def main():
args = _args().parse_args()
top = files_io.GROMACSTopologyFile(args.top)
top.read()
# We need graph to get the connected components.
g = nx.Graph()
g.add_nodes_from(top.atoms.keys())
g.add_edges_from(top.bonds.keys())
print('Number of nodes: {}'.format(g.number_of_nodes()))
print('Number of edges: {}'.format(g.number_of_edges()))
connected_components = list(nx.connected_component_subgraphs(g))
print('Number of chains: {}'.format(len(connected_components)))
chain_idx = 0
chain_names = []
chain_size = []
with open(args.output_index, 'w') as out_index:
for s in connected_components:
if s.number_of_nodes() > 3: # Do not put monomers into the groups
nodes = sorted(s.nodes())
out_index.write('[ chain_{} ]\n'.format(chain_idx))
chain_names.append('chain_{}'.format(chain_idx))
out_index.write(' '.join(map(str, nodes)))
out_index.write('\n\n')
chain_size.append(s.number_of_nodes())
chain_idx += 1
cmds = args.polystat_cmd.split() + [
'-f', args.trj, '-n', args.output_index, '-b', '5000'
]
output_file = '{}_ee_rg.csv'.format(args.output_prefix)
with open(output_file, 'w') as outd:
outd.write('# chidx ee rg n\n')
p = mp.Pool(args.nt)
_compute = functools.partial(compute, cmds)
out_data = p.map(_compute, chain_names)
for chidx, data in enumerate(out_data):
if data[0]:
try:
re_findall = re.findall(r'[0-9]+\.[0-9]+', data[0])
ee, rg = map(float, re_findall)
outd.write('{} {:.4f} {:.4f} {:.4f}\n'.format(
chidx, ee, rg, chain_size[chidx]))
except ValueError:
print(data, re_findall)
print('Saved {}'.format(output_file))
def main():
args = _args().parse_args()
top = files_io.GROMACSTopologyFile(args.in_top)
top.read()
water_molecules = [x for x in sorted(top.atoms) if top.atoms[x].chain_name == args.mol_name]
atom_ids = [(water_molecules[i:i+args.mol_size], mol_id)
for mol_id, i in enumerate(range(0, len(water_molecules), args.mol_size))]
p = mp.Pool(args.nt)
_calc_msd = functools.partial(calc_msd, args.trj, args.output_prefix)
mol_msd_data = p.map(_calc_msd, atom_ids)
np.savetxt('{}{}'.format(args.output_prefix, args.output), mol_msd_data, header='mol_id D +/-')
def main():
args = _args().parse_args()
top = files_io.GROMACSTopologyFile(args.top)
top.read()
# We need graph to get the connected components.
g = nx.Graph()
g.add_nodes_from(top.atoms.keys())
g.add_edges_from(top.bonds.keys())
print('Number of nodes: {}'.format(g.number_of_nodes()))
print('Number of edges: {}'.format(g.number_of_edges()))
connected_components = list(nx.connected_component_subgraphs(g))
print('Number of chains: {}'.format(len(connected_components)))
chain_idx = 0
with open(args.output_index, 'w') as out_index:
for s in connected_components:
if s.number_of_nodes() > 3: # Do not put monomers into the groups
nodes = sorted(s.nodes())
out_index.write('[ chain_{} ]\n'.format(chain_idx))
out_index.write(' '.join(map(str, nodes)))
out_index.write('\n\n')
chain_idx += 1
def main():
args = _args()
top = args.top
conf = args.conf
h5 = args.h5
top = files_io.GROMACSTopologyFile(top)
top.read()
conf = files_io.GROFile(conf)
conf.read()
h5 = h5py.File(h5, 'r')
species2typename = {
0: 'D',
1: 'A',
2: 'B',
3: 'C',
4: 'E',
5: 'W',
6: 'Z',
7: 'X'
}
species = h5['/particles/atoms/species/value'][-1]
chain_names = {
0: 'DIO',
4: 'DIO',
1: 'TER',
2: 'TER',
3: 'TER',
5: 'H2O',
6: 'H2O',
7: 'DUM'
}
names = {
'DIO': {
('D', ): ['D1'],
('E', ): ['E1'],
},
'TER': {
('A', 'B', 'A'): ['A1', 'B1', 'A2'],
