def test_actionlist(self):
'''test_actionlist
'''
dslist = DatasetList()
actlist = ActionList()
traj = pt.iterload("./data/Tc5b.x", "./data/Tc5b.top")
mask_list = ['@CB @CA', '@CA @H']
for mask in mask_list:
actlist.add(CA.Action_Vector(), mask, traj.top, dslist=dslist)
actlist.compute(traj)
dslist2 = pt.calc_vector(traj, mask_list)
dslist4 = va.vector_mask(traj, mask_list)
dslist3_0 = pt.calc_vector(traj, mask_list[0])
dslist3_1 = pt.calc_vector(traj, mask_list[1])
aa_eq(dslist3_0, dslist2[0])
aa_eq(dslist3_1, dslist2[1])
aa_eq(dslist3_0, dslist4[0])
aa_eq(dslist3_1, dslist4[1])
aa_eq(dslist3_0, dslist[0].values)
aa_eq(dslist3_1, dslist[1].values)
def test_actionlist(self):
'''test_actionlist
'''
dslist = DatasetList()
actlist = ActionList()
traj = pt.iterload("./data/Tc5b.x", "./data/Tc5b.top")
mask_list = ['@CB @CA', '@CA @H']
for mask in mask_list:
actlist.add(CA.Action_Vector(),
mask,
traj.top,
dslist=dslist)
actlist.compute(traj)
dslist2 = pt.calc_vector(traj, mask_list)
dslist4 = va.vector_mask(traj, mask_list)
dslist3_0 = pt.calc_vector(traj, mask_list[0])
dslist3_1 = pt.calc_vector(traj, mask_list[1])
aa_eq(dslist3_0, dslist2[0])
aa_eq(dslist3_1, dslist2[1])
aa_eq(dslist3_0, dslist4[0])
aa_eq(dslist3_1, dslist4[1])
aa_eq(dslist3_0, dslist[0].values)
aa_eq(dslist3_1, dslist[1].values)
def calcNOEfull(self, tstep=1.0, numtaus=128):
print('calculating correlation functions\n')
self.getDistances()
self.ACFfull = dict()
for (r1, a1), (r2, a2) in zip(self.list1, self.list2):
label = str(r1) + a1 + ':' + str(r2) + a2
first_in_seq = list(self.traj.top.residues)[0].index
idx1 = pt.select_atoms(
searchAT[a1] + ' & :' + str(r1 - first_in_seq), self.traj.top)
idx2 = pt.select_atoms(
searchAT[a2] + ' & :' + str(r2 - first_in_seq), self.traj.top)
pairs = list(map(list, product(idx1, idx2)))
data_vec = va.vector_mask(self.traj, pairs, dtype='ndarray')
self.ACFfull[label] = []
if len(data_vec.shape) < 3:
data_vec = [data_vec]
#self.dist[label] ?
for n, vals in enumerate(data_vec):
r = self.dist[label][n]
cosines = np.array([
np.dot(unit_vector(vals[0]), unit_vector(v)) for v in vals
])
cosines = 1.5 * cosines**2 - .5
cosines = cosines / ((r * 1e-10)**3)
tot = mycorrelate2(cosines, norm=False)
self.ACFfull[label].append(tot)
del self.traj
dump(self.ACFfull, open(get_today() + '-ACFfull.pkl', 'wb'))
print('fitting correlation functions\n')
self.ACFfull_fit = dict()
for label, acfs in self.ACFfull.items():
#print(acfs)
acf = np.mean(acfs, axis=0)
norm_fact = acf[0]
acf /= acf[0]
#print(acf)
taus, amps, nu1, nu2 = \
smooth_and_fit(acf, tstep=tstep, lp_threshold=0.15, lp_threshold_2=0.05, mintau=2., numtaus=numtaus)
self.ACFfull_fit[label] = [taus, np.array(amps) * norm_fact]
#print(taus, np.array(amps)*norm_fact)
dump(self.ACFfull_fit, open(get_today() + '-ACFfull-fit.pkl', 'wb'))
print('calculating NOE rates\n')
self.sigma_full = dict()
for label, fit in self.ACFfull_fit.items():
taus, amps = fit
taus = np.array(taus) * 1e-12
sigma = (mu_0 / (4 * pi))**2
sigma = sigma * gamma**4 * hbar**2 * .1
#sigma = sigma/(self.reff[label]*1e-10)**6
