Merge pull request #31 from MannLabs/development

Python >3.9 compatibility and some refactoring for expressiveness

directlfq/__init__.py

@@ -2,7 +2,7 @@
 
 
 __project__ = "directlfq"
-__version__ = "0.2.17"
+__version__ = "0.2.18"
 __license__ = "Apache"
 __description__ = "An open-source Python package of the AlphaPept ecosystem"
 __author__ = "Mann Labs"
@@ -13,7 +13,7 @@
     "software",
     "AlphaPept ecosystem",
 ]
-__python_version__ = ">=3.8,<3.10"
+__python_version__ = ">=3.8"
 __classifiers__ = [
     # "Development Status :: 1 - Planning",
     # "Development Status :: 2 - Pre-Alpha",

directlfq/lfq_manager.py

@@ -45,6 +45,8 @@ def run_lfq(input_file,  columns_to_add = [], selected_proteins_file :str = None
     filter_dict = load_filter_dict_if_given_as_yaml(filter_dict)
     input_file = lfqutils.add_mq_protein_group_ids_if_applicable_and_obtain_annotated_file(input_file, input_type_to_use,mq_protein_groups_txt, columns_to_add)
     input_df = lfqutils.import_data(input_file=input_file, input_type_to_use=input_type_to_use, filter_dict=filter_dict)
+
+    input_df = lfqutils.sort_input_df_by_protein_id(input_df)
     input_df = lfqutils.index_and_log_transform_input_df(input_df)
     input_df = lfqutils.remove_allnan_rows_input_df(input_df)
 

directlfq/normalization.py

@@ -407,11 +407,11 @@ def _shift_to_reference_sample(self, row_idx):
     @staticmethod
     def _calc_distance(samples_1, samples_2):
         distrib = get_fcdistrib(samples_1, samples_2)
-        distance = np.nanmedian(distrib)
-        if np.isnan(distance):
+        is_all_nan = np.all(np.isnan(distrib))
+        if is_all_nan:
             return np.nan
         else:
-            return distance
+            return np.nanmedian(distrib)
 
 
     def _update_ion_dataframe(self):

directlfq/protein_intensity_estimation.py

@@ -26,7 +26,17 @@
 
 
 def estimate_protein_intensities(normed_df, min_nonan, num_samples_quadratic, num_cores):
-    "derives protein pseudointensities from between-sample normalized data"
+    """"Performs the LFQ protein intensity estimation. The input is a normalized ion intensity dataframe. The output is the lfq protein intensity dataframe and the lfq ion intensity dataframe.
+
+    Args:
+        normed_df (pd.DataFrame): A pandas dataframe that contains the multi-index of [config.PROTEIN_ID, config.QUANT_ID]. The columns are the sample names and the values are the ion intensities. Zero value are replaced with NaNs and the dataframe is subsequently log-transformed. The dataframe needs to be sorted by the protein IDs, as the subsequent functions assume this.
+        min_nonan (int): minimum number of NaNs
+        num_samples_quadratic (int): minimum number of samples to use for the quadratic normalization
+        num_cores (int): number of cores to use for the multiprocessing. If set to 1, the processing is done sequentially.
+
+    Returns:
+        tuple[protein_intensity_df, ion_intensity_df]: protein intensity dataframe and an ion intensity dataframe. The ion intensity dataframe is only compiled if the config.COMPILE_NORMALIZED_ION_TABLE is set to True.
+    """
 
     allprots = list(normed_df.index.get_level_values(0).unique())
     LOGGER.info(f"{len(allprots)} lfq-groups total")

directlfq/utils.py

@@ -327,6 +327,9 @@ def index_and_log_transform_input_df(data_df):
 def remove_allnan_rows_input_df(data_df):
     return data_df.dropna(axis = 0, how = 'all')
 
+def sort_input_df_by_protein_id(data_df):
+    return data_df.sort_values(by = config.PROTEIN_ID,ignore_index=True)
+
 
 
 

misc/bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.2.17
+current_version = 0.2.18
 commit = True
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<build>\d+))?

