Revert corems type, add start of lipid workflow

metaMS/lcms_lipidomics_workflow.py

@@ -1,7 +1,11 @@
 from dataclasses import dataclass
 import toml
 from pathlib import Path
+import datetime
+from multiprocessing import Pool
 
+from corems.mass_spectra.input.mzml import MZMLSpectraParser
+from corems.mass_spectra.input.rawFileReader import ImportMassSpectraThermoMSFileReader
 
 @dataclass
 class LipidomicsWorkflowParameters:
@@ -33,6 +37,38 @@ class LipidomicsWorkflowParameters:
     scan_translator_path: str = None
     cores: int = 1
 
+def instantiate_lcms_obj(file_in):
+    """Instantiate a corems LCMS object from a binary file.  Pull in ms1 spectra into dataframe (without storing as MassSpectrum objects to save memory)
+
+    Parameters
+    ----------
+    file_in : str or Path
+        Path to binary file
+    verbose : bool
+        Whether to print verbose output
+
+    Returns
+    -------
+    myLCMSobj : corems LCMS object
+        LCMS object with ms1 spectra in dataframe
+    """
+    # Instantiate parser based on binary file type
+    if ".raw" in str(file_in):
+        pass
+        #parser = ImportMassSpectraThermoMSFileReader(file_in)
+
+    if ".mzML" in str(file_in):
+        parser = MZMLSpectraParser(file_in)
+
+    # Instantiate lc-ms data object using parser and pull in ms1 spectra into dataframe (without storing as MassSpectrum objects to save memory)
+    myLCMSobj = parser.get_lcms_obj(spectra="ms1")
+
+    return myLCMSobj
+
+def run_lipid_sp_ms1(file_in, out_path, params_toml, scan_translator):
+    time_start = datetime.datetime.now()
+    myLCMSobj = instantiate_lcms_obj(file_in)   
+    # TODO KRH: Add signal processing and ms1 molecular search here
 
 def run_lcms_lipidomics_workflow(
     lipidomics_workflow_paramaters_file=None,
@@ -65,4 +101,33 @@ def run_lcms_lipidomics_workflow(
     files_list = list(file_dir.glob("*.raw"))
     out_paths_list = [out_dir / f.stem for f in files_list]
 
+    # Run signal processing, get associated ms1, add associated ms2, do ms1 molecular search, and export temp results
+    cores = lipid_workflow_params.cores
+    params_toml = lipid_workflow_params.corems_toml_path
+    scan_translator = lipid_workflow_params.scan_translator_path
+    if cores == 1 or len(files_list) == 1:
+        mz_dicts = []
+        for file_in, file_out in list(zip(files_list, out_paths_list)):
+            mz_dict = run_lipid_sp_ms1(
+                file_in=str(file_in),
+                out_path=str(file_out),
+                params_toml=params_toml,
+                scan_translator=scan_translator,
+            )
+            mz_dicts.append(mz_dict)
+    elif cores > 1:
+        pool = Pool(cores)
+        args = [
+            (
+                str(file_in),
+                str(file_out),
+                params_toml,
+                scan_translator,
+            )
+            for file_in, file_out in list(zip(files_list, out_paths_list))
+        ]
+        mz_dicts = pool.starmap(run_lipid_sp_ms1, args)
+        pool.close()
+        pool.join()
+
     # TODO KRH: Add full lipidomics workflow here

requirements.txt

@@ -1,4 +1,4 @@
-corems>=3.0.0
+corems>=2.0.0
 Click>=7.1.1
 requests
 nmdc-schema>=7.0.0