Merge pull request #197 from MannLabs/revision

Release 0.6.0 Revision

Author: elena-krismer

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.5.4
+current_version = 0.6.0
 commit = True
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<build>\d+))?

HISTORY.md

@@ -1,4 +1,7 @@
 # Changelog
+# 0.6.0
+* ADD mzTAB support
+* ENH color Volcano Plot data points using list of protein names `color_list=your_protein_list`
 
 # 0.5.4
 * FIX altair version - binning of streamlit version

alphastats/DataSet.py

@@ -10,6 +10,7 @@
 from alphastats.loader.FragPipeLoader import FragPipeLoader
 from alphastats.loader.MaxQuantLoader import MaxQuantLoader
 from alphastats.loader.SpectronautLoader import SpectronautLoader
+from alphastats.loader.mzTabLoader import mzTabLoader
 
 from alphastats.DataSet_Plot import Plot
 from alphastats.DataSet_Preprocess import Preprocess
@@ -96,7 +97,7 @@ def _check_loader(self, loader):
             loader : loader
         """
         if not isinstance(
-            loader, (AlphaPeptLoader, MaxQuantLoader, DIANNLoader, FragPipeLoader, SpectronautLoader)
+            loader, (AlphaPeptLoader, MaxQuantLoader, DIANNLoader, FragPipeLoader, SpectronautLoader, mzTabLoader)
         ):
             raise LoaderError(
                 "loader must be from class: AlphaPeptLoader, MaxQuantLoader, DIANNLoader, FragPipeLoader or SpectronautLoader"

alphastats/DataSet_Plot.py

@@ -119,15 +119,16 @@ def plot_volcano(
         self,
         group1,
         group2,
-        column=None,
-        method="ttest",
-        labels=False,
-        min_fc=1,
-        alpha=0.05,
-        draw_line=True,
-        perm=100, 
-        fdr=0.05,
-        compare_preprocessing_modes=False
+        column:str=None,
+        method:str="ttest",
+        labels:bool=False,
+        min_fc:float=1.0,
+        alpha:float=0.05,
+        draw_line:bool=True,
+        perm:int=100, 
+        fdr:float=0.05,
+        compare_preprocessing_modes:bool=False,
+        color_list:list=[]
     ):
         """Plot Volcano Plot
 
@@ -142,6 +143,7 @@ def plot_volcano(
             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.
+            color_list (list): list with ProteinIDs that should be highlighted.
             compare_preprocessing_modes(bool): Will iterate through normalization and imputation modes and return a list of VolcanoPlots in different settings, Default False.
 
 
@@ -166,7 +168,8 @@ def plot_volcano(
                 alpha=alpha,
                 draw_line=draw_line,
                 perm=perm, 
-                fdr=fdr
+                fdr=fdr,
+                color_list=color_list
             )
 
             return volcano_plot.plot
@@ -241,7 +244,7 @@ def plot_intensity(
             ID (str): ProteinGroup ID
             group (str, optional): A metadata column used for grouping. Defaults to None.
             subgroups (list, optional): Select variables from the group column. Defaults to None.
-            method (str, optional):  Violinplot = "violin", Boxplot = "box", Scatterplot = "scatter". Defaults to "box".
+            method (str, optional):  Violinplot = "violin", Boxplot = "box", Scatterplot = "scatter" or "all". Defaults to "box".
             add_significance (bool, optional): add p-value bar, only possible when two groups are compared. Defaults False.
             log_scale (bool, optional): yaxis in logarithmic scale. Defaults to False.
 

alphastats/__init__.py

@@ -1,5 +1,5 @@
 __project__ = "alphastats"
-__version__ = "0.5.4"
+__version__ = "0.6.0"
 __license__ = "Apache"
 __description__ = "An open-source Python package for Mass Spectrometry Analysis"
 __author__ = "Mann Labs"

alphastats/loader/mzTabLoader.py

@@ -0,0 +1,29 @@
+from pyteomics import mztab
+from alphastats.loader.BaseLoader import BaseLoader
+
+class mzTabLoader(BaseLoader):
+    def __init__(self, file, intensity_column: str="protein_abundance_[sample]", index_column:str="accession"):
+        """Load mzTab file. Will add contamination column for further analysis.
+
+        Args:
+            file (str): path to mzTab file.
+            intensity_column (str, optional):  columns where the intensity of the proteins are given.. Defaults to "protein_abundance_[sample]".
+            index_column (str, optional): column indicating the protein groups.  Defaults to "accession".
+        """
+        self.filter_columns = []
+        self.gene_names = None
+        self.intensity_column = intensity_column
+        self.index_column = index_column
+        self.confidence_column = None
+        self.evidence_df = None
+        self.gene_names = None
+        self._load_protein_table(file=file)
+        self._add_contamination_column()
+
+
+    def _load_protein_table(self, file):
+        tables = mztab.MzTab(file)
+        self.rawinput = tables.protein_table
+        self.mztab_metadata = tables.metadata
+        self.software = tables.protein_table.search_engine[0]
+        

alphastats/plots/IntensityPlot.py

@@ -128,6 +128,12 @@ def _plot(self):
                  self.prepared_df, y=self.protein_id, x=self.group, color=self.group, labels={self.protein_id: self.y_label}
             )
 
