Update dependencies

.github/workflows/ci.yml

@@ -5,14 +5,19 @@ on:
     paths-ignore:
       - '*.md'
 
+  # Routinely check that tests pass with new versions of dependencies.
+  schedule:
+    # Every day at 17:30 UTC / 9:30 Seattle (winter) / 10:30 Seattle (summer)
+    - cron: "30 17 * * *"
+
 jobs:
   build:
 
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.8", "3.9", "3.10", "3.11"]
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
 
     steps:
     - uses: actions/checkout@v3

CHANGES.md

@@ -1,5 +1,13 @@
 # CHANGELOG
 
+## 4.0.0
+
+### Major Changes
+
+* Update dependencies, removing hardcoded pins on all dependency versions ([#38][])
+
+[#38]: https://github.com/blab/pathogen-embed/pull/38
+
 ## 3.1.0
 
 ### Features

setup.py

@@ -7,7 +7,7 @@
 
 setup(
     name='pathogen-embed',
-    version='3.1.0',
+    version='4.0.0',
     description='Reduced dimension embeddings for pathogen sequences',
     url='https://github.com/blab/pathogen-embed/',
     author='Sravani Nanduri <nandsra@cs.washington.edu> , John Huddleston <huddlej@gmail.com>',
@@ -26,12 +26,9 @@
     install_requires=['numpy',
                       'pandas',
                       "biopython",
-                      'scikit-learn >=1.3,<1.5',
-                      'umap-learn ==0.5.*',
-                      # Pin Numba at maximum supported version for the pinned umap-learn version.
-                      # For more details see:
-                      # https://numba.readthedocs.io/en/stable/reference/deprecation.html#deprecation-of-object-mode-fall-back-behaviour-when-using-jit
-                      'numba <0.59.0',
+                      'scikit-learn',
+                      'umap-learn',
+                      'numba',
                       'matplotlib',
                       'hdbscan'
                       ],
@@ -45,10 +42,11 @@
         'Intended Audience :: Science/Research',
         'License :: OSI Approved :: MIT License',
         'Programming Language :: Python :: 3',
-        'Programming Language :: Python :: 3.8',
-        'Programming Language :: Python :: 3.9',
         'Programming Language :: Python :: 3.10',
         'Programming Language :: Python :: 3.11',
+        'Programming Language :: Python :: 3.12',
+        'Programming Language :: Python :: 3.13',
+        'Programming Language :: Python :: 3.14',
     ],
     entry_points = {
         "console_scripts": [

src/pathogen_embed/pathogen_embed.py

@@ -355,7 +355,9 @@ def embed(args):
 
     warnings.simplefilter('ignore', category=NumbaDeprecationWarning)
 
+    import packaging.version
     from scipy.stats import linregress
+    import sklearn
     from sklearn.decomposition import PCA
     from sklearn.manifold import TSNE, MDS
     from umap import UMAP
@@ -540,14 +542,21 @@ def embed(args):
     elif args.command == "mds":
         embedding_class = MDS
         embedding_parameters = {
-            "dissimilarity": "precomputed",
             "n_components": n_components,
             "n_jobs": 1,
-            "n_init": 2,
+            "n_init": 1,
             "random_state" : args.random_seed,
             "normalized_stress": False,
         }
 
+        if packaging.version.parse(sklearn.__version__) >= packaging.version.parse("1.8"):
+            # Only use init and metric arguments in later versions of scikit-learn that support them:
+            # https://scikit-learn.org/stable/modules/generated/sklearn.manifold.MDS.html#sklearn.manifold.MDS
+            embedding_parameters["init"] = "classical_mds"
+            embedding_parameters["metric"] = "precomputed"
+        else:
+            embedding_parameters["dissimilarity"] = "precomputed"
+
     # Override defaults with parameter values passed through embedding parameters, if
     # possible.
     if external_embedding_parameters is not None and args.command != "pca":
@@ -634,7 +643,7 @@ def embed(args):
         fig, ax = plt.subplots(1, 1, figsize=(8, 8), dpi=300)
         boxplot = ax.boxplot(
             grouped_euclidean_distances,
-            labels=list(genetic_distance_range),
+            tick_labels=list(genetic_distance_range),
             positions=list(genetic_distance_range),
             boxprops={
                 "linewidth": 0.25,
@@ -704,7 +713,7 @@ def cluster(args):
     clusterer = hdbscan.HDBSCAN(**clustering_parameters)
 
     data_df = data_df.astype(float)
-    clusterer.fit(data_df)
+    clusterer.fit(data_df.values)
     data_df[args.label_attribute] = clusterer.labels_.astype(str)
 
     if args.output_figure is not None: