def process_folder(epoch,
folder,
trajName,
reportName,
output_filename,
top,
trajs_to_select=None):
if epoch is None:
allTrajs = glob.glob(os.path.join(folder, trajName))
full_reportName = os.path.join(folder, reportName)
else:
allTrajs = glob.glob(os.path.join(folder, epoch, trajName))
full_reportName = os.path.join(folder, epoch, reportName)
epoch = int(epoch)
allFiles = []
for traj in allTrajs:
trajNum = utilities.getTrajNum(traj)
if trajs_to_select is not None and trajNum not in trajs_to_select:
continue
if top is not None:
top_file = top.getTopologyFile(epoch, trajNum)
else:
top_file = None
report_file = full_reportName % trajNum
allFiles.append(
(traj, report_file, top_file, epoch, output_filename % trajNum))
return allFiles
def trajectory_and_snapshot_to_pdb(trajectory_path, snapshot, output_path,
topology_contents):
"""
Given an absolute path to a trajectory of Adaptive and a snapshot (MODEL) in xtc format, the function transform it
into a PDB format.
:param trajectory_path: Absolute path to a trajectory from Adaptive, in xtc format.
:type trajectory_path:str
:param snapshot: model of a trajectory that you want to transform.
:type snapshot: int
:param output_path: output path of the new pdb file.
:type output_path: str
:return: Creates a PDB file.
"""
# get the path where the adaptive simulation resides
topology_path_splited = trajectory_path.split(os.sep)
epoch = int(topology_path_splited[-2])
traj = adapt_tools.getTrajNum(topology_path_splited[-1])
trajectory = adapt_tools.getSnapshots(trajectory_path)
try:
single_model = trajectory[snapshot]
PDB = atomset.PDB()
PDB.initialise(single_model,
topology=topology_contents.getTopology(epoch, traj))
except IndexError:
exit(
"You are selecting the model {} for a trajectory that has {} models, please, reselect the model index "
"(starting from 0).".format(snapshot, len(trajectory)))
with open(output_path, "w") as fw:
fw.write("MODEL %4d\n" % (snapshot + 1))
fw.write(PDB.pdb)
fw.write("ENDMDL\n")
fw.write("END\n")
def fixReportsSymmetry(outputPath, resname, reschain, resnum, nativeStructure,
symmetries, topologies):
"""
Adds a new column in the report file with the RMSD that takes into account symmetries.
New reports are stored in the fixedReport_i where i is the number of the report
:param outputPath: Path where trajectories are found
:type outputPath: str
:param resname: Residue name of the ligand in the system pdb
:type resname: str
:param reschain: Chain name of the ligand in the system pdb
:type reschain: str
:param resnum: Residue number of the ligand in the system pdb
:type resnum: int
:param nativeStructure: Path to the native structure pdb
:type nativeStructure: str
:param symmetries: Dictionary containg the symmetries of the ligand
:type symmetries: dict
:param topologies: Topology object containing the set of topologies needed for the simulation
:type topologies: :py:class:`.Topology`
:raise IndexError: If original report file not found in output folder
"""
outputFilename = "fixedReport_%d" # move to constants?
trajName = "*traj*" # move to constants?
reportName = "*report_%d" # move to constants?
epoch = int(os.path.basename(outputPath))
trajs = glob.glob(os.path.join(outputPath, trajName))
nativePDB = atomset.PDB()
nativePDB.initialise(str(nativeStructure),
resname=resname,
chain=reschain,
resnum=resnum)
for traj in trajs:
trajNum = utilities.getTrajNum(traj)
rmsd = list(
utilities.getRMSD(traj,
nativePDB,
resname,
reschain,
resnum,
symmetries,
topology=topologies.getTopology(epoch, trajNum)))
try:
reportFilename = glob.glob(
os.path.join(outputPath, reportName) % trajNum)[0]
except IndexError:
raise IndexError("File %s not found in folder %s" %
(reportName % trajNum, outputPath))
rmsd.insert(0, "\tCorrected RMSD")
with open(reportFilename, "r") as f:
report = f.readlines()
outfile = open(os.path.join(outputPath, outputFilename % trajNum), "w")
for line, value in zip(report, rmsd):
outfile.write(line.rstrip("\n") + str(value) + "\n")
outfile.close()
def main(controlFile):
"""
Calculate the corrected rmsd values of conformation taking into account
molecule symmetries
:param controlFile: Control file
:type controlFile: str
"""
# Constants
folder = "."
