def normalize_pts(self, pts1, pts2):
#print pts1
A = pts1[:, 0:2]
B = pts2[:, 0:2]
assert len(A) == len(B)
N = A.shape[0] # total points
# find centroids
centroid_A = rmsd.centroid(A)
centroid_B = rmsd.centroid(B)
# centre the points
AA = A - centroid_A
BB = B - centroid_B
#get optimal rotation matrix
R = rmsd.kabsch(AA, BB)
# Apply transformation
AA1 = np.dot(AA, R)
angle1 = math.atan2(R[1][0], R[1][1])
pts1[:, 0:2] = AA1 + centroid_B
# rotate the rotations
pts1[:, 2] -= angle1
return rmsd.rmsd(AA1, BB)
def test_pdb(example_path="examples", threshold=0.001):
"""
A simple test for the PDB functionality
:return: True if all test passed
"""
p_atoms, P = rmsd.get_coordinates(example_path+'/ci2_1.pdb', 'pdb')
q_atoms, Q = rmsd.get_coordinates(example_path+'/ci2_2.pdb', 'pdb')
n_rmsd = rmsd.rmsd(P, Q)
Pc = rmsd.centroid(P)
Qc = rmsd.centroid(Q)
P -= Pc
Q -= Qc
k_rmsd = rmsd.kabsch_rmsd(P, Q)
q_rmsd = rmsd.quaternion_rmsd(P, Q)
if abs(n_rmsd - 26.975) > threshold:
print('Failed to calculate normal RMSD, result: {0}'.format(n_rmsd))
return False
if abs(k_rmsd - 11.777) > threshold:
print('Failed to calculate Kabsch RMSD, result: {0}'.format(k_rmsd))
return False
if abs(q_rmsd - 11.777) > threshold:
print('Failed to calculate quaternion RMSD, result: {0}'.format(q_rmsd))
return False
if abs(q_rmsd - k_rmsd) > threshold ** 2:
print('Failed to yield similar Kabsch and quaternion RMSD, result: {0} vs {1}'.format(k_rmsd, q_rmsd))
return False
return True
def rmsd_distance(atoms_list,
coordinates_list,
translation=False,
rotation=False):
N = len(atoms_list)
kernel = np.zeros((N, N))
# Lower triangular
for i in range(N):
for j in range(i):
coord_i = coordinates_list[i]
coord_j = coordinates_list[j]
# unique pairs
if translation:
coord_i = coord_i - rmsd.centroid(coord_i)
coord_j = coord_j - rmsd.centroid(coord_j)
if rotation:
kernel[i, j] = rmsd.kabsch_rmsd(coord_i, coord_j)
else:
kernel[i, j] = rmsd.rmsd(coord_i, coord_j)
kernel[j, i] = kernel[i, j]
# np.fill_diagonal(kernel, 0.0)
# iu2 = np.triu_indices(N, 1)
# il2 = np.tril_indices(N, -1)
# kernel[iu2] = kernel[il2]
return kernel
def alin(A,B): A -= rmsd.centroid(A) B -= rmsd.centroid(B) #print (B) U = rmsd.kabsch(A, B) A = np.dot(A, U) return A, B, rmsd.rmsd(A,B)
def calc_rmsd(mol1, mol2):
mol_coord1 = mol_to_coord_mat(mol1)
mol_coord2 = mol_to_coord_mat(mol2)
#Localize on centroid
mol_coord1 -= rmsd.centroid(mol_coord1)
mol_coord2 -= rmsd.centroid(mol_coord2)
return rmsd.quaternion_rmsd(mol_coord1, mol_coord2)
def CalRmsdFrame(coor_frame, ref_coor_frame):
"""
Calculate the rmsd between two frames.
