def cpmgfit_input(dir=None, binary='cpmgfit', spin_id=None, force=False):
"""Create the CPMGFit input files.
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into a directory named after the dispersion model.
@type dir: str or None
@keyword binary: The name of the CPMGFit binary file. This can include the path to the binary.
@type binary: str
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError("The relaxation dispersion experiment type has not been specified.")
# Test if the model has been set.
if not hasattr(cdp, 'model_type'):
raise RelaxError("The relaxation dispersion model has not been specified.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# The 'run.sh' script.
batch = open_write_file('batch_run.sh', dir, force)
batch.write("#! /bin/sh\n\n")
# Generate the input files for each spin.
for spin, spin_id in spin_loop(return_id=True, skip_desel=True):
# Translate the model.
function = translate_model(spin.model)
# Create the input file.
file_in = create_spin_input(function=function, spin=spin, spin_id=spin_id, dir=dir)
# The output file name.
file_out = spin_file_name(spin_id=spin_id, output=True)
# Add the file to the batch script.
batch.write("%s -grid -xmgr -f %s | tee %s\n" % (binary, file_in, file_out))
# Close the batch script, then make it executable.
batch.close()
if dir:
chmod(dir + sep + 'batch_run.sh', S_IRWXU|S_IRGRP|S_IROTH)
else:
chmod('batch_run.sh', S_IRWXU|S_IRGRP|S_IROTH)
def create(algor='LM', dir=None, force=False):
"""Create the Dasha script file 'dasha_script' for controlling the program.
@keyword algor: The optimisation algorithm to use. This can be the Levenberg-Marquardt algorithm 'LM' or the Newton-Raphson algorithm 'NR'.
@type algor: str
@keyword dir: The optional directory to place the script into.
@type dir: str or None
@keyword force: A flag which if True will cause any pre-existing file to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Determine the parameter set.
model_type = determine_model_type()
# Test if diffusion tensor data for the data_pipe exists.
if model_type != 'local_tm' and not hasattr(cdp, 'diff_tensor'):
raise RelaxNoTensorError('diffusion')
# Test if the PDB file has been loaded (for the spheroid and ellipsoid).
if model_type != 'local_tm' and cdp.diff_tensor.type != 'sphere' and not hasattr(
cdp, 'structure'):
raise RelaxNoPdbError
# Test the optimisation algorithm.
if algor not in ['LM', 'NR']:
raise RelaxError(
"The Dasha optimisation algorithm '%s' is unknown, it should either be 'LM' or 'NR'."
% algor)
# Deselect certain spins.
__deselect_spins()
# Directory creation.
if dir == None:
dir = pipes.cdp_name()
mkdir_nofail(dir, verbosity=0)
# Calculate the angle alpha of the XH vector in the spheroid diffusion frame.
if cdp.diff_tensor.type == 'spheroid':
angles.spheroid_frame()
# Calculate the angles theta and phi of the XH vector in the ellipsoid diffusion frame.
elif cdp.diff_tensor.type == 'ellipsoid':
angles.ellipsoid_frame()
# The 'dasha_script' file.
script = open_write_file(file_name='dasha_script', dir=dir, force=force)
create_script(script, model_type, algor)
script.close()
def cpmgfit_input(dir=None, binary='cpmgfit', spin_id=None, force=False):
"""Create the CPMGFit input files.
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into a directory named after the dispersion model.
@type dir: str or None
@keyword binary: The name of the CPMGFit binary file. This can include the path to the binary.
@type binary: str
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError("The relaxation dispersion experiment type has not been specified.")
# Test if the model has been set.
if not hasattr(cdp, 'model_type'):
raise RelaxError("The relaxation dispersion model has not been specified.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# The 'run.sh' script.
batch = open_write_file('batch_run.sh', dir, force)
batch.write("#! /bin/sh\n\n")
# Generate the input files for each spin.
for spin, spin_id in spin_loop(return_id=True, skip_desel=True):
# Translate the model.
function = translate_model(spin.model)
# Create the input file.
file_in = create_spin_input(function=function, spin=spin, spin_id=spin_id, dir=dir)
# The output file name.
file_out = spin_file_name(spin_id=spin_id, output=True)
# Add the file to the batch script.
batch.write("%s -grid -xmgr -f %s | tee %s\n" % (binary, file_in, file_out))
# Close the batch script, then make it executable.
batch.close()
if dir:
chmod(dir + sep + 'batch_run.sh', S_IRWXU|S_IRGRP|S_IROTH)
else:
chmod('batch_run.sh', S_IRWXU|S_IRGRP|S_IROTH)
def create(algor='LM', dir=None, force=False):
"""Create the Dasha script file 'dasha_script' for controlling the program.
@keyword algor: The optimisation algorithm to use. This can be the Levenberg-Marquardt algorithm 'LM' or the Newton-Raphson algorithm 'NR'.
@type algor: str
@keyword dir: The optional directory to place the script into.
@type dir: str or None
@keyword force: A flag which if True will cause any pre-existing file to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Determine the parameter set.
model_type = determine_model_type()
# Test if diffusion tensor data for the data_pipe exists.
if model_type != 'local_tm' and not hasattr(cdp, 'diff_tensor'):
raise RelaxNoTensorError('diffusion')
# Test if the PDB file has been loaded (for the spheroid and ellipsoid).
if model_type != 'local_tm' and cdp.diff_tensor.type != 'sphere' and not hasattr(cdp, 'structure'):
raise RelaxNoPdbError
# Test the optimisation algorithm.
if algor not in ['LM', 'NR']:
raise RelaxError("The Dasha optimisation algorithm '%s' is unknown, it should either be 'LM' or 'NR'." % algor)
# Deselect certain spins.
__deselect_spins()
# Directory creation.
if dir == None:
dir = pipes.cdp_name()
mkdir_nofail(dir, verbosity=0)
# Calculate the angle alpha of the XH vector in the spheroid diffusion frame.
if cdp.diff_tensor.type == 'spheroid':
angles.spheroid_frame()
# Calculate the angles theta and phi of the XH vector in the ellipsoid diffusion frame.
elif cdp.diff_tensor.type == 'ellipsoid':
angles.ellipsoid_frame()
# The 'dasha_script' file.
script = open_write_file(file_name='dasha_script', dir=dir, force=force)
create_script(script, model_type, algor)
script.close()
def nessy_input(file='save.NESSY', dir=None, spin_id=None, force=False):
"""Create the NESSY input files.
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
@keyword binary: The name of the CPMGFit binary file. This can include the path to the binary.
@type binary: str
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError(
"The relaxation dispersion experiment type has not been specified."
)
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# The save file.
save_file = open_write_file(file, dir, force)
# Create the NESSY data object.
data = Nessy_data(spin_id=spin_id)
# Create the NESSY file.
write_program_setup(file=save_file, dir=dir, data=data)
# Loop over the experiments.
for ei in range(data.num_exp):
write_sequence(file=save_file, data=data, ei=ei)
write_cpmg_datasets(file=save_file, data=data, ei=ei)
write_spinlock_datasets(file=save_file, data=data, ei=ei)
write_experiment_setup(file=save_file, data=data, ei=ei)
def sample(self):
"""Generate the ensembles by random sampling of the snapshots."""
# Create the directory for the ensembles, if needed.
mkdir_nofail(dir=self.results_dir + sep + "ensembles")
# Loop over the configurations.
for conf_index in range(len(self.configs)):
# Loop over each ensemble.
for ens in range(self.num_ens):
# Random sampling.
rand = []
for j in range(self.num_models):
rand.append(
randint(self.snapshot_min[conf_index],
self.snapshot_max[conf_index]))
# Print out.
print("Generating ensemble %s%s from structures %s." %
(self.configs[conf_index], ens, rand))
# The file name.
file_name = "ensembles" + sep + self.configs[
conf_index] + repr(ens) + ".pdb"
# Open the output file.
out = open(self.results_dir + sep + file_name, 'w')
# Header.
out.write("REM Structures: " + repr(rand) + "\n")
# Concatenation the files.
for j in range(self.num_models):
# The random file.
rand_name = self.snapshot_dir[
conf_index] + sep + self.configs[conf_index] + repr(
rand[j]) + ".pdb"
# Append the file.
out.write(open(rand_name).read())
# Close the file.
out.close()
def nessy_input(file='save.NESSY', dir=None, spin_id=None, force=False):
"""Create the NESSY input files.
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
@keyword binary: The name of the CPMGFit binary file. This can include the path to the binary.
@type binary: str
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
"""
# Test if the current pipe exists.
pipes.test()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError("The relaxation dispersion experiment type has not been specified.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# The save file.
save_file = open_write_file(file, dir, force)
# Create the NESSY data object.
data = Nessy_data(spin_id=spin_id)
# Create the NESSY file.
write_program_setup(file=save_file, dir=dir, data=data)
# Loop over the experiments.
for ei in range(data.num_exp):
write_sequence(file=save_file, data=data, ei=ei)
write_cpmg_datasets(file=save_file, data=data, ei=ei)
write_spinlock_datasets(file=save_file, data=data, ei=ei)
write_experiment_setup(file=save_file, data=data, ei=ei)
def catia_input(file='Fit.catia', dir=None, output_dir='output', force=False):
"""Create the CATIA input files.
