updated similarity

src/gcms_data_analysis/gcms.py

@@ -103,6 +103,7 @@ def __init__(
 
         self.list_of_all_compounds: list[str] | None = None
         self.compounds_properties: pd.DataFrame | None = None
+        self.dict_names_to_iupacs: dict[str, str] | None = None
 
         self.deriv_list_of_all_compounds: list[str] | None = None
         self.deriv_files_present: bool = False
@@ -306,11 +307,9 @@ def create_list_of_all_compounds(self):
             self.load_all_files()
         if not self.calibrations:
             self.load_calibrations()
-        all_dfs_with_comps = []
-        for file in self.files.values():
-            all_dfs_with_comps.append(file)
-        for calib in self.calibrations.values():
-            all_dfs_with_comps.append(calib)
+        all_dfs_with_comps = [f for f in self.files.values()] + [
+            f for f in self.calibrations.values()
+        ]
         # non-derivatized compounds
         all_compounds: pd.DataFrame = pd.concat(all_dfs_with_comps)
 
@@ -336,7 +335,7 @@ def create_compounds_properties(
 
         if self.dict_classes_to_codes is None:
             self.load_class_code_frac()
-        if not self.list_of_all_compounds is None:
+        if self.list_of_all_compounds is None:
             self.create_list_of_all_compounds()
         # cpdf = pd.DataFrame(index=pd.Index(self.list_of_all_compounds))
         #
@@ -373,6 +372,55 @@ def load_compounds_properties(self) -> pd.DataFrame:
             cpdf = self.create_compounds_properties()
         return self.compounds_properties
 
+    def create_dict_names_to_iupacs(self) -> dict[str, str]:
+        if self.compounds_properties is None:
+            self.load_compounds_properties()
+        self.dict_names_to_iupacs = self.compounds_properties["iupac_name"].to_dict()
+        return self.dict_names_to_iupacs
+
+    def add_iupac_to_files_and_calibrations(self):
+        """Adds the IUPAC name to each compound in the loaded files,
+        distinguishing between underivatized and derivatized compounds,
+        and updates the corresponding file dataframes."""
+        if not self.files:
+            self.load_all_files()
+        if self.compounds_properties is None:
+            self.load_compounds_properties()
+        for file in self.files.values():
+            file["iupac_name"] = file.index.map(self.dict_names_to_iupacs)
+        for file in self.calibrations.values():
+            file["iupac_name"] = file.index.map(self.dict_names_to_iupacs)
+        return self.files, self.calibrations
+
+    # def apply_calibration_to_files(self):
+    #     """Applies the appropriate calibration curve to each compound
+    #     in the loaded files, adjusting concentrations based on calibration
+    #     data, and updates the 'files' attribute with calibrated data."""
+    #     print("Info: apply_calibration_to_files: loop started")
+    #     if not self.files:
+    #         self.load_all_files()
+    #     if not self.calibrations:
+    #         self.load_calibrations()
+    #     if not self.iupac_to_files_added:
+    #         _, _ = self.add_iupac_to_files()
+
+    #     for filename, _ in self.files.items():
+    #         calibration_name = self.files_info.loc[filename, "calibration_file"]
+    #         calibration = self.calibrations[calibration_name]
+    #         if not self.is_files_deriv[filename]:
+    #             df_comps = self.compounds_properties
+    #         else:
+    #             df_comps = self.deriv_compounds_properties
+    #         file = self._apply_calib_to_file(filename, calibration, df_comps)
+    #         if Project.auto_save_to_excel:
+    #             self.save_file(file, filename)
+    #     self.calibration_to_files_applied = True
+    #     return self.files, self.is_files_deriv
+
+
+def create_tanimoto_matrix(smiles_list: list[str]):
+    pass
+
 
 def get_compound_from_pubchempy(comp_name: str) -> pcp.Compound:
     if not isinstance(comp_name, str) or comp_name.isspace():

tests/data_minimal_case/S_1.txt

@@ -7,6 +7,7 @@ Output Time	11:00:20 AM
 # of Peaks	3
 Mass	TIC
 Peak#	Ret.Time	Proc.From	Proc.To	Mass	Area	Height	A/H	Conc.	Mark	Name	Ret. Index	Area%	Height%	SI	CAS #
-1	13.703	13.580	13.900	TIC	20	20	1	0	 V 	Phenol		2.59	3.01	97	108-95-2
+1	13.703	13.580	13.900	TIC	20	20	1	0	 V 	Capric Acid		2.59	3.01	97	108-95-2
 2	20.942	20.767	21.020	TIC	200	200	1	0	 V 	Naphthalene		8.11	9.20	98	91-20-3
 3	21.426	21.373	21.500	TIC	2000	2000	1	0	 V 	Dodecane		0.36	0.57	96	112-40-3
+4	22.426	22.373	22.500	TIC	1000	1000	1	0	 V 	NotValidComp		0.36	0.57	96	112-40-3

