Improved QSAR Modeling of Anti-HIV-1 Activities by Means of the Optimized Correlation Weights of Local Graph Invariants

D.J.G. MARINO; EDUARDO ALBERTO CASTRO; A.A. TOROPOV

Cent Eur J Chem

Año: 2006 vol. 4 p. 135 - 148

Results of using descriptors calculated with the correlation weights (CWs) of local graph invariants for modeling of anti-HIV-1 potencies of two groups of reverse transcriptase inhibitors are reported. The presence of different chemical elements in molecular structure of the inhibitors and   Morgan extended connectivity values of zeroth-, first-, and second order have been examined as local graph invariants in the labeled hydrogen–filled graphs. Values of the CWs which produce as large values as possible of correlation coefficient between the numerical data on the anti-HIV-1 potencies and values of  the descriptors on the training set have been computed via Monte Carlo optimization procedure. The model of the anti-HIV-1 activity obtained with compounds of training set by means of optimization of correlation weights of presence of chemical elements together with presence of Morgan extended connectivity of first order makes up a sensible model for a satisfactory prediction of the endpoints under consideration for test set compounds.