nbdev_nbs/02_normalization.ipynb

@@ -1360,6 +1360,72 @@
     "test_that_profiles_with_noise_are_close()\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test 2: One array is entirely NaN\n",
+      "Test 1: Both arrays are non-NaN and identical\n",
+      "Test 3: Arrays with some NaN values\n",
+      "Test 4: Arrays with different values but no NaNs\n",
+      "Test 5: Empty arrays\n",
+      "All tests passed!\n"
+     ]
+    }
+   ],
+   "source": [
+    "import directlfq.normalization as lfq_norm\n",
+    "import numpy as np\n",
+    "\n",
+    "def _calc_distance(samples_1, samples_2):\n",
+    "    distrib = lfq_norm.get_fcdistrib(samples_1, samples_2)\n",
+    "    is_all_nan = np.all(np.isnan(distrib))\n",
+    "    if is_all_nan:\n",
+    "        return np.nan\n",
+    "    else:\n",
+    "        return np.nanmedian(distrib)\n",
+    "\n",
+    "def test_calc_distance():\n",
+    "    print(\"Test 2: One array is entirely NaN\")\n",
+    "    samples_1 = np.array([np.nan, np.nan, np.nan])\n",
+    "    samples_2 = np.array([1, 2, 3])\n",
+    "    assert np.isnan(lfq_norm.SampleShifterLinear._calc_distance(samples_1, samples_2)) == True, \"Test 2 failed: Expected NaN for an array entirely of NaNs\"\n",
+    "    \n",
+    "\n",
+    "    print(\"Test 1: Both arrays are non-NaN and identical\")\n",
+    "    samples_1 = np.array([1, 2, 3])\n",
+    "    samples_2 = np.array([1, 2, 3])\n",
+    "    assert np.isnan(_calc_distance(samples_1, samples_2)) == False, \"Test 1 failed: Expected a non-NaN result for identical arrays\"\n",
+    "    assert lfq_norm.SampleShifterLinear._calc_distance(samples_1, samples_2) == 0, \"Test 1 failed: Expected a distance of 0 for identical arrays\"\n",
+    "    \n",
+    "\n",
+    "    print(\"Test 3: Arrays with some NaN values\")\n",
+    "    samples_1 = np.array([1, np.nan, 3])\n",
+    "    samples_2 = np.array([13, 2, np.nan])\n",
+    "    assert np.isnan(_calc_distance(samples_1, samples_2)) == False, \"Test 3 failed: Expected a valid number even with some NaNs\"\n",
+    "    assert lfq_norm.SampleShifterLinear._calc_distance(samples_1, samples_2) == -12 , \"Test 3 failed: Expected a distance of -12\"\n",
+    "    \n",
+    "    print(\"Test 4: Arrays with different values but no NaNs\")\n",
+    "    samples_1 = np.array([1, 4, 7])\n",
+    "    samples_2 = np.array([2, 5, 8])\n",
+    "    assert lfq_norm.SampleShifterLinear._calc_distance(samples_1, samples_2) != 0, \"Test 4 failed: Expected a non-zero distance for different values\"\n",
+    "    \n",
+    "    print(\"Test 5: Empty arrays\")\n",
+    "    samples_1 = np.array([])\n",
+    "    samples_2 = np.array([])\n",
+    "    assert np.isnan(lfq_norm.SampleShifterLinear._calc_distance(samples_1, samples_2)) == True, \"Test 5 failed: Expected NaN for empty arrays\"\n",
+    "\n",
+    "    print(\"All tests passed!\")\n",
+    "\n",
+    "# Run the test function\n",
+    "test_calc_distance()"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -1422,6 +1488,18 @@
    "display_name": "directlfq",
    "language": "python",
    "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.17"
   }
  },
  "nbformat": 4,

release/one_click_linux_gui/control

@@ -1,5 +1,5 @@
 Package: directlfq
-Version: 0.2.17
+Version: 0.2.18
 Architecture: all
 Maintainer: Mann Labs <[email protected]>
 Description: directlfq

