3D Scaffold Hopping. Guaranteed.
In my opinion, software of this sort belongs on the desktop of any chemist designing bioactive molecules.
ReCore works well! It has a user-friendly interface, and the constraints are easy to set up.
The diversity of solutions from ReCore is much better than similar types of software.
ReCore proposed an elegant ring closure in a linear scaffold, not only retaining key pharmacophore features, but activity as well.
"Sounds great, however, does it work?" – Works a treat, see for yourself, here are some examples:
|Here is CSD MSC194 with the two green bonds defining the query "exit vectors":|
|Here are results (yellow), overlaid with the query:|
|Here are closely related inhibitors (in orange) superimposed with the ReCore solutions:|
|Note:||amide flip suggestion||meta substituted aromatics||a close match to a Trovirdine derivative|
|A typical ReCore query was taken from a TRH CSD crystal (TRHTRT) and torsionally modified to satisfy Olson's  idea (cp. below). To obtain more rigid results, a "pharmacophore" priority was introduced: A ring atom has to lie within the sphere depicted in red.|
|Based on the query (please note the green arrows), ReCore generates several interesting and nicely aligning cyclic structures; here are the ReCore results (colored) with the query in grey:|
|Note:||saturated ring||aromatic ring||rings are good||a tri-subst'd 6-ring|
|And, actually, the tri-substituted 6-ring was the basis for a TRH peptide mimetic by Olson  (in orange):|
|||Olson et al., . J. Med. Chem. 1995, 38, 2866-2879.|
In the original paper , the authors state:
Simple linking would not have worked in this case, as the individually bound fragments (68 and 69) cause conformational changes in the protein upon binding that would have hindered a simple linking approach. Reengineering the original fragments preserved their binding contacts responsible for potency and produced vectors for merging.
ReCore yields almost the same scaffold of the merged molecule as in the original publication. The compound resulted in a dramatically improved potency.
Using your corporate compounds as a starting point for a fragment library, your possibility of replacements would grow dramatically.
|The cutting points are defined by the exit vectors shown in green.|
|Here are the results:|
|||Erlanson, D.A. et al., . J. Med. Chem. 2004, 47, 3463 ff.|
We have seen similar results for other targets! For more application examples please see the original publication.
ReCore has been developed by Patrick Maass in Matthias Rarey's group at the ZBH, University of Hamburg, in collaboration with F. Hoffmann-La Roche AG (Basel, Switzerland).
The graphics were made available by courtesy of Martin Stahl and Tanja Schulz-Gasch at F. Hoffmann-La Roche.
ReCore requires a 3D fragment library to search in. Depending on the tool you want to use, please proceed with:
|||All ReCore spaces need to be de-compressed in two steps. Please unzip the download, and check instructions in the readme.|
|||The CSD ReCore index requires a license from CCDC.|
|||The GDB17-based index was created with help and friendly permission from Prof. Jean-Louis Reymond. Note that any publication about it must refer to the original authors (cp. Readme in the package). Further note that this index contains virtual molecules made to provide food for thought. They are laden with fluorine atoms which you may want to eliminate in a post-processing!|