A series of 74 dihydroalkoxybenzyloxopyrimidines (DABOs), a class of highly potent non-nucleoside reverse transcriptase inhibitors (NNRTIs), was retrieved from the literature and studied by receptor-dependent (RD) three-dimensional quantitative structure-activity relationship (3D-QSAR) analysis to derive RD-3D-QSAR models. The descriptors in this new method are the steric and electrostatic interaction energies of the protein-ligand complexes (per residue) simulated by molecular dynamics, an approach named Residue-Ligand Interaction Energy (ReLIE). This study was performed using a training set of 59 compounds and the MKC-442/RT complex structure as reference. The ReLIE-3D-QSAR models were constructed and evaluated by genetic algorithm (GA) and partial least squares (PLS). In the best equations, at least one term is related to one of the amino acid residues of the p51 subunit: Asn136, Asn137, Glu138, and Thr139. This fact implies the importance of interchain interaction (p66-p51) in the equations that best describe the structure-activity relationship for this class of compounds. The best equation shows q² = 0.660, SE(cv) = 0.500, r² = 0.930, and SEE = 0.226. The external predictive ability of this best model was evaluated using a test set of 15 compounds. In order to design more potent DABO analogues as anti-HIV/AIDS agents, substituents capable of interactions with residues like Ile94, Lys101, Tyr181, and Tyr188 should be selected. Also, given the importance of the conserved Asn136, this residue could become an attractive target for the design of novel NNRTIs with improved potency and increased ability to avoid the development of drug-resistant viruses.