Voprosy meditsinskoi khimii (ISSN 0042-8809)

The complex interval models application to prediction of the reverse transcriptase inhibitory activity


1. Lomonosov Moscow State Academy of Fine Chemical Technology
PubMed Id: 12189631
Year: 2002 vol: 48  issue:2  pages: 219-230
Abstract: Searching of new anti-HIV agents is still crucial now. In general, researches are looking for inhibitors of certain HIV’s vital enzymes, especially for reverse transcriptase (RT) inhibitors. Modern generation of anti-HIV agents represents non-nucleoside reverse transcriptase inhibitors (NNRTIs). They are much less toxic than nucleoside analogues and more chemically stable, thus being slower metabolized and emitted from the human body. Thus, search of new NNRTIs is actual today. Synthesis and study of new anti-HIV drugs is very expensive. So employment of the activity prediction techniques for such a search is very beneficial. This technique allows predicting the activities for newly proposed structures. It is based on the property model built by investigation of a series of known compounds with measured activity. This paper presents an approach of activity prediction based on “structure-activity” models designed to form a hypothesis about probably activity interval estimate. This hypothesis formed is based on structure descriptor domains, calculated for all energetically allowed conformers for each compound in the studied set. Tetrahydroimidazobenzodiazipenone (TIBO) derivatives and phenylethyltiazolyltiourea (PETT) derivatives illustrated the predictive power of this method. The results are consistent with experimental date and allow to predict inhibitory activity of compounds, which were not included into training set
Download PDF:
Reference: Burljaeva E.V., Tarkhov A.E., Burljaev V.V., Yurkevich A.M., Schvetz V.I., The complex interval models application to prediction of the reverse transcriptase inhibitory activity, Voprosy meditsinskoi khimii, 2002, vol: 48(2), 219-230.