use canonical normal form(s)

Author: jfennick

examples/diffdock/run_diffdock.wic

@@ -1,81 +1,81 @@
 steps:
 
-  - extract_pdbbind_refined:
-      in:
-        # https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.query.html
-        # "The query() method uses a slightly modified Python syntax by default.
-        # For example, the & and | (bitwise) operators have the precedence of their boolean cousins, and and or.
-        # This is syntactically valid Python, however the semantics are different."
-        query: !ii '(Kd_Ki == "Kd") and (value < 0.001)'
-        max_row: !ii 1 #25 # Use 1 for CI
-        convert_Kd_dG: !ii True
-      out:
-      - output_pdb_paths: !& pdbbind_pdbs
-      - output_sdf_paths: !& pdbbind_sdfs
-      - experimental_dGs: !& exp_dGs
+  - id: extract_pdbbind_refined
+    in:
+      # https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.query.html
+      # "The query() method uses a slightly modified Python syntax by default.
+      # For example, the & and | (bitwise) operators have the precedence of their boolean cousins, and and or.
+      # This is syntactically valid Python, however the semantics are different."
+      query: !ii '(Kd_Ki == "Kd") and (value < 0.001)'
+      max_row: !ii 1 #25 # Use 1 for CI
+      convert_Kd_dG: !ii True
+    out:
+    - output_pdb_paths: !& pdbbind_pdbs
+    - output_sdf_paths: !& pdbbind_sdfs
+    - experimental_dGs: !& exp_dGs
 
-  - fix_side_chain:
-      scatter: [input_pdb_path]
-      in:
-        input_pdb_path: !* pdbbind_pdbs
-      out:
-      - output_pdb_path: !& pdbbind_pdbs.pdb
+  - id: fix_side_chain
+    scatter: [input_pdb_path]
+    in:
+      input_pdb_path: !* pdbbind_pdbs
+    out:
+    - output_pdb_path: !& pdbbind_pdbs.pdb
 
-  - sanitize_ligand:
-      scatter: [input_small_mol_ligand]
-      in:
-        input_small_mol_ligand: !* pdbbind_sdfs
-      out:
-      - output_ligand: !& sanitized_sdfs
-      - valid_ligand: !& valid_ligands
+  - id: sanitize_ligand
+    scatter: [input_small_mol_ligand]
+    in:
+      input_small_mol_ligand: !* pdbbind_sdfs
+    out:
+    - output_ligand: !& sanitized_sdfs
+    - valid_ligand: !& valid_ligands
 
-  - filter_array: # remove invalid ligands from sanitized_ligand, avoid using null
-      in:
-        input_array: !* sanitized_sdfs
-        input_bool_array: !* valid_ligands
-      out:
-      - output_array: !& final_sanitized_sdfs
+  - id: filter_array # remove invalid ligands from sanitized_ligand, avoid using null
+    in:
+      input_array: !* sanitized_sdfs
+      input_bool_array: !* valid_ligands
+    out:
+    - output_array: !& final_sanitized_sdfs
 
-  - filter_array: # remove proteins corresponding to invalid ligands from sanitized_ligand
-      in:
-        input_array: !* pdbbind_pdbs.pdb
-        input_bool_array: !* valid_ligands
-      out:
-      - output_array: !& final_pdbbind_pdbs.pdb
+  - id: filter_array # remove proteins corresponding to invalid ligands from sanitized_ligand
+    in:
+      input_array: !* pdbbind_pdbs.pdb
+      input_bool_array: !* valid_ligands
+    out:
+    - output_array: !& final_pdbbind_pdbs.pdb
 
-  - filter_array: # remove dGs corresponding to invalid ligands from sanitized_ligand
-      in:
-        input_array: !* exp_dGs
-        input_bool_array: !* valid_ligands
-      out:
-      - output_array: !& final_exp_dGs
+  - id: filter_array # remove dGs corresponding to invalid ligands from sanitized_ligand
+    in:
+      input_array: !* exp_dGs
+      input_bool_array: !* valid_ligands
+    out:
+    - output_array: !& final_exp_dGs
 