@keyword file: The main CATIA execution file.
@type file: str
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
@keyword output_dir: The CATIA output directory, located within the directory specified by the dir argument. This directory will be created.
@type output_dir: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: Bool
"""
# Data checks.
check_pipe()
check_mol_res_spin_data()
check_spectra_id_setup()
check_model_type()
# Check that this is CPMG data.
for id in cdp.spectrum_ids:
if cdp.exp_type[id] != 'SQ CPMG':
raise RelaxError("Only CPMG type data is supported.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# Create the R2eff files.
write_r2eff_files(input_dir='input_r2eff', base_dir=dir, force=force)
# Create the parameter files.
write_param_files(global_file="ParamGlobal.inp",
set_file="ParamSet1.inp",
dir=dir,
force=force)
# Create the main execution file.
write_main_file(file=file, dir=dir, output_dir=output_dir, force=force)
# Create the output directory as needed by CATIA (it does not create it itself).
mkdir_nofail(dir + sep + output_dir, verbosity=0)
def catia_input(file='Fit.catia', dir=None, output_dir='output', force=False):
"""Create the CATIA input files.
@keyword file: The main CATIA execution file.
@type file: str
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
@keyword output_dir: The CATIA output directory, located within the directory specified by the dir argument. This directory will be created.
@type output_dir: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: Bool
"""
# Data checks.
pipes.test()
check_mol_res_spin_data()
check_spectra_id_setup()
check_model_type()
# Check that this is CPMG data.
for id in cdp.spectrum_ids:
if cdp.exp_type[id] != 'SQ CPMG':
raise RelaxError("Only CPMG type data is supported.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# Create the R2eff files.
write_r2eff_files(input_dir='input_r2eff', base_dir=dir, force=force)
# Create the parameter files.
write_param_files(global_file="ParamGlobal.inp", set_file="ParamSet1.inp", dir=dir, force=force)
# Create the main execution file.
write_main_file(file=file, dir=dir, output_dir=output_dir, force=force)
# Create the output directory as needed by CATIA (it does not create it itself).
mkdir_nofail(dir + sep + output_dir, verbosity=0)
def sample(self):
"""Generate the ensembles by random sampling of the snapshots."""
# Create the directory for the ensembles, if needed.
mkdir_nofail(dir=self.results_dir + sep + "ensembles")
# Loop over the configurations.
for conf_index in range(len(self.configs)):
# Loop over each ensemble.
for ens in range(self.num_ens):
# Random sampling.
rand = []
for j in range(self.num_models):
rand.append(randint(self.snapshot_min[conf_index], self.snapshot_max[conf_index]))
# Print out.
print("Generating ensemble %s%s from structures %s." % (self.configs[conf_index], ens, rand))
# The file name.
file_name = "ensembles" + sep + self.configs[conf_index] + repr(ens) + ".pdb"
# Open the output file.
out = open(self.results_dir+sep+file_name, 'w')
# Header.
out.write("REM Structures: " + repr(rand) + "\n")
# Concatenation the files.
for j in range(self.num_models):
# The random file.
rand_name = self.snapshot_dir[conf_index] + sep + self.configs[conf_index] + repr(rand[j]) + ".pdb"
# Append the file.
out.write(open(rand_name).read())
# Close the file.
out.close()
def sherekhan_input(spin_id=None, force=False, dir='ShereKhan'):
"""Create the ShereKhan input files.
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError("The relaxation dispersion experiment type has not been specified.")
# Test if the model has been set.
if not hasattr(cdp, 'model_type'):
raise RelaxError("The relaxation dispersion model has not been specified.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# Loop over the spin blocks.
cluster_index = 0
for spin_ids in loop_cluster():
# The spin containers.
spins = spin_ids_to_containers(spin_ids)
# Loop over the magnetic fields.
for exp_type, frq, ei, mi in loop_exp_frq(return_indices=True):
# Loop over the time, and count it.
time_i = 0
for time, ti in loop_time(exp_type=exp_type, frq=frq, return_indices=True):
time_i += 1
# Check that not more than one time point is returned.
if time_i > 1:
raise RelaxError("Number of returned time poins is %i. Only 1 time point is expected."%time_i)
# The ShereKhan input file for the spin cluster.
file_name = 'sherekhan_frq%s.in' % (mi+1)
if dir != None:
dir_name = dir + sep + 'cluster%s' % (cluster_index+1)
else:
dir_name = 'cluster%s' % (cluster_index+1)
file = open_write_file(file_name=file_name, dir=dir_name, force=force)
# The B0 field for the nuclei of interest in MHz (must be positive to be accepted by the server).
file.write("%.10f\n" % abs(frq / periodic_table.gyromagnetic_ratio('1H') * periodic_table.gyromagnetic_ratio('15N') / 1e6))
# The constant relaxation time for the CPMG experiment in seconds.
file.write("%s\n" % (time))
# The comment line.
file.write("# %-18s %-20s %-20s\n" % ("nu_cpmg (Hz)", "R2eff (rad/s)", "Error"))
# Loop over the spins of the cluster.
for i in range(len(spins)):
# Get the residue container.
res = return_residue(spin_ids[i])
# Name the residue if needed.
res_name = res.name
if res_name == None:
res_name = 'X'
# Initialise the lines to output (to be able to catch missing data).
lines = []
# The residue ID line.
lines.append("# %s%s\n" % (res_name, res.num))
# Loop over the dispersion points.
for offset, point in loop_offset_point(exp_type=exp_type, frq=frq, skip_ref=True):
# The parameter key.
param_key = return_param_key_from_data(exp_type=exp_type, frq=frq, offset=offset, point=point)
# No data.
if param_key not in spins[i].r2eff:
continue
# Store the data.
lines.append("%20.15g %20.13g %20.13g\n" % (point, spins[i].r2eff[param_key], spins[i].r2eff_err[param_key]))
# No data.
if len(lines) == 1:
continue
# Write out the data.
for line in lines:
file.write(line)
# Close the file.
file.close()
# Increment the cluster index.
cluster_index += 1
def write(file=None, dir=None, version='3.1', force=False):
"""Create a BMRB NMR-STAR formatted file.
@keyword file: The name of the file to create or a file object.
@type file: str or file object
@keyword dir: The optional directory to place the file into. If set to 'pipe_name', then it will be placed in a directory with the same name as the current data pipe.
@type dir: str or None
@keyword version: The NMR-STAR version to create. This can be either '2.1', '3.0', or '3.1'.
@type version: str
@keyword force: A flag which if True will allow a currently existing file to be overwritten.
@type force: bool
"""
# Test if bmrblib is installed.
if not dep_check.bmrblib_module:
raise RelaxNoModuleInstallError('BMRB library', 'bmrblib')
# Test if the current data pipe exists.
pipe_name = cdp_name()
if not pipe_name:
raise RelaxNoPipeError
# Check the file name.
if file == None:
raise RelaxError("The file name must be specified.")
# A file object.
if isinstance(file, str):
# The special data pipe name directory.
if dir == 'pipe_name':
dir = pipe_name
# Get the full file path.
file = get_file_path(file, dir)
# Fail if the file already exists and the force flag is False.
if access(file, F_OK) and not force:
raise RelaxFileOverwriteError(file, 'force flag')
# Print out.
print("Opening the file '%s' for writing." % file)
# Create the directories.
mkdir_nofail(dir, verbosity=0)
# Specific results writing function.
write_function = specific_analyses.setup.get_specific_fn('bmrb_write', ds[pipe_name].pipe_type)
# Get the info box.
info = Info_box()
# Add the relax citations.
for id, key in zip(['relax_ref1', 'relax_ref2'], ['dAuvergneGooley08a', 'dAuvergneGooley08b']):
# Alias the bib entry.
bib = info.bib[key]
# Add.
exp_info.citation(cite_id=id, authors=bib.author2, doi=bib.doi, pubmed_id=bib.pubmed_id, full_citation=bib.cite_short(doi=False, url=False), title=bib.title, status=bib.status, type=bib.type, journal_abbrev=bib.journal, journal_full=bib.journal_full, volume=bib.volume, issue=bib.number, page_first=bib.page_first, page_last=bib.page_last, year=bib.year)
# Add the relax software package.
exp_info.software(name=exp_info.SOFTWARE['relax'].name, version=version_full(), vendor_name=exp_info.SOFTWARE['relax'].authors, url=exp_info.SOFTWARE['relax'].url, cite_ids=['relax_ref1', 'relax_ref2'], tasks=exp_info.SOFTWARE['relax'].tasks)
# Execute the specific BMRB writing code.
write_function(file, version=version)
# Add the file to the results file list.
if isinstance(file, str):
add_result_file(type='text', label='BMRB', file=file)
def noe_viol(self):
"""NOE violation calculations."""
