Generate graphs using cell sets as unifying concept

.gitignore

@@ -128,8 +128,12 @@ dmypy.json
 # Pyre type checker
 .pyre/
 
+# Ontology files
+*.owl
+*.ttl
+*.nt
+
 # pycharm/mac
 .DS_Store
 .idea/
 .DS_Store
-.DS_Store

pandasaurus_cxg/anndata_analyzer.py

@@ -1,6 +1,5 @@
-from enum import Enum
-import itertools
 import os
+from enum import Enum
 from typing import List, Optional
 
 import pandas as pd
@@ -9,7 +8,6 @@
 from pandasaurus_cxg.anndata_loader import AnndataLoader
 from pandasaurus_cxg.schema.schema_loader import read_json_file
 
-
 # Check if the DEBUG environment variable is set
 debug_mode = os.getenv("DEBUG")
 
@@ -59,20 +57,22 @@ def co_annotation_report(self, disease: Optional[str] = None, enrich: bool = Fal
         """
         free_text_cell_type = [key for key, value in self._schema.items() if value]
         temp_result = []
-        cell_type_combinations = list(itertools.combinations(free_text_cell_type, 2))
-        for combination in cell_type_combinations:
-            field_name_1 = combination[1]
-            field_name_2 = combination[0]
-            if (field_name_1 in self._anndata.obs.columns
-                and field_name_2 in self._anndata.obs.columns
-            ):
-                co_oc = self._filter_data_and_remove_duplicates(field_name_1, field_name_2, disease)
-
-                if enrich:
-                    co_oc = self._enrich_co_annotation(co_oc, field_name_1, field_name_2)
-
-                AnndataAnalyzer._assign_predicate_column(co_oc, field_name_1, field_name_2)
-                temp_result.extend(co_oc.to_dict(orient="records"))
+        for field_name_2 in free_text_cell_type:
+            for field_name_1 in free_text_cell_type:
+                if (
+                    field_name_1 != field_name_2
+                    and field_name_1 in self._anndata.obs.columns
+                    and field_name_2 in self._anndata.obs.columns
+                ):
+                    co_oc = self._filter_data_and_drop_duplicates(
+                        field_name_1, field_name_2, disease
+                    )
+
+                    if enrich:
+                        co_oc = self._enrich_co_annotation(co_oc, field_name_1, field_name_2)
+
+                    AnndataAnalyzer._assign_predicate_column(co_oc, field_name_1, field_name_2)
+                    temp_result.extend(co_oc.to_dict(orient="records"))
 
         result = [
             [item for sublist in [[k, v] for k, v in record.items()] for item in sublist]
@@ -84,19 +84,6 @@ def co_annotation_report(self, disease: Optional[str] = None, enrich: bool = Fal
             columns=["field_name1", "value1", "predicate", "field_name2", "value2"],
         )
 
-    def _filter_data_and_remove_duplicates(self, field_name_1, field_name_2, disease):
-        # Filter the data based on the disease condition
-        co_oc = (
-            self._anndata.obs[
-                (self._anndata.obs["disease_ontology_term_id"].str.lower() == disease.lower())
-            ][[field_name_1, field_name_2]]
-            if disease
-            else self._anndata.obs[[field_name_1, field_name_2]]
-        )
-        # Drop duplicates
-        co_oc = co_oc.drop_duplicates().reset_index(drop=True)
-        return co_oc
-
     def enriched_co_annotation_report(self, disease: Optional[str] = None):
         """
         Generates an enriched co-annotation report based on the provided schema. The enrichment
@@ -121,13 +108,13 @@ def enriched_co_annotation_report(self, disease: Optional[str] = None):
     def _enrich_co_annotation(self, co_oc, field_name_1, field_name_2):
         enricher = AnndataEnricher(self._anndata)
         simple = enricher.simple_enrichment()
-        df = simple[simple["o"].isin(enricher.get_seed_list())][
-            ["s_label", "o_label"]
-        ].rename(columns={"s_label": field_name_1, "o_label": field_name_2})
+        df = simple[simple["o"].isin(enricher.get_seed_list())][["s_label", "o_label"]].rename(
+            columns={"s_label": field_name_1, "o_label": field_name_2}
+        )
         co_oc = pd.concat([co_oc, df], axis=0).reset_index(drop=True)
         return co_oc
 