-  - diffdock:
-      scatter: [protein_path, ligand_path]
-      scatterMethod: dotproduct
-      in:
-        protein_path: !* final_pdbbind_pdbs.pdb
-        ligand_path: !* final_sanitized_sdfs
-        samples_per_complex: !ii 20 # figure 3 left in DiffDock paper
-        inference_steps: !ii 20 # figure S11 in DiffDock paper
-      out:
-      - output_files: !& diffdock_poses
+  - id: diffdock
+    scatter: [protein_path, ligand_path]
+    scatterMethod: dotproduct
+    in:
+      protein_path: !* final_pdbbind_pdbs.pdb
+      ligand_path: !* final_sanitized_sdfs
+      samples_per_complex: !ii 20 # figure 3 left in DiffDock paper
+      inference_steps: !ii 20 # figure S11 in DiffDock paper
+    out:
+    - output_files: !& diffdock_poses
 
-  - rank_diffdock_poses:
-      scatter: [diffdock_poses]
-      in:
-        top_n_confident: !ii 1000 # if only using top_percent_confidence, then set top_n_confident to trivially high number
-        # if only want to use top_n_confident, then set top_percent_confidence to 100
-        top_percent_confidence: !ii 33 # take top third of most confident poses, see figure 3 right in DiffDock paper
-        diffdock_poses: !* diffdock_poses
-      out:
-      - output_poses: !& output_poses
+  - id: rank_diffdock_poses
+    scatter: [diffdock_poses]
+    in:
+      top_n_confident: !ii 1000 # if only using top_percent_confidence, then set top_n_confident to trivially high number
+      # if only want to use top_n_confident, then set top_percent_confidence to 100
+      top_percent_confidence: !ii 33 # take top third of most confident poses, see figure 3 right in DiffDock paper
+      diffdock_poses: !* diffdock_poses
+    out:
+    - output_poses: !& output_poses
 
-  - pose_cluster_filter:
-      scatter: [predicted_poses]
-      in:
-        predicted_poses: !* output_poses
-        centroid_cutoff: !ii 5 # if centroid of all poses are within cutoff then only keep most confident pose, requires visual inspection
+  - id: pose_cluster_filter
+    scatter: [predicted_poses]
+    in:
+      predicted_poses: !* output_poses
+      centroid_cutoff: !ii 5 # if centroid of all poses are within cutoff then only keep most confident pose, requires visual inspection
 
 wic:
   graphviz:

examples/diffdock/run_diffdock_weekly.wic

@@ -1,5 +1,5 @@
 steps:
-- run_diffdock.wic:
+  run_diffdock.wic:
 
 wic:
   steps:

examples/docking/align_protein_CA_pymol.wic

@@ -1,5 +1,5 @@
 steps:
-  - pymol_align_protein_ca:
+    pymol_align_protein_ca:
       in:
         input_1_path: !* receptor.xyz
         input_2_path: !* pose_ligand_1.pdb
@@ -9,7 +9,7 @@ steps:
         # pymol only supports saving to pdb format which does not support storing the time of each frame.
       out:
       - output_file_path: !& prod_align_protein_ca.pdb
-  - gmx_rms_nofit:
+    gmx_rms_nofit:
       in:
         input_structure_path: !* npt.gro # prod.tpr ?
         input_traj_path: !* prod_align_protein_ca.pdb
@@ -19,7 +19,7 @@ steps:
         output_xvg_path: !ii rmsd_equil_ligand_notime.txt # Use .txt to avoid plotting
       out:
       - output_xvg_path: !& rmsd_equil_ligand_notime
-  - bash_xvg:
+    bash_xvg:
       in:
         script: !ii /replace_first_column.sh # NOTE: Initial / required
         # But we can recover the time column by copying it from another xvg file.

examples/docking/assign_partial_charges.wic

@@ -6,15 +6,15 @@ inputs:
     - edam:format_3814
 
 steps:
-- convert_mol2:
+  convert_mol2:
     in:
       input_path: input_path #!* ligand_min.sdf
 #   out:
 #   - output_mol2_path: !& ligand_min.mol2
 # NOTE: If we directly convert from sdf to pdbqt, openbabel will NOT add
 # partial charges! We have to factor through mol2 format to trigger the
 # partial charge heuristics code path. This does affect the docking scores!
-- convert_pdbqt:
+  convert_pdbqt:
 #    in:
 #      input_path: !* ligand_min.mol2  # conformer.mol2
 #    out:

