Update INH/ACT/Z-SCORE plots

dashboard/assets/style.css

@@ -159,6 +159,11 @@
     align-items: center;
 }
 
+.stats-input{
+    margin-left: 10px ;
+    width: 150px
+}
+
 .btn-report {
     width: 350px;
 }

dashboard/data/combine.py

@@ -138,7 +138,6 @@ def combine_bmg_echo_data(
     df_stats: pd.DataFrame,
     plate_values: np.ndarray,
     modes: dict[Mode],
-    echo_keys: tuple[str] = ("Destination Plate Barcode", "Destination Well"),
 ) -> pd.DataFrame:
     """
     Combine Echo data with activation and inhibition values.
@@ -147,9 +146,11 @@ def combine_bmg_echo_data(
     :param df_stats: dataframe containing statistics for each plate
     :param plate_values: numpy array with activation and inhibition values - shape: (#plates, 2, 16, 24)
     :param modes: dictionary with modes for each plate
-    :param echo_keys: keys used to merge Echo data with activation and inhibition values #TODO: maybe not necessary and should be hard-coded?
     :return: dataframe with Echo data and activation and inhibition values
     """
+    PLATE = "Destination Plate Barcode"
+    WELL = "Destination Well"
+
     if modes is None:
         modes = dict()
     act_inh_dict = get_activation_inhibition_zscore_dict(df_stats, plate_values, modes)
@@ -158,17 +159,19 @@ def combine_bmg_echo_data(
         activation_inhibition_df = get_activation_inhibition_zscore_df(
             barcode, values_dict
         )
+
         dfs.append(
-            echo_df.merge(
-                activation_inhibition_df,
-                left_on=echo_keys,
-                right_on=("Barcode", "Well"),
+            activation_inhibition_df.merge(
+                echo_df, left_on=("Barcode", "Well"), right_on=(PLATE, WELL), how="left"
             )
         )
 
-    bmg_echo_combined_df = pd.concat(dfs, ignore_index=True).drop(
-        columns=["Barcode", "Well"]
+    bmg_echo_combined_df = (
+        pd.concat(dfs, ignore_index=True)
+        .drop(columns=[PLATE, WELL])
+        .rename(columns={"Barcode": PLATE, "Well": WELL})
     )
+
     return reorder_bmg_echo_columns(bmg_echo_combined_df)
 
 
@@ -178,8 +181,60 @@ def split_compounds_controls(df: pd.DataFrame) -> tuple[pd.DataFrame]:
     :param df: dataframe with compounds and controls
     :return: tuple of dataframes with compounds, positive controls and negative controls
     """
+    columns_to_drop = ["EOS", "Source Plate Barcode", "Source Well", "Actual Volume"]
     mask = df["Destination Well"].str[-2:]
     control_pos_df = df[mask == "24"]
+    control_pos_df = control_pos_df.drop(
+        columns=columns_to_drop, errors="ignore"
+    )  # no error raised if the specified column does not exist
     control_neg_df = df[mask == "23"]
+    control_neg_df = control_neg_df.drop(columns=columns_to_drop, errors="ignore")
     compounds_df = df[~mask.isin(["23", "24"])]
     return compounds_df, control_pos_df, control_neg_df
+
+
+def aggregate_well_plate_stats(
+    df: pd.DataFrame, assign_x_coords: bool = False
+) -> tuple[pd.DataFrame]:
+    """
+    Aggregates the statistics (mean and std) per plate needed for the plots.
+    self
+    :param df: dataframe with echo and bmg data combined
+    :param key: key to group by
+    :param assign_x_coords: whether to assign x coordinates to the plates
+    :return: dataframe: echo_bmg_df with additional columns useful for
+    plotting activation/inhibition/z-score and one with statistics per plate
+    """
+
+    PLATE = "Destination Plate Barcode"
+    ACTIVATION = "% ACTIVATION"
+    INHIBITION = "% INHIBITION"
+    Z_SCORE = "Z-SCORE"
+
+    stats_df = (
+        df.groupby(PLATE)[[ACTIVATION, INHIBITION, Z_SCORE]]
+        .agg(["mean", "std", "min", "max"])
+        .reset_index()
+    )
+    stats_df.columns = [
+        PLATE,
+        f"{ACTIVATION}_mean",
+        f"{ACTIVATION}_std",
+        f"{ACTIVATION}_min",
+        f"{ACTIVATION}_max",
+        f"{INHIBITION}_mean",
+        f"{INHIBITION}_std",
+        f"{INHIBITION}_min",
+        f"{INHIBITION}_max",
+        f"{Z_SCORE}_mean",
+        f"{Z_SCORE}_std",
+        f"{Z_SCORE}_min",
+        f"{Z_SCORE}_max",
+    ]
+
+    if assign_x_coords:
+        for col in [ACTIVATION, INHIBITION, Z_SCORE]:
+            stats_df = stats_df.sort_values(by=f"{col}_mean")  # sort by mean
+            stats_df[f"{col}_x"] = range(len(stats_df))  # get the x coordinates
+
+    return stats_df

dashboard/data/utils.py

@@ -1,6 +1,7 @@
 import pandas as pd
 
 
+# TODO: TO BE REMOVED???
 def generate_dummy_links_dataframe(compound_ids: list[str]) -> pd.DataFrame:
     eos = [
         f"[EOS{i+1}](https://ecbd.eu/compound/EOS{i+1})"

dashboard/pages/components.py

@@ -15,6 +15,7 @@
         dcc.Store(id="error-msg", data=""),
         html.Div(id="dummy"),
         dcc.Store(id="user-uuid", storage_type="local"),
+        dcc.Store(id="report-data-csv", storage_type="local"),
         dcc.Store(id="report-data-second-stage", storage_type="local"),
         dcc.Store(id="report-data-third-stage", storage_type="local"),
     ],