While deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here we describe a deep learning–based protein sequence design method, ProteinMPNN, with outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4%, compared to 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using X-ray crystallography, cryoEM and functional studies by rescuing previously failed designs, made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target binding proteins.

The binding and catalytic functions of proteins are generally mediated by a small number of functional residues held in place by the overall protein structure. Here, we describe deep learning approaches for scaffolding such functional sites without needing to prespecify the fold or secondary structure of the scaffold. The first approach, “constrained hallucination,” optimizes sequences such that their predicted structures contain the desired functional site. The second approach, “inpainting,” starts from the functional site and fills in additional sequence and structure to create a viable protein scaffold in a single forward pass through a specifically trained RoseTTAFold network. We use these two methods to design candidate immunogens, receptor traps, metalloproteins, enzymes, and protein-binding proteins and validate the designs using a combination of in silico and experimental tests.

There has been considerable recent progress in designing new proteins using deep-learning methods1–9. Despite this progress, a general deep-learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher-order symmetric architectures, has yet to be described. Diffusion models10,11 have had considerable success in image and language generative modelling but limited success when applied to protein modelling, probably due to the complexity of protein backbone geometry and sequence–structure relationships. Here we show that by fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. The accuracy of RFdiffusion is confirmed by the cryogenic electron microscopy structure of a designed binder in complex with influenza haemagglutinin that is nearly identical to the design model. In a manner analogous to networks that produce images from user-specified inputs, RFdiffusion enables the design of diverse functional proteins from simple molecular specifications.

While deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here we describe a deep learning based protein sequence design method, ProteinMPNN, with outstanding performance in both in silico and experimental tests. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4%, compared to 32.9% for Rosetta. Incorporation of noise during training improves sequence recovery on protein structure models, and produces sequences which more robustly encode their structures as assessed using structure prediction algorithms. We demonstrate the broad utility and high accuracy of ProteinMPNN using X-ray crystallography, cryoEM and functional studies by rescuing previously failed designs, made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target binding proteins.

There has been considerable recent progress in designing new proteins using deep learning methods. Despite this progress, a general deep learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher order symmetric architectures, has yet to be described. Diffusion models have had considerable success in image and language generative modeling but limited success when applied to protein modeling, likely due to the complexity of protein backbone geometry and sequence-structure relationships. Here we show that by fine tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding, and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold Diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of new designs. In a manner analogous to networks which produce images from user-specified inputs, RFdiffusion enables the design of diverse, complex, functional proteins from simple molecular specifications.

Deep learning generative approaches provide an opportunity to broadly explore protein structure space beyond the sequences and structures of natural proteins. Here we use deep network hallucination to generate a wide range of symmetric protein homo-oligomers given only a specification of the number of protomers and the protomer length. Crystal structures of 7 designs are very close to the computational models (median RMSD: 0.6 Å), as are 3 cryoEM structures of giant 10 nanometer rings with up to 1550 residues and C33 symmetry; all differ considerably from previously solved structures. Our results highlight the rich diversity of new protein structures that can be generated using deep learning, and pave the way for the design of increasingly complex components for nanomachines and biomaterials.

Understanding mechanisms of antibody synergy is important for vaccine design and antibody cocktail development. Examples of synergy between antibodies are well-documented, but the mechanisms underlying these relationships often remain poorly understood. The leading blood-stage malaria vaccine candidate, CyRPA, is essential for invasion of Plasmodium falciparum into human erythrocytes. Here we present a panel of anti-CyRPA monoclonal antibodies that strongly inhibit parasite growth in in vitro assays. Structural studies show that growth-inhibitory antibodies bind epitopes on a single face of CyRPA. We also show that pairs of non-competing inhibitory antibodies have strongly synergistic growth-inhibitory activity. These antibodies bind to neighbouring epitopes on CyRPA and form lateral, heterotypic interactions which slow antibody dissociation. We predict that such heterotypic interactions will be a feature of many immune responses. Immunogens which elicit such synergistic antibody mixtures could increase the potency of vaccine-elicited responses to provide robust and long-lived immunity against challenging disease targets.

Despite ongoing efforts, a highly effective vaccine against Plasmodium falciparum remains elusive. Vaccines targeting the pre-erythrocytic stages of the P. falciparum life cycle are the most advanced to date, affording moderate levels of efficacy in field trials. However, the discovery that the members of the merozoite PfRH5-PfCyRPA-PfRipr (RCR) complex are capable of inducing strain-transcendent neutralizing antibodies has renewed enthusiasm for the possibility of preventing disease by targeting the parasite during the blood stage of infection. With Phase I/II clinical trials now underway using firstgeneration vaccines against PfRH5, and more on the horizon for PfCyRPA and PfRipr, this review explores the rationale and future potential of the RCR complex as a P. falciparum vaccine target. Malaria Vaccine StatusGlobal malaria mortality and morbidity has declined sharply in recent decades. Still, in 2018 there were 228 million infections resulting in 405 000 deaths, disproportionately shouldered by the developing world [1]. Despite progress, malaria continues to be an intractable global health threat. Improved access to insecticide-impregnated bed nets, vector control, and the availability of effective antimalarial medications have been the cornerstones of global malaria control efforts and will continue to play indispensable roles. However, even optimal deployment of current tools will still leave elimination in high-transmission settings unattainable [2].A highly effective malaria vaccine would be one way to achieve further reductions in the global malaria burden. The observation that passive immunoglobulin (Ig) transfer confers immunity against malaria suggests that a malaria vaccine is conceivable [3]. One malaria vaccine candidate from GlaxoSmithKline, RTS,S/AS01, targets the circumsporozoite protein (CSP) and has now progressed beyond Phase III and into pilot implementation trials. This is a milestone for the malaria vaccine field and provides an important proof-of-principle for this approach; however, improvement is still required given that RTS,S/AS01 affords only partial protection of modest duration [4]. Nonetheless, this partial success gives strong impetus for continued effort and investment to develop a more effective next-generation malaria vaccine.Currently, the field awaits proof-of-concept for a substantially improved CSP-based vaccine, while whole-sporozoite strategies still face challenges with regard to scalability, immunogenicity in African infants, and breadth of protection [5,6]. An alternative and complementary approach would be to include a vaccine targeting the subsequent pathogenic blood stage of infection; this would have the potential to protect against malaria death, disease, and transmission. This review focuses on the recently described PfRH5-PfCyRPA-PfRipr (RCR) complex, an elongated protein trimer formed on the P. falciparum merozoite surface that binds to erythrocyte basigin, as a new and highly promising next-generation blood-stage vaccine (see Glossary) candidate (Figure 1A). I...