We propose a model Hamiltonian for describing charge transport through short homogeneous double stranded DNA molecules. We show that the hybridization of the overlapping orbitals in the base-pair stack coupled to the backbone is sufficient to predict the existence of a gap in the nonequilibrium current-voltage characteristics with a minimal number of parameters. Our results are in a good agreement with the recent finding of semiconducting behavior in short poly(G)-poly(C) DNA oligomers. In particular, our model provides a correct description of the molecular resonances which determine the quasilinear part of the current out of the gap region. DOI: 10.1103/PhysRevB.65.241314 PACS number͑s͒: 72.80.Le, 05.60.Ϫk, 87.10.ϩe, 87.14.Gg The attempt to understand the mechanism of electron motion along DNA is the source of an intense debate in the biochemical and chemical physics communities. 1 Solving this problem is an essential step for the development of DNA-based molecular electronics. New insights to this issue are brought by recent breakthroughs in direct measurements through DNA molecules. 2-8 Transport measurements through nanostructured systems are potentially capable of addressing the basic issues of the conduction properties of molecular and supramolecular aggregates. The aftermath for the realization of molecular electronics devices is straightforward. 9 It is thus not surprising that DNA molecules became the subject of an intense study concerning their potency to carry an electric current, 2-7 and to provide a scaffold for the metal assembling of highly conductive nanowires. 4,8 From a nanoelectronics perspective, the DNA possesses ideal structural and molecular-recognition properties, and the understanding of the charge transport through DNA may result in the ambitious goal of self assembling nanodevices with a definite molecular architecture. 10 The hypothesis that double stranded DNA supports charge transport as a linear chain of overlapping orbitals located on the stacked base pairs, already advanced in the early sixties, 11 received first experimental boosts only recently via long-range electron transfer measurements. 12 As far as transport through DNA is concerned, the available experiments are still controversial mainly due to the complexity of the environment and the molecule itself ͑sequence variability, 13 thermal vibrations . . . ). Concerning theory, the most reliable procedure to tackle these systems would be the ab initio quantum chemistry approach. However, massive numerical costs complicate its use for realistic biological systems. 14 To our knowledge, at the present time, only few densityfunctional-theory ͑DFT͒ calculations for DNA molecules are available. 7,15 In a parallel development particular aspects of the DNA transport phenomenology have been explained as mediated by polarons, 16 solitons, 17 electrons or holes. 1,18 Such lack of a unifying theoretical scheme calls for reproducible and unambiguous experimental results that are still a great technological challenge.Recently, Porath ...