Building an ab initio solvated DNA model using Euclidean neural networks

Accurately modeling large biomolecules such as DNA from first principles is fundamentally challenging due to the steep computational scaling of ab initio quantum chemistry methods. This limitation becomes even more prominent when modeling biomolecules in solution due to the need to include large numbers of solvent molecules. We present a machine learning electron density model based on a Euclidean neural network framework to model explicitly solvated double-stranded DNA. The neural network framework has a built-in understanding of equivariance, allowing the model to learn 3D structural information efficiently by exploiting the properties of Euclidean symmetry. By training the machine learning model using molecular fragments that sample the key DNA and solvent interactions, we show that the model predicts electron densities of arbitrary systems of solvated DNA accurately, resolves polarization effects that are neglected by classical force fields, and captures the physics of the DNA-solvent interaction at the ab initio level.