Release 1.43.7

ingest/anndata_.py

@@ -161,15 +161,29 @@ def generate_cluster_body(adata, clustering_name):
         Append clustering data to clustering file
         """
         cluster_cells = pd.DataFrame(adata.obs_names)
+        raw_cluster_body = pd.DataFrame(adata.obsm[clustering_name])
         cluster_body = pd.concat(
-            [cluster_cells, pd.DataFrame(adata.obsm[clustering_name])], axis=1
+            [cluster_cells, pd.DataFrame(
+                AnnDataIngestor.get_cluster_body_as_ndarray(raw_cluster_body)
+            )], axis=1
         )
         filename = AnnDataIngestor.set_clustering_filename(clustering_name)
         pd.DataFrame(cluster_body).to_csv(
             filename, sep="\t", mode="a", header=None, index=False
         )
         AnnDataIngestor.compress_file(filename)
 
+    @staticmethod
+    def get_cluster_body_as_ndarray(raw_body):
+        """
+        Get the cluster body as a numpy array
+        will convert on the fly as necessary
+        """
+        if raw_body.__class__.__name__ == "DataFrame":
+            return raw_body.to_numpy()
+        else:
+            return raw_body
+
     @staticmethod
     def set_clustering_filename(name):
         return f"h5ad_frag.cluster.{name}.tsv"

ingest/validation/ontologies/version.txt

@@ -1 +1 @@
-1772637743 # validation cache key
+1774285924 # validation cache key

requirements.txt

@@ -21,7 +21,7 @@ pytest==7.2.1
 coverage==7.12.0
 pytest-xdist==1.29.0
 pre-commit==1.18.1
-black==24.3.0
+black==26.3.1
 flake8==3.7.8
 pytest-cov==2.8.1
 diffxpy==0.7.4

tests/test_anndata.py

@@ -8,6 +8,8 @@
 import os
 import gzip
 import time
+import pandas as pd
+from numpy.testing import assert_array_equal
 from unittest.mock import MagicMock, patch
 
 from test_expression_files import mock_expression_load
@@ -145,6 +147,21 @@ def test_generate_cluster_body(self):
                 'did not get expected coordinates from cluster body',
             )
 
+    def test_get_cluster_body_as_ndarray(self):
+        test_df = pd.DataFrame({
+            'X': [1, 2, 3],
+            'Y': [4, 5, 6],
+        })
+        cluster_body = self.anndata_ingest.get_cluster_body_as_ndarray(test_df)
+        expected_body = [[1, 4],[2, 5],[3, 6]]
+        assert_array_equal(expected_body, cluster_body)
+        # normal test
+        # long precision is because we're looking at raw numpy array
+        raw_body = self.anndata_ingest.obtain_adata().obsm[self.cluster_name]
+        cluster_body = self.anndata_ingest.get_cluster_body_as_ndarray(raw_body)
+        assert_array_equal([16.00995445251465, -21.07384490966797], cluster_body[0].tolist())
+
+
     def test_generate_metadata_file(self):
         self.anndata_ingest.generate_metadata_file(
             self.anndata_ingest.obtain_adata(), self.metadata_filename