Merge pull request #90 from loucerac/develop

Develop
loucerac · Aug 23, 2024 · a88629b · a88629b
2 parents f9bac0c + 5d40091
commit a88629b
Show file tree

Hide file tree

Showing 4 changed files with 18 additions and 16 deletions.
diff --git a/drexml/datasets.py b/drexml/datasets.py
@@ -55,21 +55,9 @@ def load_disgenet():
         Disgenet curated dataset of gene-disease associations.
     """
 
-    url = "/".join(
-        [
-            "https:/",
-            "www.disgenet.org",
-            "static",
-            "disgenet_ap1",
-            "files",
-            "downloads",
-            "curated_gene_disease_associations.tsv.gz",
-        ]
-    )
+    fpath = get_resource_path("curated_gene_disease_associations.tsv.gz")
 
-    disgenet: pd.DataFrame = pystow.ensure_csv(
-        "drexml", "datasets", url=url, read_csv_kwargs={"sep": "\t"}
-    )
+    disgenet: pd.DataFrame = pd.read_csv(fpath, sep="\t")
 
     disgenet = disgenet.rename(
         columns={

diff --git a/drexml/resources/curated_gene_disease_associations.tsv.gz b/drexml/resources/curated_gene_disease_associations.tsv.gz
diff --git a/pyproject.toml b/pyproject.toml
@@ -21,7 +21,7 @@ dependencies = [
     "pystow>=0.5.0",
 ]
 name = "drexml"
-version = "1.1.2"
+version = "1.1.3"
 description = "(DRExM³L) Drug REpurposing using and eXplainable Machine Learning and Mechanistic Models of signal transduction\""
 readme = "README.md"
 

diff --git a/tests/test_datasets.py b/tests/test_datasets.py
@@ -9,7 +9,7 @@
 import pytest
 from pandas.errors import ParserError
 
-from drexml.datasets import get_disease_data, load_df
+from drexml.datasets import get_disease_data, load_df, load_disgenet
 
 from .test_utils import make_disease_config
 
@@ -63,3 +63,17 @@ def test_load_df_fails_feather():
     _, tmp_file = mkstemp()
     pd.DataFrame().reset_index(names="vader").to_feather(tmp_file)
     load_df(tmp_file)
+
+
+def test_load_disgenet():
+    """Unit test load_disgenet."""
+
+    disgenet = load_disgenet()
+
+    assert disgenet.shape[0] > 0
+    assert all(
+        [
+            x in ["disease_name", "disease_id", "entrez_id", "dga_score"]
+            for x in disgenet.columns
+        ]
+    )