caddforge-docs 0.3.0__py3-none-any.whl

caddforge_docs/site/examples/manifest.json

@@ -0,0 +1,158 @@
+[
+  {
+    "section": "docking",
+    "tool": "unidock",
+    "config": "cadd_docking.unidock.workflow.UniDockConfig",
+    "path": "tools/docking-unidock.yaml"
+  },
+  {
+    "section": "docking",
+    "tool": "idockya",
+    "config": "cadd_docking.idockya.workflow.IDockYaConfig",
+    "path": "tools/docking-idockya.yaml"
+  },
+  {
+    "section": "docking",
+    "tool": "gnina",
+    "config": "cadd_docking.gnina.workflow.GninaConfig",
+    "path": "tools/docking-gnina.yaml"
+  },
+  {
+    "section": "descriptors",
+    "tool": "ani-psa",
+    "config": "cadd_descriptors.ani_psa.workflow.AniPsaConfig",
+    "path": "tools/descriptors-ani-psa.yaml"
+  },
+  {
+    "section": "descriptors",
+    "tool": "phesa",
+    "config": "cadd_descriptors.phesa.workflow.PhesaConfig",
+    "path": "tools/descriptors-phesa.yaml"
+  },
+  {
+    "section": "descriptors",
+    "tool": "ani-conformers",
+    "config": "cadd_descriptors.ani_conformers.workflow.AniConformersConfig",
+    "path": "tools/descriptors-ani-conformers.yaml"
+  },
+  {
+    "section": "descriptors",
+    "tool": "cluster-molecules",
+    "config": "cadd_descriptors.cluster_molecules.workflow.ClusterMoleculesConfig",
+    "path": "tools/descriptors-cluster-molecules.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "fix-structure",
+    "config": "cadd_md.fix_structure.workflow.FixStructureConfig",
+    "path": "tools/md-fix-structure.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "center-and-align",
+    "config": "cadd_md.center_and_align.workflow.CenterAndAlignConfig",
+    "path": "tools/md-center-and-align.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "extract-frames",
+    "config": "cadd_md.extract_frames.workflow.ExtractFramesConfig",
+    "path": "tools/md-extract-frames.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "cluster-representative",
+    "config": "cadd_md.cluster_representative.workflow.ClusterRepresentativeConfig",
+    "path": "tools/md-cluster-representative.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "plain-md",
+    "config": "cadd_md.plain_md.workflow.PlainMDConfig",
+    "path": "tools/md-plain-md.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "minimize",
+    "config": "cadd_md.minimize.MinimizeConfig",
+    "path": "tools/md-minimize.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "openbpmd",
+    "config": "cadd_md.openbpmd.workflow.OpenBPMDConfig",
+    "path": "tools/md-openbpmd.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "grand",
+    "config": "cadd_md.grand.workflow.GrandConfig",
+    "path": "tools/md-grand.yaml"
+  },
+  {
+    "section": "md",
+    "tool": "inducia",
+    "config": "cadd_md.inducia.workflow.InduciaConfig",
+    "path": "tools/md-inducia.yaml"
+  },
+  {
+    "section": "fep",
+    "tool": "openfe-abfe",
+    "config": "cadd_fep.openfe_abfe.workflow.OpenfeAbfeConfig",
+    "path": "tools/fep-openfe-abfe.yaml"
+  },
+  {
+    "section": "fep",
+    "tool": "openfe-rbfe-septop",
+    "config": "cadd_fep.openfe_rbfe_septop.workflow.OpenfeRbfeSeptopConfig",
+    "path": "tools/fep-openfe-rbfe-septop.yaml"
+  },
+  {
+    "section": "qm",
+    "tool": "transfer-free-energy",
+    "config": "cadd_qm.cosmors.workflow.CosmoRSConfig",
+    "path": "tools/qm-transfer-free-energy.yaml"
+  },
+  {
+    "section": "ml",
+    "tool": "simple-gp",
+    "config": "cadd_ml.simple_gp.workflow.SimpleGPConfig",
+    "path": "tools/ml-simple-gp.yaml"
+  },
+  {
+    "section": "ml",
+    "tool": "pca",
+    "config": "cadd_ml.pca.workflow.PCAConfig",
+    "path": "tools/ml-pca.yaml"
+  },
+  {
+    "section": "ml",
+    "tool": "model-zoo",
+    "config": "cadd_ml.model_zoo.workflow.ModelZooConfig",
+    "path": "tools/ml-model-zoo.yaml"
+  },
+  {
+    "section": "ml",
+    "tool": "model-factory",
+    "config": "cadd_ml.model_factory.workflow.ModelFactoryConfig",
+    "path": "tools/ml-model-factory.yaml"
+  },
+  {
+    "section": "ml",
+    "tool": "explore-classifier",
+    "config": "cadd_ml.explore_classifier.workflow.ExploreClassifierConfig",
+    "path": "tools/ml-explore-classifier.yaml"
+  },
+  {
+    "section": "datascience",
+    "tool": "csv-scatter",
+    "config": "cadd_datascience.csv_scatter.workflow.CsvScatterConfig",
+    "path": "tools/datascience-csv-scatter.yaml"
+  },
+  {
+    "section": "datascience",
+    "tool": "distribution-diagnostics",
+    "config": "cadd_datascience.distribution_diagnostics.workflow.DistributionDiagnosticsConfig",
+    "path": "tools/datascience-distribution-diagnostics.yaml"
+  }
+]

