Human immunodeficiency virus (HIV) remains a major global health challenge and the development of an effective vaccine is still an important objective. Rhesus cytomegalovirus (RhCMV)-based vaccines expressing simian immunodeficiency virus (SIV) antigens have previously shown promising protection in rhesus macaques, but the molecular mechanisms separating protected from non-protected animals remain incompletely understood. In this study, longitudinal transcriptomic data from rhesus macaques vaccinated with a clinical ortholog RhCMV/SIV vaccine platform administrated through either one, two or three doses, were analyzed to identify gene expression signatures associated with vaccine-mediated protection. Differential expression analyses were performed to characterize vaccine-induced transcriptional responses across immunization regimens, followed by interpretable rule-based machine learning (RBML) using R.ROSETTA to identify combinations of genes capable of distinguishing protected and non-protected animals. The generated rule-based models successfully identified transcriptomic patterns associated with protection within each vaccination group (p-value <0.001). However, limited predictive transferability across groups suggested that the identified signatures primarily represented group-specific rather than universal dose-dependent responses. Despite these differences, MHC-I-related transcripts repeatedly appeared within booster-associated models, indicating that variation in antigen presentation and associated cellular immune responses may contribute to differences in vaccine outcome.