge: 0 Mul: 1 "'%(self.strucsfolder+f))
chg = lines[1].split("Charge:")[1].split()[0]
mul = lines[1].split("Mul:")[1].split()[0]
spc = []
for i in range(nat):
var = lines[conf*(nat+2)+2+i].split()
spc.append(var[0])
crd.append([float(var[1]),float(var[2]),float(var[3])])
else:
for i in range(nat):
var = lines[conf*(nat+2)+2+i].split()
crd.append([float(var[1]),float(var[2]),float(var[3])])
crds.append(crd)
# Select up to N random structures, if needed
if (nconfs>N):
list=[]
for i in range(N):
num=np.random.random_integers(0,nconfs-1)
while(num in list):
num=num=np.random.random_integers(0,nconfs-1)
list.append(num)
ncrds=[]
for i in sorted(list):
ncrds.append(crds[i])
del crds
crds=ncrds
del ncrds
# Converting list to numpy array
crds=np.asarray(crds, dtype=np.float32)
# Filter by QBC
sigma = anicv.compute_stddev_conformations(crds,spc)
sid = np.where( sigma > sig )[0]
Ntt += sigma.size
Nkt += sid.size
of.write(str(cnt+1)+' of '+str(len(files))+') file: '+ str(self.strucsfolder+f) +'\n')
of.write(' -Total: '+str(sid.size)+' of '+str(sigma.size)+' percent: '+"{:.2f}".format(100.0*sid.size/sigma.size)+'\n')
of.flush()
if sid.size > 0:
cfn = f.split('.')[0]+'_strucs.xyz'
cmts = [' '+chg+' '+mul for c in range(sid.size)]
hdt.writexyzfilewc(self.cdir+cfn,crds[sid],spc,cmts)
cnt += 1
of.write('\nGrand Total: '+ str(Nkt)+ ' of '+ str(Ntt)+' percent: '+"{:.2f}".format(100.0*Nkt/Ntt)+'\n')
of.close()