"""
coor_tmp = coor_frame.reshape(-1, 3)
ref_tmp = ref_coor_frame.reshape(-1, 3)
coor_tmp -= rmsd.centroid(coor_tmp)
ref_tmp -= rmsd.centroid(ref_tmp)
return rmsd.kabsch_rmsd(coor_tmp, ref_tmp)
def KabschTransform(MAT_B, MAT_R):
C_road = rmsd.centroid(MAT_R) # centreroid of the road points
C_back = rmsd.centroid(MAT_B) # centreroid of the back points
XYZroad_centered = MAT_R - C_road #XYZ of road after centering
XYZback_centered = MAT_B - C_back #XYZ of back after centering
R = rmsd.kabsch(
XYZroad_centered, XYZback_centered
) # the optimal rotation matrix to rotate road points to back coordinates
return C_road, C_back, R
def count_rmsd(t1, t2):
P = np.array([[float(v.x), float(v.y), float(v.z)] for v in t1.points])
Q = np.array([[float(v.x), float(v.y), float(v.z)] for v in t2.points])
# print("RMSD before translation: ", round(rmsd.kabsch_rmsd(P, Q), 6))
P -= rmsd.centroid(P)
Q -= rmsd.centroid(Q)
# print("RMSD after translation: ", round(rmsd.kabsch_rmsd(P, Q), 6))
return round(rmsd.kabsch_rmsd(P, Q), 4) # round 6 in triangles_matrix
def get_transformation_matrix_kabsch(q, p):
# compute centroids
Pc = rmsd.centroid(p)
Qc = rmsd.centroid(q)
# Kabsch algorithm for estimating R and t
T = np.identity(4)
T[:3, :3] = rmsd.kabsch(p - Pc, q - Qc)
T[:3, 3] = Pc - np.dot(T[:3, :3], Qc)
return T
def RMSDmetric(structure1, structure2):
numParticles = len(structure1) / 3
coords1 = structure1.reshape(int(numParticles), 3)
coords2 = structure2.reshape(int(numParticles), 3)
coords1 = coords1 - rmsd.centroid(coords1)
coords2 = coords2 - rmsd.centroid(coords2)
return rmsd.kabsch_rmsd(coords1, coords2)
def crmsd_kabsch(A, B):
# Translate
A -= rmsd.centroid(A)
B -= rmsd.centroid(B)
# Rotate
U = rmsd.kabsch(A, B)
A = np.dot(A, U)
return rmsd.rmsd(A, B)
def compute_similarity(site_a, site_b):
"""
Compute the similarity between two given ActiveSite instances.
Input: two ActiveSite instances
Output: the similarity between them (a floating point number)
"""
# Get strings of single letter aa residues
s_a = output_aa_string(site_a.residues)
s_b = output_aa_string(site_b.residues)
# Align strings using local alignment algorithm which relies
# on dynamic programming to compute all possible alignments and
# returns the highest scoring alignment.
# Local alignment aims to find the max alignment for substrings
# of two larger strings.
# Matches = +1
# Mismatches, gaps = +0
alignments = pairwise2.align.localxx(s_a, s_b) # perform alignment
if len(alignments) == 0: return float("inf") # return INF if no alignment found
align_a, align_b, s = alignments[0][:3] # extract first alignment
# Output indices where nucleotides in alignment match
inds_a, inds_b = match(align_a, align_b)
if len(inds_a) < 2: return float("inf")
# Create matrix of coordinates for atom CA
V = create_coord_matrix(site_a, inds_a)
W = create_coord_matrix(site_b, inds_b)
# Center and rotate Ca matrices then calculate Root-Mean-Square-Deviation (RMSD)
# It measures the average distance between backbone atoms of two
# superimposed proteins.
# The greater the RMSD, the less similar the proteins are.
# A RMSD equal to 0 represents identical proteins.
# Each protein is a matrix containing x, y, and z coordinates for each CA atom
# The rows of the two matrices are matching residues obtained from the alignment
# To minimize RMSD you must first center the coordinates on the origin so the
# two vectors can be near each other.
V -= rmsd.centroid(V)
W -= rmsd.centroid(W)
# Then find the optimal rotation for matrix W that aligns it best with V
# This is the Kabasch algorithm which works by calculating a covariance matrix
# and then finding the singular value decomposition (SVD) of the cov. matrix
# Last, find the optimal rotation matrix which is the dot product of V and W
# optimized by lowest RMSD
return rmsd.kabsch_rmsd(V,W)
