-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
260 lines (204 loc) · 8.96 KB
/
main.py
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
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
import numpy as np
import MDAnalysis as mda
from HBond import HBond
import os, time, json, sys
import networkx as nx
from itertools import groupby
from shared import rings as rg
import multiprocessing, queue
def main():
dir_source ="/home/tsingularity/Desktop/MD/SCAN/cc.ct/CC/"
filename = "traj.lammpstrj"
#----------------------必要的基本设置--------------------------
parameter = {"start": 0,
"stop": 1,
"step": 1,
"size_limit": 9,
"definition": 2,
"num_O_atoms": 129, #体系氧原子数量
}
#------------------------指定并行数----------------------------
if len(sys.argv) < 2:
print("Usage: python script.py <integer>")
sys.exit(1)
try:
parameter["np"] = int(sys.argv[1])
except ValueError:
print("The argument must be an integer.")
sys.exit(1)
st = time.time()
rings(dir_source, filename, param=parameter)
ed = time.time()
print(f"Time used: {ed-st:.2f}s")
def rings(dir_source, filename, param=None):
#-------------------------氢键定义---------------------------------
u = mda.Universe(os.path.join(dir_source, filename), topology_format="LAMMPSDUMP", format="LAMMPSDUMP")
hbonds_don = HBond(
universe=u,
donors_sel='index 126 127 128',
hydrogens_sel='type 2',
acceptors_sel='index 0:125',
d_h_cutoff=1.3,
d_a_cutoff=3.0,
h_d_a_angle_cutoff=25,
update_selections=False)
hbonds_acc = HBond(
universe=u,
donors_sel='index 0:125',
hydrogens_sel='type 2',
acceptors_sel='index 126 127 128',
d_h_cutoff=1.3,
d_a_cutoff=3.0,
h_d_a_angle_cutoff=25,
update_selections=False)
hbonds_water = HBond(
universe=u,
donors_sel='index 0:125',
hydrogens_sel='type 2',
acceptors_sel='index 0:125',
d_h_cutoff=1.3,
d_a_cutoff=3.5,
h_d_a_angle_cutoff=30,
update_selections=False)
#------------------------环统计-----------------------------------
hbonds_don.run(start=param["start"], stop=param["stop"], step=param["step"], verbose=True)
hbonds_acc.run(start=param["start"], stop=param["stop"], step=param["step"], verbose=True)
hbonds_water.run(start=param["start"], stop=param["stop"], step=param["step"], verbose=True)
hbonds = np.vstack([hbonds_don.results['hbonds'], hbonds_acc.results['hbonds'], hbonds_water.results['hbonds']])
result_dict = mpi_run(rings_stats, param, u, hbonds)
dict_rings = dict(sorted(result_dict.items()))
#-----------------------结果存为json-----------------------------
save_json(dir_source, dict_rings)
def mpi_run(func, param, u, hbonds):
num_frames = (param["stop"] - param["start"]) // param["step"]
num_processes = param["np"] # 获取可用 CPU 核心数
# 计算每个进程需要处理的帧数
frames_per_process = num_frames // num_processes
remaining_frames = num_frames % num_processes
# 用于结果收集
manager = multiprocessing.Manager()
result_dict = manager.dict()
# 创建一个线程安全的队列来存储位置和步骤信息
step_queues = [queue.Queue() for i in range(num_processes)]
positions_queues = [queue.Queue() for i in range(num_processes)]
# 将帧的索引放入队列中
start_idx = param["start"]
for i in range(num_processes):
end_idx = start_idx + frames_per_process
if i < remaining_frames:
end_idx += 1 # 将多余的帧分配给前面的进程
# 将位置和步骤信息放入队列中
for step, ts in zip(range(start_idx, end_idx, param["step"]), u.trajectory[start_idx:end_idx:param["step"]]):
positions_queues[i].put(u.select_atoms("all").positions)
step_queues[i].put(step)
start_idx = end_idx
processes = []
# 创建多个进程来处理数据
for _ in range(num_processes):
process = multiprocessing.Process(target=func,
args=(step_queues[_],
positions_queues[_],