('A', 'B', 'C'): ['A1', 'B1', 'Q2'],
('C', 'B', 'A'): ['Q1', 'B1', 'A2'],
('C', 'B', 'C'): ['Q1', 'B1', 'Q2']
},
'H2O': {
('W', ): ['W1'],
('Z', ): ['Z1']
}
}
# First set correct type
last_chain_idx = 0
chain_id = 1
old2new = {}
atidx = 1
new_atoms = {}
conf_new_atoms = {}
for at_id, sp in enumerate(species, 1):
if at_id not in top.atoms:
continue
old2new[at_id] = atidx
at_data = copy.copy(top.atoms[at_id])
new_atoms[atidx] = at_data
at_data.atom_type = species2typename[sp]
at_data.atom_id = atidx
if top.atoms[at_id].chain_idx != last_chain_idx:
last_chain_idx = top.atoms[at_id].chain_idx
at_data.chain_idx = chain_id
chain_id += 1
at_data.chain_name = chain_names[sp]
conf_new_atoms[atidx] = conf.atoms[at_id]._replace(
chain_name=chain_names[sp], chain_idx=chain_id, atom_id=atidx)
atidx += 1
atom_list = [new_atoms[x] for x in sorted(new_atoms)]
# Set names
at_id = 0
total_num = len(atom_list)
while at_id < total_num:
at_data = atom_list[at_id]
name_seq = names[at_data.chain_name]
window_size = len(name_seq.keys()[0])
at_type_seq = tuple(x.atom_type
for x in atom_list[at_id:at_id + window_size])
for i, x in enumerate(atom_list[at_id:at_id + window_size]):
x.name = name_seq[at_type_seq][i]
new_atoms[x.atom_id].atom_name = x.name
conf_new_atoms[x.atom_id] = conf_new_atoms[x.atom_id]._replace(
name=x.name)
at_id += window_size
top.atoms = new_atoms
conf.atoms = conf_new_atoms
top.write('new_{}'.format(sys.argv[1]), force=True)
conf.write('new_{}'.format(sys.argv[2]), force=True)
def main():
args = _args().parse_args()
top = files_io.GROMACSTopologyFile(args.top)
top.read()
# Remove sections
top.atomstate = {}
top.nonbonded_params = {}
# Add missing bondtypes
if 'E' not in top.bondtypes:
top.bondtypes['E'] = {'C': {'func': 1, 'params': ['0.256395', '27244.6']}}
if 'C' not in top.angletypes:
top.angletypes['C'] = {}
if 'B' not in top.angletypes['C']:
top.angletypes['C']['B'] = {}
top.angletypes['C']['B']['C'] = {'func': 8, 'params': ['2', '1.0']}
top.angletypes[('C', 'B', 'C')] = {'func': 8, 'params': ['2', '1.0']}
if 'E' not in top.dihedraltypes:
top.dihedraltypes['E'] = {}
if 'C' not in top.dihedraltypes['E']:
top.dihedraltypes['E']['C'] = {}
if 'B' not in top.dihedraltypes['E']['C']:
top.dihedraltypes['E']['C']['B'] = {}
top.dihedraltypes['E']['C']['B']['C'] = {'func': 8, 'params': ['0', '1.0']}
top.dihedraltypes[('E', 'C', 'B', 'C')] = {'func': 8, 'params': ['0', '1.0']}
for k in top.angles:
t = tuple([top.atoms[x].atom_type for x in k])
angletype = top.angletypes[t]
if not ' '.join(top.angles[k]).split(';')[0].strip():
top.angles[k] = [str(angletype['func'])] + angletype['params'] + top.angles[k]
for k in top.dihedrals:
t = tuple([top.atoms[x].atom_type for x in k])
dihedraltype = top.dihedraltypes[t]
if not ' '.join(top.dihedrals[k]).split(';')[0].strip():
top.dihedrals[k] = [str(dihedraltype['func'])] + dihedraltype['params'] + top.dihedrals[k]
conf = files_io.GROFile(args.conf)
conf.read()
h5 = h5py.File(args.h5, 'r')
species2typename = {0: 'D', 1: 'A', 2: 'B', 3: 'C', 4: 'E', 5: 'W', 6: 'Z', 7: 'X'}
species = h5['/particles/atoms/species/value'][-1]
chain_names = {
0: 'DIO',
4: 'DIO',
1: 'TER',
2: 'TER',
3: 'TER',
5: 'H2O',
6: 'H2O',
7: 'DUM'
}
names = {
'DIO': {('D',): ['D1'],
('E',): ['E1'],
},
'TER': {('A', 'B', 'A'): ['A1', 'B1', 'A2'],
('A', 'B', 'C'): ['A1', 'B1', 'C2'],
('C', 'B', 'A'): ['C1', 'B1', 'A2'],
('C', 'B', 'C'): ['C1', 'B1', 'C2']
},
'H2O': {('W', ): ['W1'],