# Redirect STDOUT to a log file.
if self.log:
sys.stdout = open(
self.results_dir + sep + "logs" + sep + "NOE_viol.log", 'w')
# Create a directory for the save files.
dir = self.results_dir + sep + "NOE_results"
mkdir_nofail(dir=dir)
# Loop over the configurations.
for config in self.configs:
# Print out.
print("\n" * 10 + "# Set up for config " + config + " #" + "\n")
# Open the results file.
out = open(self.results_dir + sep + "NOE_viol_" + config, 'w')
out_sorted = open(
self.results_dir + sep + "NOE_viol_" + config + "_sorted", 'w')
out.write("%-20s%20s\n" % ("# Ensemble", "NOE_volation"))
out_sorted.write("%-20s%20s\n" % ("# Ensemble", "NOE_volation"))
# Create the data pipe.
self.interpreter.pipe.create("noe_viol_%s" % config, "N-state")
# Read the first structure.
self.interpreter.structure.read_pdb(
"ensembles" + sep + config + "0.pdb",
dir=self.results_dir,
set_mol_name=config,
set_model_num=list(range(1, self.num_models + 1)))
# Load all protons as the sequence.
self.interpreter.structure.load_spins("@H*", ave_pos=False)
# Create the pseudo-atoms.
for i in range(len(self.pseudo)):
self.interpreter.spin.create_pseudo(
spin_name=self.pseudo[i][0],
members=self.pseudo[i][1],
averaging="linear")
self.interpreter.sequence.display()
# Read the NOE list.
self.interpreter.noe.read_restraints(file=self.noe_file)
# Set up the N-state model.
self.interpreter.n_state_model.select_model(model="fixed")
# Print out.
print("\n" * 2 + "# Set up complete #" + "\n" * 10)
# Loop over each ensemble.
noe_viol = []
for ens in range(self.num_ens):
# Print out the ensemble to both the log and screen.
if self.log:
sys.stdout.write(config + repr(ens) + "\n")
sys.stderr.write(config + repr(ens) + "\n")
# Delete the old structures and rename the molecule.
self.interpreter.structure.delete()
# Read the ensemble.
self.interpreter.structure.read_pdb(
"ensembles" + sep + config + repr(ens) + ".pdb",
dir=self.results_dir,
set_mol_name=config,
set_model_num=list(range(1, self.num_models + 1)))
# Get the atomic positions.
self.interpreter.structure.get_pos(ave_pos=False)
# Calculate the average NOE potential.
self.interpreter.minimise.calculate()
# Sum the violations.
cdp.sum_viol = 0.0
for i in range(len(cdp.ave_dist)):
if cdp.quad_pot[i][2]:
cdp.sum_viol = cdp.sum_viol + cdp.quad_pot[i][2]
# Write out the NOE violation.
noe_viol.append([cdp.sum_viol, ens])
out.write("%-20i%30.15f\n" % (ens, cdp.sum_viol))
# Save the state.
self.interpreter.results.write(file="%s_results_%s" %
(config, ens),
dir=dir,
force=True)
# Sort the NOE violations.
noe_viol.sort()
# Write the data.
for i in range(len(noe_viol)):
out_sorted.write("%-20i%20.15f\n" %
(noe_viol[i][1], noe_viol[i][0]))
def create(dir=None,
binary=None,
diff_search=None,
sims=None,
sim_type=None,
trim=None,
steps=None,
heteronuc_type=None,
atom1=None,
atom2=None,
spin_id=None,
force=False,
constraints=True):
"""Create the Modelfree4 input files.
The following files are created:
- dir/mfin
- dir/mfdata
- dir/mfpar
- dir/mfmodel
- dir/run.sh
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into a directory named after the current data pipe.
@type dir: str or None
@keyword binary: The name of the Modelfree4 binary file. This can include the path to the binary.
@type binary: str
@keyword diff_search: The diffusion tensor search algorithm (see the Modelfree4 manual for details).
@type diff_search: str
@keyword sims: The number of Monte Carlo simulations to perform.
@type sims: int
@keyword sim_type: The type of simulation to perform (see the Modelfree4 manual for details).
@type sim_type: str
@keyword trim: Trimming of the Monte Carlo simulations (see the Modelfree4 manual for details).
@type trim: int
@keyword steps: The grid search size (see the Modelfree4 manual for details).
@type steps: int
@keyword heteronuc_type: The Modelfree4 three letter code for the heteronucleus type, e.g. '15N', '13C', etc.
@type heteronuc_type: str
@keyword atom1: The name of the heteronucleus in the PDB file.
@type atom1: str
@keyword atom2: The name of the proton in the PDB file.
@type atom2: str
@keyword spin_id: The spin identification string.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
@keyword constraints: A flag which if True will result in constrained optimisation.
@type constraints: bool
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the PDB file is loaded (for the spheroid and ellipsoid).
if hasattr(cdp, 'diff_tensor'
) and not cdp.diff_tensor.type == 'sphere' and not hasattr(
cdp, 'structure'):
raise RelaxNoPdbError
# Deselect certain spins.
__deselect_spins()
# Directory creation.
if dir == None:
dir = pipes.cdp_name()
mkdir_nofail(dir, verbosity=0)
# Number of field strengths and values.
frq = []
for ri_id in cdp.ri_ids:
# New frequency.
if cdp.spectrometer_frq[ri_id] not in frq:
frq.append(cdp.spectrometer_frq[ri_id])
# The 'mfin' file.
mfin = open_write_file('mfin', dir, force)
create_mfin(mfin,
diff_search=diff_search,
sims=sims,
sim_type=sim_type,
trim=trim,
num_frq=len(frq),
frq=frq)
mfin.close()
# Open the 'mfdata', 'mfmodel', and 'mfpar' files.
mfdata = open_write_file('mfdata', dir, force)
mfmodel = open_write_file('mfmodel', dir, force)
mfpar = open_write_file('mfpar', dir, force)
# Loop over the sequence.
for spin, mol_name, res_num, res_name, id in spin_loop(spin_id,
full_info=True,
return_id=True):
# Skip deselected spins.
if not spin.select:
continue
# The 'mfdata' file.
if not create_mfdata(
mfdata, spin=spin, spin_id=id, num_frq=len(frq), frq=frq):
continue
# The 'mfmodel' file.
create_mfmodel(mfmodel,
spin=spin,
spin_id=id,
steps=steps,
constraints=constraints)
# The 'mfpar' file.
create_mfpar(mfpar,
spin=spin,
spin_id=id,
res_num=res_num,
atom1=atom1,
atom2=atom2)
# Close the 'mfdata', 'mfmodel', and 'mfpar' files.
mfdata.close()
mfmodel.close()
mfpar.close()
# The 'run.sh' script.
run = open_write_file('run.sh', dir, force)
create_run(run, binary=binary, dir=dir)
run.close()
chmod(dir + sep + 'run.sh', S_IRWXU | S_IRGRP | S_IROTH)
def __init__(self,
stage=1,
results_dir=None,
num_ens=10000,
num_models=10,
configs=None,
snapshot_dir='snapshots',
snapshot_min=None,
snapshot_max=None,
pseudo=None,
noe_file=None,
noe_norm=None,
rdc_name=None,
rdc_file=None,
rdc_spin_id1_col=None,
rdc_spin_id2_col=None,
rdc_data_col=None,
rdc_error_col=None,
bond_length=None,
bond_length_file=None,
log=None,
bucket_num=200,
lower_lim_noe=0.0,
upper_lim_noe=600.0,
lower_lim_rdc=0.0,
upper_lim_rdc=1.0):
"""Set up for the stereochemistry analysis.
@keyword stage: Stage of analysis (see the module docstring above for the options).
@type stage: int
@keyword results_dir: The optional directory to place all results files into.
@type results_dir: None or str
@keyword num_ens: Number of ensembles.
@type num_ens: int
@keyword num_models: Ensemble size.
@type num_models: int
@keyword configs: All the configurations.
@type configs: list of str
@keyword snapshot_dir: Snapshot directories (corresponding to the configurations).
@type snapshot_dir: list of str
@keyword snapshot_min: The number of the first snapshots (corresponding to the configurations).
@type snapshot_min: list of int
@keyword snapshot_max: The number of the last snapshots (corresponding to the configurations).
@type snapshot_max: list of int
@keyword pseudo: The list of pseudo-atoms. Each element is a list of the pseudo-atom name and a list of all those atoms forming the pseudo-atom. For example, pseudo = [["Q7", ["@H16", "@H17", "@H18"]], ["Q9", ["@H20", "@H21", "@H22"]]].
