Affordable Access

Publisher Website

Structure–activity relationship (SAR) of substituted 17α-acetoxyprogesterones studied with principal component analysis and neural networks using calculated physicochemical parameters

Authors
Journal
Journal of Molecular Structure THEOCHEM
0166-1280
Publisher
Elsevier
Publication Date
Volume
489
Issue
1
Identifiers
DOI: 10.1016/s0166-1280(99)00039-1
Keywords
  • Structure–Activity Relationship
  • Progestational Activity
  • 17α-Acetoxyprogesterones
  • Principal Component Analysis
  • Neural Network
Disciplines
  • Computer Science

Abstract

Abstract It was shown that the two different methods, the principal component analysis (PCA) and the neural network (NN), can classify oral progestational activity of substituted 17α-acetoxyprogesterones into two categories, high active and low active, using only calculated molecular properties. The two methods can predict the category of each molecule with a fairly high percentage of success. The NN work was slightly better than the PCA in the prediction of the category. Ionization potential, molecular hardness, net atomic charges, frontier indices were found to be useful parameters for the classification of the compounds.

There are no comments yet on this publication. Be the first to share your thoughts.