Merge branch 'main' into multicova

.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.5.4
+current_version = 0.6.2
 commit = True
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<build>\d+))?

HISTORY.md

@@ -1,11 +1,24 @@
 # Changelog
 
-# 0.6.0
+
+# 0.6.3
 * ENH download metadata template in the GUI
 * ENH multicova analysis
 * ENH filter data completeness `dataset.preprocess(data_completeness=0.7)` 
 * ADD `GenericLoader` for not supported data formats
 
+# 0.6.2
+* FIX preprocessing with VST floats and inf
+* FIX plotly display
+
+# 0.6.1
+* FIX data loading
+
+# 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
 

README.md

@@ -24,6 +24,19 @@
 
 [link]:https://alphapeptstats.readthedocs.io/en/main/
 
+<div align = center>
+<br>
+<br>
+
+[<kbd> <br> Streamlit WebApp <br> </kbd>][link_streamlit]
+
+</div>
+
+<br>
+<br>
+
+[link_streamlit]:https://mannlabs-alphapeptstats-alphastatsguialphapeptstats-qyzgwd.streamlit.app/
+
 An open-source Python package for downstream mass spectrometry downstream data analysis from the [Mann Group at the University of Copenhagen](https://www.cpr.ku.dk/research/proteomics/mann/).
 
 

alphastats/DataSet.py

@@ -11,6 +11,8 @@
 from alphastats.loader.MaxQuantLoader import MaxQuantLoader
 from alphastats.loader.SpectronautLoader import SpectronautLoader
 from alphastats.loader.GenericLoader import GenericLoader
+from alphastats.loader.mzTabLoader import mzTabLoader
+
 
 from alphastats.DataSet_Plot import Plot
 from alphastats.DataSet_Preprocess import Preprocess
@@ -102,15 +104,18 @@ def _check_loader(self, loader):
             loader : loader
         """
         if not isinstance(
+
             loader,
             (
                 AlphaPeptLoader,
                 MaxQuantLoader,
                 DIANNLoader,
                 FragPipeLoader,
                 SpectronautLoader,
-                GenericLoader
+                GenericLoader,
+                mzTabLoader
             ),
+
         ):
             raise LoaderError(
                 "loader must be from class: AlphaPeptLoader, MaxQuantLoader, DIANNLoader, FragPipeLoader or SpectronautLoader"
@@ -159,8 +164,10 @@ def create_matrix(self):
         df.columns = df.columns.str.replace(substring_to_remove, "")
         # transpose dataframe
         mat = df.transpose()
+        mat.replace([np.inf, -np.inf], np.nan, inplace=True)
         # remove proteins with only zero
         self.mat = mat.loc[:, (mat != 0).any(axis=0)]
+        self.mat = self.mat.astype(float)
         # reset preproccessing info
         self.preprocessing_info = self._save_dataset_info()
         self.preprocessed = False
@@ -219,7 +226,7 @@ def overview(self):
         dataset_overview = (
             "Attributes of the DataSet can be accessed using: \n"
             + "DataSet.rawinput:\t Raw Protein data.\n"
-            + "DataSet.mat:\tProcessed data matrix with ProteinIDs/ProteinGroups as columns and samples as rows. All computations are performed on this matrix.\n"
+            + "DataSet.mat:\t\tProcessed data matrix with ProteinIDs/ProteinGroups as columns and samples as rows. All computations are performed on this matrix.\n"
             + "DataSet.metadata:\tMetadata for the samples in the matrix. Metadata will be matched with DataSet.mat when needed (for instance Volcano Plot)."
         )
         print(dataset_overview)

alphastats/DataSet_Plot.py

@@ -122,19 +122,22 @@ def plot_umap(self, group: str = None, circle: bool = False):
         )
         return dimensionality_reduction.plot
 
+
+    @ignore_warning(RuntimeWarning)
     def plot_volcano(
         self,
         group1: Union[str, list],
         group2: Union[str, list],
-        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,
+        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
 
@@ -149,6 +152,7 @@ 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.
 
 
@@ -174,8 +178,9 @@ def plot_volcano(
                 min_fc=min_fc,
                 alpha=alpha,
                 draw_line=draw_line,
-                perm=perm,
+                perm=perm, 
                 fdr=fdr,
+                color_list=color_list
             )
 
             return volcano_plot.plot
@@ -252,7 +257,7 @@ def plot_intensity(
             protein_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/DataSet_Preprocess.py

@@ -109,11 +109,11 @@ def _imputation(self, method: str):
         logging.info("Imputing data...")
 
         if method == "mean":
-            imp = sklearn.impute.SimpleImputer(missing_values=np.nan, strategy="mean")
+            imp = sklearn.impute.SimpleImputer(missing_values=np.nan, strategy="mean", keep_empty_features=True)
             imputation_array = imp.fit_transform(self.mat.values)
 
         elif method == "median":
-            imp = sklearn.impute.SimpleImputer(missing_values=np.nan, strategy="median")
+            imp = sklearn.impute.SimpleImputer(missing_values=np.nan, strategy="median", keep_empty_features=True)
             imputation_array = imp.fit_transform(self.mat.values)
 
         elif method == "knn":
@@ -173,7 +173,7 @@ def _normalization(self, method: str):
             )
 
         elif method == "vst":
-            scaler = sklearn.preprocessing.PowerTransformer()
+            scaler = sklearn.preprocessing.PowerTransformer(standardize=False)
             normalized_array = scaler.fit_transform(self.mat.values)
 
         else:
@@ -195,13 +195,14 @@ def reset_preprocessing(self):
         self.create_matrix()
         print("All preprocessing steps are reset.")
 
