Move calculation and plot methods

protzilla/constants/constants.py

@@ -3,3 +3,13 @@
 PATH_TO_PROJECT = Path(__file__).resolve().parent.parent.parent
 PATH_TO_RUNS = Path(f"{PATH_TO_PROJECT}/user_data/runs")
 PATH_TO_WORKFLOWS = Path(f"{PATH_TO_PROJECT}/user_data/workflows")
+
+# color schemes
+PROTZILLA_DISCRETE_COLOR_SEQUENCE = [
+    "#4A536A",
+    "#87A8B9",
+    "#CE5A5A",
+    "#8E3325",
+    "#E2A46D",
+]
+PROTZILLA_DISCRETE_COLOR_OUTLIER_SEQUENCE = ["#4A536A", "#CE5A5A"]

protzilla/constants/method_mapping.py

@@ -1,15 +1,62 @@
-from ..data_preprocessing import filter_proteins, filter_samples
+from ..importing import main_data_import
+from ..data_preprocessing import filter_proteins, filter_samples, normalisation
 
 
 method_map = {
     (
-        "data_preprocessing",
-        "filter_proteins",
-        "by_low_frequency",
+        "importing",
+        "main-data-import",
+        "ms-data-import",
+    ): main_data_import.max_quant_import,
+    (
+        "data-preprocessing",
+        "filter-proteins",
+        "low-frequency-filter",
     ): filter_proteins.by_low_frequency,
     (
-        "data_preprocessing",
-        "filter_samples",
-        "by_protein_intensity_sum",
+        "data-preprocessing",
+        "filter-proteins",
+        "protein-intensity-sum-filter",
     ): filter_samples.by_protein_intensity_sum,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "median",
+    ): normalisation.by_median,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "totalsum",
+    ): normalisation.by_totalsum,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "ref-protein",
+    ): normalisation.by_reference_protein,
+}
+
+# reverse_map = {v: k for k, v in method_map.items()}
+
+
+plot_map = {
+    (
+        "data-preprocessing",
+        "filter-proteins",
+        "low-frequency-filter",
+    ): filter_proteins.by_low_frequency_plot,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "median",
+    ): normalisation.by_median_plot,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "totalsum",
+    ): normalisation.by_totalsum_plot,
+    (
+        "data_preprocessing",
+        "normalisation",
+        "ref-protein",
+    ): normalisation.by_reference_protein_plot,
 }

protzilla/data_preprocessing/filter_proteins.py

@@ -1,7 +1,28 @@
+from protzilla.data_preprocessing.plots import create_bar_plot, create_pie_plot
+
 from ..utilities.transform_dfs import long_to_wide
 
 
 def by_low_frequency(intensity_df, threshold):
+    """
+    This function filters proteins with a low frequency of occurrence from
+    a protein dataframe based on a set threshold. The threshold is defined
+    by the relative amount of samples a protein is detected in.
+
+    :param intensity_df: the dataframe that should be filtered\
+    in long format
+    :type intensity_df: pd.DataFrame
+    :param threshold: float ranging from 0 to 1. Defining the\
+    relative share of samples the proteins should be present in\
+    order to be included. Example 0.5 - all proteins with intensities\
+    equal to zero in at least 50% of samples are discarded. Default: 0.5
+    :type threshold: float
+    :return: returns a Dataframe with the samples that meet the\
+    filtering criteria, a dict with a list with names of samples\
+    that were discarded and a list with names of samples\
+    that were kept
+    :rtype: list
+    """
     min_threshold = threshold * len(intensity_df.Sample.unique())
     transformed_df = long_to_wide(intensity_df)
 
@@ -14,5 +35,30 @@ def by_low_frequency(intensity_df, threshold):
     # TODO: might be redundant to remaining_proteins
     return (
         intensity_df[~(intensity_df["Protein ID"].isin(filtered_proteins_list))],
-        dict(filtered_proteins=filtered_proteins_list),
+        dict(
+            filtered_proteins=filtered_proteins_list,
+            remaining_proteins=remaining_proteins.tolist(),
+        ),
     )
+
+
+def by_low_frequency_plot(df, result_df, current_out, graph_type):
+    if graph_type == "Pie chart":
+        fig = create_pie_plot(
+            values_of_sectors=[
+                len(current_out["remaining_proteins"]),
+                len(current_out["filtered_proteins"]),
+            ],
+            names_of_sectors=["Proteins kept", "Proteins filtered"],
+            heading="Number of Filtered Proteins",
+        )
+    if graph_type == "Bar chart":
+        fig = create_bar_plot(
+            values_of_sectors=[
+                len(current_out["remaining_proteins"]),
+                len(current_out["filtered_proteins"]),
+            ],
+            names_of_sectors=["Proteins kept", "Proteins filtered"],
+            heading="Number of Filtered Proteins",
+        )
+    return [fig]