Main indexFormer pageNext pageBeginningEnd

Tutorial B: The Application of WODCA to the Design of a Combinatorial Library

Introduction

The following example demonstrates the application of WODCA to the design of a library of 1-phenyl-3,5-dimethylpyrazole compounds (see Figure 1). Before we go into the details of the process for planning this library, an overview of the individual steps is given.
  1. Start WODCA by calling the WODCA start script wodca.sh.
  2. Define a representative compound of your library by starting the CACTVS Molecule Editor from the context menu of WODCA's structure display and export it into WODCA.
  3. Build a new synthesis tree by exporting the target compound to the CACTVS Synthesis Planner. This can be also performed within the context menu of WODCA's structure display.
  4. Choose a disconnection strategy and perform a strategic bonds search to dissect the target compound into suitable synthesis precursors.
  5. Cut a strategic bond that has a high rating to generate suitable synthesis precursors.
  6. Export the precursors generated by WODCA to the CACTVS Synthesis Planner.
  7. Define substructures for each of the starting materials by defining open sites at those positions where the precursor should be generalized. Then perform substructure searches.
  8. Search for a suitable synthesis of a library of compounds. This involves: Repeat step 4 to 8 with the precursor compounds until a promising synthesis plan has been developed which is based on available starting materials from the catalog of chemicals and which allows to build a library with a large number of compounds.

Figure 1: The lead structure of 1-phenyl-3,5-dimethylpyrazole compounds

The Synthesis Planning of a Library of 1-Phenyl-3,5-dimethylpyrazole Compounds

The following screenshots demonstrate the individual steps which are necessary to build a synthesis plan for the lead structure (Figure 1) of the library. The numbered headlines correspond to the individual steps which are described in the introduction to this chapter.

1. Start WODCA

Call the start script wodca.sh from a UNIX shell. This provides the starting screen of WODCA as shown in Figure 2.

Figure 2: WODCA's starting screen

2. Input the Target Compound

WODCA needs a completely defined target structure without any open sites or R-groups to plan the synthesis of a combinatorial library. Thus, a representative compound of this lead structure has to be given. Use the following simple rules to transform a general lead structure (Markush structure) into a specific compound:
Figure 3 demonstrates this for the 1-phenyl-3,5-dimethylpyrazole lead structure of Figure 1. For the R1-group there are no restrictions on the kinds of the substituents. Thus, hydrogen atoms are set for the five positions of the R1-group at the aromatic ring system. The R2-group and the R3-group represents alkyl or aryl groups which are replaced by methyl groups.

Figure 3: Definition of a simple representative of the lead structure

The CACTVS Molecule Editor is needed to input the target structure. If this molecule editor has not yet been started then click on the entry Tools in WODCA's menu bar, pull it down and click on Launch Molecule Editor. Another possibility is given by the use of the context menu in WODCA's structure display. Select the command Define New Target .... . After a few seconds the CACTVS Molecule Editor window appears. It allows the drawing of a new structure just with the mouse. It is not necessary to draw hydrogen atoms into your target structure since they will be added automatically to open valences. Once the target structure has been drawn it has to be transferred into the structure display of WODCA's main window. In order to achieve this, click on the File entry (Figure 4: (1)) in the menu bar of the CACTVS Molecule Editor, pull it down and click on the Export entry (Figure 4: (2)). The chemical structure is then automatically transferred into WODCA and the molecule editor window will automatically be minimized to an icon.

Figure 4: Export of a molecule from the CACTVS Molecule Editor into WODCA

3. Build a Synthesis Tree

In order to create a graphical scheme of the synthesis plan generated in this session you have to export the target compound to the CACTVS Synthesis Planner. This can be performed with the context menu in WODCA's structure display (Figure 5: (1)). The context menu appears after pressing the right mouse button in WODCA's structure display . Select the entry Start New Synthesis Plan ...(Figure 5: (2)) to perform the export of the target compound into the CACTVS Synthesis Planner. It is not necessary to separately start the CACTVS Synthesis Planner, because it is started automatically after calling the menu command.