release/one_click_linux_gui/create_installer_linux.sh

@@ -17,7 +17,7 @@ python setup.py sdist bdist_wheel
 # Setting up the local package
 cd release/one_click_linux_gui
 # Make sure you include the required extra packages and always use the stable or very-stable options!
-pip install "../../dist/directlfq-0.2.17-py3-none-any.whl[stable, gui]"
+pip install "../../dist/directlfq-0.2.18-py3-none-any.whl[stable, gui]"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==4.10

release/one_click_macos_gui/Info.plist

@@ -9,9 +9,9 @@
 	<key>CFBundleIconFile</key>
 	<string>alpha_logo.icns</string>
 	<key>CFBundleIdentifier</key>
-	<string>directlfq.0.2.17</string>
+	<string>directlfq.0.2.18</string>
 	<key>CFBundleShortVersionString</key>
-	<string>0.2.17</string>
+	<string>0.2.18</string>
 	<key>CFBundleInfoDictionaryVersion</key>
 	<string>6.0</string>
 	<key>CFBundleName</key>

release/one_click_macos_gui/create_installer_macos.sh

@@ -20,7 +20,7 @@ python setup.py sdist bdist_wheel
 
 # Setting up the local package
 cd release/one_click_macos_gui
-pip install "../../dist/directlfq-0.2.17-py3-none-any.whl[stable, gui]"
+pip install "../../dist/directlfq-0.2.18-py3-none-any.whl[stable, gui]"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==4.10
@@ -40,5 +40,5 @@ cp ../../LICENSE Resources/LICENSE
 cp ../logos/alpha_logo.png Resources/alpha_logo.png
 chmod 777 scripts/*
 
-pkgbuild --root dist/directlfq --identifier de.mpg.biochem.directlfq.app --version 0.2.17 --install-location /Applications/directlfq.app --scripts scripts directlfq.pkg
+pkgbuild --root dist/directlfq --identifier de.mpg.biochem.directlfq.app --version 0.2.18 --install-location /Applications/directlfq.app --scripts scripts directlfq.pkg
 productbuild --distribution distribution.xml --resources Resources --package-path directlfq.pkg dist/directlfq_gui_installer_macos.pkg

release/one_click_macos_gui/distribution.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="utf-8" standalone="no"?>
 <installer-script minSpecVersion="1.000000">
-    <title>directlfq 0.2.17</title>
+    <title>directlfq 0.2.18</title>
     <background mime-type="image/png" file="alpha_logo.png" scaling="proportional"/>
     <welcome file="welcome.html" mime-type="text/html" />
     <conclusion file="conclusion.html" mime-type="text/html" />

release/one_click_windows_gui/create_installer_windows.sh

@@ -17,7 +17,7 @@ python setup.py sdist bdist_wheel
 # Setting up the local package
 cd release/one_click_windows_gui
 # Make sure you include the required extra packages and always use the stable or very-stable options!
-pip install "../../dist/directlfq-0.2.17-py3-none-any.whl[stable, gui]"
+pip install "../../dist/directlfq-0.2.18-py3-none-any.whl[stable, gui]"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==4.10

release/one_click_windows_gui/directlfq_innoinstaller.iss

@@ -2,7 +2,7 @@
 ; SEE THE DOCUMENTATION FOR DETAILS ON CREATING INNO SETUP SCRIPT FILES!
 
 #define MyAppName "directlfq"
-#define MyAppVersion "0.2.17"
+#define MyAppVersion "0.2.18"
 #define MyAppPublisher "Max Planck Institute of Biochemistry and the University of Copenhagen, Mann Labs"
 #define MyAppURL "https://github.com/MannLabs/directlfq"
 #define MyAppExeName "directlfq_gui.exe"

settings.ini

@@ -13,7 +13,7 @@ author = Constantin Ammar
 author_email = [email protected]
 copyright = fast.ai
 branch = master
-version = 0.2.17
+version = 0.2.18
 min_python = 3.6
 audience = Developers
 language = English