better fix for rdkit error

ur-whitelab · Feb 28, 2025 · 7e4ab0b · 7e4ab0b
1 parent fde6a31
commit 7e4ab0b
Showing 1 changed file with 4 additions and 1 deletion.
diff --git a/paper2_LIME/RF-lime.ipynb b/paper2_LIME/RF-lime.ipynb
@@ -51,6 +51,9 @@
     ")\n",
     "#drop smile with containing 'P'\n",
     "soldata = soldata[soldata[\"SMILES\"].str.contains(\"P\") == False]\n",
+    "#drop smile where rdkit.Chem.MolFromSmiles return None\n",
+    "soldata = soldata[soldata[\"SMILES\"].apply(lambda smiles: rdkit.Chem.MolFromSmiles(smiles) is not None)]\n",
+    "\n",
     "\n",
     "features_start_at = list(soldata.columns).index(\"MolWt\")"
    ]
@@ -84,7 +87,7 @@
     ")\n",
     "\n",
     "# make subsample from pandas df\n",
-    "molecules = [mol for smi in soldata.SMILES if (mol := rdkit.Chem.MolFromSmiles(smi)) is not None]\n",
+    "molecules = [rdkit.Chem.MolFromSmiles(smi) for smi in soldata.SMILES]\n",
     "\n",
     "raw_features = []\n",
     "for e, c in zip(molecules, calc.map(molecules, quiet=True)):\n",