examples/docking/assign_partial_charges_batch.wic

@@ -1,13 +1,13 @@
 steps:
-- convert_mol2:
+  convert_mol2:
 #    in:
 #      input_path: !* ligand_min.sdf
 #    out:
 #    - output_mol2_path: !& ligand_min.mol2
 # NOTE: If we directly convert from sdf to pdbqt, openbabel will NOT add
 # partial charges! We have to factor through mol2 format to trigger the
 # partial charge heuristics code path. This does affect the docking scores!
-- convert_pdbqt:
+  convert_pdbqt:
 #    in:
 #      input_path: !* ligand_min.mol2  # conformer.mol2
 #    out:

examples/docking/autodock_vina_rescore.wic

@@ -7,55 +7,55 @@ inputs:
     format: edam:format_1476
 
 steps:
-- python_script:
-    in:
-      script: !ii ../scripts/atomselect.py
-      dockerPull: !ii jakefennick/atomselect
-      selection_string: !ii protein # Extract the protein from the last timestep
-      input_pdb_path: input_pdb_path # !* prod.pdb
+- id: python_script
+  in:
+    script: !ii ../scripts/atomselect.py
+    dockerPull: !ii jakefennick/atomselect
+    selection_string: !ii protein # Extract the protein from the last timestep
+    input_pdb_path: input_pdb_path # !* prod.pdb
 # Assign partial charges (protein)
 # NOTE: Although we only start off with one protein, the final coordinates
 # after doing MD with different ligands will be different; hence scatter
-- convert_mol2:
-- convert_pdbqt:
-    in:
-      arg1: !ii -xr # Receptor needs to be rigid
-    out:
-    - output_pdb_path: !& protein_prod.pdbqt
-- python_script:
-    in:
-      script: !ii ../scripts/atomselect.py
-      dockerPull: !ii jakefennick/atomselect
-      selection_string: !ii resname MOL # Extract the ligand from the last timestep
-      input_pdb_path: input_pdb_path # !* prod.pdb
-    # out:
-    # - output_pdb_path: !& ligand_temp.pdbqt
+- id: convert_mol2
+- id: convert_pdbqt
+  in:
+    arg1: !ii -xr # Receptor needs to be rigid
+  out:
+  - output_pdb_path: !& protein_prod.pdbqt
+- id: python_script
+  in:
+    script: !ii ../scripts/atomselect.py
+    dockerPull: !ii jakefennick/atomselect
+    selection_string: !ii resname MOL # Extract the ligand from the last timestep
+    input_pdb_path: input_pdb_path # !* prod.pdb
+  # out:
+  # - output_pdb_path: !& ligand_temp.pdbqt
 
 # It utilizes a helper PDB file to overwrite the atom element
-# types (last column) of the input PDB file 
+# types (last column) of the input PDB file
 # Input pdb file
 # ATOM   4653  C26 MOL A 286      44.880  26.370  42.560  1.00  0.00           C
 # ATOM   4654  BR  MOL A 286      48.190  27.540  40.260  1.00  0.00           B
 # Input helepr pdb file
 # ATOM     52  C26 MOL Z   1     -18.069 -34.542  22.368  1.00  0.00           C
 # ATOM     53   BR MOL Z   1     -14.786 -32.819  20.524  1.00  0.00          BR
-- fix_pdb_atom_column:
-    in:
-      input_helper_structure_path: input_ligand_pdb_path
+- id: fix_pdb_atom_column
+  in:
+    input_helper_structure_path: input_ligand_pdb_path
 # Assign partial charges (ligand)
-- convert_mol2:
-- convert_pdbqt:
-    out:
-    - output_pdb_path: !& ligand_prod.pdbqt
-- autodock_vina_rescore:
-    in:
-      input_ligand_pdbqt_path: !* ligand_prod.pdbqt
-      input_receptor_pdbqt_path: !* protein_prod.pdbqt
-      score_only: !ii True
-      #local_only: !ii True
-    out:
-    - output_log_path: !& vina_rescore.log
-    - docking_score: !& docking_rescores
+- id: convert_mol2
+- id: convert_pdbqt
+  out:
+  - output_pdb_path: !& ligand_prod.pdbqt
+- id: autodock_vina_rescore
+  in:
+    input_ligand_pdbqt_path: !* ligand_prod.pdbqt
+    input_receptor_pdbqt_path: !* protein_prod.pdbqt
+    score_only: !ii True
+    #local_only: !ii True
+  out:
+  - output_log_path: !& vina_rescore.log
+  - docking_score: !& docking_rescores
 
