Small bug fixes in LDMatrix and system_utils

CHANGELOG.md

@@ -15,6 +15,8 @@ not work well for very large datasets with millions of variants and it causes ov
 - Updated the way we determine the `pandas` chunksize when converting from `plink` tables to `zarr`.
 - Simplified the way we compute the quantization scale in `model_utils`.
 - Fixed major bug in how LD window thresholds that are passed to `plink1.9` are computed.
+- Fixed in-place `fillna` in `from_plink_table` in `LDMatrix` to conform to latest `pandas` API.
+- Update `run_shell_script` to check for and capture errors.
 
 ### Added
 

magenpy/LDMatrix.py

@@ -229,7 +229,7 @@ def from_plink_table(cls,
             row_index = snp_idx[ld_chunk['SNP_A'].values]
 
             # Fill N/A in R before storing it:
-            ld_chunk['R'].fillna(0., inplace=True)
+            ld_chunk.fillna({'R': 0.}, inplace=True)
 
             # Add LD data to the zarr array:
             if np.issubdtype(dtype, np.integer):

magenpy/utils/system_utils.py

@@ -153,9 +153,12 @@ def run_shell_script(cmd):
     """
     Run the shell script given the command prompt in `cmd`.
     :param cmd: A string with the shell command to run.
+
+    :return: The result of the shell command.
+    :raises: subprocess.CalledProcessError if the shell command fails.
     """
 
-    result = subprocess.run(cmd, shell=True, capture_output=True)
+    result = subprocess.run(cmd, shell=True, capture_output=True, check=True)
 
     if result.stderr:
         raise subprocess.CalledProcessError(
@@ -175,5 +178,5 @@ def delete_temp_files(prefix):
     for f in glob.glob(f"{prefix}*"):
         try:
             os.remove(f)
-        except Exception:
+        except (OSError, FileNotFoundError):
             continue