def test_reorder_inertia_hungarian():
# coordinates of scrambled and rotated butane
atoms = np.array(
["C", "C", "C", "C", "H", "H", "H", "H", "H", "H", "H", "H", "H", "H"]
)
p_coord = np.array(
[
[2.142e00, 1.395e00, -8.932e00],
[3.631e00, 1.416e00, -8.537e00],
[4.203e00, -1.200e-02, -8.612e00],
[5.691e00, 9.000e-03, -8.218e00],
[1.604e00, 7.600e-01, -8.260e00],
[1.745e00, 2.388e00, -8.880e00],
[2.043e00, 1.024e00, -9.930e00],
[4.169e00, 2.051e00, -9.210e00],
[3.731e00, 1.788e00, -7.539e00],
[3.665e00, -6.470e-01, -7.940e00],
[4.104e00, -3.840e-01, -9.610e00],
[6.088e00, -9.830e-01, -8.270e00],
[5.791e00, 3.810e-01, -7.220e00],
[6.230e00, 6.440e-01, -8.890e00],
]
)
q_coord = np.array(
[
[6.71454, -5.53848, -3.50851],
[6.95865, -6.22697, -2.15264],
[8.16747, -5.57632, -1.45606],
[5.50518, -6.19016, -4.20589],
[5.33617, -5.71137, -5.14853],
[7.58263, -5.64795, -4.12498],
[6.51851, -4.49883, -3.35011],
[6.09092, -6.11832, -1.53660],
[5.70232, -7.22908, -4.36475],
[7.15558, -7.26640, -2.31068],
[8.33668, -6.05459, -0.51425],
[7.97144, -4.53667, -1.29765],
[4.63745, -6.08152, -3.58986],
[9.03610, -5.68475, -2.07173],
]
)
p_coord -= rmsd.centroid(p_coord)
q_coord -= rmsd.centroid(q_coord)
review = rmsd.reorder_inertia_hungarian(atoms, atoms, p_coord, q_coord)
result_rmsd = rmsd.kabsch_rmsd(p_coord, q_coord[review])
np.testing.assert_almost_equal(0, result_rmsd, decimal=2)
def kabsch(pointsCurrent, pointsRef):
C_curr = rmsd.centroid(pointsCurrent)
C_ref = rmsd.centroid(pointsRef)
points_curr_norm = pointsCurrent - C_curr # XYZ current after centering
points_ref_norm = pointsRef - C_ref
# the optimal rotation matrix to rotate XYZcurrent to XYZref
R = rmsd.kabsch(points_curr_norm, points_ref_norm)
return (R, C_curr, C_ref)
def test_icp(A, shuffle = 1):
# Generate a random dataset
for i in range(num_tests):
print('test', i)
B = np.copy(A)
# Translate
t = np.random.rand(dim)*translation
B += t
# Rotate
R = rotation_matrix(np.random.rand(dim), np.random.rand() * rotation)
B = np.dot(R, B.T).T
# Add noise
B += np.random.randn(N, dim) * noise_sigma
# Shuffle to disrupt correspondence
if (shuffle == 1):
np.random.shuffle(B)
#B[0:100] = 0
#Bcenter = rmsd.centroid(B)
#B[0:100] = Bcenter
print("Normal squared_distance", squared_distance(A, B))
plot3d(A, B)
# Run ICP
A -= rmsd.centroid(A)
B -= rmsd.centroid(B)
T, distances, iterations = icp.icp(B, A, tolerance=0.000001)
print(T.shape)
# Make C a homogeneous representation of B
C = np.ones((N, 4))
C[:,0:3] = np.copy(B)
# Transform C
C = np.dot(T, C.T).T
C = C[:, 0:3]
print("Rotated squared_distance", squared_distance(A, C))
plot3d(A, C)
#assert np.mean(distances) < 6*noise_sigma # mean error should be small
#assert np.allclose(T[0:3,0:3].T, R, atol=6*noise_sigma) # T and R should be inverses
#assert np.allclose(-T[0:3,3], t, atol=6*noise_sigma) # T and t should be inverses
return
def generate_set(A, B):
Asize = A.shape[0]
Bsize = B.shape[0]
temp = np.zeros((max(Asize, Bsize), 3))
if (Asize < Bsize):
temp += rmsd.centroid(A)
temp[0:Asize] = A
A = temp
else:
temp += rmsd.centroid(B)
temp[0:Bsize] = B
B = temp
return A, B
def normalize_pts(pts1, pts2):
#print pts1
A = pts1[:, 0:2]
B = pts2[:, 0:2]
assert len(A) == len(B)
N = A.shape[0] # total points
# find centroids
centroid_A = rmsd.centroid(A)
centroid_B = rmsd.centroid(B)
# centre the points
AA = A - centroid_A
BB = B - centroid_B
#get optimal rotation matrix
R = rmsd.kabsch(AA, BB)
# Apply transformation
AA1 = np.dot(AA, R)
angle1 = math.atan2(R[1][0], R[1][1])
#try flippin' it
AA[:, 1] *= -1
#get optimal rotation matrix
R = rmsd.kabsch(AA, BB)
# Apply transformation
AA2 = np.dot(AA, R)#get optimal rotation matrix
angle2 = math.atan2(R[1][0], R[1][1])
#go with whichever one is better
a2diff = np.abs(pts2[:,2]-(pts1[:,2]-angle2))
a1diff = np.abs(pts2[:,2]-(pts1[:,2]+angle1))
#angle differences are fun!