@type pseudo: list of list of str and list of str
@keyword noe_file: The name of the NOE restraint file.
@type noe_file: str
@keyword noe_norm: The NOE normalisation factor (equal to the sum of all NOEs squared).
@type noe_norm: float
@keyword rdc_name: The label for this RDC data set.
@type rdc_name: str
@keyword rdc_file: The name of the RDC file.
@type rdc_file: str
@keyword rdc_spin_id1_col: The spin ID column of the first spin in the RDC file.
@type rdc_spin_id1_col: None or int
@keyword rdc_spin_id2_col: The spin ID column of the second spin in the RDC file.
@type rdc_spin_id2_col: None or int
@keyword rdc_data_col: The data column of the RDC file.
@type rdc_data_col: int
@keyword rdc_error_col: The error column of the RDC file.
@type rdc_error_col: int
@keyword bond_length: The bond length value in meters. This overrides the bond_length_file argument.
@type bond_length: float or None
@keyword bond_length_file: The file of bond lengths for each atom pair in meters. The first and second columns must be the spin ID strings and the third column must contain the data.
@type bond_length_file: float or None
@keyword log: Log file output flag (only for certain stages).
@type log: bool
@keyword bucket_num: Number of buckets for the distribution plots.
@type bucket_num: int
@keyword lower_lim_noe: Distribution plot limits.
@type lower_lim_noe: int
@keyword upper_lim_noe: Distribution plot limits.
@type upper_lim_noe: int
@keyword lower_lim_rdc: Distribution plot limits.
@type lower_lim_rdc: int
@keyword upper_lim_rdc: Distribution plot limits.
@type upper_lim_rdc: int
"""
# Initial printout.
title(file=sys.stdout, text="Stereochemistry auto-analysis")
# Safely execute the full protocol.
try:
# Execution lock.
status.exec_lock.acquire('auto stereochem analysis',
mode='auto-analysis')
# Set up the analysis status object.
status.init_auto_analysis('stereochem', type='stereochem')
status.current_analysis = 'auto stereochem analysis'
# Store all the args.
self.stage = stage
self.results_dir = results_dir
self.num_ens = num_ens
self.num_models = num_models
self.configs = configs
self.snapshot_dir = snapshot_dir
self.snapshot_min = snapshot_min
self.snapshot_max = snapshot_max
self.pseudo = pseudo
self.noe_file = noe_file
self.noe_norm = noe_norm
self.rdc_name = rdc_name
self.rdc_file = rdc_file
self.rdc_spin_id1_col = rdc_spin_id1_col
self.rdc_spin_id2_col = rdc_spin_id2_col
self.rdc_data_col = rdc_data_col
self.rdc_error_col = rdc_error_col
self.bond_length = bond_length
self.bond_length_file = bond_length_file
self.log = log
self.bucket_num = bucket_num
self.lower_lim_noe = lower_lim_noe
self.upper_lim_noe = upper_lim_noe
self.lower_lim_rdc = lower_lim_rdc
self.upper_lim_rdc = upper_lim_rdc
# Load the interpreter.
self.interpreter = Interpreter(show_script=False,
raise_relax_error=True)
self.interpreter.populate_self()
self.interpreter.on(verbose=False)
# Create the results directory.
if self.results_dir:
mkdir_nofail(self.results_dir)
# Or use the current working directory.
else:
self.results_dir = getcwd()
# Create a directory for log files.
if self.log:
mkdir_nofail(self.results_dir + sep + "logs")
# Clean up.
finally:
# Final printout.
title(file=sys.stdout,
text="Completion of the stereochemistry auto-analysis")
print_elapsed_time(time() - status.start_time)
# Finish and unlock execution.
status.auto_analysis['stereochem'].fin = True
status.current_analysis = None
status.exec_lock.release()
def __init__(self, stage=1, results_dir=None, num_ens=10000, num_models=10, configs=None, snapshot_dir='snapshots', snapshot_min=None, snapshot_max=None, pseudo=None, noe_file=None, noe_norm=None, rdc_name=None, rdc_file=None, rdc_spin_id1_col=None, rdc_spin_id2_col=None, rdc_data_col=None, rdc_error_col=None, bond_length=None, bond_length_file=None, log=None, bucket_num=200, lower_lim_noe=0.0, upper_lim_noe=600.0, lower_lim_rdc=0.0, upper_lim_rdc=1.0):
"""Set up for the stereochemistry analysis.
@keyword stage: Stage of analysis (see the module docstring above for the options).
@type stage: int
@keyword results_dir: The optional directory to place all results files into.
@type results_dir: None or str
@keyword num_ens: Number of ensembles.
@type num_ens: int
@keyword num_models: Ensemble size.
@type num_models: int
@keyword configs: All the configurations.
@type configs: list of str
@keyword snapshot_dir: Snapshot directories (corresponding to the configurations).
@type snapshot_dir: list of str
@keyword snapshot_min: The number of the first snapshots (corresponding to the configurations).
@type snapshot_min: list of int
@keyword snapshot_max: The number of the last snapshots (corresponding to the configurations).
@type snapshot_max: list of int
@keyword pseudo: The list of pseudo-atoms. Each element is a list of the pseudo-atom name and a list of all those atoms forming the pseudo-atom. For example, pseudo = [["Q7", ["@H16", "@H17", "@H18"]], ["Q9", ["@H20", "@H21", "@H22"]]].
@type pseudo: list of list of str and list of str
@keyword noe_file: The name of the NOE restraint file.
@type noe_file: str
@keyword noe_norm: The NOE normalisation factor (equal to the sum of all NOEs squared).
@type noe_norm: float
@keyword rdc_name: The label for this RDC data set.
@type rdc_name: str
@keyword rdc_file: The name of the RDC file.
@type rdc_file: str
@keyword rdc_spin_id1_col: The spin ID column of the first spin in the RDC file.
@type rdc_spin_id1_col: None or int
@keyword rdc_spin_id2_col: The spin ID column of the second spin in the RDC file.
@type rdc_spin_id2_col: None or int
@keyword rdc_data_col: The data column of the RDC file.
@type rdc_data_col: int
@keyword rdc_error_col: The error column of the RDC file.
@type rdc_error_col: int
@keyword bond_length: The bond length value in meters. This overrides the bond_length_file argument.
@type bond_length: float or None
@keyword bond_length_file: The file of bond lengths for each atom pair in meters. The first and second columns must be the spin ID strings and the third column must contain the data.
@type bond_length_file: float or None
@keyword log: Log file output flag (only for certain stages).
@type log: bool
@keyword bucket_num: Number of buckets for the distribution plots.
@type bucket_num: int
@keyword lower_lim_noe: Distribution plot limits.
@type lower_lim_noe: int
@keyword upper_lim_noe: Distribution plot limits.
@type upper_lim_noe: int
@keyword lower_lim_rdc: Distribution plot limits.
@type lower_lim_rdc: int
@keyword upper_lim_rdc: Distribution plot limits.
@type upper_lim_rdc: int
"""
# Execution lock.
status.exec_lock.acquire('auto stereochem analysis', mode='auto-analysis')
# Set up the analysis status object.
status.init_auto_analysis('stereochem', type='stereochem')
status.current_analysis = 'auto stereochem analysis'
# Store all the args.
self.stage = stage
self.results_dir = results_dir
self.num_ens = num_ens
self.num_models = num_models
self.configs = configs
self.snapshot_dir = snapshot_dir
self.snapshot_min = snapshot_min
self.snapshot_max = snapshot_max
self.pseudo = pseudo
self.noe_file = noe_file
self.noe_norm = noe_norm
self.rdc_name = rdc_name
self.rdc_file = rdc_file
self.rdc_spin_id1_col = rdc_spin_id1_col
self.rdc_spin_id2_col = rdc_spin_id2_col
self.rdc_data_col = rdc_data_col
self.rdc_error_col = rdc_error_col
self.bond_length = bond_length
self.bond_length_file = bond_length_file
self.log = log
self.bucket_num = bucket_num
self.lower_lim_noe = lower_lim_noe
self.upper_lim_noe = upper_lim_noe
self.lower_lim_rdc = lower_lim_rdc
self.upper_lim_rdc = upper_lim_rdc
# Load the interpreter.
self.interpreter = Interpreter(show_script=False, quit=False, raise_relax_error=True)
self.interpreter.populate_self()
self.interpreter.on(verbose=False)
# Create the results directory.
if self.results_dir:
mkdir_nofail(self.results_dir)
# Or use the current working directory.
else:
self.results_dir = getcwd()
# Create a directory for log files.
if self.log:
mkdir_nofail(self.results_dir + sep + "logs")
# Finish and unlock execution.
status.auto_analysis['stereochem'].fin = True
status.current_analysis = None
status.exec_lock.release()
def superimpose(self):
"""Superimpose the ensembles using fit to first in Molmol."""
