In Silico Studies toward the Discovery of New Anti-HIV Nucleoside Compounds through the Use of TOPS-MODE and 2D/3D Connectivity Indices. 2. Purine Derivatives

• Published in: Journal of chemical information and modeling Vol. 45; no. 2; pp. 502 - 514
• Main Authors: Vilar, Santiago, Estrada, Ernesto, Uriarte, Eugenio, Santana, Lourdes, Gutierrez, Yaquelin
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 01.03.2005
• Subjects: Anti-HIV Agents - chemistry; Anti-HIV Agents - pharmacology; Chemical compounds; Computational Biology; Databases, Factual; Dimensional analysis; Drug Evaluation, Preclinical; HIV; Human immunodeficiency virus; Indexes; Molecular Structure; Nucleosides - chemistry; Nucleosides - pharmacology; Purines - chemistry; Quantitative Structure-Activity Relationship
• ISSN: 1549-9596; 1549-960X
• DOI: 10.1021/ci049662o

The TOPological Substructural MOlecular DEsign (TOPS-MODE) approach has been used to predict the anti-HIV activity in MT-4 assays (Estrada et al., 2002) of a diverse range of purine-based nucleosides. A database of 206 nucleosides has been selected from the literature and a theoretical virtual screening model has been developed. The model is able of discriminating between compounds that have anti-HIV activity and those that do not, with a good classification level of 85% in the training and 82.8% in the cross-validation series. On the basis of the information generated by the model, the correct classification of practically 80% of compounds from an external prediction set has been achieved using the theoretical model. Furthermore, the contribution of a range of molecular fragments to the pharmacological action has been calculated and this could provide a powerful tool in the design of nucleoside analogues that show activity against the HIV. Finally, a QSAR model has been developed that allows quantitative data to be obtained regarding the pharmacological potency shown by this type of compound.