for i in range(0, len(a1diff)):
while (a2diff[i] > np.pi):
a2diff[i] -= 2*np.pi
while (a1diff[i] > np.pi):
a1diff[i] -= 2*np.pi
print np.degrees(a2diff), np.degrees(a1diff)
if (np.sum(np.abs(a1diff)) > np.sum(np.abs(a2diff))):
pts1[:, 0:2] = AA2 + centroid_B
# rotate the rotations (backwards, since we mirrored the thing)
pts1[:, 2] -= angle2
return rmsd.rmsd(AA2, BB)
else:
pts1[:, 0:2] = AA1 + centroid_B
# rotate the rotations
pts1[:, 2] += angle1
return rmsd.rmsd(AA1, BB)
def center_and_kabsch(P, Q):
""" returns centroid of P, Q and a rotational matrix (in subsequent analyses, you may want to subtract centroidP/Q from your point sets)
that minimizes RMSD between corresponding points in P and Q, *when applied to P* """
P = np.array(P)
Q = np.array(Q)
centroidP = rmsd.centroid(P)
centroidQ = rmsd.centroid(Q)
P = P - centroidP
Q = Q - centroidQ
return centroidP, centroidQ, rmsd.kabsch(P, Q)
def calculate_rmsd(mol_id,
geometry_id1=None,
geometry_id2=None,
heavy_atoms_only=False):
# These cause circular import errors if we put them at the top of the file
from ._girder import GirderClient
from ._calculation import GirderMolecule
# Allow the user to pass in a GirderMolecule instead of an id
if isinstance(mol_id, GirderMolecule):
mol_id = mol_id._id
if geometry_id1 is None:
cjson1 = GirderClient().get('/molecules/%s/cjson' % mol_id)
else:
cjson1 = GirderClient().get('/molecules/%s/geometries/%s/cjson' %
(mol_id, geometry_id1))
if geometry_id2 is None:
cjson2 = GirderClient().get('/molecules/%s/cjson' % mol_id)
else:
cjson2 = GirderClient().get('/molecules/%s/geometries/%s/cjson' %
(mol_id, geometry_id2))
coords1 = cjson1['atoms']['coords']['3d']
coords1_triplets = zip(coords1[0::3], coords1[1::3], coords1[2::3])
A = np.array([x for x in coords1_triplets])
coords2 = cjson2['atoms']['coords']['3d']
coords2_triplets = zip(coords2[0::3], coords2[1::3], coords2[2::3])
B = np.array([x for x in coords2_triplets])
if heavy_atoms_only:
atomic_numbers = cjson1['atoms']['elements']['number']
heavy_indices = [i for i, n in enumerate(atomic_numbers) if n != 1]
A = A.take(heavy_indices, 0)
B = B.take(heavy_indices, 0)
# Translate
A -= rmsd.centroid(A)
B -= rmsd.centroid(B)
# Rotate
U = rmsd.kabsch(A, B)
A = np.dot(A, U)
return rmsd.rmsd(A, B)
def xyz4a_position(i, lines):
template1 = np.array([[-0.93454, 0.91954, 1.11786],
[-0.01675, 0.03488, 0.58954],
[1.10621, 0.72253, -0.004],
[1.6148, 1.14817, -0.57075]])
template2 = np.array([[-0.93454, 0.91954, 1.11786],
[-0.01675, 0.03488, 0.58954],
[1.10621, 0.72253, -0.004],
[1.6148, 1.14817, -0.57075],
[-0.64719, -0.82998, -0.40926],
[-1.12252, -1.99515, -0.72338]])
amino_acid_position = []
for k in range(i, i + 4, 1):
try:
x = lines[k][30:38]
y = lines[k][38:46]
z = lines[k][46:54]
amino_acid_position.append(np.asarray([x, y, z], dtype=float))
except:
sys.exit(
"Error parsing input for the following line: \n{0:s}".format(
lines[k]))
amino_acid_position = np.asarray(amino_acid_position)
#print amino_acid_position
a_acid_p = amino_acid_position - rmsd.centroid(amino_acid_position)
template1 -= rmsd.centroid(template1)
#centroid P2 into f
template2 = template2 - rmsd.centroid(template2)
#for debug
print '*************8888888888888888'
#print template1.shape#(4,3)
#print a_acid_p.shape#(5,3)
#print template2.shape#(6.3)
rot = rmsd.kabsch(template1, a_acid_p)
#pp is the rotated martix,Rotate matrix P2 unto Q
new_position = np.dot(template2, rot)
#translation the P2 into initial position after rotation
new_position += rmsd.centroid(amino_acid_position)
C = new_position.tolist()
#print new_position
#print lenthg
#for n in range(0,6,1):
position5 = ('%8s' % str(float('%.3f' % C[4][0]))) + ('%8s' % str(
float('%.3f' % C[4][1]))) + ('%8s' % str(float('%.3f' % C[4][2])))
position6 = ('%8s' % str(float('%.3f' % C[5][0]))) + ('%8s' % str(
float('%.3f' % C[5][1]))) + ('%8s' % str(float('%.3f' % C[5][2])))
print 'print the position of position 5 and position 6'
return position5, position6
def getkabschangle(Ac, Bc):
# Ac = Ac[:,0:3:2]
# Bc = Bc[:,0:3:2]
# Manipulate
A = Ac - rmsd.centroid(Ac)
B = Bc - rmsd.centroid(Bc)
U = rmsd.kabsch(A, B)
# print(U)
degvals = (rotationMatrixToEulerAngles(U))
return degvals[1]
"""
def get_rmsd(coords, reference):
"""
Calculate the rmsd of the given coordinates compared to a reference.