caddforge_docs/site/examples/minimize/component_overrides.yaml

@@ -0,0 +1,14 @@
+# Replace indices with zero-based INPUT residue indices, not PDB numbers.
+complex_pdb_path: complex_with_cofactor_and_water.pdb
+ligand_path: ligand.sdf
+mode: pocket_restrained
+cofactors:
+  - residue_index: 200
+    chemistry_path: cofactor.sdf
+component_mobility:
+  200:
+    mobility: restrained
+    restraint_strength: moderate # k = 1000 kJ mol^-1 nm^-2, E = 0.5*k*displacement^2
+  201:
+    mobility: free # retained water; flexible geometry, no positional restraints
+run_directory: runs/minimize-components

caddforge_docs/site/examples/minimize/ligand_only.yaml

@@ -0,0 +1,13 @@
+# Replace paths with one complex and its authoritative ligand chemistry.
+complex_pdb_path: complex.pdb
+ligand_path: ligand.sdf
+lig_resname: LIG
+mode: ligand_only
+pocket_radius_angstrom: 5.0
+shell_radius_angstrom: 5.0
+backbone_restraint_strength: moderate
+shell_restraint_strength: very_strong
+tolerance_kj_mol_nm: 10.0
+max_iterations: 1000
+platform: CPU
+run_directory: runs/minimize-ligand_only

caddforge_docs/site/examples/minimize/pipeline_docking.yaml

@@ -0,0 +1,84 @@
+# Rigid-receptor GNINA docking followed by local OpenMM pose cleanup.
+# Supply GNINA on PATH and authoritative ligand chemistry in the input SDF.
+version: 1
+name: dock-extract-minimize
+
+defaults:
+  backend: local
+  run_directory: runs/dock-extract-minimize
+  registry_path: runs/job_registry.sqlite
+  export_results: false
+
+stages:
+  - id: dock
+    tool: docking/gnina
+    config:
+      receptor_path: data/receptor.pdb
+      ligand_path: data/ligands.sdf
+      autobox_ligand_path: data/reference_ligand.sdf
+      output_format: sdf
+    outputs:
+      poses: poses_dir
+      scores: best_scores_csv
+
+  - id: extract
+    type: poses.extract
+    inputs:
+      poses: dock.poses
+      table: dock.scores
+    config:
+      ligand_column: ligand
+      path_column: pose_file
+      output_name: pose_manifest.csv
+
+  - id: hydrogenate
+    type: molecules.hydrogenate
+    inputs:
+      molecules: extract.poses
+      table: extract.manifest
+    config:
+      on_template_failure: fail
+      on_failure: fail
+      add_coords: true
+
+  - id: minimize
+    tool: md/minimize
+    config:
+      protein_pdb_path: data/receptor.pdb
+      mode: sidechains_free_backbone_restrained
+      backbone_restraint_strength: moderate
+      strict_quality: false
+    inputs:
+      ligand: hydrogenate.molecules
+    scatter:
+      by: ligand_path
+      chunk_size: 1
+    outputs:
+      results: result_record
+      ligands: minimized_ligand
+
+  - id: collect
+    type: results.collect
+    inputs:
+      results: minimize.results
+    config:
+      metrics:
+        quality_passed: metrics.quality_passed
+        quality_flag_count: metrics.quality_flag_count
+        convergence: metrics.convergence_status
+        ligand_rmsd_angstrom: metrics.ligand_heavy_atom_rmsd_angstrom
+        minimized_complex_file: paths.minimized_complex
+        minimized_ligand_file: paths.minimized_ligand
+        diagnostics_file: paths.diagnostics
+      output_name: minimized_pose_metrics.csv
+
+  - id: accepted
+    type: filter.tabular
+    inputs:
+      table: collect.table
+    config:
+      filters:
+        - column: quality_passed
+          op: '=='
+          value: 1
+      output_name: accepted_minimized_poses.csv

