Tutorial: Focused Library – Core Input Floe with XLogP Filtering for Postprocessing
The Focused Library – Core Input Floe is one of several floes that can be used for generative design. The input is a molecule core than can have up to four attachment points. This floe enumerates a focused library by processing only a half-reaction on qualified reagents in the database for a particular reagent and then joining at a defined site on a core.
Enumerate Reagents for an Input Core Using a Built Database: Single Site Enumeration
This tutorial uses the following floe:
Focused Library – Core Input
In this tutorial, we will be working with only one attachment point and it will be assumed that you already have the prepared database from the previous tutorial. If you want to go through this section without preparing your own database, you can use one of the OpenEye databases provided in the Organization Data folder.
Choose the Focused Library – Core Input Floe. Click “Launch Floe” to bring up the Job Form. The parameters can be specified as below.
Input Parameters
Library Core Input (Library Core):
Click “Choose Input” to bring up the Sketcher window. Paste the SMILES string c1cc(ccc1C(=O)O)S(=O)(=O)Nc2ccc(c(c2)C#N)Oc3ccc(c(c3)C(F)(F)F)Cl into the Sketcher.
To define the enumeration site, attach the R group by hovering over the OH and pressing R on the keyboard twice or by right-clicking with the mouse. It must be R1, not a wildcard (*), that is attached to the core. The floe can handle up to four attachment points (R1, R2, R3, and R4) with different chemistries to create analogs, but a combinatorial explosion can result by doing this. Many parameters can be used to keep the library size in control. Figure 1 shows the core with the attachment point defined.
Reaction & Reagent Database (Append R4): Select the Mcule reaction and reagent database that you created previously as the Reaction & Reagent database (in the Create and Inspect a Reaction and Reagent Database from a SMILES File tutorial). You can also use one of the databases provided by OpenEye, which are found in the Organization Data folder.
Product Dataset (Products Dataset): Title the output dataset with a recognizable name.
Figure 1. Molecule core sketch with enumeration site identified as R1.
R1 Site: These are the enumeration-related parameters for R1. The reagents that will react and join with our core can be selected here. For this tutorial, we will make amides at the selected site, with the core being a carboxylic acid as one component of the Schotten-Baumann reaction, and the building blocks being the amine that will join with the core.
R1 Reagent Selection: Select Schotten-Baumann_amide:Amine. The functional group described along the reaction specification refers to the functional group that must be present in the screened building blocks.
R1 Maximum Reagents: The Analyze page does not support datasets with over 100,000 records; thus, we will cap the maximum number of reagents at 50,000. If you want to exhaustively generate all possible products, enter 0.
Focused Library Filtering Options: These parameters control filtering of the products for the focused library generation.
Filter Products: Toggle the switch to Off. For more information about OpenEye’s default filters, please click here.
Focused Library Property Generation: These options specify the computed molecular properties for the products from the focused library generation.
Compute Product Properties: To reduce effort in postprocessing, select the XlogP, TPSA, and MolWeight properties in the drop-down.
Please note that there are additional parameters under Advanced Enumeration Options and Advanced Enumeration Options that will allow you to limit the number of structures produced or to sample reagents instead of exhaustively enumerating ideas. This is particularly pertinent if you are enumerating at multiple sites at once, which presents the possibility of a combinatorial library explosion (see the next example using this floe for multiple site enumeration).
Check that the parameters in your Job Form are as shown in Figures 2 and 3, then click “Start Job” to begin the floe.
Figure 2. Input parameters of the Focused Library – Core Input Floe.
Figure 3. Input parameters of the Focused Library – Core Input Floe.
Filter Enumerated Products on the 3D & Analyze Page
Once the job is complete, activate the generated dataset by clicking on the circle with a + sign (that becomes a checkmark when activated) and navigate to the 3D & Analyze page.
Select ‘Filters’ from the Active Data Bar at the top of the page. Click on the “Choose a Filter” box and select XLogP, with a range from 3 to 5. Ensure that “Enable” is checked and press Enter on your keyboard to set the range (Figure 4).
On the Active Data Bar, select Save Records under the ‘Passing Filters’ drop-down. Select New Dataset as the destination and enter the name Filtered_Amide_Ideas, choose the “Replace Active Data” radio button, then click the “Save” button (Figure 5). It will take Orion about a minute to write this new dataset. The 3D & Analyze page will now display your filtered dataset, complete with identifiers for the building blocks that were used to build the compounds of interest should you want to order starting materials for a particular structure from the vendor (Figure 6).
Figure 4. The 3D & Analyze page showing how to apply the xLogP filter. The chosen filter allows all analogs having xLogP values between 3 and 5.
Figure 5. The “Save Records” window showing how to create a new dataset with the filtered records.
Figure 6. The 3D & Analyze page showing the spreadsheet with the filtered results. Highlighted with the red rectangle are the names of the building blocks used to enumerate the products.
For directions on how to run this floe using more than one attachment point, please see the Focused Library – Core Input with Multiple Attachment Points tutorial.