Predictive pharmacophore models were developed for a large series of piperidine- and piperazine-based CCR5 antagonists as anti-HIV-1 agents reported by Schering-Plough Research Institute in recent years. The pharmacophore models were generated using a training set consisting of 25 carefully selected antagonists based on well documented criteria. The activity spread, expressed in K(i), of training set molecules was from 0.1 to 1300 nM. The most predictive pharmacophore model (hypothesis 1), consisting of five features, namely, two hydrogen bond acceptors and three hydrophobic, had a correlation (r) of 0.920 and a root mean square of 0.879, and the cost difference between null cost and fixed cost was 44.46 bits. The model was cross-validated by randomizing the data using the CatScramble technique. The results confirmed that the pharmacophore models generated from the test set were not due to chance correlation. The best model (hypothesis 1) was validated using test set molecules (total of 78) and performed well in classifying active and inactive molecules correctly. The model was further validated by mapping onto it a diverse set of six CCR5 antagonists identified by five different pharmaceutical companies. The best model correctly predicted these compounds as being highly active. These multiple validation approaches provide confidence in the utility of the predictive pharmacophore model developed in this study as a 3D query tool in virtual screening to retrieve new chemical entities as potent CCR5 antagonists. The model can also be used in predicting biological activities of compounds prior to undertaking their costly synthesis.