def compress(self, clustering, workspace_handler, data_handler,
matrix_handler, options):
representatives = []
compression_type = self.parameters.get_value("type",
default_value="KMEDOIDS")
pdb_name = self.parameters.get_value("filename",
default_value="compressed")
pdb_path = os.path.join(workspace_handler["results"],
"%s.pdb" % pdb_name)
if compression_type == "RANDOM":
representatives = self.__naive_compression(clustering,
matrix_handler)
elif compression_type == "KMEDOIDS":
representatives = self.__kmedoids_compression(
clustering, matrix_handler)
else:
print "[ERROR Compressor::compress] The compression type does not exist (%s)" % (
self.type)
exit()
save_cluster_elements(representatives, pdb_path, data_handler, options)
return pdb_name
def run(self, clustering, postprocessing_parameters, data_handler, workspaceHandler, matrixHandler, generatedFiles):
results_place = workspaceHandler["results"]
clusters_place = workspaceHandler["clusters"]
cluster_files = []
for cluster in clustering.clusters:
output_path = os.path.join(clusters_place,
"%s.pdb"%(cluster.id))
save_cluster_elements(cluster.all_elements,
[cluster.id]*len(cluster.all_elements), # all share same cluster id
output_path,
data_handler,
postprocessing_parameters)
cluster_files.append(output_path)
# 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 run(self, clustering, postprocessing_parameters, data_handler, workspaceHandler, matrixHandler, generatedFiles):
results_place = workspaceHandler["results"]
clusters_place = workspaceHandler["clusters"]
cluster_files = []
for cluster in clustering.clusters:
output_path = os.path.join(clusters_place, "%s.pdb" % (cluster.id))
save_cluster_elements(
cluster.all_elements,
[cluster.id] * len(cluster.all_elements), # all share same cluster id
output_path,
data_handler,
postprocessing_parameters,
)
cluster_files.append(output_path)
# 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 compress(self, clustering, workspace_handler, data_handler, matrix_handler, options):
representatives = []
compression_type = self.parameters.get_value("type", default_value="KMEDOIDS")
pdb_name = self.parameters.get_value("filename", default_value="compressed")
pdb_path = os.path.join(workspace_handler["results"], "%s.pdb" % pdb_name)
if compression_type == "RANDOM":
representatives = self.__naive_compression(clustering, matrix_handler)
elif compression_type == "KMEDOIDS":
representatives = self.__kmedoids_compression(clustering, matrix_handler)
else:
print "[ERROR Compressor::compress] The compression type does not exist (%s)" % (self.type)
exit()
save_cluster_elements(representatives, pdb_path, data_handler, options)
return pdb_name
