Merge pull request #157 from AlexandrovLab/pandas2-up-v2

v0.2.0: Require Pandas and Numpy >= 2.0.0 and Python >= 3.9

Author: mdbarnesUCSD

.travis.yml

@@ -1,8 +1,8 @@
-dist: xenial
+dist: focal
 language: python
 
 python:
-  - '3.8'
+  - '3.12'
 
 cache:
   directories:

CHANGELOG.md

@@ -6,6 +6,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 ## [Unreleased]
 
+## [0.2.0] - 2025-02-11
+
+### Changed
+- Updated dependencies: Now requires **Pandas >= 2.0.0**, **NumPy >= 2.0.0**, and **Python >= 3.9**.
+- Dropped support for **Python 3.8**
+
 ## [0.1.9] - 2024-11-12
 
 ### Changed

SigProfilerAssignment/DecompositionPlots/PlotDecomposition.py

@@ -279,18 +279,18 @@ def calculate_similarities(denovo, denovo_name, est_denovo):
 
     # If matrix is 1536 context, then collapse it to 96 format
     if denovo.shape[0] == 1536:
-        index = denovo.iloc[:, 0]
-        denovo_tmp = pd.DataFrame(denovo, index=index)
-        denovo_tmp = denovo.groupby(denovo_tmp.index.str[1:8]).sum()
-        denovo = pd.DataFrame(denovo_tmp)
-        denovo = denovo.reset_index()
-    elif denovo.shape[0] == 288:
-        index = denovo.iloc[:, 0]
-        denovo_tmp = pd.DataFrame(denovo, index=index)
-        denovo_tmp = denovo.groupby(denovo_tmp.index.str[2:9]).sum()
-        denovo = pd.DataFrame(denovo_tmp)
-        denovo = denovo.reset_index()
+        denovo["IndexKey"] = denovo.iloc[:, 0].astype(str)
+        denovo["GroupKey"] = denovo["IndexKey"].str[1:8]
+        denovo_grouped = denovo.groupby("GroupKey", as_index=False).sum(numeric_only=True)
+        denovo_grouped.rename(columns={"GroupKey": "MutationType"}, inplace=True)
+        denovo = denovo_grouped.reset_index(drop=True)
 
+    elif denovo.shape[0] == 288:
+        denovo["MutationType"] = denovo["MutationType"].astype(str)  # Ensure strings
+        denovo["GroupKey"] = denovo["MutationType"].str[2:9]  # Extract substring for grouping
+        denovo_grouped = denovo.groupby("GroupKey", as_index=False).sum(numeric_only=True)
+        denovo_grouped.rename(columns={"GroupKey": "MutationType"}, inplace=True)
+        denovo = denovo_grouped.reset_index(drop=True) # updated pandas 2.0.0
     sample_names = [denovo_name]
 
     if sample_names is False:

SigProfilerAssignment/decomposition.py

@@ -103,7 +103,7 @@ def generate_sample_reconstruction(
         basis_names = (
             subset[subset["Samples"].str.contains(sample_name)].columns[1:].tolist()
         )
-        recon_tmb = subset.sum(axis=1)
+        recon_tmb = subset.select_dtypes(include=['number']).sum(axis=1)
         weights = []
         for i in range(len(basis_names)):
             weights.append(

setup.py

@@ -6,7 +6,7 @@
 if os.path.exists("dist"):
     shutil.rmtree("dist")
 
-VERSION = "0.1.9"
+VERSION = "0.2.0"
 
 
 def write_version_py(filename="SigProfilerAssignment/version.py"):
@@ -15,7 +15,7 @@ def write_version_py(filename="SigProfilerAssignment/version.py"):
 # THIS FILE IS GENERATED FROM SigProfilerAssignment SETUP.PY
 short_version = '%(version)s'
 version = '%(version)s'
-Update = 'v0.1.9: Replace PdfMerger with PdfReader and PdfWriter for pypdf 5.0.0'
+Update = 'v0.2.0: Require Pandas and Numpy >= 2.0.0 and Python >= 3.9'
 
     
     """
@@ -33,11 +33,11 @@ def write_version_py(filename="SigProfilerAssignment/version.py"):
     long_description = f.read()
 
 requirements = [
-    "scipy>=1.6.3",
-    "numpy>=1.21.2,<2.0.0",
-    "pandas>=1.2.4,<2.0.0",
-    "SigProfilerMatrixGenerator>=1.2.28",
-    "sigProfilerPlotting>=1.3.24",
+    "scipy>=1.13",
+    "numpy>=2.0.0",
+    "pandas>=2.0.0",
+    "SigProfilerMatrixGenerator>=1.3.0",
+    "sigProfilerPlotting>=1.4.0",
     "reportlab>=3.5.42",
     "pypdf>=5.0.0",
     "alive_progress>=2.4.1",
@@ -57,6 +57,7 @@ def write_version_py(filename="SigProfilerAssignment/version.py"):
     license="UCSD",
     packages=["SigProfilerAssignment"],
     install_requires=requirements,
+    python_requires=">=3.9",
     extras_require={
         "tests": [
             "pytest",