Bookkeeping of session state variables

alphastats/gui/pages/02_Import Data.py

@@ -11,6 +11,13 @@
         empty_session_state,
     )
     from alphastats.gui.utils.ui_helper import sidebar_info
+    from alphastats.DataSet import DataSet
+    from alphastats.gui.utils.analysis_helper import (
+        get_sample_names_from_software_file,
+        read_uploaded_file_into_df,
+    )
+    from alphastats.gui.utils.ui_helper import sidebar_info, StateKeys
+    from alphastats.loader.MaxQuantLoader import MaxQuantLoader
 
 except ModuleNotFoundError:
     from utils.ui_helper import sidebar_info
@@ -22,24 +29,34 @@
         empty_session_state,
     )
 
+    from utils.ui_helper import sidebar_info, StateKeys
+    from utils.analysis_helper import (
+        get_sample_names_from_software_file,
+        read_uploaded_file_into_df,
+    )
+    from alphastats import MaxQuantLoader
+    from alphastats import DataSet
+
+import pandas as pd
+import plotly.express as px
 from streamlit.runtime import get_instance
 from streamlit.runtime.scriptrunner.script_run_context import get_script_run_ctx
 
 runtime = get_instance()
 session_id = get_script_run_ctx().session_id
-session_info = runtime._session_mgr.get_session_info(session_id)
+# session_info = runtime._session_mgr.get_session_info(session_id)
 
 user_session_id = session_id
-st.session_state["user_session_id"] = user_session_id
+st.session_state[StateKeys.USER_SESSION_ID] = user_session_id
 
 if "loader" not in st.session_state:
-    st.session_state["loader"] = None
+    st.session_state[StateKeys.LOADER] = None
 
 if "gene_to_prot_id" not in st.session_state:
-    st.session_state["gene_to_prot_id"] = {}
+    st.session_state[StateKeys.GENE_TO_PROT_ID] = {}
 
 if "organism" not in st.session_state:
-    st.session_state["organism"] = 9606  # human
+    st.session_state[StateKeys.ORGANISM] = 9606  # human
 
 sidebar_info()
 

alphastats/gui/pages/03_Data Overview.py

@@ -8,42 +8,42 @@
         get_intensity_distribution_processed,
         get_sample_histogram_matrix,
     )
-    from alphastats.gui.utils.ui_helper import sidebar_info
+    from alphastats.gui.utils.ui_helper import sidebar_info, StateKeys
 
 except ModuleNotFoundError:
     from utils.overview_helper import (
         display_matrix,
         get_intensity_distribution_processed,
         get_sample_histogram_matrix,
     )
-    from utils.ui_helper import sidebar_info
+    from utils.ui_helper import sidebar_info, StateKeys
 
 sidebar_info()
 
-if "dataset" in st.session_state:
+if StateKeys.DATASET in st.session_state:
     st.markdown("## DataSet overview")
 
     c1, c2 = st.columns(2)
 
     with c1:
         st.markdown("**Intensity distribution raw data per sample**")
         st.plotly_chart(
-            st.session_state.distribution_plot.update_layout(plot_bgcolor="white"),
+            st.session_state[StateKeys.DISTRIBUTION_PLOT].update_layout(plot_bgcolor="white"),
             use_container_width=True,
         )
 
     with c2:
         st.markdown("**Intensity distribution data per sample used for analysis**")
         st.plotly_chart(
             get_intensity_distribution_processed(
-                user_session_id=st.session_state.user_session_id
+                user_session_id=st.session_state[StateKeys.USER_SESSION_ID]
             ).update_layout(plot_bgcolor="white"),
             use_container_width=True,
         )
 
     st.plotly_chart(
         get_sample_histogram_matrix(
-            user_session_id=st.session_state.user_session_id
+            user_session_id=st.session_state[StateKeys.USER_SESSION_ID]
         ).update_layout(plot_bgcolor="white"),
         use_container_width=True,
     )

alphastats/gui/utils/import_helper.py

@@ -7,6 +7,7 @@
 
 try:
     from alphastats.DataSet import DataSet
+    from alphastats.gui.utils.ui_helper import StateKeys
     from alphastats.gui.utils.analysis_helper import (
         get_sample_names_from_software_file,
         read_uploaded_file_into_df,
@@ -15,6 +16,7 @@
     from alphastats.loader.MaxQuantLoader import MaxQuantLoader
 
 except ModuleNotFoundError:
+    from utils.ui_helper import StateKeys
     from utils.analysis_helper import (
         get_sample_names_from_software_file,
         read_uploaded_file_into_df,
@@ -27,8 +29,8 @@
 def load_options():
     from alphastats.gui.utils.options import plotting_options, statistic_options
 