hbonds, u,
result_dict,
param["size_limit"],
param["definition"],
param["num_O_atoms"]))
process.start()
processes.append(process)
# 等待所有进程完成
for process in processes:
process.join()
return result_dict
def rings_stats(step_queue, positions_queue,
hbonds, u, result_dict,
size_limit, definition,
num_O_atoms) -> None:
while True:
try:
step = step_queue.get_nowait()
positions = positions_queue.get_nowait()
except queue.Empty:
break
else:
frame = hbonds[:, 0] == step
bonds = hbonds[frame, 1:4:2].astype(np.uint16)
rings_single = networkx_ring(bonds, positions, u.dimensions[:3], size_limit=size_limit, definition=definition, num_O_atoms=num_O_atoms)
result_dict[step] = rings_single
print(f"Frame: {step}\n----------------------------\n")
for k, v in rings_single.items():
print(f"size: {k}, num: {len(v)}\n")
print(v)
print("\n")
def networkx_ring(bonds, pos, celldm,
size_limit=6, definition=2,
num_O_atoms=128) -> dict:
if size_limit <= 11: # 选择C++或Python搜寻环函数
Cpp_findrings = True
else:
Cpp_findrings = False
Cpp_validrings = False # 选择C++或Python验证环函数，两者相差不多
if Cpp_findrings:
# 小环快(小于等于11元环时)
# Cpp深度优先算法（找环非常全，但很多是大环包含小环，这些是被我们排除的）
pre_rings = rg.findrings(bonds, num_O_atoms, size_limit)
else: # 图论约翰逊算法（大于11元环时快）
G = nx.Graph()
for bond in bonds:
donor_oxygen = bond[0]
acceptor_oxygen = bond[1]
G.add_edge(donor_oxygen, acceptor_oxygen)
pre_rings = sorted(nx.simple_cycles(G, length_bound=size_limit))
sum_rings = {size: np.empty((0, size), dtype=np.uint16) for size in range(3, size_limit + 1)}
groups = groupby(sorted(pre_rings, key=len), len)
unsum_cycles = {k: np.array(list(v), dtype=np.uint16) for k, v in groups} #未考虑周期性的环/链
for k, v in unsum_cycles.items(): #排除链
for ring in v:
if sumUp(pos, ring, celldm): # 周期性边界条件
sum_rings[k] = np.append(sum_rings[k], np.atleast_2d(ring), axis=0)
# 验证环
if Cpp_validrings: # validrings(sum_rings, definition) 纯C++实现的环验证函数，定义见shared/validring.h
rings_single = rg.validrings(sum_rings, definition)
else: # 与validing等价的python程序，性能方面相差无几，性能依赖于new_bond函数的性能，但已经用C++优化
[rings_single, rings_single_all] = [{size: np.empty((0, size), dtype=np.uint16) for size in range(3, size_limit + 1)} for i in range(2)]
for k, v in sum_rings.items():
tmp = []
for ring in v:
if rg.new_bond(ring, rings_single, rings_single_all, definition):
tmp.append(ring)
rings_single_all[k] = np.append(rings_single_all[k], v, axis=0)
if len(tmp) == 0:
continue
rings_single[k] = np.append(rings_single[k], np.array(tmp), axis=0)
return rings_single
def sumUp(pos, ring, celldm):
ring1 = ring.copy()
ring2 = np.roll(ring1, -1)
ds = [pos[i1, :] - pos[i2, :] for i1, i2 in zip(ring1, ring2)]
ds = np.array(ds)
bool_xp = ds[:, 0] < -0.5 * celldm[0]
bool_xn = ds[:, 0] > 0.5 * celldm[0]
bool_yp = ds[:, 1] < -0.5 * celldm[1]
bool_yn = ds[:, 1] > 0.5 * celldm[1]
bool_zp = ds[:, 2] < -0.5 * celldm[2]
bool_zn = ds[:, 2] > 0.5 * celldm[2]
for i, mask in zip(range(3), [bool_xp, bool_yp, bool_zp]):
ds[mask, i] += celldm[i]
for j, mask in zip(range(3), [bool_xn, bool_yn, bool_zn]):
ds[mask, j] -= celldm[j]
if np.max([np.abs(np.sum(ds[:, 0])), np.abs(np.sum(ds[:, 1])), np.abs(np.sum(ds[:, 2]))]) > 1.0e-3:
return False
else:
return True
def save_json(dir_source, dict_rings):
def ndarray2list(arr):
if isinstance(arr, np.ndarray):
return arr.tolist()
else:
return arr
json_data = json.dumps(dict_rings, default=ndarray2list, indent=0)
with open(dir_source+'rings.json', 'w') as fp:
fp.write(json_data)
if __name__ == "__main__":
import warnings
warnings.filterwarnings("ignore")
main()