Input:
> coords list[np.array, np.array, ...]; coordinates to be compared
vs. the reference coordinates
> reference list[np.array, np.array, ...]; reference to compare
coordinates to
Sources: https://pypi.org/project/rmsd/
"""
# translate center of geometry to the origin for reference and
# current frame
coords -= rmsd.centroid(coords)
reference -= rmsd.centroid(reference)
return rmsd.kabsch_rmsd(coords, reference)
def calc_rmsd(coords1, coords2):
"""Calculate root-mean square deviation.
Parameters
----------
coords1 : list(list(str,int,int,int))
The coordinates for the first
molecule (conformer).
coords2 : list(list(str,int,int,int))
The coordinates for the second
molecule (conformer).
Notes
-----
The coordinates are given as follows:
atom name, x coord, y coord, z coord
Example for carbon monoxide:
[[C, 0.0, 0.0, 0.0],[O, 1.0, 0.0, 0.0]]
The distances are in Angstroms.
Returns
-------
rmsd : float
The rmsd between two molecular
geometries. This rmsd is calculated
after applying centering and a
rotation matrix (calculated via
the Kabsch algorithm).
"""
# trivial check
assert len(coords1) == len(coords2)
# first turn lists into numpy arrays (otherwise
# cannot perform some rmsd operations)
# and excise atom names (not needed for rmsd)
mol1 = np.array([[coords1[i][1], coords1[i][2], coords1[i][3]] for i in range(len(coords1))])
mol2 = np.array([[coords2[i][1], coords2[i][2], coords2[i][3]] for i in range(len(coords2))])
# first center each molecule
mol1 -= rmsd.centroid(mol1)
mol2 -= rmsd.centroid(mol2)
# calculate the rotation matrix
rot_matrix = rmsd.kabsch(mol1, mol2)
# apply the rotation matrix
mol1 = np.dot(mol1, rot_matrix)
# finally get the rmsd
return rmsd.rmsd(mol1, mol2)
def test_best_fit(A, shuffle = 1):
# Generate a random dataset
for i in range(num_tests):
print('test', i)
B = np.copy(A)
# Translate
t = np.random.rand(dim)*translation
B += t
# Rotate
R = rotation_matrix(np.random.rand(dim), np.random.rand()*rotation)
B = np.dot(R, B.T).T
# Add noise
B += np.random.randn(N, dim) * noise_sigma
# Shuffle to disrupt correspondence
if (shuffle == 1):
np.random.shuffle(B)
print("Normal squared_distance", squared_distance(A, B))
plot3d(A, B)
# Find best fit transform
A -= rmsd.centroid(A)
B -= rmsd.centroid(B)
T, R1, t1 = icp.best_fit_transform(B, A)
# Make C a homogeneous representation of B
C = np.ones((N, 4))
C[:,0:3] = B
# Transform C
C = np.dot(T, C.T).T
C = C[:, 0:3]
print("Rotated squared_distance", squared_distance(A, C))
plot3d(A, C)
#assert np.allclose(C[:,0:3], A, atol=6*noise_sigma) # T should transform B (or C) to A
#assert np.allclose(-t1, t, atol=6*noise_sigma) # t and t1 should be inverses
#assert np.allclose(R1.T, R, atol=6*noise_sigma) # R and R1 should be inverses
return
def test_quaternion_rmsd_pdb():
filename_p = "ci2_1.pdb"
filename_q = "ci2_2.pdb"
filename_p = pathlib.PurePath(RESOURCE_PATH, filename_p)
filename_q = pathlib.PurePath(RESOURCE_PATH, filename_q)
p_atoms, p_coord = rmsd.get_coordinates_pdb(filename_p)
q_atoms, q_coord = rmsd.get_coordinates_pdb(filename_q)
p_center = rmsd.centroid(p_coord)
q_center = rmsd.centroid(q_coord)
p_coord -= p_center
q_coord -= q_center
result = rmsd.quaternion_rmsd(p_coord, q_coord)
np.testing.assert_almost_equal(11.7768, result, decimal=3)
def align_coordinates(coordinates):
# Align everything to first coordinate
coordinates = copy.deepcopy(coordinates)
coordinate_ref = coordinates[0]
coordinate_ref -= rmsd.centroid(coordinate_ref)
coordinates[0] = coordinate_ref
for i, coordinate in enumerate(coordinates[1:]):
coordinate -= rmsd.centroid(coordinate)
U = rmsd.kabsch(coordinate, coordinate_ref)
coordinate = np.dot(coordinate, U)
coordinates[i] = coordinate
return coordinates
def calculate_transformation_kabsch(
src_points: np.ndarray,
dst_points: np.ndarray) -> Tuple[np.array, float]:
"""
Calculates the optimal rigid transformation from src_points to
dst_points
(regarding the least squares error)
Parameters:
-----------
src_points: array
(3,N) matrix
dst_points: array
(3,N) matrix
Returns:
-----------
rotation_matrix: array
(3,3) matrix
translation_vector: array
(3,1) matrix
rmsd_value: float
"""
assert src_points.shape == dst_points.shape
if src_points.shape[0] != 3:
raise Exception(
"The input data matrix had to be transposed in order to compute transformation."