-    st.session_state["plotting_options"] = plotting_options(st.session_state)
-    st.session_state["statistic_options"] = statistic_options(st.session_state)
+    st.session_state[StateKeys.PLOTTING_OPTIONS] = plotting_options(st.session_state)
+    st.session_state[StateKeys.STATISTIC_OPTIONS] = statistic_options(st.session_state)
 
 
 def load_proteomics_data(uploaded_file, intensity_column, index_column, software):
@@ -60,16 +62,16 @@ def upload_softwarefile(software):
         select_columns_for_loaders(software=software, software_df=softwarefile_df)
 
         if (
-            "intensity_column" in st.session_state
-            and "index_column" in st.session_state
+            StateKeys.INTENSITY_COLUMN in st.session_state
+            and StateKeys.INDEX_COLUMN in st.session_state
         ):
             loader = load_proteomics_data(
                 softwarefile_df,
-                intensity_column=st.session_state.intensity_column,
-                index_column=st.session_state.index_column,
+                intensity_column=st.session_state[StateKeys.INTENSITY_COLUMN],
+                index_column=st.session_state[StateKeys.INDEX_COLUMN],
                 software=software,
             )
-            st.session_state["loader"] = loader
+            st.session_state[StateKeys.LOADER] = loader
 
 
 def upload_metadatafile(software):
@@ -80,8 +82,13 @@ def upload_metadatafile(software):
         key="metadatafile",
     )
 
-    if metadatafile_upload is not None and st.session_state.loader is not None:
-        metadatafile_df = read_uploaded_file_into_df(st.session_state.metadatafile)
+    if (
+        metadatafile_upload is not None
+        and st.session_state[StateKeys.LOADER] is not None
+    ):
+        metadatafile_df = read_uploaded_file_into_df(
+            st.session_state[StateKeys.METADATAFILE]
+        )
         # display metadata
         st.write(
             f"File successfully uploaded. Number of rows: {metadatafile_df.shape[0]}"
@@ -92,26 +99,30 @@ def upload_metadatafile(software):
 
         if select_sample_column_metadata(metadatafile_df, software):
             # create dataset
-            st.session_state["dataset"] = DataSet(
-                loader=st.session_state.loader,
+            st.session_state[StateKeys.DATASET] = DataSet(
+                loader=st.session_state[StateKeys.LOADER],
                 metadata_path=metadatafile_df,
-                sample_column=st.session_state.sample_column,
+                sample_column=st.session_state[StateKeys.SAMPLE_COLUMN],
+            )
+            st.session_state[StateKeys.METADATA_COLUMNS] = (
+                metadatafile_df.columns.to_list()
             )
-            st.session_state["metadata_columns"] = metadatafile_df.columns.to_list()
             load_options()
 
-    if st.session_state.loader is not None:
+    if st.session_state[StateKeys.LOADER] is not None:
         create_metadata_file()
         st.write(
             "Download the template file and add additional information as "
             + "columns to your samples such as disease group. "
             + "Upload the updated metadata file."
         )
 
-    if st.session_state.loader is not None:
+    if st.session_state[StateKeys.LOADER] is not None:
         if st.button("Create a DataSet without metadata"):
-            st.session_state["dataset"] = DataSet(loader=st.session_state.loader)
-            st.session_state["metadata_columns"] = ["sample"]
+            st.session_state[StateKeys.DATASET] = DataSet(
+                loader=st.session_state[StateKeys.LOADER]
+            )
+            st.session_state[StateKeys.METADATA_COLUMNS] = ["sample"]
 
             load_options()
 
@@ -170,9 +181,9 @@ def load_sample_data():
         ]
     ]
     ds.preprocess(subset=True)
-    st.session_state["loader"] = loader
-    st.session_state["metadata_columns"] = ds.metadata.columns.to_list()
-    st.session_state["dataset"] = ds
+    st.session_state[StateKeys.LOADER] = loader
+    st.session_state[StateKeys.METADATA_COLUMNS] = ds.metadata.columns.to_list()
+    st.session_state[StateKeys.DATASET] = ds
 
     load_options()
 
@@ -186,43 +197,45 @@ def import_data():
         key="software",
     )
 
-    if st.session_state.software != "<select>":
-        upload_softwarefile(software=st.session_state.software)
+    if st.session_state[StateKeys.SOFTWARE] != "<select>":
+        upload_softwarefile(software=st.session_state[StateKeys.SOFTWARE])
     if "loader" not in st.session_state:
-        st.session_state["loader"] = None
-    if st.session_state.loader is not None:
-        upload_metadatafile(st.session_state.software)
+        st.session_state[StateKeys.LOADER] = None
+    if st.session_state[StateKeys.LOADER] is not None:
+        upload_metadatafile(st.session_state[StateKeys.SOFTWARE])
 
 
 def display_loaded_dataset():
     st.info("Data was successfully imported")
     st.info("DataSet has been created")
 
