diff --git a/experiments/001_example-ligand-only-graph-subset.py b/experiments/001_example-ligand-only-graph-EGFR-subset.py similarity index 84% rename from experiments/001_example-ligand-only-graph-subset.py rename to experiments/001_example-ligand-only-graph-EGFR-subset.py index a417275..1dbfe7b 100644 --- a/experiments/001_example-ligand-only-graph-subset.py +++ b/experiments/001_example-ligand-only-graph-EGFR-subset.py @@ -5,7 +5,7 @@ # DATA -- Glob paths must be relative to the root of the repository: REPO / features PARQUET_FILES = [ - "ligand-only-graph-subsample/_output/ligand__SmilesToLigandFeaturizer__GraphLigandFeaturizer/ChEMBLDatasetProvider/*.parquet", + "ligand-only-graph-EGFR-subsample/_output/ligand__SmilesToLigandFeaturizer__GraphLigandFeaturizer/ChEMBLDatasetProvider/*.parquet", ] # Model -- specified with the full import path to the class object @@ -38,4 +38,4 @@ BOOTSTRAP_SAMPLE_RATIO = 1 # Output -VERBOSE = False \ No newline at end of file +VERBOSE = False diff --git a/features/ligand-only-graph-subsample.py b/features/ligand-only-graph-EGFR-subsample.py similarity index 81% rename from features/ligand-only-graph-subsample.py rename to features/ligand-only-graph-EGFR-subsample.py index ac0d8a6..70cfd7d 100644 --- a/features/ligand-only-graph-subsample.py +++ b/features/ligand-only-graph-EGFR-subsample.py @@ -7,7 +7,7 @@ DATASET_CLS = "kinoml.datasets.chembl.ChEMBLDatasetProvider" DATASET_KWARGS = { - "path_or_url": "https://github.com/openkinome/kinodata/releases/download/v0.2/activities-chembl28-sample100_v0.2.zip", + "path_or_url": "https://github.com/openkinome/kinodata/releases/download/v0.3/EGFR-activities-chembl29-sample.zip", } PIPELINES = { @@ -24,14 +24,10 @@ FEATURIZE_KWARGS = {"keep": False} GROUPS = [ - [ - "kinoml.datasets.groups.CallableGrouper", - {"function": "lambda measurement: measurement.system.protein.name"}, - ], # by kinase [ "kinoml.datasets.groups.CallableGrouper", {"function": "lambda measurement: type(measurement).__name__"}, ], # by measurement type ] -TRAIN_TEST_VAL_KWARGS = {"idx_train": 0.8, "idx_test": 0.1, "idx_val": 0.1} \ No newline at end of file +TRAIN_TEST_VAL_KWARGS = {"idx_train": 0.8, "idx_test": 0.1, "idx_val": 0.1} diff --git a/tests/experiments/test_model_ligand_only.sh b/tests/experiments/test_model_ligand_only.sh index afead2d..b12d4ac 100644 --- a/tests/experiments/test_model_ligand_only.sh +++ b/tests/experiments/test_model_ligand_only.sh @@ -6,8 +6,11 @@ echo "Models" echo "Ligand only" -echo "Running model for '001_example-ligand-only-morgan1024-EGFR-subset'" -python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-morgan1024-EGFR-subset.py --overwrite +echo "Running model for '001_example-ligand-only-graph-EGFR-subset'" +python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-graph-EGFR-subset.py --overwrite + +# echo "Running model for '001_example-ligand-only-morgan1024-EGFR-subset'" +# python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-morgan1024-EGFR-subset.py --overwrite # These won't run because there are not enough data points for each kinase on the sample data. @@ -18,4 +21,4 @@ python run_notebook.py experiments/torch-train-test-debug-template.ipynb experim # python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-smiles-subset.py --overwrite # echo "Running model for '001_example-ligand-only-graph-subset'" -# python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-graph-subset.py --overwrite +# python run_notebook.py experiments/torch-train-test-debug-template.ipynb experiments/001_example-ligand-only-graph-subset.py --overwrite \ No newline at end of file diff --git a/tests/features/test_featurization_ligand_only.sh b/tests/features/test_featurization_ligand_only.sh index 3b92e08..15efe2a 100644 --- a/tests/features/test_featurization_ligand_only.sh +++ b/tests/features/test_featurization_ligand_only.sh @@ -7,14 +7,13 @@ echo "Featurization" echo "Ligand only" echo "Running featurization for 'ligand-only-morgan1024-EGFR-subsample'" -python run_notebook.py features/featurize-template.ipynb features/ligand-only-morgan1024-EGFR-subsample.py --overwrite +# python run_notebook.py features/featurize-template.ipynb features/ligand-only-morgan1024-EGFR-subsample.py --overwrite echo "Running featurization for 'ligand-only-morgan1024-subsample'" -python run_notebook.py features/featurize-template.ipynb features/ligand-only-morgan1024-subsample.py --overwrite +# python run_notebook.py features/featurize-template.ipynb features/ligand-only-morgan1024-subsample.py --overwrite echo "Running featurization for 'ligand-only-onehotsmiles-subsample'" -python run_notebook.py features/featurize-template.ipynb features/ligand-only-onehotsmiles-subsample.py --overwrite - -echo "Running featurization for 'ligand-only-graph-subsample'" -python run_notebook.py features/featurize-template.ipynb features/ligand-only-graph-subsample.py --overwrite +# python run_notebook.py features/featurize-template.ipynb features/ligand-only-onehotsmiles-subsample.py --overwrite +echo "Running featurization for 'ligand-only-graph-EGFR-subsample'" +python run_notebook.py features/featurize-template.ipynb features/ligand-only-graph-EGFR-subsample.py --overwrite