Search for: All records

Abstract We introduceBoltzGen, an all-atom generative model for designing proteins and peptides across all modalities to bind a wide range of biomolecular targets. BoltzGen builds strong structural reasoning capabilities about target-binder interactions into its generative design process. This is achieved by unifying design and structure prediction, resulting in a single model that also reaches state-of-the-art folding performance. BoltzGen’s generation process can be controlled with a flexible design specification language over covalent bonds, structure constraints, binding sites, and more. We experimentally validate these capabilities in a total of eight diverse wetlab design campaigns with functional and affinity readouts across 26 targets. The experiments span binder modalities from nanobodies to disulfide-bonded peptides and include targets ranging from disordered proteins to small molecules. For instance, we test 15 nanobody and protein binder designs against each of nine novel targets with low similarity to any protein with a known bound structure. For both binder modalities, this yields nanomolar binders for 66% of targets. We release model weights, data, and both inference and training code at:https://github.com/HannesStark/boltzgen. 

The design of organic–peptide hybrids has the potential to combine our vast knowledge of protein design with small molecule engineering to create hybrid structures with complex functions. Here, we describe the computational design of a photoswitchable Ca²⁺-binding organic–peptide hybrid. The designed molecule, designated Ca²⁺-binding switch (CaBS), combines an EF-hand motif from classical Ca²⁺-binding proteins such as calmodulin with a photoswitchable group that can be reversibly isomerized between a spiropyran (SP) and merocyanine (MC) state in response to different wavelengths of light. The MC/SP group acts both as a photoswitch as well as an optical sensor of Ca²⁺binding. Photoconversion of the SP to the corresponding MC unmasks an acidic phenol, which CaBS uses as an integral part of both its Ca²⁺-binding site as well as its tertiary and quaternary structure. By design, the SP state of CaBS is monomeric, while the Ca²⁺-bound form of the MC state is an obligate dimer, with two Ca²⁺-binding sites formed at the interface of a domain-swapped dimer. Thus, light and Ca²⁺were expected to serve as an “AND gate” that powers a change in backbone structure/dynamics, oligomerization state, and fluorescence properties of the designed molecule. CaBS was designed using Rosetta and molecular dynamics simulations, and experimentally characterized by nuclear magnetic resonance, isothermal titration calorimetry, and optical titrations. These data illustrate the potential of combining small molecule engineering with de novo protein design to develop sensors whose conformation, association state, and optical properties respond to multiple environmental cues.