Update

pyproject.toml

@@ -24,17 +24,12 @@ include = [
     { path = "README.md" },
     { path = "README_SHORT.md" },
     { path = "LICENSE" },
-    { path = "doc/**" },
-    { path = "scripts/*" },
-    { path = "stpipeline/common/cdistance.pyx" },
-    { path = "stpipeline/common/*.pxd" },
-    { path = "stpipeline/common/*.so" },
+    { path = "doc/**" }
 ]
 
 [tool.poetry.dependencies]
 python = ">=3.10,<=3.12"
 argparse = "^1.4.0"
-cython = "^3.0.11"
 numpy = "^2.2.1"
 pandas = "^2.2.3"
 scipy = "^1.15.0"
@@ -48,6 +43,7 @@ seaborn = "^0.13.2"
 types-regex = "^2024.11.6.20241221"
 pandas-stubs = "^2.2.3.241126"
 dnaio = "^1.2.3"
+distance = "^0.1.3"
 
 [tool.poetry.scripts]
 st_qa = "scripts.st_qa:main"
@@ -85,8 +81,8 @@ doc = [
     ]
 
 [build-system]
-requires = ["poetry-core>=1.0.0", "setuptools", "cython", "numpy"]
-build-backend = "setuptools.build_meta"
+requires = ["poetry-core>=1.0.0"]
+build-backend = "poetry.core.masonry.api"
 
 [tool.ruff]
 # Exclude a variety of commonly ignored directories.

stpipeline/common/clustering.py

@@ -6,7 +6,7 @@
 import numpy as np
 from scipy.cluster.hierarchy import linkage, fcluster  # type: ignore
 from collections import defaultdict
-from stpipeline.common.cdistance import hamming_distance  # type: ignore
+from stpipeline.common.distance import hamming_distance
 import random
 from collections import Counter
 from typing import List, Dict, Set, Any
@@ -42,12 +42,7 @@ def _get_adj_list_adjacency(umis: List[str], allowed_mismatches: int) -> Dict[st
     Constructs an adjacency list where each UMI points to all other UMIs within
     the allowed mismatches.
     """
-    return {
-        umi: [
-            umi2 for umi2 in umis if hamming_distance(umi.encode("UTF-8"), umi2.encode("UTF-8")) <= allowed_mismatches
-        ]
-        for umi in umis
-    }
+    return {umi: [umi2 for umi2 in umis if hamming_distance(umi, umi2) <= allowed_mismatches] for umi in umis}
 
 
 def _get_connected_components_adjacency(adj_list: Dict[str, List[str]], counts: Counter[str]) -> List[List[str]]:
@@ -100,8 +95,7 @@ def _get_adj_list_directional_adjacency(
         umi: [
             umi2
             for umi2 in umis
-            if hamming_distance(umi.encode("UTF-8"), umi2.encode("UTF-8")) <= allowed_mismatches
-            and counts[umi] >= (counts[umi2] * 2) - 1
+            if hamming_distance(umi, umi2) <= allowed_mismatches and counts[umi] >= (counts[umi2] * 2) - 1
         ]
         for umi in umis
     }
@@ -139,14 +133,13 @@ def dedup_hierarchical(molecular_barcodes: List[str], allowed_mismatches: int, m
     if len(molecular_barcodes) == 2:
         return (
             molecular_barcodes
-            if hamming_distance(molecular_barcodes[0].encode("UTF-8"), molecular_barcodes[1].encode("UTF-8"))
-            <= allowed_mismatches
+            if hamming_distance(molecular_barcodes[0], molecular_barcodes[1]) <= allowed_mismatches
             else [random.choice(molecular_barcodes)]
         )
 
     def d(coord: Any) -> int:
         i, j = coord
-        return hamming_distance(molecular_barcodes[i].encode("UTF-8"), molecular_barcodes[j].encode("UTF-8"))  # type: ignore
+        return hamming_distance(molecular_barcodes[i], molecular_barcodes[j])
 
     indices = np.triu_indices(len(molecular_barcodes), 1)
     distance_matrix = np.apply_along_axis(d, 0, indices)

stpipeline/common/distance.py

@@ -0,0 +1,23 @@
+"""
+Module for distance metrics.
+"""
+import distance  # type: ignore
+
+
+def hamming_distance(a: str, b: str) -> int:
+    """
+    Calculates the Hamming distance between two strings using the `distance` library.
+
+    Args:
+        a: First string.
+        b: Second string.
+
+    Returns:
+        The Hamming distance between the two strings.
+
+    Raises:
+        ValueError: If the strings are of unequal length.
+    """
+    if len(a) != len(b):
+        raise ValueError("Strings must be of equal length to calculate Hamming distance")
+    return distance.hamming(a, b)  # type: ignore