File size: 6,240 Bytes
cc7c981
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
# Written by Mohd Ibrahim
# Technical University of Munich
# Email: ibrahim.mohd@tum.de

import numpy as np
import MDAnalysis as mda
import argparse
import warnings
warnings.filterwarnings("ignore")

np.random.seed(42)  

parser = argparse.ArgumentParser(
    description="Add capping groups ACE and NME to protein termini. "
                "Remove hydrogens before using this script")
parser.add_argument('-i', dest='in_file', type=str,
                    default='protein_noh.pdb', help='pdb file')
parser.add_argument('-o', dest='out_file', type=str,
                    default='protein_noh_cap.pdb', help='output file')

args = parser.parse_args()
in_file = args.in_file
out_file = args.out_file


def create_universe(n_atoms, name, resname, positions, resids, segid):
    u_new = mda.Universe.empty(
        n_atoms=n_atoms,
        n_residues=n_atoms,
        atom_resindex=np.arange(n_atoms),
        residue_segindex=np.arange(n_atoms),
        n_segments=n_atoms,
        trajectory=True
    )
    u_new.add_TopologyAttr('name', name)
    u_new.add_TopologyAttr('resid', resids)
    u_new.add_TopologyAttr('resname', resname)
    u_new.atoms.positions = positions
    u_new.add_TopologyAttr('segid', n_atoms * [segid])
    u_new.add_TopologyAttr('chainID', n_atoms * [segid])
    return u_new


def get_nme_pos(end_residue):
    if "OXT" in end_residue.names:
        index = np.where(end_residue.names == "OXT")[0][0]
        N_position = end_residue.positions[index]
        index_c = np.where(end_residue.names == "C")[0][0]
        carbon_position = end_residue.positions[index_c]
        vector = N_position - carbon_position
        vector /= np.sqrt(sum(vector**2))
        C_position = N_position + vector * 1.36
        return N_position, C_position
    else:
        index_o = np.where(end_residue.names == "O")[0][0]
        index_ca = np.where(end_residue.names == "CA")[0][0]
        mid_point = (end_residue.positions[index_o] +
                     end_residue.positions[index_ca]) / 2
        index_c = np.where(end_residue.names == "C")[0][0]
        vector = end_residue.positions[index_c] - mid_point
        vector /= np.sqrt(sum(vector**2))
        N_position = end_residue.positions[index_c] + 1.36 * vector
        C_position = N_position + 1.36 * vector
        return N_position, C_position


def get_ace_pos(end_residue):
    index_ca = np.where(end_residue.names == "CA")[0][0]
    index_n = np.where(end_residue.names == "N")[0][0]
    vector = end_residue.positions[index_n] - end_residue.positions[index_ca]
    vector /= np.sqrt(sum(vector**2))
    C1_position = end_residue.positions[index_n] + 1.36 * vector

    xa, ya, za = end_residue.positions[index_ca]
    xg, yg, zg = C1_position

    orientation = np.array([2 * np.random.rand() - 1,
                            2 * np.random.rand() - 1,
                            2 * np.random.rand() - 1])
    nx, ny, nz = orientation / np.sqrt(sum(orientation**2))

    x1 = xg - (xa - xg) / 2 + np.sqrt(3) * (ny * (za - zg) - nz * (ya - yg)) / 2
    y1 = yg - (ya - yg) / 2 + np.sqrt(3) * (nz * (xa - xg) - nx * (za - zg)) / 2
    z1 = zg - (za - zg) / 2 + np.sqrt(3) * (nx * (ya - yg) - ny * (xa - xg)) / 2

    x2 = xg - (xa - xg) / 2 - np.sqrt(3) * (ny * (za - zg) - nz * (ya - yg)) / 2
    y2 = yg - (ya - yg) / 2 - np.sqrt(3) * (nz * (xa - xg) - nx * (za - zg)) / 2
    z2 = zg - (za - zg) / 2 - np.sqrt(3) * (nx * (ya - yg) - ny * (xa - xg)) / 2

    C2_position = np.array([x1, y1, z1])
    O_position = np.array([x2, y2, z2])

    vector = C2_position - C1_position
    vector /= np.sqrt(sum(vector**2))
    C2_position = C1_position + 1.36 * vector

    vector = O_position - C1_position
    vector /= np.sqrt(sum(vector**2))
    O_position = C1_position + 1.36 * vector

    return C1_position, C2_position, O_position


# ----------- Main processing -----------
u = mda.Universe(in_file)
res_start = 0
segment_universes = []

for seg in u.segments:
    chain = u.select_atoms(f"segid {seg.segid}")

    # ACE
    resid_c = chain.residues.resids[0]
    end_residue = u.select_atoms(f"segid {seg.segid} and resid {resid_c}")
    c1_pos, c2_pos, o_pos = get_ace_pos(end_residue)

    # keep original mapping (C, CH3, O)
    ace_names = ["C", "CH3", "O"]
    ace_positions = [c1_pos, c2_pos, o_pos]
    resid = chain.residues.resids[0]
    ace_universe = create_universe(
        n_atoms=len(ace_positions),
        name=ace_names,
        resname=len(ace_names) * ["ACE"],
        positions=ace_positions,
        resids=resid * np.ones(len(ace_names)),
        segid=chain.segids[0]
    )

    # >>> Reorder rows only: CH3, C, O <<<
    ace_universe = mda.Merge(
        ace_universe.atoms.select_atoms("name CH3"),
        ace_universe.atoms.select_atoms("name C"),
        ace_universe.atoms.select_atoms("name O")
    )

    # NME
    resid_c = chain.residues.resids[-1]
    end_residue = u.select_atoms(f"segid {seg.segid} and resid {resid_c}")
    nme_positions = get_nme_pos(end_residue)
    nme_names = ["N", "C"]
    resid = chain.residues.resids[-1] + 2
    nme_universe = create_universe(
        n_atoms=len(nme_names),
        name=nme_names,
        resname=len(nme_names) * ["NME"],
        positions=nme_positions,
        resids=resid * np.ones(len(nme_names)),
        segid=chain.segids[0]
    )

    # Remove OXT if present
    if "OXT" in end_residue.names:
        index = np.where(end_residue.names == "OXT")[0][0]
        OXT = end_residue[index]
        Chain = u.select_atoms(f"segid {seg.segid} and not index {OXT.index}")
    else:
        Chain = u.select_atoms(f"segid {seg.segid}")

    # Merge ACE, protein, NME
    u_all = mda.Merge(ace_universe.atoms, Chain, nme_universe.atoms)

    # Renumber residues
    resids_ace = [res_start + 1] * 3
    resids_pro = np.arange(resids_ace[0] + 1,
                           Chain.residues.n_residues + resids_ace[0] + 1)
    resids_nme = [resids_pro[-1] + 1] * 2
    u_all.atoms.residues.resids = np.concatenate(
        [resids_ace, resids_pro, resids_nme]
    )
    res_start = u_all.atoms.residues.resids[-1]
    segment_universes.append(u_all)

# Join and write output
all_uni = mda.Merge(*(seg.atoms for seg in segment_universes))
all_uni.atoms.write(out_file)