# Create the output directory.
mkdir_nofail("ensembles_superimposed")
# Logging turned on.
if self.log:
log = open(
self.results_dir + sep + "logs" + sep +
"superimpose_molmol.stderr", 'w')
sys.stdout = open(
self.results_dir + sep + "logs" + sep + "superimpose.log", 'w')
# Loop over S and R.
for config in ["R", "S"]:
# Loop over each ensemble.
for ens in range(self.num_ens):
# The file names.
file_in = "ensembles" + sep + config + repr(ens) + ".pdb"
file_out = "ensembles_superimposed" + sep + config + repr(
ens) + ".pdb"
# Print out.
sys.stderr.write(
"Superimposing %s with Molmol, output to %s.\n" %
(file_in, file_out))
if self.log:
log.write(
"\n\n\nSuperimposing %s with Molmol, output to %s.\n" %
(file_in, file_out))
# Failure handling (if a failure occurred and this is rerun, skip all existing files).
if access(self.results_dir + sep + file_out, F_OK):
continue
# Open the Molmol pipe.
pipe = Popen("molmol -t -f -",
shell=True,
stdin=PIPE,
stdout=PIPE,
stderr=PIPE,
close_fds=False)
# Init all.
pipe.stdin.write("InitAll yes\n")
# Read the PDB.
pipe.stdin.write("ReadPdb " + self.results_dir + sep +
file_in + "\n")
# Fitting to mean.
pipe.stdin.write("Fit to_first 'selected'\n")
pipe.stdin.write("Fit to_mean 'selected'\n")
# Write the result.
pipe.stdin.write("WritePdb " + self.results_dir + sep +
file_out + "\n")
# End Molmol.
pipe.stdin.close()
# Get STDOUT and STDERR.
sys.stdout.write(pipe.stdout.read())
if self.log:
log.write(pipe.stderr.read())
# Close the pipe.
pipe.stdout.close()
pipe.stderr.close()
# Open the superimposed file in relax.
self.interpreter.reset()
self.interpreter.pipe.create('out', 'N-state')
self.interpreter.structure.read_pdb(file_out)
# Fix the retarded MOLMOL proton naming.
for model in cdp.structure.structural_data:
# Alias.
mol = model.mol[0]
# Loop over all atoms.
for i in range(len(mol.atom_name)):
# A proton.
if search('H', mol.atom_name[i]):
mol.atom_name[
i] = mol.atom_name[i][1:] + mol.atom_name[i][0]
# Replace the superimposed file.
self.interpreter.structure.write_pdb(
config + repr(ens) + ".pdb",
dir=self.results_dir + sep + "ensembles_superimposed",
force=True)
def rdc_analysis(self):
"""Perform the RDC part of the analysis."""
# Redirect STDOUT to a log file.
if self.log:
sys.stdout = open(self.results_dir+sep+"logs" + sep + "RDC_%s_analysis.log" % self.rdc_name, 'w')
# The dipolar constant.
d = 0.0
if self.bond_length != None:
d = 3.0 / (2.0*pi) * dipolar_constant(g13C, g1H, self.bond_length)
# Create a directory for the save files.
dir = self.results_dir + sep + "RDC_%s_results" % self.rdc_name
mkdir_nofail(dir=dir)
# Loop over the configurations.
for config in self.configs:
# Print out.
print("\n"*10 + "# Set up for config " + config + " #" + "\n")
# Open the results files.
out = open(self.results_dir+sep+"Q_factors_" + config, 'w')
out_sorted = open(self.results_dir+sep+"Q_factors_" + config + "_sorted", 'w')
out.write("%-20s%20s%20s\n" % ("# Ensemble", "RDC_Q_factor(pales)", "RDC_Q_factor(standard)"))
out_sorted.write("%-20s%20s\n" % ("# Ensemble", "RDC_Q_factor(pales)"))
# Create the data pipe.
self.interpreter.pipe.create("rdc_analysis_%s" % config, "N-state")
# Read the first structure.
self.interpreter.structure.read_pdb("ensembles_superimposed" + sep + config + "0.pdb", dir=self.results_dir, set_mol_name=config, set_model_num=list(range(1, self.num_models+1)))
# Load all spins as the sequence.
self.interpreter.structure.load_spins(ave_pos=False)
# Create the pseudo-atoms.
for i in range(len(self.pseudo)):
self.interpreter.spin.create_pseudo(spin_name=self.pseudo[i][0], members=self.pseudo[i][1], averaging="linear")
self.interpreter.sequence.display()
# Read the RDC data.
self.interpreter.rdc.read(align_id=self.rdc_file, file=self.rdc_file, spin_id1_col=self.rdc_spin_id1_col, spin_id2_col=self.rdc_spin_id2_col, data_col=self.rdc_data_col, error_col=self.rdc_error_col)
# Define the magnetic dipole-dipole relaxation interaction.
if self.bond_length != None:
self.interpreter.interatom.set_dist(spin_id1='@C*', spin_id2='@H*', ave_dist=self.bond_length)
self.interpreter.interatom.set_dist(spin_id1='@C*', spin_id2='@Q*', ave_dist=self.bond_length)
else:
self.interpreter.interatom.read_dist(file=self.bond_length_file, spin_id1_col=1, spin_id2_col=2, data_col=3)
# Set the nuclear isotope.
self.interpreter.spin.isotope(isotope='13C', spin_id='@C*')
self.interpreter.spin.isotope(isotope='1H', spin_id='@H*')
self.interpreter.spin.isotope(isotope='1H', spin_id='@Q*')
# Set up the model.
self.interpreter.n_state_model.select_model(model="fixed")
# Print out.
print("\n"*2 + "# Set up complete #" + "\n"*10)
# Loop over each ensemble.
q_factors = []
for ens in range(self.num_ens):
# Print out the ensemble to both the log and screen.
if self.log:
sys.stdout.write(config + repr(ens) + "\n")
sys.stderr.write(config + repr(ens) + "\n")
# Delete the old structures.
self.interpreter.structure.delete()
# Read the ensemble.
self.interpreter.structure.read_pdb("ensembles_superimposed" + sep + config + repr(ens) + ".pdb", dir=self.results_dir, set_mol_name=config, set_model_num=list(range(1, self.num_models+1)))
# Get the positional information, then load the CH vectors.
self.interpreter.structure.get_pos(ave_pos=False)
if self.bond_length != None:
self.interpreter.interatom.set_dist(spin_id1='@C*', spin_id2='@H*', ave_dist=self.bond_length)
else:
self.interpreter.interatom.read_dist(file=self.bond_length_file, spin_id1_col=1, spin_id2_col=2, data_col=3)
self.interpreter.interatom.unit_vectors(ave=False)
# Minimisation.
#grid_search(inc=4)
self.interpreter.minimise("simplex", constraints=False)
# Store and write out the Q-factors.
q_factors.append([cdp.q_rdc, ens])
out.write("%-20i%20.15f%20.15f\n" % (ens, cdp.q_rdc, cdp.q_rdc_norm2))
# Calculate the alignment tensor in Hz, and store it for reference.
cdp.align_tensor_Hz = d * cdp.align_tensors[0].A
cdp.align_tensor_Hz_5D = d * cdp.align_tensors[0].A_5D
# Save the state.
self.interpreter.results.write(file="%s_results_%s" % (config, ens), dir=dir, force=True)
# Sort the NOE violations.
q_factors.sort()
# Write the data.
for i in range(len(q_factors)):
out_sorted.write("%-20i%20.15f\n" % (q_factors[i][1], q_factors[i][0]))
def noe_viol(self):
"""NOE violation calculations."""
# Redirect STDOUT to a log file.
if self.log:
sys.stdout = open(self.results_dir+sep+"logs" + sep + "NOE_viol.log", 'w')
# Create a directory for the save files.
dir = self.results_dir + sep + "NOE_results"
mkdir_nofail(dir=dir)
# Loop over the configurations.
for config in self.configs:
# Print out.
print("\n"*10 + "# Set up for config " + config + " #" + "\n")
# Open the results file.
out = open(self.results_dir+sep+"NOE_viol_" + config, 'w')
out_sorted = open(self.results_dir+sep+"NOE_viol_" + config + "_sorted", 'w')
out.write("%-20s%20s\n" % ("# Ensemble", "NOE_volation"))
out_sorted.write("%-20s%20s\n" % ("# Ensemble", "NOE_volation"))
# Create the data pipe.
self.interpreter.pipe.create("noe_viol_%s" % config, "N-state")
# Read the first structure.
self.interpreter.structure.read_pdb("ensembles" + sep + config + "0.pdb", dir=self.results_dir, set_mol_name=config, set_model_num=list(range(1, self.num_models+1)))
# Load all protons as the sequence.
self.interpreter.structure.load_spins("@H*", ave_pos=False)
# Create the pseudo-atoms.
for i in range(len(self.pseudo)):
self.interpreter.spin.create_pseudo(spin_name=self.pseudo[i][0], members=self.pseudo[i][1], averaging="linear")
self.interpreter.sequence.display()