caddforge_docs/site/examples/minimize/pipeline_scatter.yaml

@@ -0,0 +1,49 @@
+# Poses must share receptor coordinates and contain explicit hydrogens.
+# Paths resolve relative to the submission working directory.
+version: 1
+name: pose-cleanup
+
+defaults:
+  backend: local
+  run_directory: runs/pose-cleanup
+  registry_path: runs/job_registry.sqlite
+  export_results: false
+
+stages:
+  - id: minimize
+    tool: md/minimize
+    config:
+      protein_pdb_path: receptor.pdb
+      ligand_path: poses.sdf
+      mode: ligand_only
+    scatter:
+      by: ligand_path
+      chunk_size: 1
+    outputs:
+      results: result_record
+      ligands: minimized_ligand
+
+  - id: collect
+    type: results.collect
+    inputs:
+      results: minimize.results
+    config:
+      metrics:
+        quality_passed: metrics.quality_passed
+        convergence: metrics.convergence_status
+        ligand_rmsd_angstrom: metrics.ligand_heavy_atom_rmsd_angstrom
+        minimized_ligand_file: paths.minimized_ligand
+
+  - id: accepted
+    type: filter.tabular
+    inputs:
+      table: collect.table
+    config:
+      filters:
+        - column: quality_passed
+          op: '=='
+          value: 1
+# Inspect rejected rows and detailed diagnostics; cleanup energy is not affinity.
+# When consuming an upstream poses.extract.poses artifact, replace ligand_path
+# with inputs: {ligand: extract.poses}. Core preserves its pose IDs in the
+# scatter manifest and joins them into the collected table automatically.

caddforge_docs/site/examples/minimize/pocket_free_shell_restrained.yaml

@@ -0,0 +1,13 @@
+# Replace paths with one complex and its authoritative ligand chemistry.
+complex_pdb_path: complex.pdb
+ligand_path: ligand.sdf
+lig_resname: LIG
+mode: pocket_free_shell_restrained
+pocket_radius_angstrom: 5.0
+shell_radius_angstrom: 5.0
+backbone_restraint_strength: moderate
+shell_restraint_strength: very_strong
+tolerance_kj_mol_nm: 10.0
+max_iterations: 1000
+platform: CPU
+run_directory: runs/minimize-pocket_free_shell_restrained

caddforge_docs/site/examples/minimize/pocket_restrained.yaml

@@ -0,0 +1,13 @@
+# Replace paths with one complex and its authoritative ligand chemistry.
+complex_pdb_path: complex.pdb
+ligand_path: ligand.sdf
+lig_resname: LIG
+mode: pocket_restrained
+pocket_radius_angstrom: 5.0
+shell_radius_angstrom: 5.0
+backbone_restraint_strength: moderate
+shell_restraint_strength: very_strong
+tolerance_kj_mol_nm: 10.0
+max_iterations: 1000
+platform: CPU
+run_directory: runs/minimize-pocket_restrained

caddforge_docs/site/examples/minimize/prepared_metal.yaml

@@ -0,0 +1,14 @@
+# Supply a scientifically validated nonperiodic Amber metal-site model.
+# This template does not generate or validate a model's physical accuracy.
+complex_pdb_path: complex_with_metal.pdb
+ligand_path: ligand.sdf
+amber_topology_path: prepared.prmtop
+amber_coordinates_path: prepared.inpcrd
+amber_mapping_path: prepared_mapping.json
+mode: sidechains_free_backbone_restrained
+metal_sites:
+  - atom_index: 1500 # zero-based INPUT atom index, not PDB serial
+    oxidation_state: 2
+    coordinating_atom_indices: [123, 234, 345, 456] # replace with actual donors
+    parameter_source: "Replace with parameter-set citation, version, and site model"
+run_directory: runs/minimize-metal

caddforge_docs/site/examples/minimize/sidechains_free_backbone_fixed.yaml

@@ -0,0 +1,13 @@
+# Replace paths with one complex and its authoritative ligand chemistry.
+complex_pdb_path: complex.pdb
+ligand_path: ligand.sdf
+lig_resname: LIG
+mode: sidechains_free_backbone_fixed
+pocket_radius_angstrom: 5.0
+shell_radius_angstrom: 5.0
+backbone_restraint_strength: moderate
+shell_restraint_strength: very_strong
+tolerance_kj_mol_nm: 10.0
+max_iterations: 1000
+platform: CPU
+run_directory: runs/minimize-sidechains_free_backbone_fixed

caddforge_docs/site/examples/minimize/sidechains_free_backbone_restrained.yaml

@@ -0,0 +1,13 @@
+# Replace paths with one complex and its authoritative ligand chemistry.
+complex_pdb_path: complex.pdb
+ligand_path: ligand.sdf
+lig_resname: LIG
+mode: sidechains_free_backbone_restrained
+pocket_radius_angstrom: 5.0
+shell_radius_angstrom: 5.0
+backbone_restraint_strength: moderate
+shell_restraint_strength: very_strong
+tolerance_kj_mol_nm: 10.0
+max_iterations: 1000
+platform: CPU
+run_directory: runs/minimize-sidechains_free_backbone_restrained

caddforge_docs/site/examples/pipelines/descriptor-model-and-candidate-filter.yaml

@@ -0,0 +1,44 @@
+# Advanced template: supply all prepared inputs and scientific runtimes.
+version: 1
+name: descriptor-ml-filter
+
+defaults:
+  backend: local
+  registry_path: runs/job_registry.sqlite
+
+stages:
+  - id: psa_descriptors
+    tool: descriptors/ani-psa
+    config: configs/ani_psa_batch.yaml
+    outputs:
+      # Expose Result.paths["results_csv"] as psa_descriptors.table.
+      table: results_csv
+
+  - id: train_gp
+    tool: ml/simple-gp
+    config:
+      train_path: data/training_descriptors.csv
+      target_column: measured_property
+      # Use the table artifact from the descriptor stage as the GP test/prediction set.
+      test_path: psa_descriptors.table
+      feature_columns: [ani_psa, heavy_atom_count]
+      kernel: rbf
+      mode: train
+      model_path: artifacts/simple_gp_model.joblib
+    outputs:
+      # Expose Result.paths["predictions_csv"] as train_gp.predictions.
+      predictions: predictions_csv
+
+  - id: promising_candidates
+    type: filter.tabular
+    inputs:
+      table: train_gp.predictions
+    config:
+      filters:
+        - column: y_pred
+          op: ">="
+          value: 7.0
+      sort_by: y_pred
+      ascending: false
+      top_n: 250
+      output_name: promising_candidates.csv