-    def _filter_data_and_remove_duplicates(self, field_name_1, field_name_2, disease):
+    def _filter_data_and_drop_duplicates(self, field_name_1, field_name_2, disease):
         # Filter the data based on the disease condition
         co_oc = (
             self._anndata.obs[
@@ -141,17 +128,15 @@ def _filter_data_and_remove_duplicates(self, field_name_1, field_name_2, disease
         return co_oc
 
     @staticmethod
-    def _remove_duplicates(data: List[List[str]]) -> List[List[str]]:
+    def _remove_duplicates(data: List[List[str]]):
+        # TODO do a clean up/rename if it is necessary
+        # Currently used only to clean up supercluster_of relations
         unique_data = []
-        unique_set = set()
 
         for sublist in data:
             if Predicate.SUPERCLUSTER_OF.value in sublist:
                 continue
-            sorted_sublist = tuple(sorted(set(sublist)))
-            if sorted_sublist not in unique_set:
-                unique_data.append(sublist)
-                unique_set.add(sorted_sublist)
+            unique_data.append(sublist)
         return unique_data
 
     @staticmethod

pandasaurus_cxg/anndata_enricher.py

@@ -1,7 +1,7 @@
 from typing import List, Optional
 
-from anndata import AnnData
 import pandas as pd
+from anndata import AnnData
 from pandasaurus.query import Query
 from pandasaurus.slim_manager import SlimManager
 
@@ -18,9 +18,9 @@ def __init__(
         anndata: AnnData,
         cell_type_field: Optional[str] = "cell_type_ontology_term_id",
         context_field: Optional[str] = "tissue_ontology_term_id",
-        ontology_list_for_slims: Optional[List[str]] = ["Cell Ontology"],
+        ontology_list_for_slims: Optional[List[str]] = None,
     ):
-        """Initialize the AnndataEnricher instance.
+        """Initialize the AnndataEnricher instance with AnnData object.
 
         Args:
 
@@ -33,8 +33,9 @@ def __init__(
                 The slim list is used in minimal_slim_enrichment and full_slim_enrichment.
                 Defaults to "Cell Ontology"
         """
+        if ontology_list_for_slims is None:
+            ontology_list_for_slims = ["Cell Ontology"]
         # TODO Do we need to keep whole anndata? Would it be enough to keep the obs only?
-        # file_path: The path to the file containing the anndata object.
         self._anndata = anndata
         self.__seed_list = self._anndata.obs[cell_type_field].unique().tolist()
         self.__enricher = Query(self.__seed_list)
@@ -55,8 +56,23 @@ def from_file_path(
         file_path: str,
         cell_type_field: Optional[str] = "cell_type_ontology_term_id",
         context_field: Optional[str] = "tissue_ontology_term_id",
-        ontology_list_for_slims: Optional[List[str]] = ["Cell Ontology"],
+        ontology_list_for_slims: Optional[List[str]] = None,
     ):
+        """Initialize the AnndataEnricher instance with file path.
+
+        Args:
+
+            file_path: The path to the file containing the anndata object.
+            cell_type_field: The cell type information in the anndata object.
+                Defaults to "cell_type_ontology_term_id".
+            context_field: The context information in the anndata object.
+                Defaults to "tissue_ontology_term_id".
+            ontology_list_for_slims: The ontology list for generating the slim list.
+                The slim list is used in minimal_slim_enrichment and full_slim_enrichment.
+                Defaults to "Cell Ontology"
+        """
+        if ontology_list_for_slims is None:
+            ontology_list_for_slims = ["Cell Ontology"]
         return AnndataEnricher(
             AnndataLoader.load_from_file(file_path),
             cell_type_field,
@@ -152,3 +168,6 @@ def validate_slim_list(self, slim_list):
 
     def get_seed_list(self):
         return self.__seed_list
+
+    def get_anndata(self):
+        return self._anndata