# Read the NOE list.
self.interpreter.noe.read_restraints(file=self.noe_file)
# Set up the N-state model.
self.interpreter.n_state_model.select_model(model="fixed")
# Print out.
print("\n"*2 + "# Set up complete #" + "\n"*10)
# Loop over each ensemble.
noe_viol = []
for ens in range(self.num_ens):
# Print out the ensemble to both the log and screen.
if self.log:
sys.stdout.write(config + repr(ens) + "\n")
sys.stderr.write(config + repr(ens) + "\n")
# Delete the old structures and rename the molecule.
self.interpreter.structure.delete()
# Read the ensemble.
self.interpreter.structure.read_pdb("ensembles" + sep + config + repr(ens) + ".pdb", dir=self.results_dir, set_mol_name=config, set_model_num=list(range(1, self.num_models+1)))
# Get the atomic positions.
self.interpreter.structure.get_pos(ave_pos=False)
# Calculate the average NOE potential.
self.interpreter.calc()
# Sum the violations.
cdp.sum_viol = 0.0
for i in range(len(cdp.ave_dist)):
if cdp.quad_pot[i][2]:
cdp.sum_viol = cdp.sum_viol + cdp.quad_pot[i][2]
# Write out the NOE violation.
noe_viol.append([cdp.sum_viol, ens])
out.write("%-20i%30.15f\n" % (ens, cdp.sum_viol))
# Save the state.
self.interpreter.results.write(file="%s_results_%s" % (config, ens), dir=dir, force=True)
# Sort the NOE violations.
noe_viol.sort()
# Write the data.
for i in range(len(noe_viol)):
out_sorted.write("%-20i%20.15f\n" % (noe_viol[i][1], noe_viol[i][0]))
def write_r2eff_files(input_dir=None, base_dir=None, force=False):
"""Create the CATIA R2eff input files.
@keyword input_dir: The special directory for the R2eff input files.
@type input_dir: str
@keyword base_dir: The base directory to place the files into.
@type base_dir: str
@keyword force: A flag which if True will cause a pre-existing file to be overwritten.
@type force: bool
"""
# Create the directory for the R2eff files for each field and spin.
dir = base_dir + sep + input_dir
mkdir_nofail(dir, verbosity=0)
# Determine the isotope information.
isotope = None
for spin in spin_loop(skip_desel=True):
if hasattr(spin, 'isotope'):
if isotope == None:
isotope = spin.isotope
elif spin.isotope != isotope:
raise RelaxError("CATIA only supports one spin type.")
if isotope == None:
raise RelaxError("The spin isotopes have not been specified.")
# Isotope translation.
if isotope == '1H':
isotope = 'H1'
elif isotope == '13C':
isotope = 'C13'
elif isotope == '15N':
isotope = 'N15'
# Loop over the frequencies.
for frq, mi in loop_frq(return_indices=True):
# The frequency string in MHz.
frq_string = int(frq * 1e-6)
# The set files.
file_name = "data_set_%i.inp" % frq_string
set_file = open_write_file(file_name=file_name,
dir=base_dir,
force=force)
id = frq_string
set_file.write("ID=%s\n" % id)
set_file.write("Sfrq = %s\n" % frq_string)
set_file.write("Temperature = %s\n" % 0.0)
set_file.write("Nucleus = %s\n" % isotope)
set_file.write("Couplednucleus = %s\n" % 'H1')
set_file.write("Time_equil = %s\n" % 0.0)
set_file.write("Pwx_cp = %s\n" % 0.0)
set_file.write("Taub = %s\n" % 0.0)
set_file.write("Time_T2 = %s\n" % cdp.relax_time_list[0])
set_file.write("Xcar = %s\n" % 0.0)
set_file.write("Seqfil = %s\n" % 'CW_CPMG')
set_file.write("Minerror = %s\n" % "(2.%;0.5/s)")
set_file.write("Basis = (%s)\n" % "Iph_7")
set_file.write("Format = (%i;%i;%i)\n" % (0, 1, 2))
set_file.write("DataDirectory = %s\n" % (dir + sep))
set_file.write("Data = (\n")
# Loop over the spins.
for spin, mol_name, res_num, res_name, spin_id in spin_loop(
full_info=True, return_id=True, skip_desel=True):
# The file.
file_name = "spin%s_%i.cpmg" % (spin_id.replace('#', '_').replace(
':', '_').replace('@', '_'), frq_string)
spin_file = open_write_file(file_name=file_name,
dir=dir,
force=force)
# Write the header.
spin_file.write("# %18s %20s %20s\n" %
("nu_cpmg(Hz)", "R2(1/s)", "Esd(R2)"))
# Loop over the dispersion points.
for offset, point, oi, di in loop_offset_point(
exp_type=EXP_TYPE_CPMG_SQ, frq=frq, return_indices=True):
# The key.
key = return_param_key_from_data(exp_type=EXP_TYPE_CPMG_SQ,
frq=frq,
offset=offset,
point=point)
# No data.
if key not in spin.r2eff:
continue
# Write out the data.
spin_file.write("%20.15f %20.15f %20.15f\n" %
(point, spin.r2eff[key], spin.r2eff_err[key]))
# Close the file.
spin_file.close()
# Add the file name to the set.
catia_spin_id = "%i%s" % (res_num, spin.name)
set_file.write(" [%s;%s];\n" % (catia_spin_id, file_name))
# Terminate the set file.
set_file.write(")\n")
set_file.close()
def write_r2eff_files(input_dir=None, base_dir=None, force=False):
"""Create the CATIA R2eff input files.
@keyword input_dir: The special directory for the R2eff input files.
@type input_dir: str
@keyword base_dir: The base directory to place the files into.
@type base_dir: str
@keyword force: A flag which if True will cause a pre-existing file to be overwritten.
@type force: bool
"""
# Create the directory for the R2eff files for each field and spin.
dir = base_dir + sep + input_dir
mkdir_nofail(dir, verbosity=0)
# Determine the isotope information.
isotope = None
for spin in spin_loop(skip_desel=True):
if hasattr(spin, 'isotope'):
if isotope == None:
isotope = spin.isotope
elif spin.isotope != isotope:
raise RelaxError("CATIA only supports one spin type.")
if isotope == None:
raise RelaxError("The spin isotopes have not been specified.")
# Isotope translation.
if isotope == '1H':
isotope = 'H1'
elif isotope == '13C':
isotope = 'C13'
elif isotope == '15N':
isotope = 'N15'
# Loop over the frequencies.
for frq, mi in loop_frq(return_indices=True):
# The frequency string in MHz.
frq_string = int(frq*1e-6)
# The set files.
file_name = "data_set_%i.inp" % frq_string
set_file = open_write_file(file_name=file_name, dir=base_dir, force=force)
id = frq_string
set_file.write("ID=%s\n" % id)
set_file.write("Sfrq = %s\n" % frq_string)
set_file.write("Temperature = %s\n" % 0.0)
set_file.write("Nucleus = %s\n" % isotope)
set_file.write("Couplednucleus = %s\n" % 'H1')
set_file.write("Time_equil = %s\n" % 0.0)
set_file.write("Pwx_cp = %s\n" % 0.0)
set_file.write("Taub = %s\n" % 0.0)
set_file.write("Time_T2 = %s\n"% cdp.relax_time_list[0])
set_file.write("Xcar = %s\n" % 0.0)
set_file.write("Seqfil = %s\n" % 'CW_CPMG')
set_file.write("Minerror = %s\n" % "(2.%;0.5/s)")
set_file.write("Basis = (%s)\n" % "Iph_7")
set_file.write("Format = (%i;%i;%i)\n" % (0, 1, 2))
set_file.write("DataDirectory = %s\n" % (dir+sep))
set_file.write("Data = (\n")
# Loop over the spins.
for spin, mol_name, res_num, res_name, spin_id in spin_loop(full_info=True, return_id=True, skip_desel=True):
# The file.
file_name = "spin%s_%i.cpmg" % (spin_id.replace('#', '_').replace(':', '_').replace('@', '_'), frq_string)
spin_file = open_write_file(file_name=file_name, dir=dir, force=force)
# Write the header.
spin_file.write("# %18s %20s %20s\n" % ("nu_cpmg(Hz)", "R2(1/s)", "Esd(R2)"))
# Loop over the dispersion points.
for offset, point, oi, di in loop_offset_point(exp_type=EXP_TYPE_CPMG_SQ, frq=frq, return_indices=True):
# The key.
key = return_param_key_from_data(exp_type=EXP_TYPE_CPMG_SQ, frq=frq, offset=offset, point=point)
# No data.
if key not in spin.r2eff:
continue
# Write out the data.
spin_file.write("%20.15f %20.15f %20.15f\n" % (point, spin.r2eff[key], spin.r2eff_err[key]))
# Close the file.
spin_file.close()
# Add the file name to the set.
catia_spin_id = "%i%s" % (res_num, spin.name)
set_file.write(" [%s;%s];\n" % (catia_spin_id, file_name))
# Terminate the set file.
set_file.write(")\n")
set_file.close()
def rdc_analysis(self):
"""Perform the RDC part of the analysis."""