+    @ignore_warning(RuntimeWarning)
     def _compare_preprocessing_modes(self, func, params_for_func) -> list:
         dataset = self
-        imputation_methods = ["mean", "median", "knn"]
-        normalization_methods = ["zscore", "quantile", "vst"]
-        preprocessing_modes = list(
-            itertools.product(normalization_methods, imputation_methods)
-        )
+        imputation_methods = ["mean", "median", "knn", "randomforest"]
+        normalization_methods = ["vst","zscore", "quantile" ]
+
+        preprocessing_modes = list(itertools.product(normalization_methods, imputation_methods))
+
 
         results_list = []
 
@@ -211,9 +212,8 @@ def _compare_preprocessing_modes(self, func, params_for_func) -> list:
         for preprocessing_mode in preprocessing_modes:
             # reset preprocessing
             dataset.reset_preprocessing()
-            print(
-                f"Normalization {preprocessing_mode[0]}, Imputation {str(preprocessing_mode[1])}"
-            )
+            print(f"Normalization {preprocessing_mode[0]}, Imputation {str(preprocessing_mode[1])}")
+            dataset.mat.replace([np.inf, -np.inf], np.nan, inplace=True)
 
             dataset.preprocess(
                 subset=True,
@@ -223,6 +223,8 @@ def _compare_preprocessing_modes(self, func, params_for_func) -> list:
 
             res = func(**params_for_func)
             results_list.append(res)
+
+            print("\t")
 
         return results_list
 
@@ -305,15 +307,17 @@ def preprocess(
         if subset:
             self.mat = self._subset()
 
+
         if data_completeness> 0:
             self._remove_na_values(cut_off=data_completeness)
 
-        if log2_transform:
+        if log2_transform and self.preprocessing_info.get("Log2-transformed") is False:
             self._log2_transform()
 
         if normalization is not None:
             self._normalization(method=normalization)
-
+            self.mat = self.mat.replace([np.inf, -np.inf], np.nan)
+
         if imputation is not None:
             self._imputation(method=imputation)
 

alphastats/__init__.py

@@ -1,5 +1,5 @@
 __project__ = "alphastats"
-__version__ = "0.5.4"
+__version__ = "0.6.2"
 __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/DimensionalityReduction.py

@@ -134,6 +134,7 @@ def _plot(self, sample_names, group_color):
             labels=self.labels,
             color=group_color,
             hover_data=[components[self.dataset.sample]],
+            template="simple_white+alphastats_colors"
         )
 
         # rename hover_data_0 to sample

alphastats/plots/IntensityPlot.py

@@ -3,9 +3,30 @@
 import numpy as np
 import plotly.graph_objects as go
 import plotly.express as px
+import plotly
 
 from alphastats.plots.PlotUtils import plotly_object, PlotUtils
 
+plotly.io.templates["alphastats_colors"] = plotly.graph_objects.layout.Template(
+    layout=plotly.graph_objects.Layout(
+        paper_bgcolor="rgba(0,0,0,0)",
+        plot_bgcolor="rgba(0,0,0,0)",
+        colorway=[
+            "#009599",
+            "#005358",
+            "#772173",
+            "#B65EAF",  # pink
+            "#A73A00",
+            "#6490C1",
+            "#FF894F",
+            "#2B5E8B",
+            "#A87F32",
+        ],
+    )
+)
+
+plotly.io.templates.default = "simple_white+alphastats_colors"
+
 
 class IntensityPlot(PlotUtils):
     def __init__(
@@ -122,29 +143,26 @@ def _prepare_data(self):
     def _plot(self):
         if self.method == "violin":
             fig = px.violin(
-                self.prepared_df,
-                y=self.protein_id,
-                x=self.group,
-                color=self.group,
-                labels={self.protein_id: self.y_label},
+                self.prepared_df, y=self.protein_id, x=self.group, color=self.group, labels={self.protein_id: self.y_label},
+                template="simple_white+alphastats_colors"
             )
 
         elif self.method == "box":
             fig = px.box(
-                self.prepared_df,
-                y=self.protein_id,
-                x=self.group,
-                color=self.group,
-                labels={self.protein_id: self.y_label},
+                self.prepared_df, y=self.protein_id, x=self.group, color=self.group, labels={self.protein_id: self.y_label},
+                template="simple_white+alphastats_colors"
             )
 
         elif self.method == "scatter":
             fig = px.scatter(
-                self.prepared_df,
-                y=self.protein_id,
-                x=self.group,
-                color=self.group,
-                labels={self.protein_id: self.y_label},
+                 self.prepared_df, y=self.protein_id, x=self.group, color=self.group, labels={self.protein_id: self.y_label},
+                 template="simple_white+alphastats_colors"
+            )
+
+        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", template= "simple_white+alphastats_colors"
             )
 
         else:

alphastats/plots/PlotUtils.py

@@ -3,6 +3,27 @@
 import plotly.graph_objects as go
 
 
+plotly.io.templates["alphastats_colors"] = plotly.graph_objects.layout.Template(
+    layout=plotly.graph_objects.Layout(
+        paper_bgcolor="rgba(0,0,0,0)",
+        plot_bgcolor="rgba(0,0,0,0)",
+        colorway=[
+            "#009599",
+            "#005358",
+            "#772173",
+            "#B65EAF",  # pink
+            "#A73A00",
+            "#6490C1",
+            "#FF894F",
+            "#2B5E8B",
+            "#A87F32",
+        ],
+    )
+)
+
+plotly.io.templates.default = "simple_white+alphastats_colors"
+
+
 class PlotUtils:
     def __init__(self) -> None:
         pass