def run(self):
path_file = self.path_file
if isinstance(path_file, basestring):
t = BioPaxFile()
if path_file.endswith(".gz"):
handle = gzip.GzipFile(path_file)
t.load(handle)
else:
t.load(path_file)
path_file = t
paths = []
for key,value in path_file.pathways().iteritems():
add = False
if self.pathways is None or value in self.pathways:
add = True
elif key in self.pathways:
add = True
else:
a_name = re_namesplit.split(key)[-1]
if a_name in self.pathways:
add = True
if add:
paths.append(key)
for subnet in path_file.toNet(paths):
gr = networkx.MultiDiGraph()
for k, v in subnet.meta.items():
if k != 'type':
gr.graph[k] = v
subnet.to_graph(gr)
name = re_namesplit.split(gr.graph['url'])[-1]
if self.outdir is not None:
handle = open(os.path.join(self.outdir, name + ".xgmml"), "w")
elif self.singlepath is not None:
handle = open(self.singlepath, "w")
else:
handle = sys.stdout
convert.write_xgmml(gr, handle)
if self.singlepath is not None:
handle.close()
return
def run(self):
path_file = self.path_file
if isinstance(path_file, basestring):
t = BioPaxFile()
if path_file.endswith(".gz"):
handle = gzip.GzipFile(path_file)
t.load(handle)
else:
t.load(path_file)
path_file = t
paths = []
for key, value in path_file.pathways().iteritems():
add = False
if self.pathways is None or value in self.pathways:
add = True
elif key in self.pathways:
add = True
else:
a_name = re_namesplit.split(key)[-1]
if a_name in self.pathways:
add = True
if add:
paths.append(key)
for subnet in path_file.toNet(paths):
gr = networkx.MultiDiGraph()
for k, v in subnet.meta.items():
if k != 'type':
gr.graph[k] = v
subnet.to_graph(gr)
name = re_namesplit.split(gr.graph['url'])[-1]
if self.outdir is not None:
handle = open(os.path.join(self.outdir, name + ".xgmml"), "w")
elif self.singlepath is not None:
handle = open(self.singlepath, "w")
else:
handle = sys.stdout
convert.write_xgmml(gr, handle)
if self.singlepath is not None:
handle.close()
return
def main_build(args):
gr = networkx.MultiDiGraph()
paths = pathway_opener( list( (args.pathways[i],args.pathways[i+1]) for i in range(0, len(args.pathways),2) ) )
builer = GraphBuilder(args)
type_count = {}
interaction_count = {}
duplicate_edges = 0
for cur_path in paths:
log("Scanning: %s" % (cur_path.name))
cur_gr = cur_path.read()
if args.organism is None or cur_gr.graph.get('organism', args.organism) == args.organism:
fix_gr = builer.fix_graph(cur_gr)
for node in fix_gr.node:
if node in gr.node:
if 'type' in gr.node[node] and 'type' in fix_gr.node[node] and gr.node[node]['type'] != fix_gr.node[node]['type']:
error("%s failure: Mismatch Node Type: %s :%s --> %s" % (cur_path.name, node, gr.node[node]['type'], fix_gr.node[node]['type'] ))
if args.rename_nonprotein or args.all:
#because 'protein' is a default node type, if we see something not protein, then change the node to match
if gr.node[node]['type'] == 'protein':
gr.node[node]['type'] = fix_gr.node[node]['type']
else:
log("Merging: %s" % node)
gr.add_node(node, attr_dict=fix_gr.node[node])
for src, dst, data in fix_gr.edges(data=True):
interaction = data['interaction']
src_node_type = fix_gr.node[src].get('type', None)
dst_node_type = fix_gr.node[dst].get('type', None)
if src in gr.node and dst in gr.node:
has_edge = False
if dst in gr.edge[src]:
for i in gr.edge[src][dst]:
if gr.edge[src][dst][i]['interaction'] == interaction:
has_edge = True
if not has_edge:
if not (args.remove_self or args.all) or src != dst:
gr.add_edge(src, dst, attr_dict=data )
else:
log("Removing self loop: %s" % (src))
else:
duplicate_edges += 1
connect_list = networkx.connected_components(networkx.Graph(gr))
rm_list = []
for group in connect_list:
if len(group) < args.min_subgraph:
rm_list.extend(group)
for r in rm_list:
gr.remove_node(r)
log("+---------------------------+")
log("|Node Count: %15d|" % (len(gr.nodes())))
log("|Edge Count: %15d|" % (len(gr.edges())))
log("|Duplicate Edges: %10d|" % (duplicate_edges))
log("|Connected Components: %5d|" % (networkx.number_connected_components(networkx.Graph(gr))))
log("+---------------------------+")
if args.output:
handle = open(args.output, "w")
else:
handle = sys.stdout
for n_type in type_count:
log("Node Type %s: %d" % (n_type, type_count[n_type]))
if args.spf:
network_convert.write_spf(gr, handle)
elif args.sif:
network_convert.write_sif(gr, handle)
else:
network_convert.write_xgmml(gr, handle)
handle.close()
if gr is not None:
if args.out_spf is not None:
if args.out_spf == "-":
ohandle = sys.stdout
else:
ohandle = open(args.out_spf, "w")
convert.write_spf(gr, ohandle,
node_type_field=args.node_type_field, node_type_default=args.node_type_default,
edge_type_field=args.edge_type_field, edge_type_default=args.edge_type_default
)
if args.out_xgmml is not None:
if args.out_xgmml == "-":
ohandle = sys.stdout
else:
ohandle = open(args.out_xgmml, "w")
convert.write_xgmml(gr, ohandle)
if args.out_gmt:
if args.out_gmt == "-":
ohandle = sys.stdout
else:
ohandle = open(args.out_gmt, "w")
pathmap = {}
for node in gr.node:
if 'pathway' in gr.node[node]:
for p in gr.node[node]['pathway']:
if p not in pathmap:
pathmap[p] = [node]
else:
pathmap[p].append(node)
for path in pathmap:
handle.close()
if args.file_mode:
for path in pathways:
handle = open(path)
gr = convert.read_gpml(handle)
handle.close()
if translate_table:
re_map = {}
for n in gr.node:
if 'db_xref' in gr.node[n]:
relabel = find_translation(gr.node[n]['db_xref'])
if relabel:
re_map[n] = relabel
gr = nx.relabel_nodes(gr, re_map)
convert.write_xgmml(gr, sys.stdout)
else:
for path in pathways:
print "Getting", path
grs = []
if args.biopax:
pathdata_str = server.getPathwayAs(fileType="owl", pwId=path)
pathdata_xml = base64.b64decode(pathdata_str)
handle = open( os.path.join(args.outdir, path + ".owl"), "w" )
handle.write(pathdata_xml)
handle.close()
b = biopax.BioPax()
b.parse(pathdata_xml)
grs = b.toNet()
else:
handle.close()
if args.file_mode:
for path in pathways:
handle = open(path)
gr = convert.read_gpml(handle)
handle.close()
if translate_table:
re_map = {}
for n in gr.node:
if 'db_xref' in gr.node[n]:
relabel = find_translation(gr.node[n]['db_xref'])
if relabel:
re_map[n] = relabel
gr = nx.relabel_nodes(gr, re_map)
convert.write_xgmml(gr, sys.stdout)
else:
for path in pathways:
print "Getting", path
grs = []
if args.biopax:
pathdata_str = server.getPathwayAs(fileType="owl", pwId=path)
pathdata_xml = base64.b64decode(pathdata_str)
handle = open(os.path.join(args.outdir, path + ".owl"), "w")
handle.write(pathdata_xml)
handle.close()
b = biopax.BioPax()
b.parse(pathdata_xml)
grs = b.toNet()
else: