NMA-tune: Generating Highly Designable and Dynamics Aware Protein Backbones

Abstract

Protein’s backbone flexibility is a crucial property that heavily influences its functionality. Recent work in the field of protein diffusion probabilistic modelling has leveraged Normal Mode Analysis (NMA) and, for the first time, introduced informa- tion about large scale protein motion into the gen- erative process. However, obtaining molecules with both the desired dynamics and designable quality has proven challenging. In this work, we present NMA-tune, a new method that intro- duces the dynamics information to the protein design stage. NMA-tune uses a trainable com- ponent to condition the backbone generation on the lowest normal mode of oscillation. We imple- ment NMA-tune as a plug-and-play extension to RFdiffusion, show that the proportion of samples with high quality structure and the desired dy- namics is improved as compared to other methods without the trainable component, and we show the presence of the targeted modes in the Molecular Dynamics simulations.