2D and 3D QSAR using kNN-MFA method of N-(3-(4-benzylpiperidin-1-yl)propyl)-N,N ' - diphenylureas as CCR5 antagonists as anti-HIV-1 agents
Ajay B. Bedadurge, Anwar R. Shaikh · Journal of computational methods in molecular design · 2013
Quantitative structure‐activity relationship (QSAR) analysis for recently synthesized N-[3-(4-benzylpi peridin-1yl)propyl]-N,N ’ -diphenylureas derivatives was studied for their CC R5 antagonists as anti-HIV-1 agents [1]. The statistically significant 2D-QSAR model (r 2 = 0.9493; q 2 = 0.7653; F test = 42.09; r 2 se = 0.1672; q 2 se = 0.3597; pred_r 2 = 0.5311; pred_r 2 se = 0.5001) were developed using molecular design suite (VLifeMDS 4.2). The study was performed with 20 compounds (data set) using rando m selection and manual selection methods used for t he division of the data set into training and test set. Multiple linear regression (MLR) methodology with stepwise (SW) forward-backward variable selection method was used for building the QSAR models. The results of the 2 D-QSAR models were further compared with 3D-QSAR models ge nerated by kNN-MFA, (k-Nearest Neighbor Molecular Field Analysis). The statistical significant model (q 2 = 0.4644; q 2 se = 04751; pred_r 2 = 0.4332; pred_r 2 se = 0.4890) were developed using molecular design suite (VLifeM DS 4.2) these investigating the substitutional requ irements for the favorable anti-HIV-1 agents. The results derive d may be useful in further designing novel N,N’-dip henylurea derivatives as CCR5 antagonists prior to synthesis.