简介：

Recent studies have demonstrated the importance of evaluating multiple types of descriptors and statistical methods in creating a predictive QSAR model for a set of molecules. Almond™ offers a new generation of molecular descriptors and a statistical workbench with which to rapidly create and test structure-activity relationship hypotheses.

Traditional 3D-QSAR methods require a two-step preparation to analyze a set of molecules: 1) identification of the functional groups essential for binding; and 2) superimposition of the molecules based on alignment of these groups. This preparation is the most time consuming aspect of the method, and is a source of user bias. By creating 3D QSAR models without the need to superimpose molecules, Almond is much faster and more objective yet retains the predictive power of more complicated methods.

Because Almond and its descriptors represent the internal geometrical relationship of pharmacophoric regions, they are useful for predicting pharmacodynamic properties (i.e., specific receptor/ligand interactions). This is in contrast to other descriptors more suited for pharmacokinetic properties such as logP, permeability, solubility, etc.

下载说明：

Almond Brochure (463k)
Almond Scientific References (90k)

Key Benefits

• Because no initial alignment is required, Almond's 3D-QSAR analysis can be done in a fraction of the time required by other methodologies.
• GRIND descriptors are alignment-independent, fast to compute, compact to store, and readily interpretable through interactive graphic tools.
• Unlike classic autocorrelation methods, Almond's descriptors allow the original information to be reconstructed, making it possible to interpret both the descriptors and the models generated from them in terms of the molecules in the analysis.
• Almond's risk of generating an overfitted model is less than for other 3D-QSAR methods because the variables-to-compound ratio is much smaller as a result of trimming the number of MIF variables.