)
src_points = src_points.transpose()
dst_points = dst_points.transpose()
src_points_centered = src_points - rmsd.centroid(src_points)
dst_points_centered = dst_points - rmsd.centroid(dst_points)
rotation_matrix = rmsd.kabsch(src_points_centered, dst_points_centered)
rmsd_value = rmsd.kabsch_rmsd(src_points_centered, dst_points_centered)
translation_vector = rmsd.centroid(dst_points) - np.matmul(
rmsd.centroid(src_points), rotation_matrix)
return create_homogenous(rotation_matrix.transpose(),
translation_vector.transpose()), rmsd_value
def myRMSDmetric(arr1, arr2):
"""
This function is built under the assumption that the space dimension is 3!!!
Requirement from sklearn radius_neighbors_graph: The callable should take two arrays as input and return one value indicating the distance between them.
Input: One row from reshaped XYZ trajectory as number of steps times nDOF
Inside: Reshape to XYZ format and apply rmsd as r=rmsd(X[i], X[j])
Output: rmsd distance
"""
nParticles = len(arr1) / 3
assert (nParticles == int(nParticles))
X1 = arr1.reshape(int(nParticles), 3)
X2 = arr2.reshape(int(nParticles), 3)
X1 = X1 - rmsd.centroid(X1)
X2 = X2 - rmsd.centroid(X2)
return rmsd.kabsch_rmsd(X1, X2)
def normalize_pts_old(pts1, pts2, win):
#print pts1
A = pts1[:, 0:2]
B = pts2[:, 0:2]
assert len(A) == len(B)
N = A.shape[0] # total points
# find centroids
centroid_A = rmsd.centroid(A)
centroid_B = rmsd.centroid(B)
# centre the points
AA = A - centroid_A
BB = B - centroid_B
#get optimal rotation matrix
R = rmsd.kabsch(AA, BB)
# Apply transformation
AA1 = np.dot(AA, R)
angle1 = math.atan2(R[1][0], R[1][1])
i=0
for pt in pts1:
pt_graphics = Point(int(pt[0]-centroid_A[0]+centroid_B[0]), int(pt[1]-centroid_A[1]+centroid_B[1]))
cir = Circle(pt_graphics, 8)
cir.setOutline('green')
cir.draw(win)
line = Line(pt_graphics, Point(pt_graphics.x+8*np.cos(-pt[2]), pt_graphics.y+8*np.sin(-pt[2])))
line.draw(win)
text = Text(pt_graphics, str(i))
i += 1
text.draw(win)
pts1[:, 0:2] = AA1 + centroid_B
# rotate the rotations
pts1[:, 2] -= angle1
return rmsd.rmsd(AA1, BB)
def test_centroid():
a1 = np.array([-19.658, 17.18, 25.163], dtype=float)
a2 = np.array([-20.573, 18.059, 25.88], dtype=float)
a3 = np.array([-22.018, 17.551, 26.0], dtype=float)
atms = np.asarray([a1, a2, a3])
centroid = rmsd.centroid(atms)
assert 3 == len(centroid)
np.testing.assert_array_almost_equal([-20.7496, 17.5966, 25.6810],
centroid,
decimal=3)
def calc_rmsd(self):
"""
Calculate normal, translated, and rotated RMSD.
"""
tmp_1, tmp_2 = self.coord_1.copy(), self.coord_2.copy()
self.rmsd_normal = rmsd.rmsd(self.coord_1, self.coord_2)
# Manipulate recenter
self.coord_1 -= rmsd.centroid(self.coord_1)
self.coord_2 -= rmsd.centroid(self.coord_2)
self.rmsd_translate = rmsd.rmsd(self.coord_1, self.coord_2)
# Rotate
rotation_matrix = rmsd.kabsch(self.coord_1, self.coord_2)
self.coord_1 = np.dot(self.coord_1, rotation_matrix)
self.rmsd_rotate = rmsd.rmsd(self.coord_1, self.coord_2)
self.coord_1, self.coord_2 = tmp_1, tmp_2
def main(self):
"""
main object to run DSR as command line program
"""
dbatoms = []
# The database content:
import atomhandling
basefilename = filename_wo_ending(self.res_file)
if not basefilename:
print('*** Illegal option ***')
sys.exit()
if len(self.reslist) == 0:
print("*** The input file is empty. Can not proceed! ***")
sys.exit()
find_atoms = atomhandling.FindAtoms(self.reslist)
rle = ResListEdit(self.reslist, find_atoms)
dsrp = DSRParser(self.reslist)
self.fragment = dsrp.fragment
restraints = self.gdb.get_restraints(self.fragment) # this is only executed once
db_residue_string = self.gdb.get_resi(self.fragment)
dbatoms = self.gdb.get_atoms(self.fragment, self.invert) # only the atoms of the dbentry as list