temp = [a * t for a, t in zip(amps, taus)]
self.sigma_full[label] = sigma * np.sum(temp)
dump(self.sigma_full, open(get_today() + '-sigma-full.pkl', 'wb'))
def RDC_from_S(eigx, eigz, Aa, Ar, trajfiles, topfile, rdc_types, masks, step=1, seg=None):
results = dict()
for n, tr in enumerate(trajfiles):
temp =dict()
traj = pt.iterload(tr, topfile, frame_slice=(0, -1, step))
if seg==None:
residues = [r.index+1 for r in traj.topology.residues if r.name!='PRO' and r.index>0]
else:
residues = [r.index+1 for r in traj.topology.residues
if r.name!='PRO' and r.index>=1 and r.index+1>=seg[0] and r.index+1<=seg[1]]
#get vectors from traj
for m, el in enumerate(rdc_types):
if seg==None:
indices1 = pt.select_atoms(masks[m][0]+' & !(:1) & !(:PRO)', traj.top)
indices2 = pt.select_atoms(masks[m][1]+' & !(:1) & !(:PRO)', traj.top)
else:
indices1 = np.array([
pt.select_atoms(masks[m][0]+' & :'+str(res)+' & !(:PRO)', traj.top)[0]
for res in residues #range(seg[0], seg[1]+1)
if len(pt.select_atoms(masks[m][0]+' & :'+str(res)+' & !(:PRO)', traj.top))>0])
indices2 = np.array([
pt.select_atoms(masks[m][1]+' & :'+str(res)+' & !(:PRO)', traj.top)[0]
for res in residues #range(seg[0], seg[1]+1)
if len(pt.select_atoms(masks[m][1]+' & :'+str(res)+' & !(:PRO)', traj.top))>0])
if el=='CAHA':
indices2 = indices1+1
pairs = np.array(list(zip(indices1, indices2)))
data_vec = va.vector_mask(traj, pairs, dtype='ndarray')
#calculate theta and phi angles
thetas = [[angle_between(vec, eigz) for vec in veclist] for veclist in data_vec]
projs = [[projection(vec, eigz) for vec in veclist] for veclist in data_vec]
phis = [[angle_between(vec, eigx) for vec in veclist] for veclist in projs]
Dmax = -(gammas[masks[m][0]]*gammas[masks[m][1]]*h*mu_0)/(8*pi**3*bond[masks[m][0]+masks[m][1]]**3)
temp[el] = {residues[idx]:np.mean([Dmax*(0.5*(3*np.cos(th)**2-1)*Aa+0.75*Ar*np.sin(th)**2*np.cos(2*phis[idx][f]) )
for f, th in enumerate(ths)])
for idx, ths in enumerate(thetas)}
results[n] = temp
with open('rdc_'+str(n)+'.pkl', 'wb') as tf:
dump(temp, tf)
return results
def calcACFreff(self):
self.ACFreff = dict()
#self.test=dict()
for (r1, a1), (r2, a2) in zip(self.list1, self.list2):
label = str(r1) + a1 + ':' + str(r2) + a2
first_in_seq = list(self.traj.top.residues)[0].index
idx1 = pt.select_atoms(
searchAT[a1] + ' & :' + str(r1 - first_in_seq), self.traj.top)
idx2 = pt.select_atoms(
searchAT[a2] + ' & :' + str(r2 - first_in_seq), self.traj.top)
pairs = list(map(list, product(idx1, idx2)))
data_vec = va.vector_mask(self.traj, pairs, dtype='ndarray')
if len(data_vec.shape) < 3:
data_vec = [data_vec]
self.ACFreff[label] = [
pt.timecorr(vals,
vals,
order=2,
tstep=1,
tcorr=len(vals),
norm=False,
dtype='ndarray') for vals in data_vec
]
#for testing only: calculate the ACF without pytraj.timecorr
#self.test[label] = []
#for vals in data_vec:
#vals2 = np.array([unit_vector(v) for v in vals])
#x = vals2[:, 0]
#y = vals2[:, 1]
#z = vals2[:, 2]
#x2 = mycorrelate2(x**2, norm=False)
#y2 = mycorrelate2(y**2, norm=False)
#z2 = mycorrelate2(z**2, norm=False)
#xy = mycorrelate2(x*y, norm=False)
#yz = mycorrelate2(y*z, norm=False)
