QSAR of the testosterone binding globulin affinity by means of correlation weighting of local invariants of the graph of atomic orbitals

Ivan Raska, Andrey Toropov

Research output: Contribution to journalArticle

Abstract

Numerical values of the testosterone binding globulin affinity have been modeled as a mathematical function of molecular structure in two versions of molecular structure elucidation: first, by hydrogen-filled molecular graphs (HFG); second, by the so-called graphs of atomic orbitals (GAO). Increased orders of Morgan extended connectivity in the HFG and GAO have been examined as local invariants. Using optimisation of the correlation weights of the above-mentioned invariants, quantitative structure-activity relationships (QSAR) have been obtained. Best statistical characteristics of these QSARs are derived in the case of the Morgan extended connectivity of first order in the GAO. They are as follows: n = 11, r2 = 0.6540, s = 0.824, F = 17 (training set); n = 9, r2 = 0.8791, s = 0.388, F = 51 (test set).

Original languageEnglish
Pages (from-to)6830-6835
Number of pages6
JournalBioorganic and Medicinal Chemistry
Volume13
Issue number24
DOIs
Publication statusPublished - Dec 15 2005

Keywords

  • Graph of atomic orbitals
  • QSAR
  • Testosterone binding globulin affinity

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Organic Chemistry
  • Drug Discovery
  • Pharmaceutical Science

Fingerprint Dive into the research topics of 'QSAR of the testosterone binding globulin affinity by means of correlation weighting of local invariants of the graph of atomic orbitals'. Together they form a unique fingerprint.

  • Cite this