Merge pull request #126 from MannLabs/multicova

add option for compare preprocessing

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.4.1
+current_version = 0.4.2
 commit = True
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<build>\d+))?

.github/workflows/python-package.yml

@@ -31,6 +31,7 @@ jobs:
         python -m pip install --upgrade pip
         pip install coverage
         pip install wheel
+        pip install dictdiffer
     - name: Test installation
       run: |
         pip install -e.

HISTORY.md

@@ -1,5 +1,9 @@
 # Changelog
 
+# 0.4.2
+* ADD option compare_preprocessing_modes
+* update to streamlit 1.19 with new caching functions
+
 # 0.4.1
 * FIX bug when drawing FDR lines
 * ADD functionionality `dataset.reset_preprocessing()` to reset all preprocessing steps

alphastats/DataSet_Plot.py

@@ -125,7 +125,8 @@ def plot_volcano(
         alpha=0.05,
         draw_line=True,
         perm=100, 
-        fdr=0.05
+        fdr=0.05,
+        compare_preprocessing_modes=False
     ):
         """Plot Volcano Plot
 
@@ -136,31 +137,38 @@ def plot_volcano(
             method (str): "anova", "wald", "ttest", "SAM" Defaul ttest.
             labels (bool): Add text labels to significant Proteins, Default False.
             alpha(float,optional): p-value cut off.
-            min_fc (float): Minimum fold change
+            min_fc (float): Minimum fold change.
             draw_line(boolean): whether to draw cut off lines.
             perm(float,optional): number of permutations when using SAM as method. Defaults to 100.
             fdr(float,optional): FDR cut off when using SAM as method. Defaults to 0.05.
+            compare_preprocessing_modes(bool): Will iterate through normalization and imputation modes and return a list of VolcanoPlots in different settings, Default False.
 
 
         Returns:
             plotly.graph_objects._figure.Figure: Volcano Plot
         """
 
-        volcano_plot = VolcanoPlot(
-            dataset=self,
-            group1=group1,
-            group2=group2,
-            column=column,
-            method=method,
-            labels=labels,
-            min_fc=min_fc,
-            alpha=alpha,
-            draw_line=draw_line,
-            perm=perm, 
-            fdr=fdr
-        )
+        if compare_preprocessing_modes:
+            params_for_func = locals()
+            results = self._compare_preprocessing_modes(func=VolcanoPlot,params_for_func=params_for_func)
+            return results
+
+        else:
+            volcano_plot = VolcanoPlot(
+                dataset=self,
+                group1=group1,
+                group2=group2,
+                column=column,
+                method=method,
+                labels=labels,
+                min_fc=min_fc,
+                alpha=alpha,
+                draw_line=draw_line,
+                perm=perm, 
+                fdr=fdr
+            )
 
-        return volcano_plot.plot
+            return volcano_plot.plot
 
     def plot_correlation_matrix(self, method="pearson"):
         """Plot Correlation Matrix
@@ -224,6 +232,7 @@ def plot_intensity(
         method="box",
         add_significance=False,
         log_scale=False,
+        compare_preprocessing_modes=False
     ):
         """Plot Intensity of individual Protein/ProteinGroup
 
@@ -238,6 +247,11 @@ def plot_intensity(
         Returns:
             plotly.graph_objects._figure.Figure: Plotly Plot
         """
+        if compare_preprocessing_modes:
+            params_for_func = locals()
+            results = self._compare_preprocessing_modes(func=IntensityPlot,params_for_func=params_for_func)
+            return results
+
         intensity_plot = IntensityPlot(
             dataset = self,
             protein_id=protein_id,

alphastats/DataSet_Preprocess.py

@@ -7,6 +7,7 @@
 import sklearn.impute
 from alphastats.utils import ignore_warning
 from sklearn.experimental import enable_iterative_imputer
+import itertools
 
