Merge pull request OpenMS#41 from axelwalter/main

Framework for TOPP tool workflows

assets/topp-workflow-default-params.json

@@ -0,0 +1,4 @@
+{
+    "mzML_files": [
+    ]
+}

environment.yml

@@ -11,7 +11,7 @@ dependencies:
   - mono==6.12.0.90
   - pip: # dependencies only available through pip
       # streamlit dependencies
-    - streamlit==1.28.0
+    - streamlit==1.29.0
     - streamlit-plotly-events==0.0.6
     - streamlit-aggrid==0.3.4.post3
     - captcha==0.5.0

pages/2_Simple_Workflow.py

@@ -1,7 +1,7 @@
 import streamlit as st
 
 from src.common import page_setup, save_params, show_table
-from src import workflow
+from src import simpleworkflow
 from src.captcha_ import captcha_control
 
 # Page name "workflow" will show mzML file selector in sidebar
@@ -40,7 +40,7 @@
 # Get a dataframe with x and y dimensions via time consuming (sleep) cached function
 # If the input has been given before, the function does not run again
 # Input x from local variable, input y from session state via key
-df = workflow.generate_random_table(xdimension, st.session_state["example-y-dimension"])
+df = simpleworkflow.generate_random_table(xdimension, st.session_state["example-y-dimension"])
 
 # Display dataframe via custom show_table function, which will render a download button as well
 show_table(df, download_name="random-table")

pages/5_TOPP-Workflow.py

@@ -0,0 +1,26 @@
+import streamlit as st
+from src.common import page_setup
+from src.Workflow import Workflow
+
+# The rest of the page can, but does not have to be changed
+if __name__ == "__main__":
+
+    params = page_setup()
+
+    wf = Workflow()
+
+    st.title(wf.name)
+
+    t = st.tabs(["📁 **File Upload**", "⚙️ **Configure**", "🚀 **Run**", "📊 **Results**"])
+    with t[0]:
+        wf.show_file_upload_section()
+
+    with t[1]:
+        wf.show_parameter_section()
+
+    with t[2]:
+        wf.show_execution_section()
+
+    with t[3]:
+        wf.show_results_section()
+

requirements.txt

@@ -1,7 +1,7 @@
 # the requirements.txt file is intended for deployment on streamlit cloud and if the simple container is built
 # note that it is much more restricted in terms of installing third-parties / etc.
 # preferably use the batteries included or simple docker file for local hosting
-streamlit==1.28.0
+streamlit==1.29.0
 streamlit-plotly-events==0.0.6
 streamlit-aggrid==0.3.4.post3
 pandas==2.1.2

