298 histogram of imputed and original values

protzilla/constants/workflow_meta.json

@@ -498,6 +498,15 @@
                 "Protein ID"
               ],
               "default": "None"
+            },
+            "visual_transformation": {
+              "name": "Visual Transformation:",
+              "type": "categorical",
+              "categories": [
+                "linear",
+                "log10"
+              ],
+              "default": "linear"
             }
           },
           {
@@ -545,6 +554,15 @@
                 "Protein ID"
               ],
               "default": "None"
+            },
+            "visual_transformation": {
+              "name": "Visual Transformation:",
+              "type": "categorical",
+              "categories": [
+                "linear",
+                "log10"
+              ],
+              "default": "linear"
             }
           },
           {
@@ -592,6 +610,15 @@
                 "Protein ID"
               ],
               "default": "None"
+            },
+            "visual_transformation": {
+              "name": "Visual Transformation:",
+              "type": "categorical",
+              "categories": [
+                "linear",
+                "log10"
+              ],
+              "default": "linear"
             }
           },
           {
@@ -642,6 +669,15 @@
                 "Protein ID"
               ],
               "default": "None"
+            },
+            "visual_transformation": {
+              "name": "Visual Transformation:",
+              "type": "categorical",
+              "categories": [
+                "linear",
+                "log10"
+              ],
+              "default": "linear"
             }
           },
           {
@@ -689,6 +725,15 @@
                 "Protein ID"
               ],
               "default": "None"
+            },
+            "visual_transformation": {
+              "name": "Visual Transformation:",
+              "type": "categorical",
+              "categories": [
+                "linear",
+                "log10"
+              ],
+              "default": "linear"
             }
           },
           {

protzilla/data_preprocessing/imputation.py

@@ -262,7 +262,6 @@ def by_normal_distribution_sampling(
         transformed_df = long_to_wide(intensity_df)
         # iterate over all protein groups
         for protein_grp in transformed_df.columns:
-
             number_of_nans = transformed_df[protein_grp].isnull().sum()
 
             # don't impute values if there not enough values (> 1) to sample from
@@ -327,50 +326,116 @@ def by_normal_distribution_sampling(
 
 
 def by_knn_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
 def by_normal_distribution_sampling_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
 def by_simple_imputer_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
 def by_min_per_sample_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
 def by_min_per_protein_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
 def by_min_per_dataset_plot(
-    df, result_df, current_out, graph_type, graph_type_quantities, group_by
+    df,
+    result_df,
+    current_out,
+    graph_type,
+    graph_type_quantities,
+    group_by,
+    visual_transformation,
 ):
     return _build_box_hist_plot(
-        df, result_df, graph_type, graph_type_quantities, group_by
+        df,
+        result_df,
+        graph_type,
+        graph_type_quantities,
+        group_by,
+        visual_transformation,
     )
 
 
@@ -379,7 +444,7 @@ def number_of_imputed_values(input_df, result_df):
 
 
 def _build_box_hist_plot(
-    df, result_df, graph_type, graph_type_quantities, group_by
+    df, result_df, graph_type, graph_type_quantities, group_by, visual_transformation
 ) -> list[Figure]:
     """
     This function creates two visualisations:
@@ -391,24 +456,40 @@ def _build_box_hist_plot(
 
     2. a graph summarising the amount of
     filtered proteins.
-
     """
+
+    intensity_name_df = df.columns[3]
+    intensity_name_result_df = result_df.columns[3]
+
+    imputed_df = result_df.copy()
+
+    imputed_df[intensity_name_result_df] = list(
+        map(
+            lambda x, y: y if np.isnan(x) else np.nan,
+            df[intensity_name_df],
+            result_df[intensity_name_result_df],
+        )
+    )
+
     if graph_type == "Boxplot":
         fig1 = create_box_plots(
             dataframe_a=df,
-            dataframe_b=result_df,
-            name_a="Before Imputation",
-            name_b="After Imputation",
+            dataframe_b=imputed_df,
+            name_a="Original Values",
+            name_b="Imputed Values",
             heading="Distribution of Protein Intensities",
             group_by=group_by,
+            visual_transformation=visual_transformation,
         )
     elif graph_type == "Histogram":
         fig1 = create_histograms(
             dataframe_a=df,
-            dataframe_b=result_df,
-            name_a="Before Imputation",
-            name_b="After Imputation",
+            dataframe_b=imputed_df,
+            name_a="Original Values",
+            name_b="Imputed Values",
             heading="Distribution of Protein Intensities",
+            visual_transformation=visual_transformation,
+            overlay=True,
         )
 
     values_of_sectors = [

protzilla/data_preprocessing/plots.py

@@ -1,9 +1,12 @@
+import numpy as np
 import pandas as pd
 import plotly.express as px
 import plotly.graph_objects as go
 from plotly.graph_objects import Figure
 from plotly.subplots import make_subplots
 
+from protzilla.data_preprocessing.plots_helper import generate_log_tics
+
 from ..constants.colors import (
     PROTZILLA_DISCRETE_COLOR_OUTLIER_SEQUENCE,
     PROTZILLA_DISCRETE_COLOR_SEQUENCE,
@@ -116,6 +119,7 @@ def create_box_plots(
     y_title="",
     x_title="",
     group_by: str = "None",
+    visual_transformation="linear",
 ) -> Figure:
     """
     A function to create a boxplot for visualisation
@@ -204,6 +208,8 @@ def create_box_plots(
             "yanchor": "top",
         },
     )
+    if visual_transformation == "log10":
+        fig.update_yaxes(type="log")
     fig.update_yaxes(rangemode="tozero")
     return fig
 
@@ -216,6 +222,8 @@ def create_histograms(
     heading="",
     y_title="",
     x_title="",
+    visual_transformation="linear",
+    overlay=False,
 ) -> Figure:
     """
     A function to create a histogram for visualisation
@@ -239,25 +247,71 @@ def create_histograms(
     :type y_title: str
     :param x_title: Optional x axis title for graphs.
     :type x_title: str
+    :param overlay: Specifies whether to draw one Histogram with overlay or two separate histograms
+    :type overlay: bool
+    :param visual_transformation: Visual transformation of the y-axis data.
+    :type visual_transformation: str
     :return: returns a pie or bar chart of the data
     :rtype: Figure (plotly object)
     """
+    if visual_transformation not in {"linear", "log10"}:
+        raise ValueError(
+            f"""visual_transformation parameter  must be "linear" or
+                "log10" but is {visual_transformation}"""
+        )
+
     intensity_name_a = dataframe_a.columns[3]
     intensity_name_b = dataframe_b.columns[3]
-    fig = make_subplots(rows=1, cols=2)
+
+    intensities_a = dataframe_a[intensity_name_a]
+    intensities_b = dataframe_b[intensity_name_b]
+
+    if visual_transformation == "log10":
+        intensities_a = intensities_a.apply(np.log10)
+        intensities_b = intensities_b.apply(np.log10)
+
+    min_value = min(min(intensities_a), min(intensities_b))
+    max_value = max(max(intensities_a), max(intensities_b))
+
+    binsize_factor = 0.0005 if visual_transformation == "linear" else 0.02
+
     trace0 = go.Histogram(
-        x=dataframe_a[intensity_name_a],
+        x=intensities_a,
         marker_color=PROTZILLA_DISCRETE_COLOR_SEQUENCE[0],
         name=name_a,
+        xbins=dict(
+            start=min_value,
+            end=max_value,
+            size=(max_value - min_value) * binsize_factor,
+        ),
     )
     trace1 = go.Histogram(
-        x=dataframe_b[intensity_name_b],
+        x=intensities_b,
         marker_color=PROTZILLA_DISCRETE_COLOR_SEQUENCE[1],
         name=name_b,
+        xbins=dict(
+            start=min_value,
+            end=max_value,
+            size=(max_value - min_value) * binsize_factor,
+        ),
     )
-    fig.add_trace(trace0, 1, 1)
-    fig.add_trace(trace1, 1, 2)
-    fig.update_layout(bargap=0.2)
+    if not overlay:
+        fig = make_subplots(rows=1, cols=2)
+        fig.add_trace(trace0, 1, 1)
+        fig.add_trace(trace1, 1, 2)
+        if visual_transformation == "log10":
+            fig.update_layout(
+                xaxis=generate_log_tics(0, max_value),
+                xaxis2=generate_log_tics(0, max_value),
+            )
+    else:
+        fig = go.Figure()
+        fig.add_trace(trace0)
+        fig.add_trace(trace1)
+        fig.update_layout(barmode="overlay")
+        fig.update_traces(opacity=0.75)
+        if visual_transformation == "log10":
+            fig.update_layout(xaxis=generate_log_tics(0, max_value))
 
     fig.update_layout(
         xaxis_title=x_title,
@@ -276,6 +330,7 @@ def create_histograms(
         },
     )
     fig.update_yaxes(rangemode="tozero")
+
     return fig