def ShowNodeClasses(doc, verbose=0):
#nodes=xpath.find('/graphml/graph/nnode',doc)
nodes = doc.findall('/graphml/graph/nnode')
node_classes = {}
for node in nodes:
node_id = None
node_class = None
if node.attributes:
for attName, attValue in node.attributes.items():
if attName == 'id':
node_id = attValue
node_class_uri = re.sub(r'/[^/]+$', '', node_id)
for cnode in node.childNodes:
if cnode.nodeName == 'data':
if cnode.attributes:
for attName, attValue in cnode.attributes.items():
if attName == 'key' and attValue == 'class':
node_class = xml_utils.DOM_NodeText(cnode)
node_classes[node_class] = node_class_uri
print >> sys.stderr, 'Node classes (%d):' % len(node_classes)
for node_class in sorted(node_classes.keys()):
print >> sys.stderr, '%18s:\t[%s]' % (node_class,
node_classes[node_class])
return node_classes
def DTPathGraphml2TIDs(doc, fout, verbose):
#tgt_nodes=xpath.find('/graphml/graph/nnode/data[@key="class" and text()="gene"]/parent::node()',doc)
tgt_nodes = doc.findall(
'/graphml/graph/nnode/data[@key="class" and text()="gene"]/parent::node()'
)
n_tid = 0
for node in tgt_nodes:
tid = None
atts = node.attributes
for cnode in node.childNodes:
if cnode.attributes:
for attName, attValue in cnode.attributes.items():
if attName == 'key' and attValue == 'label':
txt = xml_utils.DOM_NodeText(cnode)
try:
tid = re.sub(r'\(.*\)', '', txt)
except Exception, e:
print >> sys.stderr, 'Error (Exception): %s' % e
if tid == None: continue
fout.write('%s\n' % tid)
n_tid += 1
def ShowEdgeClasses(doc, verbose=0):
#edges=xpath.find('/graphml/graph/edge',doc)
edges = doc.findall('/graphml/graph/edge')
edge_classes = {}
for edge in edges:
source_id = None
source_class = None
target_id = None
target_class = None
if edge.attributes:
for attName, attValue in edge.attributes.items():
if attName == 'source':
source_id = attValue
source_class_uri = re.sub(r'/[^/]+$', '', source_id)
if attName == 'target':
target_id = attValue
target_class_uri = re.sub(r'/[^/]+$', '', target_id)
for cnode in edge.childNodes:
if cnode.nodeName == 'data':
if cnode.attributes:
for attName, attValue in cnode.attributes.items():
if attName == 'key' and attValue == 'uri':
edge_class = xml_utils.DOM_NodeText(cnode)
if not edge_classes.has_key(edge_class):
edge_classes[edge_class] = set()
edge_classes[edge_class].add(
tuple(
sorted(
[source_class_uri, target_class_uri])))
print >> sys.stderr, 'Edge classes (%d):' % len(edge_classes)
for edge_class in sorted(edge_classes.keys()):
edge_class_name = re.sub(r'^.*/', '', edge_class)
print >> sys.stderr, '%18s:\t[%s]' % (edge_class_name, edge_class)
for source, target in edge_classes[edge_class]:
source_name = re.sub(r'^.*/', '', source)
target_name = re.sub(r'^.*/', '', target)
print >> sys.stderr, '\t\t%s : %s' % (source_name, target_name)
return edge_classes
def AnnotateDTPathGraphmlCompounds(doc, fin_csv, verbose=0):
'''Annotate graphml with additional data from CSV file.'''
