def main():
parser = argparse.ArgumentParser(description='Creating Clustering Info Summary')
parser.add_argument('proteosafe_parameters', help='proteosafe_parameters')
parser.add_argument('networking_pairs_results_file', help='networking_pairs_results_file')
parser.add_argument('networking_pairs_results_file_filtered', help='networking_pairs_results_file_filtered')
parser.add_argument('networking_pairs_results_file_filtered_classic_output', help='networking_pairs_results_file_filtered_classic_output')
args = parser.parse_args()
param_obj = ming_proteosafe_library.parse_xml_file(open(args.proteosafe_parameters))
top_k_val = 10
max_component_size = 0
if "TOPK" in param_obj:
top_k_val = int(param_obj["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in param_obj:
max_component_size = int(param_obj["MAXIMUM_COMPONENT_SIZE"][0])
G = molecular_network_filtering_library.loading_network(args.networking_pairs_results_file, hasHeaders=True)
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph_with_headers(G, args.networking_pairs_results_file_filtered)
molecular_network_filtering_library.output_graph(G, args.networking_pairs_results_file_filtered_classic_output)
def main():
usage()
output_file_path = sys.argv[2]
input_file_path = sys.argv[4]
params_file_path = sys.argv[5]
top_k_val = 10
max_component_size = 0
params = ming_proteosafe_library.parse_xml_file(open(params_file_path, "r"))
if "TOPK" in params:
top_k_val = int(params["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in params:
max_component_size = int(params["MAXIMUM_COMPONENT_SIZE"][0])
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_file_path, hasHeaders=True)
#Returning None means that there are no edges in the output
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph(G, output_file_path)
def main():
parser = argparse.ArgumentParser(description='Creating GraphML')
parser.add_argument('input_pairs', help='input_pairs')
parser.add_argument('input_clusterinfosummary',
help='input_clusterinfosummary')
parser.add_argument('input_librarysearch', help='input_librarysearch')
parser.add_argument('output_graphml', help='output_graphml')
parser.add_argument('--input_analoglibrarysearch',
help='input_analoglibrarysearch')
parser.add_argument('--input_pairsfolder', help='input_pairsfolder')
args = parser.parse_args()
#Doing other filtering
G = molecular_network_filtering_library.loading_network(args.input_pairs,
hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(
G, args.input_clusterinfosummary)
molecular_network_filtering_library.add_library_search_results_to_graph(
G, args.input_librarysearch)
if args.input_pairsfolder is not None:
all_pairs_files = glob.glob(os.path.join(args.input_pairsfolder, "*"))
for additional_pairs_file in all_pairs_files:
print("Adding Additional Edges", additional_pairs_file)
molecular_network_filtering_library.add_additional_edges(
G, additional_pairs_file)
if args.input_analoglibrarysearch is not None:
molecular_network_filtering_library.add_library_search_results_to_graph(
G, args.input_librarysearch, annotation_prefix="Analog:")
nx.write_graphml(G, args.output_graphml, infer_numeric_types=True)
def main():
usage()
output_file_path = sys.argv[2]
input_file_path = sys.argv[4]
params_file_path = sys.argv[5]
top_k_val = 10
max_component_size = 0
params = ming_proteosafe_library.parse_xml_file(open(
params_file_path, "r"))
if "TOPK" in params:
top_k_val = int(params["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in params:
max_component_size = int(params["MAXIMUM_COMPONENT_SIZE"][0])
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_file_path,
hasHeaders=True)
#Returning None means that there are no edges in the output
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph(G, output_file_path)
def main():
input_pairs = sys.argv[1]
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_pairs, hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(G, sys.argv[2])
molecular_network_filtering_library.add_library_search_results_to_graph(G, sys.argv[3])
nx.write_graphml(G, sys.argv[4])
def main():
input_pairs = sys.argv[1]