# the atomtypes of the dbentry as list e.g. ['C', 'N', ...]
db_atom_types = atomhandling.get_atomtypes(dbatoms)
sf = atomhandling.SfacTable(self.reslist, db_atom_types)
sfac_table = sf.set_sfac_table() # from now on this sfac table is set
resi = Resi(dsrp, db_residue_string, find_atoms)
# line where the dsr command is found in the resfile:
if dsrp.cf3_active:
from cf3fit import CF3
cf3 = CF3(rle, find_atoms, self.reslist, self.fragment, sfac_table,
basefilename, dsrp, resi, self.res_file, self.options)
if self.fragment == 'cf3':
cf3.cf3(afix='130')
if self.fragment == 'cf6':
cf3.cf3(afix='120')
if self.fragment == 'cf9':
cf3.cf9()
print('\nFinished...')
sys.exit()
# checks have to be after CF3, CF6 etc.
self.gdb.check_consistency(self.fragment)
self.gdb.check_db_atom_consistency(self.fragment)
self.gdb.check_db_restraints_consistency(self.fragment)
self.gdb.check_sadi_consistence(self.fragment)
if dsrp.occupancy:
rle.set_free_variables(dsrp.occupancy)
restraints = remove_resi(restraints)
# corrects the atom type according to the previous defined global sfac table:
dbatoms = atomhandling.set_final_db_sfac_types(db_atom_types, dbatoms, sfac_table)
if not dsrp.unit_line:
print('*** No UNIT instruction in res file found! Can not proceed! ***')
print('Inserting {} into res File.'.format(self.fragment))
if self.invert:
print('Fragment inverted.')
print('Source atoms: {}'.format(', '.join(dsrp.source)))
print('Target atoms: {}'.format(', '.join(dsrp.target)))
shx = ShelxlRefine(self.reslist, basefilename, find_atoms, self.options)
shx.backup_shx_file()
# several checks if the atoms in the dsr command line are consistent
atomhandling.check_source_target(dsrp.source, dsrp.target, dbatoms)
num = atomhandling.NumberScheme(self.reslist, dbatoms, dsrp)
# returns also the atom names if residue is active
fragment_numberscheme = num.get_fragment_number_scheme()
print('Fragment atom names: {}'.format(', '.join(fragment_numberscheme)))
dfix_head = ''
if dsrp.dfix:
restr = Restraints(self.fragment, self.gdb)
dfix_12 = restr.get_formated_12_dfixes()
dfix_13 = restr.get_formated_13_dfixes()
flats = restr.get_formated_flats()
restraints = dfix_12 + dfix_13 + flats
# ##########Not using SHELXL for fragment fit: ###########
print("--- Using fast fragment fit ---")
if self.options.target_coords:
target_coords = chunks(self.options.target_coords, 3)
else:
# {'C1': ['1.123', '0.7456', '3.245']}
target_coordinates = find_atoms.get_atomcoordinates(dsrp.target)
target_coords = [target_coordinates[key] for key in dsrp.target]
# Uppercase is important here to avoid KeyErrors in source_atoms generation
atnames = self.gdb.get_atomnames(self.fragment, uppercase=True)
source_atoms = dict(zip(atnames, self.gdb.get_coordinates(self.fragment, cartesian=True, invert=self.invert)))
# Coordinates only from the source, not the entire fragment:
source_coords = [source_atoms[x] for x in dsrp.source]
target_coords = [frac_to_cart(x, rle.get_cell()) for x in target_coords]
from rmsd import fit_fragment
# The source and target atom coordinates are fitted first. Then The complete fragment
# is rotated and translated to the target position as calculated before.
# parameter cartiesian has to be false here:
fragment_coords = self.gdb.get_coordinates(self.fragment, cartesian=False, invert=self.invert)
fitted_fragment, rmsd = fit_fragment(fragment_coords,
source_atoms=source_coords,
target_atoms=target_coords)
