def batch_processing(op):
# walk through every subdir, find all pdb files
import os
extension = '.pdb'
pdb_path = {}
for root, sub_folders, files in os.walk(op['pdb_dir']):
for file_t in files:
if not file_t.endswith(extension): continue
pdb_id = file_t[:len(file_t) - len(extension)]
assert (pdb_id + extension) == file_t
assert pdb_id not in pdb_path # the pdb_id must be unique
pdb_path[pdb_id] = os.path.join(root, file_t)
if 'pdb_id_selected' in op:
pdb_path = {
_: pdb_path[_]
for _ in (set(pdb_path.keys()) & set(op['pdb_id_selected']))
}
print 'generating maps for ', len(pdb_path), 'structures'
import copy
ts = {}
for pdb_id in pdb_path:
for spacing in op['spacing_s']:
for resolution in op['resolution_s']:
op_t = copy.deepcopy(op)
op_t['pdb_id'] = pdb_id
op_t['pdb_file'] = pdb_path[pdb_id]
assert 'resolution' not in op_t
op_t['resolution'] = resolution
assert 'spacing' not in op_t
op_t['spacing'] = spacing
ts[uuid.uuid4()] = {'func': SP.convert, 'kwargs': {'op': op_t}}
import aitom.parallel.multiprocessing.util as TPMU
cre_s = TPMU.run_batch(ts)
re = {}
for cre in cre_s:
pdb_id = cre['result']['pdb_id']
resolution = cre['result']['resolution']
spacing = cre['result']['spacing']
if pdb_id not in re: re[pdb_id] = {}
if spacing not in re[pdb_id]: re[pdb_id][spacing] = {}
assert resolution not in re[pdb_id][spacing]
re[pdb_id][spacing][resolution] = cre['result']
return re
def batch_processing(op):
"""
automatically scan pdb files and convert them to density maps, and save in to a pickle file.
This is done in parallel Alternatively, save in matlab format that is same as Bsoft.pdb2em_batch_convert_test().
"""
import os
# walk through every subdir, find all pdb files
extension = '.pdb'
pdb_path = {}
for root, sub_folders, files in os.walk(op['pdb_dir']):
for file_t in files:
if not file_t.endswith(extension):
continue
pdb_id = file_t[:len(file_t) - len(extension)]
assert (pdb_id + extension) == file_t
assert pdb_id not in pdb_path # the pdb_id must be unique
pdb_path[pdb_id] = os.path.join(root, file_t)
if 'pdb_id_selected' in op:
pdb_path = {
_: pdb_path[_]
for _ in (set(pdb_path.keys()) & set(op['pdb_id_selected']))
}
print('generating maps for ', len(pdb_path), 'structures')
import copy
ts = {}
for pdb_id in pdb_path:
for spacing in op['spacing_s']:
for resolution in op['resolution_s']:
op_t = copy.deepcopy(op)
op_t['pdb_id'] = pdb_id
op_t['pdb_file'] = pdb_path[pdb_id]
assert 'resolution' not in op_t
op_t['resolution'] = resolution
assert 'spacing' not in op_t
op_t['spacing'] = spacing
ts[uuid.uuid4()] = {'func': SP.convert, 'kwargs': {'op': op_t}}
import aitom.parallel.multiprocessing.util as TPMU
cre_s = TPMU.run_batch(ts, worker_num=0, verbose=True)
re = {}
for cre in cre_s:
#print(cre)
pdb_id = cre['result']['pdb_id']
resolution = cre['result']['resolution']
spacing = cre['result']['spacing']
if pdb_id not in re:
re[pdb_id] = {}
if spacing not in re[pdb_id]:
re[pdb_id][spacing] = {}
assert resolution not in re[pdb_id][spacing]
re[pdb_id][spacing][resolution] = cre['result']
return re