Combinatorial technology and core hopping for lead discovery and optimization
The Advantages of Computational Lead Optimization
The virtual chemical space that chemists are interested in is too large to be synthesized and screened, even using modern methods of combinatorial chemistry and robotic synthesis. Therefore, there is a real need for efficient and reliable methods to rationally select the optimal library members for synthesis. Additionally, once a promising lead compound is discovered, different core scaffolds as well as side-chain substitutions must be enumerated and examined to evaluate relative binding affinities towards a particular target. Accurate ligand-receptor scoring coupled with intelligent and efficient combinatorial docking and core-hopping methods can accelerate lead optimization and aid in designing the optimal, focused compound library for further synthesis.
Schrödinger's CombiGlide can flexibly vary the core or side-chain substitutions, creating virtual combinatorial libraries that may be screened for leads, identify novel scaffolds, or generate focused libraries in support of lead optimization efforts.
CombiGlide identifies the most effective reagent combinations to produce focused libraries that have the highest likelihood of binding tightly to the target protein. CombiGlide dramatically reduces the overwhelming combinatorial space down to manageable library sizes by selecting and ranking reagents.
CombiGlide performs core hopping beginning with a lead compound/receptor complex and searches among candidate core structures to identify alternate cores exhibiting similar binding modes as the lead that can accommodate the side chains in their optimal orientation.
CombiGlide analyzes the chemical features of the virtual library in terms of their pharmacophore properties to generate insights into the chemical functionalities the receptor prefers at each core position, which can be used to guide the selection of reagents in further experimental and computational explorations.
CombiGlide can optionally filter compounds using predicted ADME properties, eliminating from consideration compounds that may bind well but exhibit undesirable pharmacokinetic profiles.
CombiGlide guides a user through the chosen workflow with an intuitive step-by-step interface. The Maestro interface provides helpful structural visualization and analysis tools.
Citations and Acknowledgements
Schrödinger Release 2016-4: CombiGlide, Schrödinger, LLC, New York, NY, 2016.
"Structure based inhibitor design targeting glycogen phosphorylase B. Virtual screening, synthesis, biochemical and biological assessment of novel N-acyl-β-d-glucopyranosylamines"Parmenopoulou V et al., Bioorg Med Chem, 2014, 22(17), 4810-25
"Structure tuning of pyrazolylpyrrole derivatives as ERK inhibitors utilizing dual tools; 3D-QSAR and side-chain hopping"Kim MH, Chung JY, Ryu JS, Hah JM, Bioorg Med Chem Lett, 2011, 21(16), 4900-4
One of CombiGlide's strengths is the ability to incorporate synthetic feasibility into combinatorial design by using actual reagents (obtained from available-chemical files or corporate databases) as the basis for specifying the side chains to be sampled at the various positions on the core. This is useful for studies where the intent is to synthesize and test the combinatorial library selected by CombiGlide.
CombiGlide can also be used to gain insight into the characteristics of the receptor by observing which functional groups, at which positions on a core, result in strong binding. This can be useful even if there is no intent to synthesize the selected library.
The CombiGlide Diverse Side-chain Collection contains a collection of representative functional groups commonly found in pharmaceuticals with linkers of variable length. All reasonable ionization and tautomeric states for the side chains are included. When used with CombiGlide, this collection will allow the user to effectively explore an active site and to identify types of functional groups that are likely to interact with the binding site in an energetically favorable manner.
Version 1.2 of the Collection includes two sets, containing 817 and 146 diverse reagents. The latter is a subset of the former, for use when coarser sampling is desired. In Version 1.2, the reagent files contain chemical features for use in chemical-feature analysis, first released in CombiGlide version 1.5 (Schrödinger Suite 2007).
A script ("lib_subset") to create a user-specified subset of the collection and prepare it for use with CombiGlide is also provided.