 
 class Preprocess:
@@ -165,6 +166,34 @@ def reset_preprocessing(self):
         # reset all preprocessing steps
         self.create_matrix()
         print("All preprocessing steps are reset.")
+
+    def _compare_preprocessing_modes(self, func, params_for_func):
+        dataset = self
+        imputation_methods = ["mean", "median", "knn"]
+        normalization_methods = ["zscore", "quantile", "vst"]
+        preprocessing_modes = list(itertools.product(normalization_methods, imputation_methods))
+
+        results_list = []
+
+        del params_for_func["compare_preprocessing_modes"]
+        params_for_func["dataset"] = params_for_func.pop("self")
+
+        for preprocessing_mode in preprocessing_modes:
+            # reset preprocessing
+            dataset.reset_preprocessing()
+            print(f"Normalization {preprocessing_mode[0]}, Imputation {str(preprocessing_mode[1])}")
+
+            dataset.preprocess(
+                subset=True,
+                normalization = preprocessing_mode[0],
+                imputation = preprocessing_mode[1]
+            )
+
+            res = func(**params_for_func)
+            results_list.append(res)
+
+        return results_list
+
 
     @ignore_warning(RuntimeWarning)
     def preprocess(

alphastats/__init__.py

@@ -1,5 +1,5 @@
 __project__ = "alphastats"
-__version__ = "0.4.1"
+__version__ = "0.4.2"
 __license__ = "Apache"
 __description__ = "An open-source Python package for Mass Spectrometry Analysis"
 __author__ = "Mann Labs"

alphastats/gui/AlphaPeptStats.py

@@ -10,6 +10,15 @@
 
 # centering with streamlit is not really centered
 
+from streamlit.runtime import get_instance
+from streamlit.runtime.scriptrunner.script_run_context import get_script_run_ctx
+
+runtime = get_instance()
+session_id = get_script_run_ctx().session_id
+session_info = runtime._session_mgr.get_session_info(session_id)
+
+user_session_id = session_id
+st.session_state["user_session_id"] = user_session_id
 
 img_center = """
 <head>  

alphastats/gui/pages/02_Import Data.py

@@ -18,6 +18,7 @@
     from alphastats import DataSet
 
 
+
 import pandas as pd
 import plotly.express as px
 
@@ -140,15 +141,14 @@ def select_sample_column_metadata(df, software):
 
 def upload_softwarefile(software):
 
-    st.file_uploader(
+    softwarefile = st.file_uploader(
         print_software_import_info(software=software),
-        key="softwarefile",
         type=["csv", "tsv", "txt", "hdf"],
     )
 
-    if st.session_state.softwarefile is not None:
+    if softwarefile is not None:
 
-        softwarefile_df = read_uploaded_file_into_df(st.session_state.softwarefile)
+        softwarefile_df = read_uploaded_file_into_df(softwarefile)
         # display head a protein data
 
         check_software_file(softwarefile_df, software)
@@ -302,6 +302,10 @@ def empty_session_state():
         del st.session_state[key]
     st.empty()
 
+    from streamlit.runtime import get_instance
+    from streamlit.runtime.scriptrunner.script_run_context import get_script_run_ctx
+    user_session_id = get_script_run_ctx().session_id
+    st.session_state["user_session_id"] = user_session_id
 
 sidebar_info()
 

alphastats/gui/pages/03_Data Overview.py

@@ -3,13 +3,13 @@
 import plotly.express as px
 
 
-@st.cache
-def convert_df(df):
+@st.cache_data
+def convert_df(df, user_session_id = st.session_state.user_session_id):
     return df.to_csv().encode("utf-8")
 
 
-@st.cache
-def get_display_matrix():
+@st.cache_data
+def get_display_matrix(user_session_id = st.session_state.user_session_id):
 
     processed_df = pd.DataFrame(
         st.session_state.dataset.mat.values,

alphastats/gui/pages/04_Analysis.py

@@ -75,8 +75,8 @@ def download_figure(obj, format, plotting_library="plotly"):
     st.download_button(label="Download as " + format, data=buffer, file_name=filename)
 
 
-@st.cache
-def convert_df(df):
+@st.cache_data
+def convert_df(df, user_session_id = st.session_state.user_session_id):
     return df.to_csv().encode("utf-8")
 
 

alphastats/gui/pages/06_Results.py

@@ -22,8 +22,8 @@ def save_plotly(plot, format):
     )
 
