def save_representatives(
representatives,
pdb_name,
workspace_handler,
trajectory_holder,
do_merged_files_have_correlative_models,
write_frame_number_instead_of_correlative_model_number,
keep_remarks=False):
"""
Saves a pdb file containing the most representative elements of the clustering.
@param representatives: A list of the representative elements of the clustering we want to extract.
@param workspace_handler: The workspace handler of this run.
@param trajectory_holder: The trajectory handler for this run or an array with pdb file paths.
@param do_merged_files_have_correlative_models: When merging, output file will have models from 0 to M, where M is the total number
of frames of the merged file.
@param write_frame_number_instead_of_model_number: When extracting frames, extract those models which number coincides with the
frame numbers in 'representatives'. Otherwise, extract those models which position coincide with the frame number in
'representatives'.
"""
results_directory = workspace_handler["results"]
# Merge pdbs (in order)
temporary_merged_trajectory_path = os.path.join(
workspace_handler["tmp"], "tmp_merged_trajectory.pdb")
#===========================================================
# THIS DOES NOT WORK IF USING DCD FILES
# merge_pdbs(trajectory_holder,
# temporary_merged_trajectory_path,
# do_merged_files_have_correlative_models)
# TEMPORARY HACK TO OVERCOME DCD MERGING BUG
merged_pdb = trajectory_holder.getMergedStructure()
prody.writePDB(temporary_merged_trajectory_path, merged_pdb)
#==========================================================
# Extract frames from the merged pdb
file_handler_in = open(temporary_merged_trajectory_path, "r")
file_handler_out = open(
os.path.join(results_directory, "%s.pdb" % pdb_name), "w")
pdb_tools.extract_frames_from_trajectory_sequentially(
file_handler_in=file_handler_in,
number_of_frames=pdb_tools.get_number_of_frames(
temporary_merged_trajectory_path),
file_handler_out=file_handler_out,
frames_to_save=representatives,
write_frame_number_instead_of_correlative_model_number=
write_frame_number_instead_of_correlative_model_number,
keep_header=keep_remarks)
file_handler_in.close()
file_handler_out.close()
return os.path.join(results_directory, "%s.pdb" % pdb_name)
def save_all_clusters(my_params, pdb_params, workspaceHandler, trajectoryHandler, clustering, generatedFiles, timer):
timer.start("Save clusters")
#Parameters
keep_remarks = my_params["keep_remarks"] if "keep_remarks" in my_params else False
keep_frame_number = my_params["keep_frame_number"] if "keep_frame_number" in my_params else False
# Places
results_place = workspaceHandler["results"]
clusters_place = workspaceHandler["clusters"]
tmp_place = workspaceHandler["tmp"]
#===========================================================
# THIS DOES NOT WORK IF USING DCD OR STRUCTS
# The real job
# input_path = os.path.join(tmp_place, "tmp_merged_trajectory.pdb")
# merge_pdbs(pdb_params, input_path)
# TEMPORARY HACK TO OVERCOME DCD MERGING BUG
merged_pdb = trajectoryHandler.getMergedStructure()
input_path = os.path.join(tmp_place, "tmp_merged_trajectory.pdb")
prody.writePDB(input_path, merged_pdb)
#==========================================================
number_of_frames = get_number_of_frames(input_path)
cluster_files = []
for cluster in clustering.clusters:
output_path = os.path.join(clusters_place, "%s.pdb"%(cluster.id))
cluster_files.append(output_path)
file_handler_in = open(input_path,"r")
file_handler_out = open(output_path,"w")
extract_frames_from_trajectory_sequentially(file_handler_in,
number_of_frames,
file_handler_out,
cluster.all_elements,
keep_header = keep_remarks,
write_frame_number_instead_of_correlative_model_number=keep_frame_number)
file_handler_in.close()
file_handler_out.close()
# Add all bz2 files to a tar file
tar_path = os.path.join(results_place,"clusters.tar.gz")
tar = tarfile.open(tar_path, "w:gz")
for comp_file in cluster_files:
tar.add(comp_file, os.path.basename(comp_file))
tar.close()
timer.stop("Save clusters")
generatedFiles.append({"description":"Clusters",
"path":os.path.abspath(tar_path),
