A theoretical study of the vibrational characteristics of six DNA codons (i.e., at the sequence forms 5 -GGX-3 and 5 -XGG-3 ; X = A, C, and T) in gas phase is presented.The simulation results show that the spectra of the codons in the near infrared regime can be classified into five semi-distinct spectral sub-regimes, namely: two end regions and one region that separates the two sequence identifier regions with features that arise primarily from coupling between the GG bases and other bases (T, C, and A). This study reveals a number of phenomenological trends that have scientific relevance to spectroscopic characterization of DNA molecules. For example, spectra in the high-frequency end-regions arise primarily from vibrational modes associated with the O-H bonds at the 5 end or 3 end of the codons and the spectra in the low-frequency end-regions are formed primarily by the vibrational modes of the backbone. Conversely, in the lower-frequency side, the sequence-identifier region is composed of the C-H bond stretch vibrations in the planes of the corresponding DNA bases, and in the higher-frequency side, sequence-identifier region is composed of the N-H bond stretch vibrations in the planes of the corresponding DNA bases. In addition, the sequence-identifier dividing region almost exclusively contains vibrational modes due to coupling between the G nucleotides and other bases. As will be shown, all the vibrational modes in sequence identifier regions are localized at the corresponding DNA bases and exhibit a definable dependence on the sequence form of the codons under study.