Typo fix

README.md

@@ -26,8 +26,8 @@ Contents
 
 The repository is organised in the following way:
 - `data`
-    -  `training_database.csv`: Contains chemist scoring data in the `chemist_score` column, in addition to pre-calculated synthetic accessibility scores using the SAScore and SCScore. A 1 corresponds to a molecule that is easy-to-synthesise and a 0 corresponds to one which is difficult-to-synthesise.
-    - `chenist_scores.json`: Contains chemist scoring in `.JSON` format. Molecules are in InChi format, in addition the chemist scores in the `synthesisable` column. A 1 corresponds to a molecule that is easy-to-synthesise and a 0 corresponds to one which is difficult-to-synthesise.
+    -  `training_database.csv`: Contains chemist scoring data in the `chemist_score` column, in addition to pre-calculated synthetic accessibility scores using the SAScore and SCScore. 1 corresponds to a molecule that is easy-to-synthesise and a 0 corresponds to one which is difficult-to-synthesise.
+    - `chenist_scores.json`: Contains chemist scoring in `.JSON` format. Molecules are in InChi format, in addition the chemist scores in the `synthesisable` column. 1 corresponds to a molecule that is easy-to-synthesise and a 0 corresponds to one which is difficult-to-synthesise.
     - `chemist_data`: Folder containing all classification data collected from chemists.
         - `chemist_name_.csv/.json`: Contains all the chemist scores obtained from the website.
     - `training_mols.json`: Contains all molecules provided to chemists in InChi format, in addition to 3D coordinates generated by the ETKDG algorithm.

notebooks/database_analysis.ipynb

@@ -209,9 +209,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 441,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The correlation between the SAScore and SCScore is: 0.2732653645310837\n"
+     ]
+    }
+   ],
    "source": [
     "# Correlation between the SAScore and SCScore\n",
     "def plot_figure4(recalculate_scores=False):\n",
@@ -246,7 +254,51 @@
     "    g.ax_joint.legend_.remove()\n",
     "    g.ax_joint.tick_params('both', labelsize='medium')\n",
     "    return g\n",
-    "fig4 = plot_figure4(recalculate_scores=True)"
+    "fig4 = plot_figure4(recalculate_scores=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 467,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<AxesSubplot:xlabel='scs_norm', ylabel='Count'>"
+      ]
+     },
+     "execution_count": 467,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "plt.clf()\n",
+    "fig, ax = plt.subplots()\n",
+    "sns.histplot(df_training[df_training['chemist_score'] == 1]['scs_norm'], color='r', ax=ax)\n",
+    "sns.histplot(df_training[df_training['chemist_score'] == 0]['scs_norm'], color='b', ax=ax)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 468,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "execution_count": 468,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "fig"
    ]
   },
   {