Multidimensional QSAR: Moving from three- to five-dimensional concepts
✍ Scribed by Vedani, Angelo ;Dobler, Max
- Publisher
- John Wiley and Sons
- Year
- 2002
- Tongue
- English
- Weight
- 184 KB
- Volume
- 21
- Category
- Article
- ISSN
- 1611-020X
No coin nor oath required. For personal study only.
✦ Synopsis
Quantitative structure-activity relationships (QSAR) is an area of computational research which correlates structural features and quantities such as the binding affinity, toxic potential, or pharmacokinetic properties of an existing or a hypothetical molecule. This correlation may be obtained by analyzing topology and fields of the molecules of interest or by simulating their spatial interactions with biological receptors or models thereof. While 3D-QSAR simulations allow for a specific interaction scheme with the virtual receptor, they presume the knowledge of the bioactive conformation of the ligand molecules and require a sophisticated guess about manifestation and magnitude of the associated induced fit ± the adaptation of the receptor binding pocket to the individual ligand topology.
To reduce the human bias ± i.e. having to choose the bioactive conformer and to select an induced-fit model ± in QSAR studies, we have extended a concept developed at our laboratory (software Quasar) to 4-D (by allowing for a multiple representation of the ligand molecules: conformation, orientation, protonation scheme, stereochemistry) and more recently to 5-D (by allowing for a multiple representation of induced-fit hypotheses). During the simulated evolution the correct entities are identified by the genetic algorithm. The current concept allows for up to 64 representations per ligand and up to six different induced-fit protocols: a subtle, individual, ligand-dependent mechanism, simulated by means of existing fields (steric, electrostatic, hydrogen bond), energy minimization, the lipophilicity potential or, in the simplest case, a linear adaptation. The genetic algorithm cannot only select among the provided six scenarios, but also build any linear or anisotropic combination thereof.
📜 SIMILAR VOLUMES