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_pairs,
hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(
G, sys.argv[2])
molecular_network_filtering_library.add_library_search_results_to_graph(
G, sys.argv[3])
nx.write_graphml(G, sys.argv[4], infer_numeric_types=True)
def create_graphml(input_pairs,
input_clusterinfosummary,
input_librarysearch,
input_analoglibrarysearch,
input_pairsfolder,
output_graphml,
collapse_ion_edges=False):
logger = logging_utils.get_logger(__name__)
# Doing other filtering
logger.debug("Creating network")
G = molecular_network_filtering_library.loading_network(input_pairs,
hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(
G, input_clusterinfosummary)
molecular_network_filtering_library.add_library_search_results_to_graph(
G, input_librarysearch)
# mark all nodes as feature or ion identity nodes (constants.NODE.TYPE_ATTRIBUTE)
logger.debug("Mark all node types")
ion_network_utils.mark_all_node_types(G)
# add analogs
if input_analoglibrarysearch is not None:
logger.debug("Add analog library search results")
molecular_network_filtering_library.add_library_search_results_to_graph(
G, input_analoglibrarysearch, annotation_prefix="Analog:")
# add additional edges - e.g. ion identity edges between different ion species of the same molecule
if input_pairsfolder is not None:
all_pairs_files = glob.glob(os.path.join(input_pairsfolder, "*"))
logger.debug("Adding additional edges from files: " +
str(len(all_pairs_files)))
for additional_pairs_file in all_pairs_files:
logger.debug("Adding Additional Edges from " +
str(additional_pairs_file))
molecular_network_filtering_library.add_additional_edges(
G, additional_pairs_file)
# collapse all ion identity networks, each into a single node
if collapse_ion_edges:
logger.debug("Collapsing additional edges of type: " +
CONST.EDGE.ION_TYPE)
try:
G = ion_network_utils.collapse_ion_networks(G)
except:
logger.debug("Failed collapsing")
pass
# export graphml
logger.info("Writing graphml: " + output_graphml)
nx.write_graphml(G, output_graphml, infer_numeric_types=True)
def main():
input_pairs = sys.argv[1]
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_pairs, hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(G, sys.argv[2])
molecular_network_filtering_library.add_library_search_results_to_graph(G, sys.argv[3])
folder_for_additional_pairs = sys.argv[4]
all_pairs_files = ming_fileio_library.list_files_in_dir(folder_for_additional_pairs)
for additional_pairs_file in all_pairs_files:
print("Adding Additional Edges", additional_pairs_file)
molecular_network_filtering_library.add_additional_edges(G, additional_pairs_file)
nx.write_graphml(G, sys.argv[5], infer_numeric_types=True)
def test_network_topology():
print("test topology")
# Getting pairs data
pairs_filename = "pairs.tsv"
if not os.path.exists(pairs_filename):
pairs_data_url = "https://gnps.ucsd.edu/ProteoSAFe/DownloadResultFile?task=b8b5f2f0581c46f1b61ce047d3bbad16&block=main&file=pairs/pairs.tsv"
os.system('wget "{}" -O pairs.tsv'.format(pairs_data_url))
pairs_df = pd.read_csv(pairs_filename, sep="\t", nrows=10000)
pairs_df.to_csv(pairs_filename, sep="\t", index=False)
G = molecular_network_filtering_library.loading_network(
pairs_filename, hasHeaders=True, edgetype="Spec2Vec")
molecular_network_filtering_library.filter_top_k(G, 10)
molecular_network_filtering_library.filter_component_additive(G, 100)
nx.write_graphml(G, "additive.graphml")
def main():
input_pairs = sys.argv[1]
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_pairs,
hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(
G, sys.argv[2])
molecular_network_filtering_library.add_library_search_results_to_graph(
G, sys.argv[3])
folder_for_additional_pairs = sys.argv[4]
all_pairs_files = ming_fileio_library.list_files_in_dir(
folder_for_additional_pairs)
for additional_pairs_file in all_pairs_files:
print("Adding Additional Edges", additional_pairs_file)
molecular_network_filtering_library.add_additional_edges(
G, additional_pairs_file)
nx.write_graphml(G, sys.argv[5], infer_numeric_types=True)
