Added remove_special_characters method (#63)

* Updated version to 0.1.11

* Added remove_special_characters method

pandasaurus_cxg/graph_generator/graph_generator.py

@@ -22,6 +22,7 @@
     colour_mapping,
     find_and_rotate_center_layout,
     generate_subgraph,
+    remove_special_characters,
     select_node_with_property,
 )
 from pandasaurus_cxg.graph_generator.graph_predicates import (
@@ -129,7 +130,7 @@ def generate_rdf_graph(self):
             for k, v in inner_dict.items():
                 if k == "subcluster_of":
                     continue
-                self.graph.add((resource, self.ns[k], Literal(v)))
+                self.graph.add((resource, self.ns[remove_special_characters(k)], Literal(v)))
 
         # add relationship between each resource based on their predicate in the co_annotation_report
         # subcluster = self.ns["subcluster_of"]
@@ -139,7 +140,7 @@ def generate_rdf_graph(self):
         for _uuid, inner_dict in grouped_dict_uuid.items():
             resource = self.ns[_uuid]
             for ik, iv in inner_dict.get("subcluster_of", {}).items():
-                predicate = URIRef(self.ns + ik)
+                predicate = self.ns[remove_special_characters(ik)]
                 for s, _, _ in self.graph.triples((None, predicate, Literal(iv))):
                     self.graph.add((resource, subcluster, s))
 

pandasaurus_cxg/graph_generator/graph_generator_utils.py

@@ -1,3 +1,5 @@
+import re
+
 import networkx as nx
 from rdflib import OWL, RDF, RDFS, BNode, Graph, Literal, Namespace, URIRef
 
@@ -120,3 +122,7 @@ def select_node_with_property(graph: Graph, _property: str, value: str):
         return [str(s) for s in graph.subjects(predicate=RDFS.label, object=Literal(value))]
     else:
         return [str(s) for s in graph.subjects(predicate=ns[_property], object=Literal(value))]
+
+
+def remove_special_characters(input_string: str) -> str:
+    return re.sub(r"[^a-zA-Z0-9_]", "", input_string.replace(" ", "_"))

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pandasaurus-cxg"
-version = "0.1.10"
+version = "0.1.11"
 description = "Ontology enrichment tool for CxG standard AnnData files."
 authors = ["Ugur Bayindir <ugur@ebi.ac.uk>"]
 license = "http://www.apache.org/licenses/LICENSE-2.0"

test/graph_generator/test_graph_generator_utils.py

@@ -1,4 +1,5 @@
 import networkx as nx
+import pytest
 from rdflib import OWL, RDF, RDFS, BNode, Graph, Literal, Namespace, URIRef
 
 from pandasaurus_cxg.graph_generator.graph_generator_utils import (
@@ -7,6 +8,7 @@
     add_outgoing_edges_to_subgraph,
     find_and_rotate_center_layout,
     generate_subgraph,
+    remove_special_characters,
     select_node_with_property,
 )
 from pandasaurus_cxg.graph_generator.graph_predicates import (
@@ -183,3 +185,17 @@ def test_select_node_with_property_predicate():
 
     assert len(result) == 1
     assert "http://example.org/subject2" in result
+
+
+@pytest.mark.parametrize(
+    "input_string, expected_output",
+    [
+        ("Hello World!", "Hello_World"),
+        ("123abc$%^", "123abc"),
+        ("!@#$%^&*()_", "_"),
+        ("_This_is_a_test_", "_This_is_a_test_"),
+        ("", ""),
+    ],
+)
+def test_remove_special_characters(input_string, expected_output):
+    assert remove_special_characters(input_string) == expected_output