('Z', ): ['Z1']
}
}
# First set correct type
last_chain_idx = 0
chain_id = 1
old2new = {}
atidx = 1
new_atoms = {}
conf_new_atoms = {}
for at_id, sp in enumerate(species, 1):
if at_id not in top.atoms:
continue
old2new[at_id] = atidx
at_data = copy.copy(top.atoms[at_id])
new_atoms[atidx] = at_data
at_data.atom_type = species2typename[sp]
at_data.atom_id = atidx
if top.atoms[at_id].chain_idx != last_chain_idx:
last_chain_idx = top.atoms[at_id].chain_idx
at_data.chain_idx = chain_id
chain_id += 1
at_data.chain_name = chain_names[sp]
conf_new_atoms[atidx] = conf.atoms[at_id]._replace(
chain_name=chain_names[sp], chain_idx=chain_id, atom_id=atidx)
atidx += 1
atom_list = [new_atoms[x] for x in sorted(new_atoms)]
# Set names
at_id = 0
total_num = len(atom_list)
while at_id < total_num:
at_data = atom_list[at_id]
name_seq = names[at_data.chain_name]
window_size = len(name_seq.keys()[0])
at_type_seq = tuple(x.atom_type for x in atom_list[at_id:at_id+window_size])
for i, x in enumerate(atom_list[at_id:at_id+window_size]):
x.name = name_seq[at_type_seq][i]
new_atoms[x.atom_id].atom_name = x.name
conf_new_atoms[x.atom_id] = conf_new_atoms[x.atom_id]._replace(name=x.name)
at_id += window_size
top.atoms = new_atoms
conf.atoms = conf_new_atoms
# Replace atom name C to Q (keep the type) this is because for backmapping all atoms has to be
# unique
for at_data in top.atoms.values():
if at_data.name.startswith('C'):
at_data.name = at_data.name.replace('C', 'Q')
for at_id, at_data in conf.atoms.items():
if at_data.name.startswith('C'):
conf.atoms[at_id] = at_data._replace(name=at_data.name.replace('C', 'Q'))
top.write(args.out_top, force=True)
conf.write(args.out_conf, force=True)
from md_libs import files_io
import sys
import networkx as nx
cg_top = files_io.GROMACSTopologyFile(sys.argv[1])
at_top = files_io.GROMACSTopologyFile(sys.argv[2])
cg_top.read()
at_top.read()
g = nx.Graph()
for b1, b2 in cg_top.bonds:
g.add_edge(b1, b2)
cg_degrees = g.degree()
new_res_names = {}
for at_id in cg_top.atoms:
at_data = cg_top.atoms[at_id]
if at_id in cg_degrees:
if at_data.chain_name == 'DIO':
if cg_degrees[at_id] == 1:
new_res_names[at_data.chain_idx] = 'EI1'
elif cg_degrees[at_id] == 2:
new_res_names[at_data.chain_idx] = 'EI2'
ter_atoms = range(1001, 4000, 3)
for t in ter_atoms:
ts = t, t + 1, t + 2
ts_names = tuple(map(lambda x: x.name, map(cg_top.atoms.get, ts)))
seq2name = {
('Q1', 'B1', 'Q2'): 'TE2',
('PA', 'PL', 'RB', 'PB'): 'N1L2',
('PA', 'PL', 'RB', 'RB'): 'N1D', # [3,5,7,7]
('RA', 'PL', 'PB', 'PB'): 'N2T', # [6,5,4,4]
('RA', 'PL', 'RB', 'PB'): 'N2L1', # [6,5,4,7] or [6,5,7,4]
('RA', 'PL', 'PB', 'RB'): 'N2L2', # [6,5,4,7] or [6,5,7,4]
('RA', 'PL', 'RB', 'RB'): 'N2D' # [6,5,7,7]
}
window_size = [len(x) for x in res_type2name]
if len(window_size) != window_size.count(window_size[0]):
raise RuntimeError('Wrong res_type2name, window size has to be the same')
print('Window size: {}'.format(window_size))
window_size = window_size[0]
args = parser.parse_args()
in_top = files_io.GROMACSTopologyFile(args.in_top)
in_top.read()
at_ids = sorted(in_top.atoms.keys())
for i in range(0, len(at_ids), window_size):
chunk_at_ids = at_ids[i:i+window_size]
chunk_types = tuple(in_top.atoms[x].atom_type for x in chunk_at_ids)
new_res_name = res_type2name[chunk_types]
print((chunk_types, new_res_name))
for at_id in chunk_at_ids:
in_top.atoms[at_id].chain_name = new_res_name
in_top.write(args.out_top)