Figure 5: Transfer of the target compound into the CACTVS Synthesis Planner

The target compound is then displayed in the CACTVS Synthesis Planner as the top of the synthesis tree (Figure 6).

Figure 6: Display of the target compound in the CACTVS Synthesis Planner

4. Choose a Disconnection Strategy and Perform a Search for Strategic Bonds

The analysis of strategic bonds is started with the disconnection strategy Carbon-Heteroatom Bonds to find suitable precursors for the target structure. Therefore, we go to the Disconnection menu in WODCA's menu-bar (Figure 7: (1)) and click on the Strategic Bonds ... command. The strategic bonds window of WODCA appears (Figure 7: (2)). We choose the disconnection strategy Carbon-Heteroatom Bonds on the right-hand side of the window (Figure 7: (3)) and press the Search button (Figure 7: (4)). WODCA perceives two strategic bonds (Figure 7: (5)). One bond is rated with the maximum value of »100«, a second bond is rated relative to this bond with the value of »87« which is also quite favorable. Having two highly rated strategic bonds promises to give suitable precursors.

Figure 7: Evaluation of strategic bonds

5. Cut a Strategic Bond

You can now follow the sequence of strategic bonds suggested by the rating of WODCA or you can change this order. Furthermore, it is possible to require several bonds to be simultaneously broken. The disconnection of strategic bonds directly leads to synthesis precursors, since WODCA adds suitable atoms to the open valences obtained on heterolysis of a bond. In our example, we decide to follow the suggestions of WODCA and break both bonds at the same time. Therefore, we insert the ranks of these two bonds into the Set of Bonds entry (bond 1 and bond 2 in our example), separate the ranks by an empty space and then press the Cut Bond button (Figure 7: (6)).
WODCA now generates two precursors for the target structure: phenylhydrazine and 2,4-pentanedione in its enol form (Figure 8).

Figure 8: Dissection of target compound

The tautomerization of this enol form to its more stable keto form is performed via the context menu of the structure display (press the right mouse button on the Precursor(s) display, Figure 9: (1)) and select the Tautomerize Precursor command (Figure 9: (2)).

Figure 9: Tautomerization of precursors

6. Export of Precursors to the CACTVS Synthesis Planner

Now, we attach both precursors to our synthesis plan. This is also performed with the help of the context menu in the strategic bond display (press the right mouse button on the Precursor(s) display, Figure 10: (1), and select the command Attach to Synthesis Plan, Figure 10: (2)).

Figure 10: Adding synthesis precursors to the synthesis tree

A second level is built at the synthesis tree handled by the CACTVS Synthesis Planner. One of the precursor compounds in this new level is marked as active (yellow box). The active precursor is always synchronized with the compound in WODCA's Strategic Bonds display. Therefore, after the attachment of the precursors to the synthesis tree, one compound - in our case 2,4-pentanedione - is automatically exported to the WODCA main program and an identity search is automatically performed by WODCA (Figure 11).

Figure 11: Building a synthesis tree

8. Substructure Searches

Now, we want to perform a substructure search with the active compound loaded in WODCA. Therefore, we go to the Searches menu in WODCA's menu bar (Figure 12: (1)) and click on the Substructure Search ... command (Figure 12: (2)).

Figure 12: Performing a substructure search in WODCA

The window of the CACTVS Substructure Search tool appears with the query compound loaded in the structure display. (Figure 13: (1)).

Figure 13: The CACTVS Substructure Search tool

This representative precursor for our lead structure is used to define a substructure with open sites. We delete hydrogen atoms at all positions where a generalization of the structure to R groups is desired in order to create a suitable substructure of our phenylhydrazine precursor. Therefore, it is necessary to switch the CACTVS Substructure Search tool from drawing mode into eraser mode by pressing the rubber icon in the upper right corner of the window (Figure 14: (1)). Then, we click once with the left mouse button on each hydrogen atom marked with a red box to delete it and to define an open valence (Figure 14: (2)). Afterwards, we press the Search button (Figure 14: (3)) to perform the substructure search in the catalog of chemicals.