+        elif self.method == "all":
+            fig = px.violin(
+                 self.prepared_df, y=self.protein_id, x=self.group, color=self.group, labels={self.protein_id: self.y_label},
+                 box=True, points="all"
+            )
+
         else:
             raise ValueError(
                 f"{self.method} is not available."

alphastats/plots/VolcanoPlot.py

@@ -15,7 +15,8 @@ def __init__(
         column=None, method=None, 
         labels=None, min_fc=None, 
         alpha=None, draw_line=None, 
-        plot=True, perm=100, fdr=0.05
+        plot=True, perm=100, fdr=0.05,
+        color_list=[]
     ):  
         self.dataset = dataset
         self.group1 = group1
@@ -31,6 +32,7 @@ def __init__(
         self.res = None
         self.pvalue_column = None
         self.perm=perm
+        self.color_list = color_list
         self._check_input()
 
         if plot:
@@ -266,6 +268,10 @@ def _annotate_result_df(self):
 
         value = ["down", "up"]
         self.res["color"] = np.select(condition, value, default="non_sig")   
+
+        if len(self.color_list) > 0:
+            self.res["color"] = np.where(self.res[self.dataset.index_column].isin(self.color_list), 
+                                          "color", "no_color")   
 
 
     def _add_labels_plot(self):
@@ -327,6 +333,17 @@ def _draw_fdr_line(self):
             line_shape='spline',
             showlegend=False)
         )
+    
+    def _color_data_points(self):
+         # update coloring
+        if len(self.color_list) == 0:
+            color_dict = {"non_sig": "#404040", "up": "#B65EAF", "down": "#009599"}
+    
+        else:
+            color_dict = {"no_color": "#404040", "color": "#B65EAF"}
+        
+        self.plot = self._update_colors_plotly(self.plot, color_dict=color_dict)
+
 
 
     def _plot(self):
@@ -339,8 +356,7 @@ def _plot(self):
         )
 
         # update coloring
-        color_dict = {"non_sig": "#404040", "up": "#B65EAF", "down": "#009599"}
-        self.plot = self._update_colors_plotly(self.plot, color_dict=color_dict)
+        self._color_data_points()
 
         if self.labels:
             self._add_labels_plot()

docs/Introduction.md

@@ -7,6 +7,12 @@
 
 AlphaPeptStats is an open-source package for analyzing mass spectrometry-based proteomics data.
 
-AlphaPeptStats was developed to simplify and standardize the process of analyzing complex datasets. Hereby AlphaPeptStats supports proteomics data generated by `AlphaPept`, `DIA-NN`, `FragPipe`, `MaxQuant` and `Spectronaut`. The tool allows a structured workflow from importing data, preprocessing data to visualization.
+AlphaPeptStats was developed to simplify and standardize the process of analyzing complex datasets. Hereby AlphaPeptStats supports proteomics data generated by `AlphaPept`, `DIA-NN`, `FragPipe`, `MaxQuant` and `Spectronaut` and quantiative proteomics results in `mzTab` format. The tool allows a structured workflow from importing data, preprocessing data to visualization.
 
 AlphaPeptStats was developed by the [Mann Group at the University of Copenhagen](https://www.biochem.mpg.de/mann) and is freely available with an [Apache License](LICENSE.txt). External Python packages (available in the [requirements](requirements) folder) have their own licenses, which can be consulted on their respective websites.
+
+The workflow consists of:
+1. Import of proteomics data 
+2. Creation of the DataSet, consisting of the imported proteomics data and metadata
+3. Data preprocessing (optional)
+4. Statistical Analysis and Visualization

docs/api_reference/loader.rst

@@ -32,3 +32,9 @@ SpectronautLoader
 .. automodule:: alphastats.loader.SpectronautLoader
     :members:
     :undoc-members:
+
+mzTabLoader
+~~~~~~~~~~~~~~~~~~~~~~
+.. automodule:: alphastats.loader.mzTabLoader
+    :members:
+    :undoc-members: