add pdb variants

README.md

@@ -20,8 +20,26 @@ configs:
     path: "kiba.parquet"
   - split: pdbbind_2020_general
     path: "pdbbind-2020-general.parquet"
+  - split: pdbbind_2020_general_equal_only
+    path: "pdbbind-2020-general-equal-only.parquet"
+  - split: pdbbind_2020_general_ic50
+    path: "pdbbind-2020-general-ic50.parquet"
+  - split: pdbbind_2020_general_ic50_equal_only
+    path: "pdbbind-2020-general-ic50-equal-only.parquet"
+  - split: pdbbind_2020_general_ki
+    path: "pdbbind-2020-general-ki.parquet"
+  - split: pdbbind_2020_general_ki_equal_only
+    path: "pdbbind-2020-general-ki-equal-only.parquet"
+  - split: pdbbind_2020_general_kd
+    path: "pdbbind-2020-general-kd.parquet"
+  - split: pdbbind_2020_general_kd_equal_only
+    path: "pdbbind-2020-general-kd-equal-only.parquet"
   - split: pdbbind_2020_refined
     path: "pdbbind-2020-refined.parquet"
+  - split: pdbbind_2020_refined_ki
+    path: "pdbbind-2020-refined-ki.parquet"
+  - split: pdbbind_2020_refined_kd
+    path: "pdbbind-2020-refined-kd.parquet"
   - split: pdbbind_2013_core
     path: "pdbbind-2013-core.parquet"
   - split: bindingdb_ic50

pdbbind-2020-general-all-equal-only.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fef459d2749f013156d662f48d5ed5dad750ca8831a363dce2dd8489530329f
+size 5453039

pdbbind-2020-general-ic50-equal-only.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fc16d8d88204a7ac8ef7c8b85f251baddf426bee20947a06e7c43eef60958bf
+size 5044228

pdbbind-2020-general-ic50.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:657cfe5bc026994d1a3c7057e08bd423b37e0ca8dcdf218d70fa977df1b4317b
+size 2384720

pdbbind-2020-general-kd-equal-only.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eb18ec1359bc837a9702b2d819c79452396767a4378ba803ff395488c5760fb
+size 600

pdbbind-2020-general-kd.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36f4e9185898548b64127f2a36cf59ee6c50054e36d35fa413be001627c7ceed
+size 1455342

pdbbind-2020-general-ki-equal-only.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eb18ec1359bc837a9702b2d819c79452396767a4378ba803ff395488c5760fb
+size 600

pdbbind-2020-general-ki.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d6b9bcc6459491e77e010c5bc29964b6afd832fd653b37a78cf7feb9cdd44a4
+size 800636

pdbbind-2020-general.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cea6a908cbe5614b789951c85e1e48c643d2a6971548d7dadcf8b62e59d1aa50
-size 5353522
+oid sha256:8fef459d2749f013156d662f48d5ed5dad750ca8831a363dce2dd8489530329f
+size 5453039

pdbbind-2020-refined-kd.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf1ca4d57baa9d492a856e1ab5f27e627d0f5f881c4816073b756ca4e5d5a4d8
+size 683573

pdbbind-2020-refined-ki.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51a089b05f82c4006ce56d5a0723df2a5ebc6c467d3f8c73f5825d61b3ad5f20
+size 605185

pdbbind-2020-refined.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b1fe14ffba1c00750d99f815ad7cccf1c49c2f7eff16e324d0cedcb51d3baa9
-size 1417508
+oid sha256:0ac490c51782e49789c4813ec273d0ce2a2c37de7f99081f9494081db679c918
+size 1496256

standardize_data.ipynb

@@ -148,31 +148,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Processing PDBbind data: 100%|██████████| 195/195 [00:03<00:00, 49.05it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Saved 195 entries to pdbbind_2013_core.parquet\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import os\n",
     "from pathlib import Path\n",
@@ -200,6 +178,13 @@
     "    pdbbind_dir = Path(pdbbind_dir).expanduser()\n",
     "    parser = PDB.PDBParser(QUIET=True)\n",
     "    data = []\n",
+    "    data_ic50 = []\n",
+    "    data_ki = []\n",
+    "    data_kd = []\n",
+    "    data_equal_only = []\n",
+    "    data_ic50_equal_only = []\n",
+    "    data_ki_equal_only = []\n",
+    "    data_kd_equal_only = []\n",
     "\n",
     "    # Read the index file\n",
     "    df_index = pd.read_csv(index_file, sep='\\s+', header=None, comment= \"#\", usecols=[0,1,2,3,4,6,7],\n",
@@ -210,9 +195,12 @@
     "\n",
     "    # Use tqdm for progress tracking\n",
     "    with tqdm(total=total_entries, desc=\"Processing PDBbind data\") as pbar:\n",
+    "        # Create separate lists for different binding types\n",
+    "        \n",
     "        for _, row in df_index.iterrows():\n",
     "            pdb_id = row['PDB_ID']\n",
     "            subdir = pdbbind_dir / pdb_id\n",
+    "            kd_ki_value = row['Kd/Ki']\n",
     "\n",
     "            if subdir.is_dir():\n",
     "                # Process protein\n",
@@ -231,37 +219,70 @@
     "                        # Get binding affinity\n",
     "                        neg_log_10_affinity = row['-logKd/Ki']\n",
     "\n",
-    "                        # Add to data list\n",
-    "                        data.append({\n",
+    "                        # Create entry dictionary\n",
+    "                        entry = {\n",
     "                            'smiles_can': smiles,\n",
     "                            'seq': sequence,\n",
-    "                            'neg_log_10_affinity_M': neg_log_10_affinity\n",
-    "                        })\n",
+    "                            'neg_log_10_affinity_M': neg_log_10_affinity,\n",
+    "                            'affinity_uM': 10**(6-neg_log_10_affinity)\n",
+    "                        }\n",
+    "                        \n",
+    "                        # Add to appropriate list based on binding type\n",
+    "                        data.append(entry)  # Add to the combined list\n",
+    "                        \n",
+    "                        if kd_ki_value.startswith('IC50'):\n",
+    "                            data_ic50.append(entry)\n",
+    "                            if \"=\" in kd_ki_value:\n",
+    "                                data_ic50_equal_only.append(entry)\n",
+    "                                data_equal_only.append(entry)\n",
+    "                        elif kd_ki_value.startswith('Ki'):\n",
+    "                            data_ki.append(entry)\n",
+    "                            if \"=\" in kd_ki_value:\n",
+    "                                data_ic50_equal_only.append(entry)\n",
+    "                                data_equal_only.append(entry)\n",
+    "                        elif kd_ki_value.startswith('Kd'):\n",
+    "                            data_kd.append(entry)\n",
+    "                            if \"=\" in kd_ki_value:\n",
+    "                                data_ic50_equal_only.append(entry)\n",
+    "                                data_equal_only.append(entry)\n",
+    "\n",
     "\n",
     "            pbar.update(1)\n",
     "\n",
-    "    return pd.DataFrame(data)\n",
+    "    # Create dataframes for each binding type\n",
+    "    df_all = pd.DataFrame(data)\n",
+    "    df_ic50 = pd.DataFrame(data_ic50)\n",
+    "    df_ki = pd.DataFrame(data_ki)\n",
+    "    df_kd = pd.DataFrame(data_kd)\n",
+    "    df_all_equal_only = pd.DataFrame(data)\n",
+    "    df_ic50_equal_only = pd.DataFrame(data_ic50_equal_only)\n",
+    "    df_ki_equal_only = pd.DataFrame(data_ki_equal_only)\n",
+    "    df_kd_equal_only = pd.DataFrame(data_kd_equal_only)\n",
+    "    \n",
+    "    return df_all, df_ic50, df_ki, df_kd, df_all_equal_only, df_ic50_equal_only, df_ki_equal_only, df_kd_equal_only\n",
     "\n",
     "# Process data from PDBbind refined set\n",
-    "pdbbind_refined_dir = \"path/to/PDBbind_v2020_refined\"\n",
+    "pdbbind_refined_dir = \"/path/to/refined-set\"\n",
     "index_refined_file = \"/path/to/INDEX_refined_data.2020\"\n",
-    "df_refined = process_pdbbind_data(pdbbind_refined_dir, index_refined_file)\n",
+    "df_refined_all, df_refined_ic50, df_refined_ki, df_refined_kd, df_refined_all_equal_only, df_refined_ic50_equal_only, df_refined_ki_equal_only, df_refined_kd_equal_only = process_pdbbind_data(pdbbind_refined_dir, index_refined_file)\n",
+    "\n",
+    "df_refined_all.to_parquet('pdbbind-2020-refined.parquet', index=False)\n",
+    "df_refined_ic50.to_parquet('pdbbind-2020-refined-ic50.parquet', index=False)\n",
+    "df_refined_ki.to_parquet('pdbbind-2020-refined-ki.parquet', index=False)\n",
+    "df_refined_kd.to_parquet('pdbbind-2020-refined-kd.parquet', index=False)\n",
+    "df_refined_all_equal_only.to_parquet('pdbbind-2020-refined-all-equal-only.parquet', index=False)\n",
+    "df_refined_ic50_equal_only.to_parquet('pdbbind-2020-refined-ic50-equal-only.parquet', index=False)\n",
+    "df_refined_ki_equal_only.to_parquet('pdbbind-2020-refined-ki-equal-only.parquet', index=False)\n",
+    "df_refined_kd_equal_only.to_parquet('pdbbind-2020-refined-kd-equal-only.parquet', index=False)\n",
+    "\n",
     "\n",
     "# Process data from PDBbind general set\n",
     "pdbbind_general_dir = \"/path/to/PDBbind_v2020_other_PL\"\n",
     "index_general_file = \"/path/to/INDEX_general_PL_data.2020\"\n",
-    "df_general = process_pdbbind_data(pdbbind_general_dir, index_general_file)\n",
-    "\n",
-    "# Combine dataframes\n",
-    "df_combined = pd.concat([df_refined, df_general], ignore_index=True)\n",
-    "\n",
-    "# Remove duplicates (if any) and reset index\n",
-    "df_combined = df_combined.drop_duplicates().reset_index(drop=True)\n",
-    "\n",
-    "# Save to CSV\n",
-    "output_file = \"data/pdbbind_2020_combined.csv\"\n",
-    "df_combined.to_csv(output_file, index=False)\n",
-    "print(f\"Saved {len(df_combined)} entries to {output_file}\")"
+    "df_general_all, df_general_ic50, df_general_ki, df_general_kd, df_general_all_equal_only, df_general_ic50_equal_only, df_general_ki_equal_only, df_general_kd_equal_only = process_pdbbind_data(pdbbind_general_dir, index_general_file)\n",
+    "df_general_all.to_parquet('pdbbind-2020-general.parquet', index=False)\n",
+    "df_general_ki.to_parquet('pdbbind-2020-general-ki.parquet', index=False)\n",
+    "df_general_kd.to_parquet('pdbbind-2020-general-kd.parquet', index=False)\n"
    ]
   },
   {