Update CSVs and conformation analysis script

BioinfoMachineLearning · Dec 12, 2024 · 4351e99 · 4351e99
1 parent 2252abf
commit 4351e99
Show file tree

Hide file tree

Showing 7 changed files with 989 additions and 496 deletions.
diff --git a/configs/scripts/evaluate_baseline_protein_conformation_changes.yaml b/configs/scripts/evaluate_baseline_protein_conformation_changes.yaml
@@ -1,5 +1,5 @@
 method: dynamicbind # the method to evaluate - NOTE: must be one of (`dynamicbind`, `neuralplexer`, `flowdock`)
-dataset: posebusters_benchmark # the dataset to use - NOTE: must be one of (`posebusters_benchmark`, `astex_diverse`, `dockgen`, `casp15`)
+dataset: posebusters_benchmark # the dataset to use - NOTE: must be one of (`posebusters_benchmark`, `astex_diverse`, `dockgen`)
 input_protein_structure_dir: ${oc.env:PROJECT_ROOT}/data/${dataset}_set/${dataset}_holo_aligned_predicted_structures # the input protein structure directory to parse
 predicted_protein_structure_dir: ${oc.env:PROJECT_ROOT}/forks/${resolve_method_title:${method}}/inference/${method}_${dataset}_outputs_1 # the predicted protein structure directory to parse
 reference_structure_dir: ${oc.env:PROJECT_ROOT}/data/${dataset}_set/ # the reference protein and ligand structure directory to parse

diff --git a/forks/DiffDock/inference/diffdock_dockgen_inputs.csv b/forks/DiffDock/inference/diffdock_dockgen_inputs.csv
diff --git a/forks/FlowDock/inference/flowdock_dockgen_inputs.csv b/forks/FlowDock/inference/flowdock_dockgen_inputs.csv
diff --git a/forks/FlowDock/inference/flowdock_posebusters_benchmark_inputs.csv b/forks/FlowDock/inference/flowdock_posebusters_benchmark_inputs.csv
diff --git a/forks/NeuralPLexer/inference/neuralplexer_dockgen_inputs.csv b/forks/NeuralPLexer/inference/neuralplexer_dockgen_inputs.csv
diff --git a/forks/NeuralPLexer/inference/neuralplexer_posebusters_benchmark_inputs.csv b/forks/NeuralPLexer/inference/neuralplexer_posebusters_benchmark_inputs.csv
diff --git a/scripts/evaluate_baseline_protein_conformation_changes.py b/scripts/evaluate_baseline_protein_conformation_changes.py
@@ -36,17 +36,19 @@ def main(cfg: DictConfig):
         desc="Evaluating protein conformation changes",
     ):
         pdb_id = os.path.basename(item).split("_holo_aligned")[0]
+        reference_protein_suffix = "_processed" if cfg.dataset == "dockgen" else ""
+        reference_ligand_file_ext = ".pdb" if cfg.dataset == "dockgen" else ".sdf"
 
         # Parse filepaths
         input_protein_structure_filepath = os.path.join(
             cfg.input_protein_structure_dir,
             f"{pdb_id}_holo_aligned_predicted_protein.pdb",
         )
         reference_protein_structure_filepath = os.path.join(
-            cfg.reference_structure_dir, pdb_id, f"{pdb_id}_protein.pdb"
+            cfg.reference_structure_dir, pdb_id, f"{pdb_id}_protein{reference_protein_suffix}.pdb"
         )
         reference_ligand_structure_filepath = os.path.join(
-            cfg.reference_structure_dir, pdb_id, f"{pdb_id}_ligand.sdf"
+            cfg.reference_structure_dir, pdb_id, f"{pdb_id}_ligand{reference_ligand_file_ext}"
         )
 
         if cfg.method == "dynamicbind":