def create_graphml(input_pairs, input_clusterinfosummary, input_librarysearch,
input_analoglibrarysearch, input_pairsfolder,
output_graphml):
#Doing other filtering
G = molecular_network_filtering_library.loading_network(input_pairs,
hasHeaders=True)
molecular_network_filtering_library.add_clusterinfo_summary_to_graph(
G, input_clusterinfosummary)
molecular_network_filtering_library.add_library_search_results_to_graph(
G, input_librarysearch)
if input_pairsfolder is not None:
all_pairs_files = glob.glob(os.path.join(input_pairsfolder, "*"))
for additional_pairs_file in all_pairs_files:
print("Adding Additional Edges", additional_pairs_file)
molecular_network_filtering_library.add_additional_edges(
G, additional_pairs_file)
if input_analoglibrarysearch is not None:
molecular_network_filtering_library.add_library_search_results_to_graph(
G, input_analoglibrarysearch, annotation_prefix="Analog:")
nx.write_graphml(G, output_graphml, infer_numeric_types=True)
def main():
parser = argparse.ArgumentParser(
description='Creating Clustering Info Summary')
parser.add_argument('proteosafe_parameters', help='proteosafe_parameters')
parser.add_argument('networking_pairs_results_file',
help='networking_pairs_results_file')
parser.add_argument('networking_pairs_results_file_filtered',
help='networking_pairs_results_file_filtered')
parser.add_argument(
'networking_pairs_results_file_filtered_classic_output',
help='networking_pairs_results_file_filtered_classic_output')
args = parser.parse_args()
param_obj = ming_proteosafe_library.parse_xml_file(
open(args.proteosafe_parameters))
top_k_val = 10
max_component_size = 0
if "TOPK" in param_obj:
top_k_val = int(param_obj["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in param_obj:
max_component_size = int(param_obj["MAXIMUM_COMPONENT_SIZE"][0])
G = molecular_network_filtering_library.loading_network(
args.networking_pairs_results_file, hasHeaders=True)
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph_with_headers(
G, args.networking_pairs_results_file_filtered)
molecular_network_filtering_library.output_graph(
G, args.networking_pairs_results_file_filtered_classic_output)
def main():
usage()
output_file_path = sys.argv[2]
input_file_path = sys.argv[4]
params_file_path = sys.argv[5]
pairsinfo_path = sys.argv[6]
top_k_val = 10
max_component_size = 0
params = parse_xml_file(open(params_file_path, "r"))
if "TOPK" in params:
top_k_val = int(params["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in params:
max_component_size = int(params["MAXIMUM_COMPONENT_SIZE"][0])
full_command = pairsinfo_path + " " + "--outfile " + output_file_path + " --input-file " + input_file_path
print full_command
os.system(full_command)
if not os.path.exists(output_file_path):
print("No Network Pairs, adjust clustering accordingly")
open(output_file_path, "w")
exit(0)
#Doing other filtering
G = molecular_network_filtering_library.loading_network(output_file_path)
#Returning None means that there are no edges in the output
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph(G, output_file_path)
def main():
usage()
output_file_path = sys.argv[2]
input_file_path = sys.argv[4]
params_file_path = sys.argv[5]
pairsinfo_path = sys.argv[6]
top_k_val = 10
max_component_size = 0
params = parse_xml_file(open(params_file_path, "r"))
if "TOPK" in params:
top_k_val = int(params["TOPK"][0])
if "MAXIMUM_COMPONENT_SIZE" in params:
max_component_size = int(params["MAXIMUM_COMPONENT_SIZE"][0])
full_command = pairsinfo_path + " " + "--outfile " + output_file_path + " --input-file " + input_file_path
print full_command
os.system(full_command)
if not os.path.exists(output_file_path):
print("No Network Pairs, adjust clustering accordingly")
open(output_file_path, "w")
exit(0)
#Doing other filtering
G = molecular_network_filtering_library.loading_network(output_file_path)
#Returning None means that there are no edges in the output
if G == None:
exit(0)
molecular_network_filtering_library.filter_top_k(G, top_k_val)
molecular_network_filtering_library.filter_component(G, max_component_size)
molecular_network_filtering_library.output_graph(G, output_file_path)