Feature/v1.5.8 (#97)

* Fixed bug in CSV export; quality_mask probes were NaN instead of 1.0 * pyparsing error on circleci
FoxoTech · Nov 4, 2021 · f59c00f · f59c00f
1 parent 53865b3
commit f59c00f
Show file tree

Hide file tree

Showing 7 changed files with 14 additions and 4 deletions.
diff --git a/Pipfile b/Pipfile
@@ -19,6 +19,7 @@ pytest-mock = "*"
 coverage = "*"
 
 [packages]
+pyparsing = "<3"
 numpy = "*" # "==1.16.3"
 pandas = ">=1.3.0"
 statsmodels = "*" # "==0.9.0"

diff --git a/docs/release-history.md b/docs/release-history.md
@@ -1,5 +1,12 @@
 # Release History
 
+## v1.5.8
+- Fixed bug in CSV export; quality_mask (probes to be excluded from pickles) were NaN instead of 1.0
+
+## v1.5.7
+- Merely a maintenance release to deal with some dependency conflicts in the python science stack.
+    Pandas version 1.3+ now supported.
+
 ## v1.5.6
 - completely rewritten and updated documentation, including a more extensive tutorial
 - updated all manifests to include probe-to-locus mapping for two genome builds

diff --git a/methylprep/processing/pipeline.py b/methylprep/processing/pipeline.py
@@ -855,7 +855,7 @@ def export(self, output_path):
                 'noob_unmeth': self.__quality_mask_excluded_probes['noob_unmeth']
                 })
         if 'quality_mask' in this.columns:
-            this['quality_mask'] = this['quality_mask'].fillna(0)
+            this['quality_mask'] = this['quality_mask'].fillna(1)
         # noob columns contain NANs now because of sesame (v1.4.0 to v1.4.5); v1.4.6+ CSVs contain all data, but pickles are filtered.
         #try:
         #    self.__data_frame['noob_meth'] = self.__data_frame['noob_meth'].astype(int, copy=False)

diff --git a/methylprep/processing/preprocess.py b/methylprep/processing/preprocess.py
@@ -292,7 +292,7 @@ def _apply_sesame_quality_mask(data_container):
     cgs = pd.DataFrame( np.zeros((len(data_container.man.index), 1)), index=data_container.man.index, columns=['quality_mask'])
     snps = pd.DataFrame( np.zeros((len(data_container.snp_man.index), 1)), index=data_container.snp_man.index, columns=['quality_mask'])
     df = pd.concat([cgs, snps])
-    df.loc[ df.index.isin(probes), 'quality_mask'] = np.nan #converted to 0 during export
+    df.loc[ df.index.isin(probes), 'quality_mask'] = np.nan #converted to 1 during export
     #LOGGER.info(f"DEBUG quality_mask: {df.shape}, {df['quality_mask'].isna().sum()} nan from {probes.shape} probes")
     return df
 

diff --git a/methylprep/version.py b/methylprep/version.py
@@ -2,4 +2,4 @@
 # 1) we don't load dependencies by storing it in __init__.py
 # 2) we can import it in setup.py for the same reason
 # 3) we can import it into your module module
-__version__ = '1.5.7'
+__version__ = '1.5.8'
diff --git a/requirements.txt b/requirements.txt
@@ -1,4 +1,5 @@
-numpy
+pyparsing<3
+snumpy
 pandas>=1.3.0
 scipy
 statsmodels

diff --git a/setup.py b/setup.py
@@ -36,6 +36,7 @@
     include_package_data=True,
     package_data={"":["*.txt.gz"]},
     install_requires=[
+        'pyparsing > 3.0',
         'numpy',
         'pandas >=1.3.0',
         'scipy',