n_in = 0
n_out = 0
errtxt = None
if not doc: return None
### Read CSV file containing target data:
csvReader = csv.DictReader(fin_csv,
fieldnames=None,
restkey=None,
restval=None,
dialect='excel',
delimiter=',',
quotechar='"')
try:
csvrow = csvReader.next() ## must do this to get fieldnames
n_in += 1
except:
errtxt = ('ERROR: bad ifile: %s' % fin_csv.name)
return doc
print >> sys.stderr, 'DEBUG: fieldnames=', csvReader.fieldnames
cid_tag = 'cid'
synonyms_tag = 'synonyms'
for tag in (cid_tag, synonyms_tag):
if tag not in csvReader.fieldnames:
print >> sys.stderr, 'ERROR: %s not in fieldnames' % tag
return doc
cid2synonyms = {}
n_synonyms_total = 0
while True:
try:
csvrow = csvReader.next()
n_in += 1
cid = int(csvrow[cid_tag])
synonyms = csvrow[synonyms_tag].split(';')
cid2synonyms[cid] = synonyms
n_synonyms_total += len(synonyms)
except:
break
print >> sys.stderr, 'DEBUG: cpd ids read: %d ; total synonyms: %d' % (
len(cid2synonyms), n_synonyms_total)
#Find cpd nodes in Graphml:
#cpd_nodes=xpath.find('/graphml/graph/nnode/data[@key="class" and text()="pubchem_compound"]/parent::node()',doc)
cpd_nodes = doc.findall(
'/graphml/graph/nnode/data[@key="class" and text()="pubchem_compound"]/parent::node()'
)
n_cid = 0
for node in cpd_nodes:
cid = None
atts = node.attributes
for cnode in node.childNodes:
if cnode.attributes:
for attName, attValue in cnode.attributes.items():
if attName == 'key' and attValue == 'label':
txt = xml_utils.DOM_NodeText(cnode)
try:
cid = int(re.sub(r'\(.*\)', '', txt))
except Exception, e:
print >> sys.stderr, 'Error (Exception): %s' % e
if cid == None: continue
n_cid += 1
if cid2synonyms.has_key(cid):
namenode = doc.createElement('data')
namenode.setAttribute('key', 'synonyms')
namenode.appendChild(
doc.createTextNode('; '.join(cid2synonyms[cid])))
node.appendChild(namenode)
#Also revise cpd label:
#cpd_label_node=xpath.find('data[@key="label"]',node)[0]
cpd_label_node = node.find('data[@key="label"]')
ReplaceNodeText(cpd_label_node, cid2synonyms[cid][0])
xml_str = node.toprettyxml()
xml_str = re.sub(r'\s*[\n\r]+', r'\n', xml_str)
xml_str = re.sub(r'<nnode', '<node', xml_str)
xml_str = re.sub(r'/nnode>', '/node>', xml_str)
#print >>sys.stderr, 'DEBUG: %s'%xml_str
else:
print >> sys.stderr, 'DEBUG: cid %d not in csv' % cid
def AnnotateDTPathGraphmlTargets(doc, fin_csv, verbose=0):
'''Annotate graphml with additional data from CSV file.'''
n_in = 0
n_out = 0
errtxt = None
if not doc:
return None
### Read CSV file containing target data:
csvReader = csv.DictReader(fin_csv,
fieldnames=None,
restkey=None,
restval=None,
dialect='excel',
delimiter=',',
quotechar='"')
try:
csvrow = csvReader.next() ## must do this to get fieldnames
n_in += 1
except:
errtxt = ('ERROR: bad ifile: %s' % fin_csv.name)
return doc
print >> sys.stderr, 'DEBUG: fieldnames=', csvReader.fieldnames
tgt_id_tag = 'protein_accession'
tgt_gene_tag = 'gene_symbol'
tgt_name_tag = 'pref_name'
for tag in (tgt_id_tag, tgt_name_tag, tgt_gene_tag):
if tag not in csvReader.fieldnames:
print >> sys.stderr, 'ERROR: %s not in fieldnames' % tag
return doc
tgt_id2name = {}
tgt_gene2id = {}
while True:
try:
csvrow = csvReader.next()
n_in += 1
tgt_id = csvrow[tgt_id_tag]
tgt_name = csvrow[tgt_name_tag]
tgt_gene = csvrow[tgt_gene_tag]
tgt_gene2id[tgt_gene] = tgt_id
tgt_id2name[tgt_id] = tgt_name
except:
break
print >> sys.stderr, 'DEBUG: protein ids/names read: %d' % len(tgt_id2name)
#for tgt_id in tgt_id2name.keys():
# print >>sys.stderr, 'DEBUG: protein id: %s'%tgt_id
#gene_nodes = xpath.find('/graphml/graph/nnode/data[@key="class" and text()="gene"]/parent::node()', doc)
gene_nodes = doc.findall(
'/graphml/graph/nnode/data[@key="class" and text()="gene"]/parent::node()'
)
print >> sys.stderr, 'DEBUG: gene count: %d' % len(gene_nodes)
for node in gene_nodes:
tgt_id = None
atts = node.attributes
for cnode in node.childNodes:
if cnode.attributes:
for attName, attValue in cnode.attributes.items():
if attName == 'key' and attValue == 'label':
tgt_gene = xml_utils.DOM_NodeText(cnode)
if tgt_gene2id.has_key(tgt_gene):
tgt_id = tgt_gene2id[tgt_gene]
print >> sys.stderr, 'DEBUG: gene = "%s" ; tgt = "%s"' % (
tgt_gene, tgt_id)
if tgt_id and tgt_id2name.has_key(tgt_id):
namenode = doc.createElement('data')
namenode.setAttribute('key', 'name')
namenode.appendChild(doc.createTextNode(tgt_id2name[tgt_id]))
node.appendChild(namenode)
xml_str = node.toprettyxml()
xml_str = re.sub(r'\s*[\n\r]+', r'\n', xml_str)
#xml_str=re.sub(r'<nnode (.*?)</nnode>',r'<node \1</node>',xml_str,re.DOTALL)
xml_str = re.sub(r'<nnode', '<node', xml_str)
xml_str = re.sub(r'/nnode>', '/node>', xml_str)
#print >>sys.stderr, 'DEBUG: %s'%xml_str
return doc