#xz = mycorrelate2(x*z, norm=False)
#tot = (x2+y2+z2+2*xy+2*xz+2*yz)
#tot /= tot[0]
#tot = 1.5*(tot)-.5
dump(self.ACFreff, open(get_today() + '-ACFreff.pkl', 'wb'))
def test_ired_vector(self):
'''test mask as a list of strings or as a 2D array of integers
'''
parm_dir = os.path.join(cpptraj_test_dir, 'Test_IRED',
'1IEE_A_prot.prmtop')
trajin_dir = os.path.join(cpptraj_test_dir, 'Test_IRED',
'1IEE_A_test.mdcrd')
traj = pt.iterload(trajin_dir, parm_dir)
# get a list of mask from cpptraj input
maskes = []
lines = None
n_indices_cpp = []
with open('data/ired.in', 'r') as fh:
lines = fh.readlines()
for line in lines:
if 'vector' in line and 'ired' in line:
# example: vector v100 @1541 ired @1542
sline = line.split()
mask = ' '.join((sline[2], sline[4]))
n_indices_cpp.append(int(sline[2][1:]) - 1)
maskes.append(mask)
h_indices_cpp = [i + 1 for i in n_indices_cpp]
# calcuate vector from a list of strings
data_vec = va.vector_mask(traj, maskes)
# calcuate vector from a 2d array of integers
nh_indices = np.array(list(zip(n_indices_cpp, h_indices_cpp)))
data_vec_2 = va.vector_mask(traj, nh_indices)
# re-create cpptraj input to run cpptraj
txt = ''.join(lines)
# add parm and trajin lines
txt = 'parm ' + parm_dir + '\n' + \
'trajin ' + trajin_dir + '\n' + \
txt
state = pt.datafiles.load_cpptraj_output(txt, dtype='state')
state.run()
cpp_data = state.datasetlist
cpp_vectors = cpp_data.grep('vector', mode='dtype').values
cpp_matired = cpp_data.grep('matrix', mode='dtype')['matired']
# assert between pytraj's data_vec and cpptraj's cpp_vectors
aa_eq(data_vec, cpp_vectors)
# from a 2D array of integers
aa_eq(data_vec_2, cpp_vectors)
# test ired vector with ired matrix
# open file
with open('data/ired_reduced.in', 'r') as fh:
text = ''.join(fh.readlines())
state2 = pt.load_batch(traj, text)
state2.run()
data = pt.ired_vector_and_matrix(traj, nh_indices, order=2)
data_vec_3 = data[0]
assert len(data_vec_3) == 126, 'must have 126 vectors'
matired = data[1]
# TODO: know why??
matired /= matired[0, 0]
aa_eq(data_vec_3, cpp_vectors)
assert pt.tools.rmsd(matired.flatten(),
cpp_matired.values) < 1E-6, 'matired'
def __init__(self,
top,
trajs,
rotfit=False,
mask='@CA',
stride=1,
n_blocks=1,
bframes=50000,
skip=25000):
#read sequence
if trajs[0][-3:] == 'dcd':
traj = pt.iterload(trajs[0], top)
else:
#filename, top_name = get_fn(trajs[0]), get_fn(top)
m_traj = md.load(trajs[0], top=top)
traj = pt.Trajectory(xyz=m_traj.xyz.astype('f8'), top=top)
self.sequence = {res.index + 1: res.name for res in traj.top.residues}
self.top = top
self.trajs = trajs
n_frames = len(traj)
self.n_indices = pt.select_atoms('@N & !(:1) & !(:PRO)', traj.top)
self.h_indices = self.n_indices + 1
self.nh_pairs = np.array(list(zip(self.n_indices, self.h_indices)))
res_list = sorted(
[k for k, val in self.sequence.items() if val != 'PRO' and k != 1])
self.acf = dict()
self.S2 = dict()
if n_blocks == 1:
for tt, t in enumerate(self.trajs):
if t[-3:] == 'dcd':
traj = pt.load(t, top, stride=stride)
else:
#filename, top_name = get_fn(t), get_fn(top)
m_traj = md.load(t, top=top)
traj = pt.Trajectory(xyz=m_traj.xyz.astype('f8'), top=top)
if rotfit == True:
#print('RMSD fitting on '+mask)
_ = traj.superpose(ref=-1, mask='@CA')
data_vec = va.vector_mask(traj, self.nh_pairs, dtype='ndarray')
self.acf[tt] = {
res_list[n]: pt.timecorr(vals,
vals,
order=2,
tstep=1,
tcorr=len(vals),
norm=False,
dtype='ndarray')
for n, vals in enumerate(data_vec)
}
self.S2[tt] = {
res_list[n]: S2(vals)
for n, vals in enumerate(data_vec)
}
else:
index = 0
bcount = 0
for tt, t in enumerate(self.trajs):
print(tt + 1, '/', len(self.trajs))
while bcount < n_blocks:
first_frame = bcount * skip
last_frame = first_frame + bframes
if t[-3:] == 'dcd':
traj = pt.load(t,
top,
frame_indices=slice(
first_frame, last_frame, stride))
else:
#filename, top_name = get_fn(t), get_fn(top)
m_traj = md.load(t, top=top)
m_traj = m_traj[first_frame:last_frame:stride]
traj = pt.Trajectory(xyz=m_traj.xyz.astype('f8'),
top=top)
if rotfit == True:
_ = traj.superpose(ref=-1, mask='@CA')
data_vec = va.vector_mask(traj,
self.nh_pairs,
dtype='ndarray')
self.acf[index] = {
res_list[n]: pt.timecorr(vals,
vals,
order=2,
tstep=1,
tcorr=len(vals),
norm=False,
dtype='ndarray')
for n, vals in enumerate(data_vec)
}
self.S2[index] = {
res_list[n]: S2(vals)
for n, vals in enumerate(data_vec)
}
bcount += 1
index += 1
bcount = 0
def test_ired_vector(self):
'''test mask as a list of strings or as a 2D array of integers
'''
parm_dir = os.path.join(cpptraj_test_dir, 'Test_IRED',
'1IEE_A_prot.prmtop')
trajin_dir = os.path.join(cpptraj_test_dir, 'Test_IRED',
'1IEE_A_test.mdcrd')
traj = pt.iterload(trajin_dir, parm_dir)
# get a list of mask from cpptraj input
maskes = []
lines = None
n_indices_cpp = []
with open('data/ired.in', 'r') as fh:
lines = fh.readlines()
for line in lines:
if 'vector' in line and 'ired' in line:
# example: vector v100 @1541 ired @1542
sline = line.split()
mask = ' '.join((sline[2], sline[4]))
n_indices_cpp.append(int(sline[2][1:]) - 1)
maskes.append(mask)
h_indices_cpp = [i + 1 for i in n_indices_cpp]
# calcuate vector from a list of strings
data_vec = va.vector_mask(traj, maskes)
# calcuate vector from a 2d array of integers
nh_indices = np.array(list(zip(n_indices_cpp, h_indices_cpp)))
data_vec_2 = va.vector_mask(traj, nh_indices)
# re-create cpptraj input to run cpptraj
txt = ''.join(lines)
# add parm and trajin lines
txt = 'parm ' + parm_dir + '\n' + \
'trajin ' + trajin_dir + '\n' + \
txt
state = pt.datafiles.load_cpptraj_output(txt, dtype='state')
state.run()
cpp_data = state.datasetlist
cpp_vectors = cpp_data.grep('vector', mode='dtype').values
cpp_matired = cpp_data.grep('matrix', mode='dtype')['matired']
# assert between pytraj's data_vec and cpptraj's cpp_vectors
aa_eq(data_vec, cpp_vectors)
# from a 2D array of integers
aa_eq(data_vec_2, cpp_vectors)
# test ired vector with ired matrix
# open file
with open('data/ired_reduced.in', 'r') as fh:
text = ''.join(fh.readlines())
state2 = pt.load_batch(traj, text)
state2.run()
data = pt.ired_vector_and_matrix(traj, nh_indices, order=2)
data_vec_3 = data[0]
assert len(data_vec_3) == 126, 'must have 126 vectors'
matired = data[1]