 
-@st.cache
-def convert_df(df):
+@st.cache_data
+def convert_df(df, user_session_id = st.session_state.user_session_id):
     return df.to_csv().encode("utf-8")
 
 

alphastats/plots/VolcanoPlot.py

@@ -11,7 +11,11 @@
 
 class VolcanoPlot(PlotUtils):
     def __init__(
-        self, dataset, group1, group2, column=None, method=None, labels=None, min_fc=None, alpha=None, draw_line=None, plot=True, perm=100, fdr=0.05
+        self, dataset, group1, group2, 
+        column=None, method=None, 
+        labels=None, min_fc=None, 
+        alpha=None, draw_line=None, 
+        plot=True, perm=100, fdr=0.05
     ):  
         self.dataset = dataset
         self.group1 = group1

docs/conf.py

@@ -23,7 +23,7 @@
 author = "Elena Krismer"
 
 # The full version, including alpha/beta/rc tags
-release = "0.4.1"
+release = "0.4.2"
 
 
 # -- General configuration ---------------------------------------------------

package.Dockerfile

@@ -0,0 +1,20 @@
+FROM python:3.9-slim-buster
+
+# Set the working directory
+WORKDIR /app
+
+# Install any necessary system dependencies here
+RUN apt-get update && apt-get install -y \
+    libpq-dev \
+    gcc \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN git clone https://github.com/MannLabs/alphapeptstats.git .
+# Install Python packages and any other dependencies
+RUN pip3 install -e.
+
+# Copy the Python package files into the container
+COPY alphapeptstats /app/alphapeptstats
+
+# Set the entrypoint command to the package's setup.py file
+ENTRYPOINT ["python", "-m", "setup.py"]

release/one_click_linux_gui/control

@@ -1,5 +1,5 @@
 Package: alphastats
-Version: 0.4.1
+Version: 0.4.2
 Architecture: all
 Maintainer: MannLabs
 Description: alphastats

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/alphastats-0.4.1-py3-none-any.whl"
+pip install "../../dist/alphastats-0.4.2-py3-none-any.whl"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==5.8

release/one_click_macos_gui/Info.plist

@@ -9,9 +9,9 @@
 	<key>CFBundleIconFile</key>
 	<string>alphapeptstats_logo.icns</string>
 	<key>CFBundleIdentifier</key>
-	<string>alphastats.0.4.1</string>
+	<string>alphastats.0.4.2</string>
 	<key>CFBundleShortVersionString</key>
-	<string>0.4.1</string>
+	<string>0.4.2</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/alphastats-0.4.1-py3-none-any.whl"
+pip install "../../dist/alphastats-0.4.2-py3-none-any.whl"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==5.8

release/one_click_windows_gui/alphastats_innoinstaller.iss

@@ -2,7 +2,7 @@
 ; SEE THE DOCUMENTATION FOR DETAILS ON CREATING INNO SETUP SCRIPT FILES!
 
 #define MyAppName "AlphaPeptStats"
-#define MyAppVersion "0.4.1"
+#define MyAppVersion "0.4.2"
 #define MyAppPublisher "MannLabs"
 #define MyAppURL "https://github.com/MannLabs/alphapeptstats"
 #define MyAppExeName "alphastats_gui.exe"

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/alphastats-0.4.1-py3-none-any.whl"
+pip install "../../dist/alphastats-0.4.2-py3-none-any.whl"
 
 # Creating the stand-alone pyinstaller folder
 pip install pyinstaller==5.8

requirements.txt

@@ -12,7 +12,7 @@ tqdm>=4.64.0
 diffxpy==0.7.4
 anndata==0.8.0
 umap-learn==0.5.3
-streamlit==1.17.0
+streamlit==1.19.0
 tables==3.7.0
 numpy==1.23.5
 numba==0.56.4

setup.py

@@ -25,7 +25,7 @@ def create_pip_wheel():
     requirements = get_requirements()
     setuptools.setup(
         name="alphastats",
-        version="0.4.1",
+        version="0.4.2",
         license="Apache",
         description="An open-source Python package for Mass Spectrometry Analysis",
         long_description=get_long_description(),