Fix method name parsing

BioinfoMachineLearning · Dec 18, 2024 · f8bf3b7 · f8bf3b7
1 parent 341da74
commit f8bf3b7
Show file tree

Hide file tree

Showing 6 changed files with 66 additions and 30 deletions.
diff --git a/notebooks/astex_method_interaction_analysis_plotting.ipynb b/notebooks/astex_method_interaction_analysis_plotting.ipynb
@@ -20,6 +20,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import copy\n",
     "import os\n",
     "import re\n",
     "import shutil\n",
@@ -279,14 +280,19 @@
     "    }\n",
     ")\n",
     "\n",
-    "for method in baseline_methods:\n",
+    "for method in copy.deepcopy(baseline_methods):\n",
     "    for repeat_index in range(1, max_num_repeats_per_method + 1):\n",
     "        method_title = method_mapping[method]\n",
     "\n",
-    "        if not os.path.exists(f\"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5\"):\n",
-    "            v1_baseline = method == \"diffdockv1\"\n",
-    "            vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "        v1_baseline = method == \"diffdockv1\"\n",
+    "        vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "\n",
+    "        vina_suffix = f\"_{vina_binding_site_method}\" if \"_\" in method else \"\"\n",
+    "        method = method.split(\"_\")[0]\n",
     "\n",
+    "        if not os.path.exists(\n",
+    "            f\"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5\"\n",
+    "        ):\n",
     "            with open_dict(cfg):\n",
     "                cfg.method = method\n",
     "                cfg.repeat_index = repeat_index\n",
@@ -358,7 +364,7 @@
     "\n",
     "                # NOTE: we iteratively save the interaction dataframes to an HDF5 file\n",
     "                with pd.HDFStore(\n",
-    "                    f\"{method}_astex_diverse_interaction_dataframes_{repeat_index}.h5\"\n",
+    "                    f\"{method}{vina_suffix}_astex_diverse_interaction_dataframes_{repeat_index}.h5\"\n",
     "                ) as store:\n",
     "                    for i, df in enumerate(astex_protein_ligand_interaction_dfs):\n",
     "                        store.put(f\"df_{i}\", df)"

diff --git a/notebooks/astex_method_interaction_analysis_plotting.py b/notebooks/astex_method_interaction_analysis_plotting.py
@@ -5,6 +5,7 @@
 # #### Import packages
 
 # %%
+import copy
 import os
 import re
 import shutil
@@ -209,14 +210,19 @@ def create_temp_pdb_with_only_molecule_type_residues(
     }
 )
 
-for method in baseline_methods:
+for method in copy.deepcopy(baseline_methods):
     for repeat_index in range(1, max_num_repeats_per_method + 1):
         method_title = method_mapping[method]
 
-        if not os.path.exists(f"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5"):
-            v1_baseline = method == "diffdockv1"
-            vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+        v1_baseline = method == "diffdockv1"
+        vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+
+        vina_suffix = f"_{vina_binding_site_method}" if "_" in method else ""
+        method = method.split("_")[0]
 
+        if not os.path.exists(
+            f"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5"
+        ):
             with open_dict(cfg):
                 cfg.method = method
                 cfg.repeat_index = repeat_index
@@ -288,7 +294,7 @@ def create_temp_pdb_with_only_molecule_type_residues(
 
                 # NOTE: we iteratively save the interaction dataframes to an HDF5 file
                 with pd.HDFStore(
-                    f"{method}_astex_diverse_interaction_dataframes_{repeat_index}.h5"
+                    f"{method}{vina_suffix}_astex_diverse_interaction_dataframes_{repeat_index}.h5"
                 ) as store:
                     for i, df in enumerate(astex_protein_ligand_interaction_dfs):
                         store.put(f"df_{i}", df)

diff --git a/notebooks/dockgen_method_interaction_analysis_plotting.ipynb b/notebooks/dockgen_method_interaction_analysis_plotting.ipynb
@@ -20,6 +20,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import copy\n",
     "import os\n",
     "import re\n",
     "import shutil\n",
@@ -292,14 +293,19 @@
     "    }\n",
     ")\n",
     "\n",
-    "for method in baseline_methods:\n",
+    "for method in copy.deepcopy(baseline_methods):\n",
     "    for repeat_index in range(1, max_num_repeats_per_method + 1):\n",
     "        method_title = method_mapping[method]\n",
     "\n",
-    "        if not os.path.exists(f\"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5\"):\n",
-    "            v1_baseline = method == \"diffdockv1\"\n",
-    "            vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "        v1_baseline = method == \"diffdockv1\"\n",
+    "        vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "\n",
+    "        vina_suffix = f\"_{vina_binding_site_method}\" if \"_\" in method else \"\"\n",
+    "        method = method.split(\"_\")[0]\n",
     "\n",
+    "        if not os.path.exists(\n",
+    "            f\"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5\"\n",
+    "        ):\n",
     "            with open_dict(cfg):\n",
     "                cfg.method = method\n",
     "                cfg.repeat_index = repeat_index\n",
@@ -371,7 +377,7 @@
     "\n",
     "                # NOTE: we iteratively save the interaction dataframes to an HDF5 file\n",
     "                with pd.HDFStore(\n",
-    "                    f\"{method}_dockgen_interaction_dataframes_{repeat_index}.h5\"\n",
+    "                    f\"{method}{vina_suffix}_dockgen_interaction_dataframes_{repeat_index}.h5\"\n",
     "                ) as store:\n",
     "                    for i, df in enumerate(dockgen_protein_ligand_interaction_dfs):\n",
     "                        store.put(f\"df_{i}\", df)"

diff --git a/notebooks/dockgen_method_interaction_analysis_plotting.py b/notebooks/dockgen_method_interaction_analysis_plotting.py
@@ -5,6 +5,7 @@
 # #### Import packages
 
 # %%
+import copy
 import os
 import re
 import shutil
@@ -222,14 +223,19 @@ def create_temp_pdb_with_only_molecule_type_residues(
     }
 )
 
-for method in baseline_methods:
+for method in copy.deepcopy(baseline_methods):
     for repeat_index in range(1, max_num_repeats_per_method + 1):
         method_title = method_mapping[method]
 
-        if not os.path.exists(f"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5"):
-            v1_baseline = method == "diffdockv1"
-            vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+        v1_baseline = method == "diffdockv1"
+        vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+
+        vina_suffix = f"_{vina_binding_site_method}" if "_" in method else ""
+        method = method.split("_")[0]
 
+        if not os.path.exists(
+            f"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5"
+        ):
             with open_dict(cfg):
                 cfg.method = method
                 cfg.repeat_index = repeat_index
@@ -301,7 +307,7 @@ def create_temp_pdb_with_only_molecule_type_residues(
 
                 # NOTE: we iteratively save the interaction dataframes to an HDF5 file
                 with pd.HDFStore(
-                    f"{method}_dockgen_interaction_dataframes_{repeat_index}.h5"
+                    f"{method}{vina_suffix}_dockgen_interaction_dataframes_{repeat_index}.h5"
                 ) as store:
                     for i, df in enumerate(dockgen_protein_ligand_interaction_dfs):
                         store.put(f"df_{i}", df)

diff --git a/notebooks/posebusters_method_interaction_analysis_plotting.ipynb b/notebooks/posebusters_method_interaction_analysis_plotting.ipynb
@@ -20,6 +20,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import copy\n",
     "import os\n",
     "import re\n",
     "import shutil\n",
@@ -289,14 +290,19 @@
     "    }\n",
     ")\n",
     "\n",
-    "for method in baseline_methods:\n",
+    "for method in copy.deepcopy(baseline_methods):\n",
     "    for repeat_index in range(1, max_num_repeats_per_method + 1):\n",
     "        method_title = method_mapping[method]\n",
     "\n",
-    "        if not os.path.exists(f\"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5\"):\n",
-    "            v1_baseline = method == \"diffdockv1\"\n",
-    "            vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "        v1_baseline = method == \"diffdockv1\"\n",
+    "        vina_binding_site_method = method.split(\"_\")[-1] if \"_\" in method else \"p2rank\"\n",
+    "\n",
+    "        vina_suffix = f\"_{vina_binding_site_method}\" if \"_\" in method else \"\"\n",
+    "        method = method.split(\"_\")[0]\n",
     "\n",
+    "        if not os.path.exists(\n",
+    "            f\"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5\"\n",
+    "        ):\n",
     "            with open_dict(cfg):\n",
     "                cfg.method = method\n",
     "                cfg.repeat_index = repeat_index\n",
@@ -370,7 +376,7 @@
     "\n",
     "                # NOTE: we iteratively save the interaction dataframes to an HDF5 file\n",
     "                with pd.HDFStore(\n",
-    "                    f\"{method}_posebusters_benchmark_interaction_dataframes_{repeat_index}.h5\"\n",
+    "                    f\"{method}{vina_suffix}_posebusters_benchmark_interaction_dataframes_{repeat_index}.h5\"\n",
     "                ) as store:\n",
     "                    for i, df in enumerate(posebusters_protein_ligand_interaction_dfs):\n",
     "                        store.put(f\"df_{i}\", df)"

diff --git a/notebooks/posebusters_method_interaction_analysis_plotting.py b/notebooks/posebusters_method_interaction_analysis_plotting.py
@@ -5,6 +5,7 @@
 # #### Import packages
 
 # %%
+import copy
 import os
 import re
 import shutil
@@ -219,14 +220,19 @@ def create_temp_pdb_with_only_molecule_type_residues(
     }
 )
 
-for method in baseline_methods:
+for method in copy.deepcopy(baseline_methods):
     for repeat_index in range(1, max_num_repeats_per_method + 1):
         method_title = method_mapping[method]
 
-        if not os.path.exists(f"{method}_{dataset}_interaction_dataframes_{repeat_index}.h5"):
-            v1_baseline = method == "diffdockv1"
-            vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+        v1_baseline = method == "diffdockv1"
+        vina_binding_site_method = method.split("_")[-1] if "_" in method else "p2rank"
+
+        vina_suffix = f"_{vina_binding_site_method}" if "_" in method else ""
+        method = method.split("_")[0]
 
+        if not os.path.exists(
+            f"{method}{vina_suffix}_{dataset}_interaction_dataframes_{repeat_index}.h5"
+        ):
             with open_dict(cfg):
                 cfg.method = method
                 cfg.repeat_index = repeat_index
@@ -300,7 +306,7 @@ def create_temp_pdb_with_only_molecule_type_residues(
 
                 # NOTE: we iteratively save the interaction dataframes to an HDF5 file
                 with pd.HDFStore(
-                    f"{method}_posebusters_benchmark_interaction_dataframes_{repeat_index}.h5"
+                    f"{method}{vina_suffix}_posebusters_benchmark_interaction_dataframes_{repeat_index}.h5"
                 ) as store:
                     for i, df in enumerate(posebusters_protein_ligand_interaction_dfs):
                         store.put(f"df_{i}", df)