for id_cluster, ids in clusters.iteritems():
output.write("%d:%d:%8.5f:%d:%s\n" %
(id_cluster, len(ids), gso_data[ids[0]].scoring,
ids[0], 'lightdock_%d.pdb' % ids[0]))
log.info("Cluster result written to %s file" %
CLUSTER_REPRESENTATIVES_FILE)
if __name__ == '__main__':
try:
# Parse command line
args = parse_command_line()
# Read LightDock output data
gso_data = read_lightdock_output(args.gso_output_file)
# Sort the glowworms data by scoring
sorted_data = sorted(gso_data, key=lambda k: k.scoring, reverse=True)
# Get the Glowworm ids sorted by their scoring
sorted_ids = [g.id_glowworm for g in sorted_data]
# Calculate the different clusters
clusters = clusterize(sorted_ids)
# Write clustering information
write_cluster_info(clusters, gso_data)
except Exception, e:
log.error('Clustering has failed. Please see error:')
def parse_command_line():
parser = argparse.ArgumentParser(prog='lgd_rank_swarm')
parser.add_argument("num_swarms",
help="number of swarms to consider",
type=int,
metavar="num_swarms")
parser.add_argument("steps",
help="steps to consider",
type=int,
metavar="steps")
return parser.parse_args()
if __name__ == '__main__':
try:
current_path = os.getcwd()
args = parse_command_line()
for swarm_id in range(args.num_swarms):
os.chdir(f'{DEFAULT_SWARM_FOLDER}{swarm_id}')
result_file_name = GSO_OUTPUT_FILE % args.steps
lightdock_output = read_lightdock_output(result_file_name)
for g in lightdock_output:
g.id_swarm = swarm_id
g.pdb_file = f'lightdock_{g.id_glowworm}.pdb'
write_ranking_to_file(lightdock_output, order_by="scoring")
os.chdir(current_path)
except IOError:
log.warning("Either num_clusters or steps not found. Exiting...")
raise SystemExit()
"""Writes the clustering result"""
with open(CLUSTER_REPRESENTATIVES_FILE, 'w') as output:
for id_cluster, ids in clusters.iteritems():
output.write("%d:%d:%8.5f:%d:%s\n" % (id_cluster, len(ids), gso_data[ids[0]].scoring,
ids[0], 'lightdock_%d.pdb' % ids[0]))
log.info("Cluster result written to %s file" % CLUSTER_REPRESENTATIVES_FILE)
if __name__ == '__main__':
try:
# Parse command line
args = parse_command_line()
# Read LightDock output data
gso_data = read_lightdock_output(args.gso_output_file)
# Sort the glowworms data by scoring
sorted_data = sorted(gso_data, key=lambda k: k.scoring, reverse=True)
# Get the Glowworm ids sorted by their scoring
sorted_ids = [g.id_glowworm for g in sorted_data]
# Calculate the different clusters
clusters = clusterize(sorted_ids)
# Write clustering information
write_cluster_info(clusters, gso_data)
except Exception, e:
log.error('Clustering has failed. Please see error:')
if not os.path.isfile(EVALUATION_FILE):
raise SystemExit("%s file not found" % EVALUATION_FILE)
# Get contacts and rmsds
contacts, rmsds = read_rmsd_and_contacts_data(EVALUATION_FILE)
swarm_id = args.swarm_id
log.info("cluster %d" % swarm_id)
solutions = []
if args.result_file:
result_file_name = os.path.join(DEFAULT_SWARM_FOLDER + str(swarm_id), args.result_file)
else:
result_file_name = os.path.join(DEFAULT_SWARM_FOLDER + str(swarm_id), GSO_OUTPUT_FILE % args.steps)
scoring_file_name = os.path.join(DEFAULT_SWARM_FOLDER + str(swarm_id), SCORING_FILE)
results = read_lightdock_output(result_file_name)
for result in results:
result.id_cluster = swarm_id
try:
result.rmsd = rmsds[result.id_cluster][result.id_glowworm]
result.contacts = contacts[result.id_cluster][result.id_glowworm]
result.pdb_file = LIGHTDOCK_PDB_FILE % result.id_glowworm
except:
pass
solutions.append(result)
if args.rmsd_file:
log.info('Previous RMSD matrix found. Loading...')
rmsd_matrix_file = os.path.join(DEFAULT_SWARM_FOLDER + str(swarm_id),
CLUSTER_DEFAULT_NAME +
DEFAULT_RMSD_EXTENSION + NUMPY_FILE_SAVE_EXTENSION)
atoms, residues, chains = parse_complex_from_file(structure)
structures.append({'atoms': atoms, 'residues': residues, 'chains': chains, 'file_name': structure})
log.info("%s atoms, %s residues read." % (len(atoms), len(residues)))
receptor = Complex.from_structures(structures)
# Ligand
structures = []
for structure in get_lightdock_structures(args.ligand_structures):
log.info("Reading %s ligand PDB file..." % structure)
atoms, residues, chains = parse_complex_from_file(structure)
structures.append({'atoms': atoms, 'residues': residues, 'chains': chains, 'file_name': structure})
log.info("%s atoms, %s residues read." % (len(atoms), len(residues)))
ligand = Complex.from_structures(structures)
# Read ranking file
glowworms = read_lightdock_output(args.lightdock_ranking_file, initial=1, final=args.top)
# Destination path is the same as the lightdock output
destination_path = os.path.dirname(args.lightdock_ranking_file)
# If normal modes used, need to read them
nmodes_rec = nmodes_lig = None
nm_path = os.path.abspath(os.path.dirname(args.receptor_structures))
# Check NM file for receptor
nm_rec_file = os.path.join(nm_path, DEFAULT_REC_NM_FILE + '.npy')
if os.path.exists(nm_rec_file):
nmodes_rec = read_nmodes(nm_rec_file)
# Check NM file for ligand
nm_lig_file = os.path.join(nm_path, DEFAULT_LIG_NM_FILE + '.npy')
if os.path.exists(nm_lig_file):
nmodes_lig = read_nmodes(nm_lig_file)
structures = []
for structure in get_lightdock_structures(args.ligand_structures):
log.info("Reading %s ligand PDB file..." % structure)
atoms, residues, chains = parse_complex_from_file(structure)
structures.append({
'atoms': atoms,
'residues': residues,
'chains': chains,
'file_name': structure
})
log.info("%s atoms, %s residues read." % (len(atoms), len(residues)))
ligand = Complex.from_structures(structures)
# Read ranking file
glowworms = read_lightdock_output(args.lightdock_ranking_file,
initial=1,
final=args.top)
# Destination path is the same as the lightdock output
destination_path = os.path.dirname(args.lightdock_ranking_file)
# If normal modes used, need to read them
nmodes_rec = nmodes_lig = None
nm_path = os.path.abspath(os.path.dirname(args.receptor_structures))
# Check NM file for receptor
nm_rec_file = os.path.join(nm_path, DEFAULT_REC_NM_FILE + '.npy')
if os.path.exists(nm_rec_file):
nmodes_rec = read_nmodes(nm_rec_file)
# Check NM file for ligand
nm_lig_file = os.path.join(nm_path, DEFAULT_LIG_NM_FILE + '.npy')
if os.path.exists(nm_lig_file):