# Moving back to the position of the first atom to have a reference:
import numpy as np
from rmsd import centroid
# I have to make sure that I use the centroid of the correct atoms from target and source,
# otherwise the fragment is shifted to a wrong position.
# The third atom from the fragment e.g. has to be the third from the fragment to get
# the correct centroid:
center_difference = centroid(np.array(target_coords)) - \
centroid(np.array([list(fitted_fragment)[atnames.index(dsrp.source[x])]
for x in range(len(source_coords))]))
# finishing shift to correct centroid:
fitted_fragment += center_difference
# Or even lower than 0.1?
if rmsd < 0.1:
print('Fragment fit successful with RMSD of: {:8.3}'.format(rmsd))
else:
print('*** Fragment fit might have failed with RMSD of: {:8.3} ***'.format(rmsd))
fitted_fragment = [cart_to_frac(x, rle.get_cell()) for x in fitted_fragment]
afix_entry = []
e2s = Elem_2_Sfac(sfac_table)
for at, coord, atype in zip(fragment_numberscheme, fitted_fragment, db_atom_types):
sfac_num = str(e2s.elem_2_sfac(atype))
if dsrp.occupancy:
occ = float(dsrp.occupancy)
else:
occ = 11.0
afix_entry.append(isoatomstr.format(at, sfac_num, coord[0], coord[1], coord[2], occ, 0.03))
afix_entry = "\n".join(afix_entry)
new_atomnames = list(reversed(fragment_numberscheme))
same_resi = ''
if not dsrp.resiflag:
restraints = rename_restraints_atoms(new_atomnames, self.gdb.get_atomnames(self.fragment), restraints)
else:
restraints = resi.format_restraints(restraints)
# SADI\n
same_resi = ["SAME_{} {} > {}\n".format(resi.get_residue_class, new_atomnames[-1], new_atomnames[0])]
# Adds a "SAME_resiclass firstatom > lastatom" to the afix:
if not self.options.rigid_group:
restraints += same_resi
# if dsrp.resiflag: # <- Or should I do this?
restraints += ["SIMU 0.04 0.08 1"]
if not options.external_restr:
restraints = remove_duplicate_restraints(self.reslist, restraints, resi.get_residue_class)
restraints = wrap_headlines(restraints)
dfx_file_name = ''
if dsrp.part:
afix_entry = "PART {} {}\n".format(dsrp.part, dsrp.occupancy) + afix_entry + "\nPART 0"
if dsrp.resiflag:
afix_entry = 'RESI {} {}\n{}\nRESI 0'.format(resi.get_residue_class, resi.get_resinumber, afix_entry)
if self.options.rigid_group:
afix_entry = 'AFIX 9\n' + afix_entry
if options.external_restr and not self.rigid:
pname, ext = os.path.splitext(basefilename + '.dfix')
if dsrp.dfix:
dfx_file_name = pname + "_dfx" + ext
else:
dfx_file_name = pname + ext
dfx_file_name = write_dbhead_to_file(dsrp, dfx_file_name, restraints, resi.get_residue_class,
resi.get_resinumber)
if dsrp.resiflag:
restraints = 'REM Restraints for residue {}:\n+{}\n' \
.format(resi.get_residue_class, dfx_file_name)
else:
restraints = 'REM Restraints for DSR fragment:\n+{}\n' \
.format(dfx_file_name)
if self.options.rigid_group:
afix_entry += '\nAFIX 0\n'
# Adds the origin of restraints and fragment to res file:
import textwrap
source = textwrap.wrap("REM Restraints for Fragment {}, {} from: {}. "
"Please cite https://doi.org/10.1107/S1600576718004508".format(
self.fragment,
self.gdb.get_fragment_name(self.fragment),
self.gdb.get_src(self.fragment)),
width=74, subsequent_indent='REM ')
source = '\n'.join(source) + '\n'
# check if restraints already inserted:
for line in self.reslist:
try:
if line.split()[4] == self.fragment + ',':
source = ''
break
except IndexError:
continue
# + 'AFIX 0\n' before hklf seems to be not needed after shelx-2013:
self.reslist[dsrp.hklf_line - 1] = self.reslist[dsrp.hklf_line - 1] + afix_entry + '\n'
if not self.rigid:
self.reslist[dsrp.unit_line] = self.reslist[dsrp.unit_line] + source + ''.join(restraints)
# write to file:
self.rl.write_resfile(self.reslist, '.res')
if dsrp.command == 'REPLACE':
print("Replace mode active\n")
self.rl = ResList(self.res_file)
reslist = self.rl.get_res_list()
self.reslist, find_atoms = atomhandling.replace_after_fit(self.rl, reslist, resi,
fragment_numberscheme, rle.get_cell())
self.rl.write_resfile(self.reslist, '.res')
os.remove(shx.backup_file)
[2.0, 1.5]])
# Same "molecule"
B = np.array([[1.0, 1.0],
[1.0, 2.0],
[2.0, 1.5]])
B *= 1.4
# Translate
B -= 3
# Rotate
B = np.dot(B, rotation_matrix(90))
print("Normal RMSD", rmsd.rmsd(A, B))
save_plot(A, B, "plot_beginning.png")
# Manipulate
A -= rmsd.centroid(A)
B -= rmsd.centroid(B)
print("Centered RMSD", rmsd.rmsd(A, B))
save_plot(A, B, "plot_centered.png")
U = rmsd.kabsch(A, B)
A = np.dot(A, U)
print("Translated RMSD", rmsd.rmsd(A, B))
save_plot(A, B, "plot_translated.png")