Figure 14: Defining the substructure query

The substructure prepared as shown in Figure 14 provides 23 hits in the FLUKA catalog that contains 16,769 compounds. This is indicated in the Match List icon of WODCA's main window (Figure 15: (1)). We click once on the Match List icon with the left mouse button to display these hits. Now, all chemical compounds are transferred to the CACTVS Structure Browser (Figure 15: (2)) which is started automatically.

Figure 15: Display of a match list of phenylhydrazine derivatives from a catalog of chemicals

In a similar manner, a substructure search for generalized variations of the 2,4-pentanedione-precursor is performed. In a first step, the diketone precursor is activated by a single mouse click in the synthesis tree and exported to the WODCA main program (Figure 16: (1)). In a second step, the substructure search is initiated by the Searches menu in WODCA's menu bar (Figure 16: (2)). To find out which 1,3-diketones with alkyl or aryl substituents at position 1 and 3 are contained in the catalog of available starting materials, open all sites at the two methyl groups of the 2,4-pentanedione precursor by deleting all hydrogen atoms marked with a red box (Figure 16: (3)). This corresponds to putting R2 and R3 groups at the end of the 1,3-diketone substructure, but it already restricts these R2 and R3 groups to have carbon atoms at their junction to the 1,3-diketone substructure. After pressing the Search button (Figure 16: (4)) the substructure search in the catalog of chemicals is performed.

Figure 16: Substructure search with the 1,3-diketone query

The result of this substructure search is shown in Figure 17. Twenty 1,3-diketones are found in the catalog of chemicals. Click on the Match List icon in WODCA's main window to view these hits (Figure 17: (1)). The CACTVS Structure Browser is then automatically started to display all the structures from the match list (Figure 17: (2)).

Figure 17: Display of a match list of 1,3-diketone derivatives from a catalog of chemicals

If one is also interested in acyl derivatives, in carboxylic acids, or in amides the substructure query has to be changed. It is possible to define a set of atom types which are allowed or which are forbidden for a certain position in the substructure query. This can be done by means of atom lists. The following example shows the definition of a substructure query which is qualified to find such compounds in a catalog of chemicals by a substructure search.
Again, the diketone precursor is activated by a single mouse click in the synthesis tree and is thereby exported to the WODCA main program (Figure 18: (1)). The substructure search is initiated (Figure 18: (2)) and the substructure query is defined. In a first step, the CACTVS Substructure Search tool is switched to »eraser mode« (Figure 18: (3)) and all hydrogen atoms marked with a red box are deleted with a single mouse click (Figure 18: (4)).

Figure 18: Preparing a new substructure query

After the hydrogen atoms are deleted, the atom list for each substituent in 1,3-position has to be defined. Therefore, the CACTVS Substructure Search tool is switched back to »pencil mode« (Figure 19: (1)) and a double click on one of the carbon atoms marked with a red box is performed (Figure 19: (2)). This opens a dialog box for the specification of atom types. Set the Atom Type to List and enter »!C« to exclude carbon atoms from the list of elements which are allowed at the atom position considered (Figure 19: (3)). Then, press the Set button to close this dialog box (Figure 19: (4)) and repeat this procedure for the other carbon atom to finish the definition of the substructure query.
It is also possible to exclude more than one elements from an atom list, e.g. by specifying »!O !N«. If more than one element is defined, the element symbols have to be separated by an empty space.

Figure 19: Specification of atom lists

An atom list is indicated by the letter L instead of an atom symbol (Figure 20: (1)). The substructure search is then started by pressing the Search button (Figure 20: (2)).

Figure 20: Substructure query with atom lists

The result of this substructure search is displayed after a single mouse click on the Match List icon in WODCA's main window is performed (Figure 21: (1)). The CACTVS Structure Browser is then started to display all 32 structures from the match list (Figure 21: (2)).

Figure 21: Result of a different substructure search

Main indexFormer pageNext pageBeginningEnd

Last change: 2000-06-29
Webmaster: matthias.pfoertner@chemie.uni-erlangen.de