# TODO: know why??
matired /= matired[0, 0]
aa_eq(data_vec_3, cpp_vectors)
assert pt.tools.rmsd(matired.flatten(),
cpp_matired.values) < 1E-6, 'matired'
def RDC(trajfiles, topfile, rdc_types, masks, step=1, seg=None):
results = dict()
for n, tr in enumerate(trajfiles):
traj = pt.iterload(tr, topfile, frame_slice=(0, -1, step))
if seg==None:
residues = [r.index+1 for r in traj.topology.residues if r.name!='PRO' and r.index>0]
else:
residues = [r.index+1 for r in traj.topology.residues
if r.name!='PRO' and r.index>=1 and r.index+1>=seg[0] and r.index+1<=seg[1]]
temp = dict()
Aa = []
Ar = []
eigx = []
eigy = []
eigz = []
alpha = []
gamma = []
beta = []
for f, frame in enumerate(traj):
pt.write_traj("temp.pdb", frame, top=traj.top, overwrite=True)
if seg==None:
os.system("pales -pdb temp.pdb > pales.out")
else:
os.system("pales -pdb temp.pdb -s1 "+str(seg[0])+" -sN "+str(seg[1])+" > pales.out")
os.system("grep \"DATA EIGENVALUES (Sxx_d,Syy_d,Szz_d)\" pales.out > eigenvalues.out")
with open('eigenvalues.out', 'r') as tf:
data = tf.readlines()
data = data[0].split()
Sxx_d=float(data[-3])
Syy_d=float(data[-2])
Szz_d=float(data[-1])
Aa.append(Szz_d)
Ar.append((2./3.)*(Sxx_d-Syy_d))
os.system("grep \"DATA EULER_ANGLES\" pales.out > euler.out")
with open('euler.out', 'r') as tf:
data = tf.readlines()
data = data[0].split()
alpha.append(float(data[-3]))
beta.append(float(data[-2]))
gamma.append(float(data[-1]))
os.system("grep \"DATA EIGENVECTORS X_AXIS\" pales.out > eigenvalues1.out")
with open('eigenvalues1.out', 'r') as tf:
data = tf.readlines()
data = data[0].split()
eigx.append([float(data[-3]), float(data[-2]), float(data[-1])])
os.system("grep \"DATA EIGENVECTORS Y_AXIS\" pales.out > eigenvalues2.out")
with open('eigenvalues2.out', 'r') as tf:
data = tf.readlines()
data = data[0].split()
eigy.append([float(data[-3]), float(data[-2]), float(data[-1])])
os.system("grep \"DATA EIGENVECTORS Z_AXIS\" pales.out > eigenvalues3.out")
with open('eigenvalues3.out', 'r') as tf:
data = tf.readlines()
data = data[0].split()
eigz.append([float(data[-3]), float(data[-2]), float(data[-1])])
R = np.array(Ar)/np.array(Aa)
#get vectors from traj
for m, el in enumerate(rdc_types):
if seg==None:
indices1 = pt.select_atoms(masks[m][0]+' & !(:1) & !(:PRO)', traj.top)
indices2 = pt.select_atoms(masks[m][1]+' & !(:1) & !(:PRO)', traj.top)
else:
indices1 = np.array([
pt.select_atoms(masks[m][0]+' & :'+str(res)+' & !(:PRO)', traj.top)[0]
for res in residues #range(seg[0], seg[1]+1)
if len(pt.select_atoms(masks[m][0]+' & :'+str(res)+' & !(:PRO)', traj.top))>0])
indices2 = np.array([
pt.select_atoms(masks[m][1]+' & :'+str(res)+' & !(:PRO)', traj.top)[0]
for res in residues #range(seg[0], seg[1]+1)
if len(pt.select_atoms(masks[m][1]+' & :'+str(res)+' & !(:PRO)', traj.top))>0])
if el=='CAHA':
indices2 = indices1+1
pairs = np.array(list(zip(indices1, indices2)))
data_vec = va.vector_mask(traj, pairs, dtype='ndarray')
#calculate theta and phi angles
thetas = [[angle_between(vec, zaxis) for vec, zaxis in zip(veclist, eigz)] for veclist in data_vec]
projs = [[projection(vec, zaxis) for vec, zaxis in zip(veclist, eigz)] for veclist in data_vec]
phis = [[angle_between(vec, xaxis) for vec, xaxis in zip(veclist, eigx)] for veclist in projs]
#calculate RDC values
#rPQcub = [np.nanmean([(np.linalg.norm(v)*1e-10)**3 for v in veclist]) for veclist in data_vec]
#Dmax = [-(gammas[masks[m][0]]*gammas[masks[m][1]]*h*mu_0)/(8*pi**3*rPQ) for rPQ in rPQcub]
Dmax = -(gammas[masks[m][0]]*gammas[masks[m][1]]*h*mu_0)/(8*pi**3*bond[masks[m][0]+masks[m][1]]**3)
# temp[el] = {residues[idx]:[Dmax[idx]*(0.5*(3*np.cos(th)**2-1)*Aa[f]+0.75*Ar[f]*np.sin(th)**2*np.cos(2*phis[idx][f]) )
# for f, th in enumerate(ths)]
# for idx, ths in enumerate(thetas)}
temp[el] = {residues[idx]:np.mean([Dmax*(0.5*(3*np.cos(th)**2-1)*Aa[f]+0.75*Ar[f]*np.sin(th)**2*np.cos(2*phis[idx][f]) )
for f, th in enumerate(ths)])
for idx, ths in enumerate(thetas)}
results[n] = temp
with open('rdc_'+str(n)+'.pkl', 'wb') as tf:
dump(temp, tf)
return results