updates

scripts/lpls/rpy_go.py

@@ -45,7 +45,7 @@ def goterms_from_gene(genelist, ontology=['BP'], garbage = ['IEA', 'ISS', 'ND'])
     _ONTOLOGIES = ['BP', 'CC', 'MF']
     assert(scipy.all([(code in _CODES) for code in garbage]))
     assert(scipy.all([(ont in _ONTOLOGIES) for ont in ontology]))
-    
+    have_these = rpy.r('as.list(GOTERM)').keys()
     goterms = {}
     for gene in genelist:
         goterms[gene] = []
@@ -53,8 +53,12 @@ def goterms_from_gene(genelist, ontology=['BP'], garbage = ['IEA', 'ISS', 'ND'])
         #print info
         if info:
             for term, desc in info.items():
+                if term not in have_these:
+                    print "GO miss:"
+                    print term
                 if desc['Ontology'] in ontology and desc['Evidence'] not in garbage:
                         goterms[gene].append(term)
+
     return goterms
 
 def genego_matrix(goterms, tmat, gene_ids, term_ids, func=min):
@@ -97,7 +101,12 @@ def goterm2desc(gotermlist):
     return term2desc
 
 def parents_dag(go_terms, ontology=['BP']):
-    """ Returns a list of lists representation of a GO DAG parents of goterms."""
+    """ Returns a list of lists representation of a GO DAG parents of goterms.
+
+    make the networkx graph by:
+        G = networkx.Digraph()
+        G = networkx.from_dict_of_lists(edge_dict, G)
+    """
     try:
         rpy.r.library("GOstats")
     except:
@@ -105,6 +114,11 @@ def parents_dag(go_terms, ontology=['BP']):
     assert(go_terms[0][:3]=='GO:')
 
     # go valid namespace
-    go_env = {'BP':rpy.r.BPPARENTS, 'MF':rpy.r.MFPARENTS, 'CC': rpy.r.CCPARENTS}
-    
-    
+    go_env = {'BP':rpy.r.GOBPPARENTS, 'MF':rpy.r.GOMFPARENTS, 'CC': rpy.r.GOCCPARENTS}
+    graph = rpy.r.GOGraph(go_terms, go_env[ontology[0]])
+    edges = rpy.r.edges(graph)
+    edges.pop('all')
+    edge_dict = {}
+    for head, nei in edges.items():
+        edge_dict[head] = nei.values()
+    return edge_dict