# Redirect STDOUT to a log file.
if self.log:
sys.stdout = open(
self.results_dir + sep + "logs" + sep +
"RDC_%s_analysis.log" % self.rdc_name, 'w')
# The dipolar constant.
d = 0.0
if self.bond_length != None:
d = 3.0 / (2.0 * pi) * dipolar_constant(
periodic_table.gyromagnetic_ratio('13C'),
periodic_table.gyromagnetic_ratio('1H'), self.bond_length)
# Create a directory for the save files.
dir = self.results_dir + sep + "RDC_%s_results" % self.rdc_name
mkdir_nofail(dir=dir)
# Loop over the configurations.
for config in self.configs:
# Print out.
print("\n" * 10 + "# Set up for config " + config + " #" + "\n")
# Open the results files.
out = open(self.results_dir + sep + "Q_factors_" + config, 'w')
out_sorted = open(
self.results_dir + sep + "Q_factors_" + config + "_sorted",
'w')
out.write("%-20s%20s%20s\n" % ("# Ensemble", "RDC_Q_factor(pales)",
"RDC_Q_factor(standard)"))
out_sorted.write("%-20s%20s\n" %
("# Ensemble", "RDC_Q_factor(pales)"))
# Create the data pipe.
self.interpreter.pipe.create("rdc_analysis_%s" % config, "N-state")
# Read the first structure.
self.interpreter.structure.read_pdb(
"ensembles_superimposed" + sep + config + "0.pdb",
dir=self.results_dir,
set_mol_name=config,
set_model_num=list(range(1, self.num_models + 1)))
# Load all spins as the sequence.
self.interpreter.structure.load_spins(ave_pos=False)
# Create the pseudo-atoms.
for i in range(len(self.pseudo)):
self.interpreter.spin.create_pseudo(
spin_name=self.pseudo[i][0],
members=self.pseudo[i][1],
averaging="linear")
self.interpreter.sequence.display()
# Read the RDC data.
self.interpreter.rdc.read(align_id=self.rdc_file,
file=self.rdc_file,
spin_id1_col=self.rdc_spin_id1_col,
spin_id2_col=self.rdc_spin_id2_col,
data_col=self.rdc_data_col,
error_col=self.rdc_error_col)
# Define the magnetic dipole-dipole relaxation interaction.
if self.bond_length != None:
self.interpreter.interatom.set_dist(spin_id1='@C*',
spin_id2='@H*',
ave_dist=self.bond_length)
self.interpreter.interatom.set_dist(spin_id1='@C*',
spin_id2='@Q*',
ave_dist=self.bond_length)
else:
self.interpreter.interatom.read_dist(
file=self.bond_length_file,
spin_id1_col=1,
spin_id2_col=2,
data_col=3)
# Set the nuclear isotope.
self.interpreter.spin.isotope(isotope='13C', spin_id='@C*')
self.interpreter.spin.isotope(isotope='1H', spin_id='@H*')
self.interpreter.spin.isotope(isotope='1H', spin_id='@Q*')
# Set up the model.
self.interpreter.n_state_model.select_model(model="fixed")
# Print out.
print("\n" * 2 + "# Set up complete #" + "\n" * 10)
# Loop over each ensemble.
q_factors = []
for ens in range(self.num_ens):
# Print out the ensemble to both the log and screen.
if self.log:
sys.stdout.write(config + repr(ens) + "\n")
sys.stderr.write(config + repr(ens) + "\n")
# Delete the old structures.
self.interpreter.structure.delete()
# Read the ensemble.
self.interpreter.structure.read_pdb(
"ensembles_superimposed" + sep + config + repr(ens) +
".pdb",
dir=self.results_dir,
set_mol_name=config,
set_model_num=list(range(1, self.num_models + 1)))
# Get the positional information, then load the CH vectors.
self.interpreter.structure.get_pos(ave_pos=False)
if self.bond_length != None:
self.interpreter.interatom.set_dist(
spin_id1='@C*',
spin_id2='@H*',
ave_dist=self.bond_length)
else:
self.interpreter.interatom.read_dist(
file=self.bond_length_file,
spin_id1_col=1,
spin_id2_col=2,
data_col=3)
self.interpreter.interatom.unit_vectors(ave=False)
# Minimisation.
#minimise.grid_search(inc=4)
self.interpreter.minimise.execute("simplex", constraints=False)
# Store and write out the Q factors.
q_factors.append([cdp.q_rdc_norm_squared_sum, ens])
out.write("%-20i%20.15f%20.15f\n" %
(ens, cdp.q_rdc_norm_squared_sum,
cdp.q_rdc_norm_squared_sum))
# Calculate the alignment tensor in Hz, and store it for reference.
cdp.align_tensor_Hz = d * cdp.align_tensors[0].A
cdp.align_tensor_Hz_5D = d * cdp.align_tensors[0].A_5D
# Save the state.
self.interpreter.results.write(file="%s_results_%s" %
(config, ens),
dir=dir,
force=True)
# Sort the NOE violations.
q_factors.sort()
# Write the data.
for i in range(len(q_factors)):
out_sorted.write("%-20i%20.15f\n" %
(q_factors[i][1], q_factors[i][0]))
def write(file=None, dir=None, version='3.1', force=False):
"""Create a BMRB NMR-STAR formatted file.
@keyword file: The name of the file to create or a file object.
@type file: str or file object
@keyword dir: The optional directory to place the file into. If set to 'pipe_name', then it will be placed in a directory with the same name as the current data pipe.
@type dir: str or None
@keyword version: The NMR-STAR version to create. This can be either '2.1', '3.0', or '3.1'.
@type version: str
@keyword force: A flag which if True will allow a currently existing file to be overwritten.
@type force: bool
"""
# Test if bmrblib is installed.
if not dep_check.bmrblib_module:
raise RelaxNoModuleInstallError('BMRB library', 'bmrblib')
# Test if the current data pipe exists.
pipe_name = cdp_name()
if not pipe_name:
raise RelaxNoPipeError
# Check the file name.
if file == None:
raise RelaxError("The file name must be specified.")
# A file object.
if isinstance(file, str):
# The special data pipe name directory.
if dir == 'pipe_name':
dir = pipe_name
# Get the full file path.
file = get_file_path(file, dir)
# Fail if the file already exists and the force flag is False.
if access(file, F_OK) and not force:
raise RelaxFileOverwriteError(file, 'force flag')
# Print out.
print("Opening the file '%s' for writing." % file)
# Create the directories.
mkdir_nofail(dir, verbosity=0)
# Get the info box.
info = Info_box()
# Add the relax citations.
for id, key in zip(['relax_ref1', 'relax_ref2'],
['dAuvergneGooley08a', 'dAuvergneGooley08b']):
# Alias the bib entry.
bib = info.bib[key]
# Add.
exp_info.citation(cite_id=id,
authors=bib.author2,
doi=bib.doi,
pubmed_id=bib.pubmed_id,
full_citation=bib.cite_short(doi=False, url=False),
title=bib.title,
status=bib.status,
type=bib.type,
journal_abbrev=bib.journal,
journal_full=bib.journal_full,
volume=bib.volume,
issue=bib.number,
page_first=bib.page_first,
page_last=bib.page_last,
year=bib.year)
# Add the relax software package.
exp_info.software(name=exp_info.SOFTWARE['relax'].name,
version=version_full(),
vendor_name=exp_info.SOFTWARE['relax'].authors,
url=exp_info.SOFTWARE['relax'].url,
cite_ids=['relax_ref1', 'relax_ref2'],
tasks=exp_info.SOFTWARE['relax'].tasks)
# Execute the specific BMRB writing code.
api = return_api(pipe_name=pipe_name)
api.bmrb_write(file, version=version)
# Add the file to the results file list.
if isinstance(file, str):
add_result_file(type='text', label='BMRB', file=file)
def superimpose(self):
"""Superimpose the ensembles using fit to first in Molmol."""
