Parametrically guided design of beta barrels and transmembrane nanopores using deep learning

Abstract

Francis Crick’s global parameterization of coiled coil geometry has been widely useful for guiding design of new protein structures and functions. However, design guided by similar global parameterization of beta barrel structures has been less successful, likely due to the deviations from ideal barrel geometry required to maintain inter-strand hydrogen bonding without introducing backbone strain. Instead, beta barrels have been designed using 2D structural blueprints; while this approach has successfully generated new fluorescent proteins, transmembrane nanopores, and other structures, it requires expert knowledge and provides only indirect control over the global shape. Here we show that the simplicity and control over shape and structure provided by parametric representations can be generalized beyond coiled coils by taking advantage of the rich sequence-structure relationships implicit in RoseTTAFold based design methods. Starting from parametrically generated barrel backbones, both RFjoint inpainting and RFdiffusion readily incorporate backbone irregularities necessary for proper folding with minimal deviation from the idealized barrel geometries. We show that for beta barrels across a broad range of beta sheet parameterizations, these methods achieve high in silico and experimental success rates, with atomic accuracy confirmed by an X-ray crystal structure of a novel barrel topology, and de novo designed 12, 14, and 16 stranded transmembrane nanopores with conductances ranging from 200 to 500 pS. By combining the simplicity and control of parametric generation with the high success rates of deep learning based protein design methods, our approach makes the design of proteins where global shape confers function, such as beta barrel nanopores, more precisely specifiable and accessible. Significance: De novo beta barrel proteins have previously been designed using “blueprint” based methods which require expert knowledge of the rules of folding and provide only indirect control of the overall barrel shape by specifying structural features such as glycine kinks and beta bulges. The barrel shape can be directly modeled using global parametric methods, but to date such methods have not succeeded in generating folded proteins, likely due to the absence of such structural features. Here, we describe methods that combine the simplicity and control of parametric barrel specification with the high success rates of deep learning based protein design methods to successfully design new beta barrel folds of different and pre-specified sizes, including both soluble designs and transmembrane nanopores.