-    st.markdown(f"*Preview:* Raw data from {st.session_state.dataset.software}")
-    st.dataframe(st.session_state.dataset.rawinput.head(5))
+    st.markdown(
+        f"*Preview:* Raw data from {st.session_state[StateKeys.DATASET].software}"
+    )
+    st.dataframe(st.session_state[StateKeys.DATASET].rawinput.head(5))
 
     st.markdown("*Preview:* Metadata")
-    st.dataframe(st.session_state.dataset.metadata.head(5))
+    st.dataframe(st.session_state[StateKeys.DATASET].metadata.head(5))
 
     st.markdown("*Preview:* Matrix")
 
     df = pd.DataFrame(
-        st.session_state.dataset.mat.values,
-        index=st.session_state.dataset.mat.index.to_list(),
+        st.session_state[StateKeys.DATASET].mat.values,
+        index=st.session_state[StateKeys.DATASET].mat.index.to_list(),
     ).head(5)
 
     st.dataframe(df)
 
 
 def save_plot_sampledistribution_rawdata():
-    df = st.session_state.dataset.rawmat
+    df = st.session_state[StateKeys.DATASET].rawmat
     df = df.unstack().reset_index()
     df.rename(
-        columns={"level_1": st.session_state.dataset.sample, 0: "Intensity"},
+        columns={"level_1": st.session_state[StateKeys.DATASET].sample, 0: "Intensity"},
         inplace=True,
     )
-    st.session_state["distribution_plot"] = px.violin(
-        df, x=st.session_state.dataset.sample, y="Intensity"
+    st.session_state[StateKeys.DISTRIBUTION_PLOT] = px.violin(
+        df, x=st.session_state[StateKeys.DATASET].sample, y="Intensity"
     )
 
 
@@ -233,12 +246,12 @@ def empty_session_state():
     for key in st.session_state.keys():
         del st.session_state[key]
     st.empty()
-    st.session_state["software"] = "<select>"
+    st.session_state[StateKeys.SOFTWARE] = "<select>"
 
     from streamlit.runtime.scriptrunner.script_run_context import get_script_run_ctx
 
     user_session_id = get_script_run_ctx().session_id
-    st.session_state["user_session_id"] = user_session_id
+    st.session_state[StateKeys.USER_SESSION_ID] = user_session_id
 
 
 def check_software_file(df, software):
@@ -348,17 +361,17 @@ def select_sample_column_metadata(df, software):
         submitted = st.form_submit_button("Create DataSet")
 
     if submitted:
-        if len(df[st.session_state.sample_column].to_list()) != len(
-            df[st.session_state.sample_column].unique()
+        if len(df[st.session_state[StateKeys.SAMPLE_COLUMN]].to_list()) != len(
+            df[st.session_state[StateKeys.SAMPLE_COLUMN]].unique()
         ):
             st.error("Sample names have to be unique.")
             st.stop()
         return True
 
 
 def create_metadata_file():
-    dataset = DataSet(loader=st.session_state.loader)
-    st.session_state["metadata_columns"] = ["sample"]
+    dataset = DataSet(loader=st.session_state[StateKeys.LOADER])
+    st.session_state[StateKeys.METADATA_COLUMNS] = ["sample"]
     metadata = dataset.metadata
     buffer = io.BytesIO()
 

alphastats/gui/utils/ui_helper.py

@@ -58,3 +58,29 @@ def img_to_bytes(img_path):
     # img_bytes = Path(img_path).read_bytes()
     # encoded = base64.b64encode(img_bytes).decode()
     return encoded_string.decode()
+
+
+class StateKeys:
+    ## 02_Data Import
+    # on 1st run
+    ORGANISM = "organism"
+    GENE_TO_PROT_ID = "gene_to_prot_id"
+    USER_SESSION_ID = "user_session_id"
+    LOADER = "loader"
+    SOFTWARE = "software"
+    # on sample run (function load_sample_data), removed on new session click
+    DATASET = "dataset"  # functions upload_metadatafile
+    PLOTTING_OPTIONS = "plotting_options"  # function load_options
+    STATISTIC_OPTIONS = "statistic_options"  # function load_options
+    DISTRIBUTION_PLOT = (
+        "distribution_plot"  # function save_plot_sampledistribution_rawdata
+    )
+    METADATA_COLUMNS = (
+        "metadata_columns"  # function create_metadata_file, upload_metadatafile
+    )
+    # on data upload
+    INTENSITY_COLUMN = "intensity_column"
+    INDEX_COLUMN = "index_column"
+    # on metadata upload
+    METADATAFILE = "metadatafile"
+    SAMPLE_COLUMN = "sample_column"