tests/data_minimal_case/S_2.txt

@@ -7,6 +7,7 @@ Output Time	11:00:20 AM
 # of Peaks	3
 Mass	TIC
 Peak#	Ret.Time	Proc.From	Proc.To	Mass	Area	Height	A/H	Conc.	Mark	Name	Ret. Index	Area%	Height%	SI	CAS #
-1	13.703	13.580	13.900	TIC	40	40	1	0	 V 	Phenol		2.59	3.01	97	108-95-2
+1	13.703	13.580	13.900	TIC	40	40	1	0	 V 	Capric Acid		2.59	3.01	97	108-95-2
 2	20.942	20.767	21.020	TIC	400	400	1	0	 V 	Naphthalene		8.11	9.20	98	91-20-3
 3	21.426	21.373	21.500	TIC	4000	4000	1	0	 V 	Dodecane		0.36	0.57	96	112-40-3
+4	24.426	26.373	21.500	TIC	1000	1000	1	0	 V 	Almost Oleic Acid		0.36	0.57	96	112-40-3

tests/test_project_class.py

@@ -12,7 +12,7 @@
 proj = Project(
     folder_path=folder_path,
     auto_save_to_excel=False,
-    # compounds_to_rename_in_files={"phenol": "catechol"},
+    compounds_to_rename_in_files={"almost oleic acid": "oleic acid"},
 )
 
 # check a couple of defaults
@@ -33,25 +33,87 @@
 
 fic.calibration_file = fil.calibration_file  # this cannot be updated automatically
 assert_frame_equal(fil, fic, check_exact=False, atol=1e-5, rtol=1e-5)
-# print(fil.columns)
-# print(fic.columns)
-# print(fil.index)
-# print(fic.index)
-# print(fil==fic)
+
 # %%
 files = proj.load_all_files()
 # %%
 ccf = proj.load_class_code_frac()
 
 # %%
 cal = proj.load_calibrations()
-print(cal)
 # %%
 lac = proj.create_list_of_all_compounds()
 
 # %%
-cpc = proj.create_compounds_properties(update_saved_files_info=False)
+cpc = proj.create_compounds_properties(update_saved_files_info=True)
 cpl = proj.load_compounds_properties()
-
 assert_frame_equal(cpc, cpl, check_exact=False, atol=1e-5, rtol=1e-5, check_dtype=False)
 # %%
+dni = proj.create_dict_names_to_iupacs()
+assert "oleic acid" in dni.keys()
+assert "notvalidcomp" in dni.keys()
+assert "decanoic acid" in dni.values()
+# %%
+files_iupac, calibration_iupac = proj.add_iupac_to_files_and_calibrations()
+# %%
+import pandas as pd
+from rdkit import Chem
+from rdkit.Chem import DataStructs
+from rdkit.Chem.AllChem import (
+    GetMorganFingerprintAsBitVect,
+)  # pylint: disable=no-name-in-module
+
+
+def create_tanimoto_similarity_dict(
+    comp_smiles: str, calib_smiless: list[str]
+) -> dict[str, list[float]]:
+
+    mols_comp = Chem.MolFromSmiles(comp_smiles)
+    mols_cal = [
+        Chem.MolFromSmiles(smi) for smi in calib_smiless if isinstance(smi, str)
+    ]
+
+    # Generate fingerprints from molecule objects, skipping None entries created from invalid SMILES
+    fps_comp = GetMorganFingerprintAsBitVect(mols_comp, 2, nBits=1024)
+
+    fps_cal = [
+        GetMorganFingerprintAsBitVect(mol, 2, nBits=1024)
+        for mol in mols_cal
+        if mol is not None
+    ]
+
+    # perform Tanimoto similarity betwenn the first and all other compounds
+    similarity = DataStructs.BulkTanimotoSimilarity(fps_comp, fps_cal)
+    # create a df with results
+    return similarity
+
+
+calib_comp_iupacs = proj.calibrations["cal_minimal"].index.tolist()
+calib_smiless = proj.compounds_properties.loc[
+    calib_comp_iupacs, "canonical_smiles"
+].tolist()
+tanimoto_similarity_df: pd.DataFrame = pd.DataFrame(
+    index=proj.compounds_properties["iupac_name"],
+    columns=calib_comp_iupacs,
+)
+for iupac, smiles in zip(
+    proj.compounds_properties["iupac_name"],
+    proj.compounds_properties["canonical_smiles"],
+):
+    if isinstance(smiles, str):
+        sim = create_tanimoto_similarity_dict(smiles, calib_smiless)
+        tanimoto_similarity_df.loc[iupac, :] = sim
+# %%
+compounds_properties = proj.compounds_properties
+calib_comp_iupacs = proj.calibrations["cal_minimal"]["iupac_name"].tolist()
+tanimoto_similarity_df: pd.DataFrame = pd.DataFrame(
+    index=compounds_properties.iupac_name.tolist(),
+    columns=proj.calibrations["cal_minimal"]["iupac_name"].tolist(),
+)
+for comp in compounds_properties.iupac_name.tolist():
+    print(comp)
+    s = create_tanimoto_similarity_dict(comp, calib_comp_iupacs)
+    tanimoto_similarity_df.loc[comp, :] = s
+# %%
+
+# %%