Assessing Similarity and Diversity of Combinatorial Libraries by Spatial Autocorrelation Functions and Neural Networks

摘要

The recent availability of high throughput techniques to find new lead structures by random screening has increased the need for a continuous supply of novel compounds. One way of fulfilling this need is by preparation of chemical libraries through combinatorial chemistry. The repetitive and systematic application of a synthesis principle to a number of chemical building blocks can produce large populations of different structures.