Understanding Nucleic Acid Charge Transport Properties using Multiscale Modelling Techniques

Abstract

Understanding charge transport in molecular systems is of fundamental interest in rapidly growing field of molecular electronics as well as to understand biological signal transfer processes which are accompanied by the spatial transport of charge. Here, we describe the charge transport properties of various nucleic acid systems under different external conditions using both incoherent hopping mechanism and coherent tunneling mechanism as described in the framework of Marcus-Hush formalism and Landauer formalism, respectively. Double stranded DNA (dsDNA) and dsRNA hold great promises in molecular electronics. We characterize the charge transport properties of dsRNA for different sequences and compare them with similar sequences of dsDNA in two extreme charge transport regimes – incoherent charge hopping regime and coherent electron transport regime. We find that the relative conductance of A-form dsRNA and B-form dsDNA depends on the mechanism of charge transport. This is attributed to various structural differences in dsDNA and dsRNA. We also study the effect of stretching and propose a method to detect the conformational changes using electrical measurements. Despite the twist-stretch coupling of dsRNA and dsDNA being different under external force, dsRNA shows similar structural polymorphism to dsDNA under different pulling protocols. Our atomistic MD simulations show that overstretching dsRNA along the 3’ ends (OS3) leads to the emergence of S-RNA whereas overstretching along the 5’ ends (OS5) leads to melting of dsRNA. Small-molecule ligands such as DNA intercalators strongly perturb the physico-chemical properties of dsDNA, thus find applications in cancer therapy, molecular imaging, and sensing. We examine the variation in the electrical conductance of dsDNA upon a drug intercalation for two different intercalators, ethidium and daunomycin and find that the DNA conductance increase by almost one order of magnitude upon drug intercalation. To study the electrical properties of every possible dsDNA sequence is a next-to-impossible task, we have also developed an ML-based model to predict the transfer integral values between any two DNA/RNA nucleobases of any given relative orientations. Using the Coulomb matrix representation which encodes the atomic identities and coordinates of the DNA base pairs to prepare the input dataset, we train a Neural Network (NN) model which can predict the electronic couplings between dsDNA base pairs with any structural orientation.