Large descriptor spaces QSAR/QSPR applications

QSAR/QSPR Applications in Large Descriptors Spaces... 

The development of QSAR/QSPR on large descriptor spaces started some time ago. One of the first widely applicable programs for such modeling was CODESSA (Comprehensive DEscriptors for Structural and Statistical Analysis), developed by one of us (M.K.) and Victor Lobanov, a graduate student at that time, in collaboration with Professor Alan R. Katritzky at the University of Florida. In 1994, the first version of the program was published and assessed for the treatment of a variety of chemical and physical properties of compounds and heterocyclic compounds, in particular (1995CSR279, 1997JCICS405). 

QSAR/QSPR APPLICATIONS IN LARGE DESCRIPTORS SPACES... 

In QSAR and QSPR studies, the standard ways of removing redundancy from large numbers of topological and topographical indices include principal component analysis, chi-squared analysis, and multiple regression analysis (MRA). Most QSAR and QSPR applications deal with very small datasets, and so the dimensionality does not cause a problem for PCA or chi-squared analysis. MRA does not impose any restrictions on the type and number of descriptors. The selection process is based on two principles, namely, to cover as much of parametric space as possible (principle of variance) while choosing independent descriptors (principle of orthogonality). 

---