Update the RMSD plotting script

BioinfoMachineLearning · Aug 12, 2024 · a17d91a · a17d91a
1 parent 60489ab
commit a17d91a
Showing 1 changed file with 33 additions and 5 deletions.
diff --git a/src/data/components/plot_dataset_rmsd.py b/src/data/components/plot_dataset_rmsd.py
@@ -105,6 +105,7 @@ def plot_dataset_rmsd(
     filtered_ids_to_keep_file: Optional[str] = None,
     filtered_ids_to_skip: Optional[Set[str]] = None,
     is_casp_dataset: bool = False,
+    public_plots: bool = True,
     accurate_rmsd_threshold: float = 4.0,
     accurate_tm_score_threshold: float = 0.7,
 ):
@@ -119,6 +120,7 @@ def plot_dataset_rmsd(
     :param filtered_ids_to_keep_file: File containing IDs of sequences to keep.
     :param filtered_ids_to_skip: Set of IDs of sequences to skip.
     :param is_casp_dataset: Whether the dataset is a CASP dataset.
+    :param public_plots: Whether to save the public versions of the plots.
     :param accurate_rmsd_threshold: RMSD threshold for accurate predictions.
     :param accurate_tm_score_threshold: TM-score threshold for accurate predictions.
     """
@@ -135,7 +137,12 @@ def plot_dataset_rmsd(
 
     dataset_rows = []
 
-    for pred_pdb_file in tqdm(os.listdir(pred_pdb_dir), desc=f"Plotting RMSD for {dataset_name}"):
+    dataset_suffix = " (Public)" if is_casp_dataset and public_plots else ""
+
+    for pred_pdb_file in tqdm(
+        os.listdir(pred_pdb_dir),
+        desc=f"Plotting RMSD for {dataset_name}{dataset_suffix}",
+    ):
         pdb_id = os.path.splitext(os.path.basename(pred_pdb_file))[0].split("_holo")[0]
 
         if filtered_ids_to_keep is not None and pdb_id not in filtered_ids_to_keep:
@@ -193,10 +200,10 @@ def plot_dataset_rmsd(
         / dataset_df.shape[0]
     )
     logging.info(
-        f"For the {dataset_name} dataset, {accurate_predictions_percent * 100:.2f}% of the predictions have RMSD < {accurate_rmsd_threshold} and TM-score > {accurate_tm_score_threshold}."
+        f"For the {dataset_name}{dataset_suffix} dataset, {accurate_predictions_percent * 100:.2f}% of the predictions have RMSD < {accurate_rmsd_threshold} and TM-score > {accurate_tm_score_threshold}."
     )
 
-    plot_dir = Path(output_dir) / ("public_plots" if is_casp_dataset else "plots")
+    plot_dir = Path(output_dir) / ("public_plots" if is_casp_dataset and public_plots else "plots")
     plot_dir.mkdir(exist_ok=True)
 
     plt.clf()
@@ -280,11 +287,32 @@ def main(cfg: DictConfig):
         ),
         usalign_exec_path=cfg.usalign_exec_path,
         filtered_ids_to_skip={
-            "T1170"
-        },  # NOTE: We don't score this target due to CASP internal parsing issues
+            "T1127v2",
+            "T1146",
+            "T1170",
+            "T1181",
+            "T1186",
+        },  # NOTE: We don't score `T1170` due to CASP internal parsing issues
         is_casp_dataset=True,
+        public_plots=True,
     )
 
+    # plot_dataset_rmsd(
+    #     "CASP15 Set",
+    #     os.path.join(cfg.data_dir, "casp15_set", "predicted_structures"),
+    #     os.path.join(cfg.data_dir, "casp15_set", "targets"),
+    #     os.path.join(
+    #         cfg.data_dir,
+    #         "casp15_set",
+    #     ),
+    #     usalign_exec_path=cfg.usalign_exec_path,
+    #     filtered_ids_to_skip={
+    #         "T1170",
+    #     },  # NOTE: We don't score `T1170` due to CASP internal parsing issues
+    #     is_casp_dataset=True,
+    #     public_plots=False,
+    # )
+
 
 if __name__ == "__main__":
     main()