outputFilename = "fixedReport_%d"
trajName = "*traj*"
reportName = "*report_%d"
# end constants
resname, nativeFilename, symmetries, rmsdColInReport = readControlFile(
controlFile)
nativePDB = atomset.PDB()
nativePDB.initialise(nativeFilename, resname=resname)
allFolders = os.listdir(folder)
epochs = [epoch for epoch in allFolders if epoch.isdigit()]
for epoch in epochs:
print("Epoch", epoch)
os.chdir(epoch)
allTrajs = glob.glob(trajName)
for traj in allTrajs:
rmsds = utilities.getRMSD(traj, nativePDB, resname, symmetries)
trajNum = utilities.getTrajNum(traj)
try:
reportFilename = glob.glob(reportName % trajNum)[0]
except IndexError:
raise IndexError("File %s not found in folder %s" %
(reportName % trajNum, epoch))
reportFile = np.loadtxt(reportFilename, ndmin=2)
if rmsdColInReport > 0 and rmsdColInReport < reportFile.shape[1]:
reportFile[:, rmsdColInReport] = rmsds
fixedReport = reportFile
else:
fixedReport = extendReportWithRmsd(reportFile, rmsds)
# print(fixedReport)
np.savetxt(outputFilename % trajNum, fixedReport, fmt=b'%.4f')
os.chdir("..")
def main(trajectory_name, path, n_processors, imaging):
epochs = utilities.get_epoch_folders(path)
to_process = []
pool = mp.Pool(n_processors)
trajectory_glob = trajectory_name + "_*"
for epoch in epochs:
with open(os.path.join(epoch, "topologyMapping.txt")) as f:
top_map = f.read().rstrip().split(":")
for traj in glob.glob(os.path.join(path, epoch, trajectory_glob)):
traj_num = utilities.getTrajNum(traj)
to_process.append(
(top_map[traj_num - 1], traj, epoch, traj_num, imaging))
pool.map(process_traj, to_process)
pool.close()
pool.terminate()
def get_epoch_traj_num(filename):
# assumes trajectories come from an Adaptive simulation
path, traj_name = os.path.split(filename)
try:
epoch = int(os.path.split(path)[-1])
except ValueError:
# if for some reason epoch number can't be inferred, assume first
# epoch
epoch = 0
try:
traj_num = utilities.getTrajNum(traj_name)
except ValueError:
# if for some reason trajectory number can't be inferred, assume
# first trajectory
traj_num = 1
return epoch, traj_num
def mapReference(ref, trajs, topology_content):
refPDB = atomset.PDB()
refPDB.initialise(ref, type="PROTEIN", topology=topology_content)
avgStruct = {
atomID: refPDB.atoms[atomID].getAtomCoords()
for atomID in refPDB.atoms
}
snapshotsTot = []
for traj in trajs:
snapshots = utilities.getSnapshots(traj)
traj_num = utilities.getTrajNum(traj)
epoch = int(os.path.abspath(traj).split(os.sep)[-2])
snapshots = utilities.getSnapshots(traj)
for snapshot in snapshots:
PDB = atomset.PDB()
PDB.initialise(snapshot,
type="PROTEIN",
topology=topology_content.getTopology(
epoch, traj_num))
snapshotsTot.append(PDB)
return avgStruct, snapshotsTot
def extractAvgPDB(trajs, topology_content):
nSnapshots = 0
avgStruct = {}
snapshotsTot = []
for traj in trajs:
snapshots = utilities.getSnapshots(traj)
traj_num = utilities.getTrajNum(traj)
epoch = int(os.path.abspath(traj).split(os.sep)[-2])
for snapshot in snapshots:
nSnapshots += 1
PDB = atomset.PDB()
PDB.initialise(snapshot,
type="PROTEIN",
topology=topology_content.getTopology(
epoch, traj_num))
snapshotsTot.append(PDB)
for atomID, atom in PDB.atoms.items():
if atomID in avgStruct:
avgStruct[atomID] += (atom.getAtomCoords() -
avgStruct[atomID]) / nSnapshots
else:
avgStruct[atomID] = atom.getAtomCoords()
return avgStruct, snapshotsTot