Icon

02_​R-Group_​Decomposition

III. Find how many molecules with each combination of the two selected R-groups are in the dataset
  1. Execute all preceding nodes in the workflow first.

  2. Explore the View of the "Pick R-Groups for visualization" component by clicking on the magnifying glass.

  3. Configure the Pivot node using flow variables carrying the column names with the selected R-groups:
    - Available flow variables can be found in the output of the preceding node (Pick R-groups for visualization component) in the Flow Variables Tab
    - Open configuration dialog of the Pivoting node
    - Groups Tab: include one column (e.g. R1); Pivots Tab: include one column (e.g. R2)
    - Manual Aggregation Tab: Add column with pIC50 values and aggregate their count
    - Tick to "enable hiliting" option in the bottom of the dialog
    - Flow Variables Tab: choose Rgroup1 for InclList variable in the grouByColumns; choose Rgroup2 for InclList variable in the pivotColumns;
    - In the Advanced Settings: Set column name to “pivot name”

  4. Execute the Pivoting node and explore the output table

II. Perform R-Group Decomposition
  1. Execute all preceding nodes

  2. Connect an RDKit R-Group Decomposition node to the original Table Reader

  3. Connect the Table Row to Variable Node to it

  4. Configure the RDKit R-Group Decomposition node to use the variable MCS for the smarts_value in the Flow Variables Tab

  5. Connect it to the RDKit From Molecule node

I. Compute Maximum Common Substructure (MCS)
  1. Execute the Table Reader node.

  2. Use the RDKit MCS node to compute the MCS for the molecules in the dataset.
    Configure the node:

    - to match a fraction of the molecules (e.g. 0.8);
    - so the ring atoms match only ring atoms
    - to match complete rings only
    - to match valences.

  3. Execute the node and explore its output in the bottom pane

  4. Connect the output port of the RDKit MCS node with the Table Row to Variable node.

02_R-Group-Decomposition

This is an exercise workflow to perform R-Group Decomposition.
To complete the workflow follow the instructions in the annotation boxes below.

Bonus Exercise: Drawing a core with the Molecule Sketcher instead of generating it with MCS.
  1. Open the View of the Molecule Sketcher

  2. Draw a core

  3. Close and apply settings temporarily

  4. Explore the output of the sketcher node.


Visualize the results of the R-group decomposition

Execute all preceding nodes in the workflow first.

The view should have 4 elements:

  1. The pivoted table in which the row headers represent the first R-group and the column headers represent the second R-group. The Table View (JavaScript) node "pivot table" is already configured. Execute the node and open the interactive view (Right Click > Execute and Open Views). Explore the Table.

  2. The images of the molecules displayed like tiles. The images are available from the first (top) output of the Render to Images metanode. Use a new Tile View (JavaScript) node and configure it:
    - Options Tab: fix number of tiles per row to 1; choose a title column to show compound_chembl_id; include columns "pIC50" and "Mol_svg" to display;
    - Interactivity Tab: set initial page size to 1; enable "Show selected rows only";
    - Formatters Tab: Enable global number format to 2 decimal places.

  3. Image of the MCS. The Tile View (JavaScript) node "core scaffold" is already configured. Execute the node and explore the view.

  4. The table with the images of the R-groups of selected molecules. The images are available from the second (bottom) output of the Render to Images metanode. Use a new Table View (JavaScript) node and configure it:
    - Options Tab: select the column with compound_chembl_id and the columns with the svg images of the R-groups to display;
    - Interactivity Tab: enable "Show selected rows only".

Create a component encapsulating these Views:

  1. Select the Table View (JavaScript) node with the pivot table, the Row Sampling node, the Tile View (JavaScript) node with the image of the molecule, the Tile View (JavaScript) node with the MCS image, and the Table View (JavaScript) node with the images of the R-groups;

  2. Right click on one of the selected nodes and chose "Create Component";

  3. Execute the Component and explore the view (click the magnifying glass).


Configure the layout of component:

  1. Right click the component, select "Component → Open Layout Editor"

  2. Rearrange the elements:

    • the pivoted table at the top

    • the image of the molecule and the image of the MCS side by side below the pivoted table

    • the table with R-groups in the bottom

  3. Apply the changes and open the views once again to explore the changes in the layout.

Connect the RDKit R-Group Decomposition node to the output of the Table Reader. Connect the core drawn with the Sketcher component to the second input. Investigate the output.

Molecular Sketcher
Pivot
Embed images for the R-groups (SVG)
core
RDKit From Molecule
moleculesr-groups
Render to Images
corescaffold
Tile View (JavaScript)
RDKit R-Group Decomposition
input.table
Table Reader
pivot table
Table View (JavaScript)
first row
Row Sampling
Pick R-groups for visualization
Table Row to Variable

Nodes

Extensions

Links