MOE™: Molecular Operating Environment

Fragment-Based Design

MOE has a unified framework for fragment-based and de novo design applications which can be applied to both ligand- and structure-based design projects.  MOE's pharmacophore applications are integrated into the fragment based design protocol which provides an effective methodology for generating novel structures while preserving key interactions.

  • Grow, link, transform and replace ligand scaffolds (with or without the receptor)
  • Refine coordinates in a (flexible) active site and calculate binding scores
  • Apply MOE pharmacophore filters for preserving interactions or functional groups
  • Automatically apply 2D and 3D descriptors, QSAR and fingerprint constraints
  • Search MOE .mdb or Omega .oeb files directly and create in-house fragment databases

MOE - Fragment Based Design Replace scaffolds [Grimshaw 2010] for fast follow-on compounds by incorporating innovative linear, cyclic or fused scaffold arrangements.  Select multiple exit vectors and perform simultaneous searches to generate diversity oriented scaffolds.  Exchange ring bonds to identify and incorporate fused ring arrangements.  Specify optional connection vectors to increase the scope of scaffold arrangements and allow for cyclization at multiple attachment points.

Combine fragments of bound ligands (crystal structures or docking poses) to create novel structures in the context of a receptor active site [Pierce 2004].  Ligand fragments are exchanged between all pairs of ligands along all overlaid "crossover¡± bonds.  Crossover bonds may be restricted to bias the generated structures towards synthetic accessibility.

MOE - Fragment Based Design Grow ligands or elaborate on a key fragment at one or more positions in the context of an active siteLink fragments to connect two or more independent fragments.  Specify optional or essential connection points for increased flexibility.

Perform small isosteric changes to a molecule by applying reaction transformations rules (.rxn) that are generated using standard 2D sketchers. CCG provides a database of 170+ medicinal chemistry transformations.  Custom transformations can be easily added to the database and multiple iterations can be automatically applied for building combinatorial combinations or evolving transformations over multiple generations to produce more complex fragment and bioisosteres.

Filter the generated structures using any of MOE's hundreds of descriptors.  Apply SMARTS pattern matches to include or exclude substructures present in molecules.  Apply QSAR, fingerprint or pharmacophore models as additional filters.  Use shape based pharmacophore queries for ligand based design projects.  Estimate the "synthetic feasibility" using the MOE rsynth descriptor.  Refine structures in a (flexible) pocket and apply scoring functions to estimate binding affinity.

Create in-house fragment databases using MOE's SD Pipeline Tools.  Enumerate protonation states and filter input SD files to obtain good starting structures for fragmentation.  Keep only structures that are leadlike, non-reactive, etc.  Fragment molecules using a combination of algorithms and generate conformations.  Use commercial or customized in-house reagents as input to a reaction engine to generate R-group or scaffold libraries.


References

[Clark 2008] Clark, A.M., Labute, P.;SD File Processing with MOE Pipeline Tools.; JCCG (2008)

[Deschenes 2007] Deschenes, A., Sourial, E.;Ligand Scaffold Replacement using MOE Pharmacophore Tools.; JCCG (2007)

[Pierce 2004] Pierce, A.C., Rao, G., Bemis, G.W.; BREED: Generating Novel Inhibitors through Hybridization of Known Ligands. Application to CDK2, P38 and HIV Protease; J. Med. Chem. 47 (2004) 2768-2775

[Grimshaw 2010] Grimshaw, S.; Scaffold Replacement in MOE; JCCG (2010)