A comparison is made of a number of computationally efficient molecular indices with a view to the screening of very large virtual data sets of molecules. The use of Bayesian regularized neural networks is discussed, and their virtue in eliminating the need for validation sets, and potentially even test sets, is emphasized. The concept of a virtual receptor is introduced, and this is illustrated by the results of screening a database of 40 000 molecules.
|Number of pages
|Journal of Chemical Information and Computer Sciences
|Published - 1999