Molecular similarity, analog selection

Basak, S.C. and Grunwald, G.D. (1994). Molecular Similarity and Risk Assessment Analog Selection and Property Estimation Using Graph Invariants. SAR QSAR Environ.Res., 2, 289-307. 

In this chapter, we will discuss the utihty of HiQSAR studies using these five classes of descriptors. We will also show the utility of molecular similarity and our recently developed tailored similarity approach in the clustering of chemical libraries and selection of analogs. 

In drug design, one uses the molecular similarity concept to select analogs of interesting leads. Such selections are usually made based on user-defined methods. Therefore, such methods are always biased. We have come up with the idea of tailored similarity , where the property of interest is used to create a structure space that is relevant to that specific property. The results indicate that such tailored methods outperform arbitrary, user-defined QMSA techniques. 

---