src/Workflow.py

@@ -0,0 +1,79 @@
+import streamlit as st
+from .workflow.WorkflowManager import WorkflowManager
+
+class Workflow(WorkflowManager):
+    # Setup pages for upload, parameter, execution and results.
+    # For layout use any streamlit components such as tabs (as shown in example), columns, or even expanders.
+    def __init__(self) -> None:
+        # Initialize the parent class with the workflow name.
+        super().__init__("TOPP Workflow", st.session_state["workspace"])
+
+    def upload(self)-> None:
+        t = st.tabs(["MS data", "Example with fallback data"])
+        with t[0]:
+            # Use the upload method from StreamlitUI to handle mzML file uploads.
+            self.ui.upload_widget(key="mzML-files", name="MS data", file_type="mzML")
+        with t[1]:
+            # Example with fallback data (not used in workflow)
+            self.ui.upload_widget(key="image", file_type="png", fallback="assets/OpenMS.png")
+
+    def configure(self) -> None:
+        # Allow users to select mzML files for the analysis.
+        self.ui.select_input_file("mzML-files", multiple=True)
+
+        # Create tabs for different analysis steps.
+        t = st.tabs(
+            ["**Feature Detection**", "**Adduct Detection**", "**SIRIUS Export**", "**Python Custom Tool**"]
+        )
+        with t[0]:
+            # Parameters for FeatureFinderMetabo TOPP tool.
+            self.ui.input_TOPP("FeatureFinderMetabo")
+        with t[1]:
+            # A single checkbox widget for workflow logic.
+            self.ui.input_widget("run-adduct-detection", False, "Adduct Detection")
+            # Paramters for MetaboliteAdductDecharger TOPP tool.
+            self.ui.input_TOPP("MetaboliteAdductDecharger")
+        with t[2]:
+            # Paramters for SiriusExport TOPP tool
+            self.ui.input_TOPP("SiriusExport")
+        with t[3]:
+            # Generate input widgets for a custom Python tool, located at src/python-tools.
+            # Parameters are specified within the file in the DEFAULTS dictionary.
+            self.ui.input_python("example")
+
+    def execution(self) -> None:
+        # Get mzML input files from self.params.
+        # Can be done without file manager, however, it ensures everything is correct.
+        in_mzML = self.file_manager.get_files(self.params["mzML-files"])
+
+        # Log any messages.
+        self.logger.log(f"Number of input mzML files: {len(in_mzML)}")
+
+        # Prepare output files for feature detection.
+        out_ffm = self.file_manager.get_files(in_mzML, "featureXML", "feature-detection")
+
+        # Run FeatureFinderMetabo tool with input and output files.
+        self.executor.run_topp(
+            "FeatureFinderMetabo", input_output={"in": in_mzML, "out": out_ffm}
+        )
+
+        # Check if adduct detection should be run.
+        if self.params["run-adduct-detection"]:
+
+            # Run MetaboliteAdductDecharger for adduct detection, with disabled logs.
+            # Without a new file list for output, the input files will be overwritten in this case.
+            self.executor.run_topp(
+                "MetaboliteAdductDecharger", {"in": out_ffm, "out_fm": out_ffm}, write_log=False
+            )
+
+        # Example for a custom Python tool, which is located in src/python-tools.
+        self.executor.run_python("example", {"in": in_mzML})
+
+        # Prepare output file for SiriusExport.
+        out_se = self.file_manager.get_files("sirius.ms", set_results_dir="sirius-export")
+        self.executor.run_topp("SiriusExport", {"in": self.file_manager.get_files(in_mzML, collect=True),
+                                                "in_featureinfo": self.file_manager.get_files(out_ffm, collect=True),
+                                                "out": out_se})
+
+    def results(self) -> None:
+        st.warning("Not implemented yet.")

src/python-tools/.gitignore

@@ -0,0 +1 @@
+__pycache__

src/python-tools/example.py

@@ -0,0 +1,65 @@
+import json
+import sys
+
+############################
+# default paramter values #
+###########################
+#
+# Mandatory keys for each parameter
+# key: a unique identifier
+# value: the default value
+#
+# Optional keys for each parameter
+# name: the name of the parameter
+# hide: don't show the parameter in the parameter section (e.g. for input/output files)
+# options: a list of valid options for the parameter
+# min: the minimum value for the parameter (int and float)
+# max: the maximum value for the parameter (int and float)
+# step_size: the step size for the parameter (int and float)
+# help: a description of the parameter
+# widget_type: the type of widget to use for the parameter (default: auto)
+# advanced: whether or not the parameter is advanced (default: False)
+
+DEFAULTS = [
+    {"key": "in", "value": [], "help": "Input files for Python Script.", "hide": True},
+    {"key": "out", "value": [], "help": "Output files for Python Script.", "hide": True},
+    {
+        "key": "number-slider",
+        "name": "number of features",
+        "value": 6,
+        "min": 2,
+        "max": 10,
+        "help": "How many features to consider.",
+        "widget_type": "slider",
+        "step_size": 2,
+    },
+    {
+        "key": "selectbox-example",
+        "value": "a",
+        "options": ["a", "b", "c"],
+    },
+    {
+        "key": "adavanced-input",
+        "value": 5,
+        "step_size": 5,
+        "help": "An advanced example parameter.",
+        "advanced": True,
+    },
+    {
+        "key": "checkbox", "value": True
+    }
+]
+
+def get_params():
+    if len(sys.argv) > 1:
+        with open(sys.argv[1], "r") as f:
+            return json.load(f)
+    else:
+        return {}
+
+if __name__ == "__main__":
+    params = get_params()
+    # Add code here:
+    print("Writing stdout which will get logged...")
+    print("Parameters for this example Python tool:")
+    print(json.dumps(params, indent=4))

src/workflow/.gitignore

@@ -0,0 +1 @@
+__pycache__