caddforge_docs/site/examples/pipelines/dock-extract-and-minimize-poses.yaml

@@ -0,0 +1,85 @@
+# Advanced template: supply all prepared inputs and scientific runtimes.
+# Rigid-receptor GNINA docking followed by local OpenMM pose cleanup.
+# Supply GNINA on PATH and authoritative ligand chemistry in the input SDF.
+version: 1
+name: dock-extract-minimize
+
+defaults:
+  backend: local
+  run_directory: runs/dock-extract-minimize
+  registry_path: runs/job_registry.sqlite
+  export_results: false
+
+stages:
+  - id: dock
+    tool: docking/gnina
+    config:
+      receptor_path: data/receptor.pdb
+      ligand_path: data/ligands.sdf
+      autobox_ligand_path: data/reference_ligand.sdf
+      output_format: sdf
+    outputs:
+      poses: poses_dir
+      scores: best_scores_csv
+
+  - id: extract
+    type: poses.extract
+    inputs:
+      poses: dock.poses
+      table: dock.scores
+    config:
+      ligand_column: ligand
+      path_column: pose_file
+      output_name: pose_manifest.csv
+
+  - id: hydrogenate
+    type: molecules.hydrogenate
+    inputs:
+      molecules: extract.poses
+      table: extract.manifest
+    config:
+      on_template_failure: fail
+      on_failure: fail
+      add_coords: true
+
+  - id: minimize
+    tool: md/minimize
+    config:
+      protein_pdb_path: data/receptor.pdb
+      mode: sidechains_free_backbone_restrained
+      backbone_restraint_strength: moderate
+      strict_quality: false
+    inputs:
+      ligand: hydrogenate.molecules
+    scatter:
+      by: ligand_path
+      chunk_size: 1
+    outputs:
+      results: result_record
+      ligands: minimized_ligand
+
+  - id: collect
+    type: results.collect
+    inputs:
+      results: minimize.results
+    config:
+      metrics:
+        quality_passed: metrics.quality_passed
+        quality_flag_count: metrics.quality_flag_count
+        convergence: metrics.convergence_status
+        ligand_rmsd_angstrom: metrics.ligand_heavy_atom_rmsd_angstrom
+        minimized_complex_file: paths.minimized_complex
+        minimized_ligand_file: paths.minimized_ligand
+        diagnostics_file: paths.diagnostics
+      output_name: minimized_pose_metrics.csv
+
+  - id: accepted
+    type: filter.tabular
+    inputs:
+      table: collect.table
+    config:
+      filters:
+        - column: quality_passed
+          op: '=='
+          value: 1
+      output_name: accepted_minimized_poses.csv

caddforge_docs/site/examples/pipelines/docking-screen-and-rescore.yaml

@@ -0,0 +1,55 @@
+# Advanced template: supply all prepared inputs and scientific runtimes.
+version: 1
+name: docking-screen-rescore
+
+defaults:
+  backend: local
+  registry_path: runs/job_registry.sqlite
+  job_name: docking-screen-rescore
+
+stages:
+  - id: unidock_screen
+    tool: docking/unidock
+    config: configs/unidock.yaml
+    scatter:
+      # Split the ligand_multisdf_path field from configs/unidock.yaml into independent shards.
+      by: ligand_multisdf_path
+      chunk_size: 50000
+    outputs:
+      # Expose Result.paths["best_scores_csv"] as unidock_screen.scores.
+      scores: best_scores_csv
+      # Expose Result.paths["poses_dir"] as unidock_screen.poses.
+      poses: poses_dir
+
+  - id: top_unidock_scores
+    type: filter.tabular
+    inputs:
+      # Read the score table produced by the previous stage.
+      table: unidock_screen.scores
+    config:
+      sort_by: best_score
+      ascending: true
+      top_n: 1000
+      output_name: top_unidock_scores.csv
+
+  - id: retain_unidock_poses
+    type: filter.molecules
+    inputs:
+      # Filter the pose directory/path collection from UniDock.
+      molecules: unidock_screen.poses
+      # Use the filtered table from filter.tabular; this built-in always emits filtered_table.
+      ids: top_unidock_scores.filtered_table
+    config:
+      # The values in this column are matched against pose metadata or filenames.
+      id_column: ligand
+
+  - id: gnina_rescore
+    tool: docking/gnina
+    config: configs/gnina_rescore.yaml
+    inputs:
+      # Bind the retained pose path collection to GninaConfig.ligand_paths.
+      ligand_paths: retain_unidock_poses.molecules
+    outputs:
+      # Expose Gnina result paths for downstream stages.
+      scores: best_scores_csv
+      poses: poses_dir