 wic:
   graphviz:
@@ -81,7 +81,7 @@ wic:
       wic:
         graphviz:
           label: Correct atom\nelement types
-    (6, convert_mol2): 
+    (6, convert_mol2):
       wic:
         graphviz:
           label: Assign Partial\nCharges

examples/docking/convert_ligand_mol2_to_pdbqt_mdanalysis.wic

@@ -1,33 +1,33 @@
 steps:
-  - python3_mol2_to_pdbqt:
-      in:
-        script: !ii /mol2_to_pdbqt.py # NOTE: Initial / required
-        input_mol2_path: !* conformer.mol2
-      # out:
-      # - output_pdb_path: !& ligand_periods.pdbqt
-  - bash_pdb:
-      in:
-        # MDAnalysis adds a . to the final atomtype column, which causes autodock to crash.
-        script: !ii /pdbqt_remove_trailing_period.sh # NOTE: Initial / required
-#        input_pdb_path: !* ligand_periods.pdbqt
-#      out:
-#      - output_pdb_path: !& ligand_keywords.pdbqt
-  - bash_pdb:
-      in:
-      # Remove the MODEL and ENDMDL keywords.
-      # Even if there is actually only one model, autodock simply looks for MODEL and ENDMDL.
-      # Unfortunately, all of the errors crash with the same cryptic message:
-      # Parse error on line ... in file ".../ligand.pdbqt": Unknown or inappropriate tag
-        script: !ii /pdbqt_remove_keywords.sh # NOTE: Initial / required
-#        input_pdb_path: !* ligand_keywords.pdbqt
-#      out:
-#      - output_pdb_path: !& ligand_temp.pdbqt
-  - bash_pdb:
-      in:
-      # Add ROOT, ENDROOT, TORSDOF keywords (if necessary).
-      # Unfortunately, all of the errors crash with the same cryptic message:
-      # Parse error on line ... in file ".../ligand.pdbqt": Unknown or inappropriate tag
-        script: !ii /pdbqt_remove_flex.sh # NOTE: Initial / required
+  - id: python3_mol2_to_pdbqt
+    in:
+      script: !ii /mol2_to_pdbqt.py # NOTE: Initial / required
+      input_mol2_path: !* conformer.mol2
+    # out:
+    # - output_pdb_path: !& ligand_periods.pdbqt
+  - id: bash_pdb
+    in:
+      # MDAnalysis adds a . to the final atomtype column, which causes autodock to crash.
+      script: !ii /pdbqt_remove_trailing_period.sh # NOTE: Initial / required
+#      input_pdb_path: !* ligand_periods.pdbqt
+#    out:
+#    - output_pdb_path: !& ligand_keywords.pdbqt
+  - id: bash_pdb
+    in:
+    # Remove the MODEL and ENDMDL keywords.
+    # Even if there is actually only one model, autodock simply looks for MODEL and ENDMDL.
+    # Unfortunately, all of the errors crash with the same cryptic message:
+    # Parse error on line ... in file ".../ligand.pdbqt": Unknown or inappropriate tag
+      script: !ii /pdbqt_remove_keywords.sh # NOTE: Initial / required
+#      input_pdb_path: !* ligand_keywords.pdbqt
+#    out:
+#    - output_pdb_path: !& ligand_temp.pdbqt
+  - id: bash_pdb
+    in:
+    # Add ROOT, ENDROOT, TORSDOF keywords (if necessary).
+    # Unfortunately, all of the errors crash with the same cryptic message:
+    # Parse error on line ... in file ".../ligand.pdbqt": Unknown or inappropriate tag
+      script: !ii /pdbqt_remove_flex.sh # NOTE: Initial / required
 #        input_pdb_path: !* ligand_temp.pdbqt
-      out:
-      - output_pdb_path: !& ligand_rigid.pdbqt
+    out:
+    - output_pdb_path: !& ligand_rigid.pdbqt