# Create the output directory.
mkdir_nofail("ensembles_superimposed")
# Logging turned on.
if self.log:
log = open(self.results_dir+sep+"logs" + sep + "superimpose_molmol.stderr", 'w')
sys.stdout = open(self.results_dir+sep+"logs" + sep + "superimpose.log", 'w')
# Loop over S and R.
for config in ["R", "S"]:
# Loop over each ensemble.
for ens in range(self.num_ens):
# The file names.
file_in = "ensembles" + sep + config + repr(ens) + ".pdb"
file_out = "ensembles_superimposed" + sep + config + repr(ens) + ".pdb"
# Print out.
sys.stderr.write("Superimposing %s with Molmol, output to %s.\n" % (file_in, file_out))
if self.log:
log.write("\n\n\nSuperimposing %s with Molmol, output to %s.\n" % (file_in, file_out))
# Failure handling (if a failure occurred and this is rerun, skip all existing files).
if access(self.results_dir+sep+file_out, F_OK):
continue
# Open the Molmol pipe.
pipe = Popen("molmol -t -f -", shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE, close_fds=False)
# Init all.
pipe.stdin.write("InitAll yes\n")
# Read the PDB.
pipe.stdin.write("ReadPdb " + self.results_dir+sep+file_in + "\n")
# Fitting to mean.
pipe.stdin.write("Fit to_first 'selected'\n")
pipe.stdin.write("Fit to_mean 'selected'\n")
# Write the result.
pipe.stdin.write("WritePdb " + self.results_dir+sep+file_out + "\n")
# End Molmol.
pipe.stdin.close()
# Get STDOUT and STDERR.
sys.stdout.write(pipe.stdout.read())
if self.log:
log.write(pipe.stderr.read())
# Close the pipe.
pipe.stdout.close()
pipe.stderr.close()
# Open the superimposed file in relax.
self.interpreter.reset()
self.interpreter.pipe.create('out', 'N-state')
self.interpreter.structure.read_pdb(file_out)
# Fix the retarded MOLMOL proton naming.
for model in cdp.structure.structural_data:
# Alias.
mol = model.mol[0]
# Loop over all atoms.
for i in range(len(mol.atom_name)):
# A proton.
if search('H', mol.atom_name[i]):
mol.atom_name[i] = mol.atom_name[i][1:] + mol.atom_name[i][0]
# Replace the superimposed file.
self.interpreter.structure.write_pdb(config + repr(ens) + ".pdb", dir=self.results_dir+sep+"ensembles_superimposed", force=True)
def sherekhan_input(spin_id=None, force=False, dir='ShereKhan'):
"""Create the ShereKhan input files.
@keyword spin_id: The spin ID string to restrict the file creation to.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into the current directory.
@type dir: str or None
"""
# Test if the current pipe exists.
check_pipe()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the experiment type has been set.
if not hasattr(cdp, 'exp_type'):
raise RelaxError("The relaxation dispersion experiment type has not been specified.")
# Test if the model has been set.
if not hasattr(cdp, 'model_type'):
raise RelaxError("The relaxation dispersion model has not been specified.")
# Directory creation.
if dir != None:
mkdir_nofail(dir, verbosity=0)
# Loop over the spin blocks.
cluster_index = 0
for spin_ids in loop_cluster():
# The spin containers.
spins = spin_ids_to_containers(spin_ids)
# Loop over the magnetic fields.
for exp_type, frq, ei, mi in loop_exp_frq(return_indices=True):
# Loop over the time, and count it.
time_i = 0
for time, ti in loop_time(exp_type=exp_type, frq=frq, return_indices=True):
time_i += 1
# Check that not more than one time point is returned.
if time_i > 1:
raise RelaxError("Number of returned time poins is %i. Only 1 time point is expected."%time_i)
# The ShereKhan input file for the spin cluster.
file_name = 'sherekhan_frq%s.in' % (mi+1)
if dir != None:
dir_name = dir + sep + 'cluster%s' % (cluster_index+1)
else:
dir_name = 'cluster%s' % (cluster_index+1)
file = open_write_file(file_name=file_name, dir=dir_name, force=force)
# The B0 field for the nuclei of interest in MHz (must be positive to be accepted by the server).
file.write("%.10f\n" % abs(frq / periodic_table.gyromagnetic_ratio('1H') * periodic_table.gyromagnetic_ratio('15N') / 1e6))
# The constant relaxation time for the CPMG experiment in seconds.
file.write("%s\n" % (time))
# The comment line.
file.write("# %-18s %-20s %-20s\n" % ("nu_cpmg (Hz)", "R2eff (rad/s)", "Error"))
# Loop over the spins of the cluster.
for i in range(len(spins)):
# Get the residue container.
res = return_residue(spin_ids[i])
# Name the residue if needed.
res_name = res.name
if res_name == None:
res_name = 'X'
# Initialise the lines to output (to be able to catch missing data).
lines = []
# The residue ID line.
lines.append("# %s%s\n" % (res_name, res.num))
# Loop over the dispersion points.
for offset, point in loop_offset_point(exp_type=exp_type, frq=frq, skip_ref=True):
# The parameter key.
param_key = return_param_key_from_data(exp_type=exp_type, frq=frq, offset=offset, point=point)
# No data.
if param_key not in spins[i].r2eff:
continue
# Store the data.
lines.append("%20.15g %20.13g %20.13g\n" % (point, spins[i].r2eff[param_key], spins[i].r2eff_err[param_key]))
# No data.
if len(lines) == 1:
continue
# Write out the data.
for line in lines:
file.write(line)
# Close the file.
file.close()
# Increment the cluster index.
cluster_index += 1
def create(dir=None, binary=None, diff_search=None, sims=None, sim_type=None, trim=None, steps=None, heteronuc_type=None, atom1=None, atom2=None, spin_id=None, force=False, constraints=True):
"""Create the Modelfree4 input files.
The following files are created:
- dir/mfin
- dir/mfdata
- dir/mfpar
- dir/mfmodel
- dir/run.sh
@keyword dir: The optional directory to place the files into. If None, then the files will be placed into a directory named after the current data pipe.
@type dir: str or None
@keyword binary: The name of the Modelfree4 binary file. This can include the path to the binary.
@type binary: str
@keyword diff_search: The diffusion tensor search algorithm (see the Modelfree4 manual for details).
@type diff_search: str
@keyword sims: The number of Monte Carlo simulations to perform.
@type sims: int
@keyword sim_type: The type of simulation to perform (see the Modelfree4 manual for details).
@type sim_type: str
@keyword trim: Trimming of the Monte Carlo simulations (see the Modelfree4 manual for details).
@type trim: int
@keyword steps: The grid search size (see the Modelfree4 manual for details).
@type steps: int
@keyword heteronuc_type: The Modelfree4 three letter code for the heteronucleus type, e.g. '15N', '13C', etc.
@type heteronuc_type: str
@keyword atom1: The name of the heteronucleus in the PDB file.
@type atom1: str
@keyword atom2: The name of the proton in the PDB file.
@type atom2: str
@keyword spin_id: The spin identification string.
@type spin_id: str
@keyword force: A flag which if True will cause all pre-existing files to be overwritten.
@type force: bool
@keyword constraints: A flag which if True will result in constrained optimisation.
@type constraints: bool
"""
# Test if the current pipe exists.
pipes.test()
# Test if sequence data is loaded.
if not exists_mol_res_spin_data():
raise RelaxNoSequenceError
# Test if the PDB file is loaded (for the spheroid and ellipsoid).
if hasattr(cdp, 'diff_tensor') and not cdp.diff_tensor.type == 'sphere' and not hasattr(cdp, 'structure'):
raise RelaxNoPdbError
# Deselect certain spins.
__deselect_spins()
# Directory creation.
if dir == None:
dir = pipes.cdp_name()
mkdir_nofail(dir, verbosity=0)
# Number of field strengths and values.
frq = []
for ri_id in cdp.ri_ids:
# New frequency.
if cdp.spectrometer_frq[ri_id] not in frq:
frq.append(cdp.spectrometer_frq[ri_id])
# The 'mfin' file.
mfin = open_write_file('mfin', dir, force)
create_mfin(mfin, diff_search=diff_search, sims=sims, sim_type=sim_type, trim=trim, num_frq=len(frq), frq=frq)
mfin.close()
# Open the 'mfdata', 'mfmodel', and 'mfpar' files.
mfdata = open_write_file('mfdata', dir, force)
mfmodel = open_write_file('mfmodel', dir, force)
mfpar = open_write_file('mfpar', dir, force)
# Loop over the sequence.
for spin, mol_name, res_num, res_name, id in spin_loop(spin_id, full_info=True, return_id=True):
# Skip deselected spins.
if not spin.select:
continue
# The 'mfdata' file.
if not create_mfdata(mfdata, spin=spin, spin_id=id, num_frq=len(frq), frq=frq):
continue
# The 'mfmodel' file.
create_mfmodel(mfmodel, spin=spin, spin_id=id, steps=steps, constraints=constraints)
# The 'mfpar' file.
create_mfpar(mfpar, spin=spin, spin_id=id, res_num=res_num, atom1=atom1, atom2=atom2)
# Close the 'mfdata', 'mfmodel', and 'mfpar' files.
mfdata.close()
mfmodel.close()
mfpar.close()
# The 'run.sh' script.
run = open_write_file('run.sh', dir, force)
create_run(run, binary=binary, dir=dir)
run.close()
chmod(dir + sep+'run.sh', S_IRWXU|S_IRGRP|S_IROTH)