"type":"compressed_pdb"})
def save_all_clusters(clustering, my_params, workspaceHandler,
trajectoryHandler, generatedFiles):
#Parameters
keep_remarks = my_params.get_value("keep_remarks", default_value=False)
keep_frame_number = my_params.get_value("keep_frame_number",
default_value=False)
# Places
results_place = workspaceHandler["results"]
clusters_place = workspaceHandler["clusters"]
tmp_place = workspaceHandler["tmp"]
merged_pdb = trajectoryHandler.getMergedStructure()
input_path = os.path.join(tmp_place, "tmp_merged_trajectory.pdb")
prody.writePDB(input_path, merged_pdb)
number_of_frames = get_number_of_frames(input_path)
cluster_files = []
for cluster in clustering.clusters:
output_path = os.path.join(clusters_place, "%s.pdb" % (cluster.id))
cluster_files.append(output_path)
file_handler_in = open(input_path, "r")
file_handler_out = open(output_path, "w")
extract_frames_from_trajectory_sequentially(
file_handler_in,
number_of_frames,
file_handler_out,
cluster.all_elements,
keep_header=keep_remarks,
write_frame_number_instead_of_correlative_model_number=
keep_frame_number)
file_handler_in.close()
file_handler_out.close()
# Add all bz2 files to a tar file
tar_path = os.path.join(results_place, "clusters.tar.gz")
tar = tarfile.open(tar_path, "w:gz")
for comp_file in cluster_files:
tar.add(comp_file, os.path.basename(comp_file))
tar.close()
generatedFiles.append({
"description": "Clusters",
"path": os.path.abspath(tar_path),
"type": "compressed_pdb"
})
def save_cluster_handler(self, data):
data = convert_to_utf8(json.loads(data))
# print "DATA", data
path = os.path.join(data["paths"]["tmp"], "tmp_merged_trajectory.pdb")
# try:
# print "PRIM PATH",path
file_handler_in = open(path,"r")
file_handler_out = open("results/tmp/cluster.pdb","w")
extract_frames_from_trajectory_sequentially(file_handler_in,
get_number_of_frames(path),
file_handler_out,
data["elements"],
keep_header=True,
write_frame_number_instead_of_correlative_model_number=True)
file_handler_in.close()
file_handler_out.close()
self.wfile.write('{"path":"results/tmp/cluster.pdb"}')
#######################################################################################################################
# Plot clusters
#######################################################################################################################
plot_clusters(os.path.join(base_dir, "clusters.svg"), all_metrics, scores, scores.index(most_negative))
#######################################################################################################################
# Store the elements of the most negative cluster
#######################################################################################################################
# This works because we have only one traj. Merging before should be mandatory.
trajectory_path = os.path.join(os.getcwd(), traj_pdb)
in_handler = open(trajectory_path,"r")
most_negative_path = os.path.join(base_dir,"most_negative_cluster.pdb")
out_handler = open(most_negative_path,"w")
extract_frames_from_trajectory_sequentially(file_handler_in = in_handler,
number_of_frames = get_number_of_frames(trajectory_path),
file_handler_out = out_handler,
frames_to_save = most_negative_cluster.all_elements,
keep_header = True,
write_frame_number_instead_of_correlative_model_number = True)
in_handler.close()
out_handler.close()
#######################################################################################################################
# Pick 5 well-separated representatives
#######################################################################################################################
working_directory = os.path.join(base_dir, "cluster_to_5")
cluster_to_5_script_path = os.path.join(base_dir, 'scripts', CLUSTERING_TO_5_SCRIPT)
params = load_dic_in_json(cluster_to_5_script_path)
params['global']['workspace']['base'] = working_directory
params['data']['files'] = [most_negative_path]
params['data']['matrix']['parameters']['dist_fit_selection'] = "name CA"
params['data']['matrix']['parameters']['body_selection'] = ligand_description
