20240801 complete fred docking

production/covid_moonshot_fragalysis_processing/combined_sdf_from_cache.py

@@ -15,7 +15,7 @@ def get_args():
         "-i",
         "--input_cache",
         type=Path,
-        help="Path to input cache generated by the asapdiscovery-data package",
+        help="Path to input cache generated by the asapdiscovery-data package. This shouldn't contain any json files except for the cache.json file.",
     )
     parser.add_argument(
         "--flatten",

production/covid_moonshot_fragalysis_processing/prep_cache_for_docking.py

@@ -0,0 +1,69 @@
+"""This script fixes some bespoke errors with the prepped fragalysis cache"""
+from pathlib import Path
+from asapdiscovery.modeling.protein_prep import PreppedComplex
+from asapdiscovery.data.backend.openeye import sdf_string_to_oemol
+from asapdiscovery.data.schema.ligand import Ligand
+import argparse
+from asapdiscovery.data.util.logging import FileLogger
+import rdkit
+
+def get_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "-i",
+        "--input_cache",
+        type=Path,
+        help="Path to input cache generated by the asapdiscovery-data package.",
+    )
+    return parser.parse_args()
+
+def main():
+    args = get_args()
+    cache = args.input_cache
+    logger = FileLogger(
+        "prep_cache_for_docking", cache, logfile="prep_cache_for_docking.log"
+    ).getLogger()
+    if not cache.exists():
+        raise FileNotFoundError(f"Cache file {cache} not found")
+
+    # fix ligand in P0097
+    prepped_directories = list(cache.glob("*P0097*"))
+    logger.info(f"Found {len(prepped_directories)} prepped directories in the cache: '{cache}'")
+    for prepped_directory in prepped_directories:
+        ligand = prepped_directory / "MAT-POS-5d65ec79-1.sdf"
+        if not ligand.exists():
+            raise FileNotFoundError(f"Ligand file {ligand} not found")
+        ligand = Ligand.from_sdf(ligand)
+        # get rid of 2nd ligand
+        rdmol = ligand.to_rdkit()
+
+        # delete second fragment if it exists
+        frags = [frag for frag in rdkit.Chem.GetMolFrags(rdmol)]
+        if len(frags) == 2:
+            frag_to_delete = frags[1]
+            editable = rdkit.Chem.EditableMol(rdmol)
+
+            # need to change index as we delete
+            # since the indices are already sorted, each time we delete one, the next will
+            # have a lower index
+            for i, idx in enumerate(frag_to_delete):
+                idx = idx - i
+                editable.RemoveAtom(idx)
+            new_mol = editable.GetMol()
+            rdkit.Chem.rdmolops.SanitizeMol(new_mol, sanitizeOps=rdkit.Chem.rdmolops.SANITIZE_ALL)
+            sdf_str = rdkit.Chem.rdmolfiles.MolToMolBlock(new_mol)
+            oemol = sdf_string_to_oemol(sdf_str)
+            ligand = Ligand.from_oemol(oemol, compound_name=ligand.compound_name)
+            ligand.to_sdf(prepped_directory / "MAT-POS-5d65ec79-1.sdf")
+
+        # overwrite old json file
+        json_file = list(prepped_directory.glob('*.json'))[0]
+        prepped_complex = PreppedComplex.from_json_file(json_file)
+        prepped_complex.ligand = ligand
+        prepped_complex.to_json_file(json_file)
+
+    # Mpro-P2243_0B
+    # Mpro-P2214_0B
+
+if __name__ == "__main__":
+    main()

science/20240403_multi_pose_docking_v2/20240430_analyze_cross_docking_results/20240502_analyze_and combine_results_csv.ipynb

@@ -389,14 +389,6 @@
     "len(possible_pairs)"
    ]
   },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "7ffaa579-ffa1-4e2b-b8f4-28f109e4ae09",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
   {
    "cell_type": "markdown",
    "id": "5669a910-a36e-453c-a0de-6756a92071ea",