An enumerative algorithm for de novo design of proteins with diverse pocket structures

Efficient Generation of Protein Pockets with PocketGen

Designing protein-binding proteins is critical for drug discovery. However, the AI-based design of such proteins is challenging due to the complexity of ligand-protein interactions, the flexibility of ligand molecules and amino acid side chains, and sequence-structure dependencies. We introduce PocketGen, a deep generative model that simultaneously produces both the residue sequence and atomic structure of the protein regions where ligand interactions occur. PocketGen ensures consistency between sequence and structure by using a graph transformer for structural encoding and a sequence refinement module based on a protein language model. The bilevel graph transformer captures interactions at multiple scales, including atom, residue, and ligand levels. To enhance sequence refinement, PocketGen integrates a structural adapter into the protein language model, ensuring that structure-based predictions align with sequence-based predictions. PocketGen can generate high-fidelity protein pockets with superior binding affinity and structural validity. It operates ten times faster than physics-based methods and achieves a 95% success rate, defined as the percentage of generated pockets with higher binding affinity than reference pockets. Additionally, it attains an amino acid recovery rate exceeding 64%.

Navigating the landscape of enzyme design: from molecular simulations to machine learning

Global environmental issues and sustainable development call for new technologies for fine chemical synthesis and waste valorization. Biocatalysis has attracted great attention as the alternative to the traditional organic synthesis. However, it is challenging to navigate the vast sequence space to identify those proteins with admirable biocatalytic functions. The recent development of deep-learning based structure prediction methods such as AlphaFold2 reinforced by different computational simulations or multiscale calculations has largely expanded the 3D structure databases and enabled structure-based design. While structure-based approaches shed light on site-specific enzyme engineering, they are not suitable for large-scale screening of potential biocatalysts. Effective utilization of big data using machine learning techniques opens up a new era for accelerated predictions. Here, we review the approaches and applications of structure-based and machine-learning guided enzyme design. We also provide our view on the challenges and perspectives on effectively employing enzyme design approaches integrating traditional molecular simulations and machine learning, and the importance of database construction and algorithm development in attaining predictive ML models to explore the sequence fitness landscape for the design of admirable biocatalysts.

Neural network extrapolation to distant regions of the protein fitness landscape

Machine learning (ML) has transformed protein engineering by constructing models of the underlying sequence-function landscape to accelerate the discovery of new biomolecules. ML-guided protein design requires models, trained on local sequence-function information, to accurately predict distant fitness peaks. In this work, we evaluate neural networks' capacity to extrapolate beyond their training data. We perform model-guided design using a panel of neural network architectures trained on protein G (GB1)-Immunoglobulin G (IgG) binding data and experimentally test thousands of GB1 designs to systematically evaluate the models' extrapolation. We find each model architecture infers markedly different landscapes from the same data, which give rise to unique design preferences. We find simpler models excel in local extrapolation to design high fitness proteins, while more sophisticated convolutional models can venture deep into sequence space to design proteins that fold but are no longer functional. We also find that implementing a simple ensemble of convolutional neural networks enables robust design of high-performing variants in the local landscape. Our findings highlight how each architecture's inductive biases prime them to learn different aspects of the protein fitness landscape and how a simple ensembling approach makes protein engineering more robust.

Novel structure of secreted small molecular weight antigen Mtb12 from Mycobacterium tuberculosis

Mtb12, a small protein secreted by Mycobacterium tuberculosis, is known to elicit immune responses in individuals infected with the pathogen. It serves as an antigen recognized by the host's immune system. Due to its immunogenic properties and pivotal role in tuberculosis (TB) pathogenesis, Mtb12 is considered a promising candidate for TB diagnosis and vaccine development. However, the structural and functional properties of Mtb12 are largely unexplored, representing a significant gap in our understanding of M. tuberculosis biology. In this study, we present the first structure of Mtb12, which features a unique tertiary configuration consisting of four beta strands and four alpha helices. Structural analysis reveals that Mtb12 has a surface adorned with a negatively charged pocket adjacent to a central cavity. The features of these structural elements and their potential effects on the function of Mtb12 warrant further exploration. These findings offer valuable insights for vaccine design and the development of diagnostic tools.

Unlocking the potential of enzyme engineering via rational computational design strategies

Enzymes play a pivotal role in various industries by enabling efficient, eco-friendly, and sustainable chemical processes. However, the low turnover rates and poor substrate selectivity of enzymes limit their large-scale applications. Rational computational enzyme design, facilitated by computational algorithms, offers a more targeted and less labor-intensive approach. There has been notable advancement in employing rational computational protein engineering strategies to overcome these issues, it has not been comprehensively reviewed so far. This article reviews recent developments in rational computational enzyme design, categorizing them into three types: structure-based, sequence-based, and data-driven machine learning computational design. Case studies are presented to demonstrate successful enhancements in catalytic activity, stability, and substrate selectivity. Lastly, the article provides a thorough analysis of these approaches, highlights existing challenges and potential solutions, and offers insights into future development directions.

D-allose, a typical rare sugar: properties, applications, and biosynthetic advances and challenges

D-allose, a C-3 epimer of D-glucose and an aldose-ketose isomer of D-allulose, exhibits 80% of sucrose's sweetness while being remarkably low in calories and nontoxic, making it an appealing sucrose substitute. Its diverse physiological functions, particularly potent anticancer and antitumor effects, render it a promising candidate for clinical treatment, garnering sustained attention. However, its limited availability in natural sources and the challenges associated with chemical synthesis necessitate exploring biosynthetic strategies to enhance production. This overview encapsulates recent advancements in D-allose's physicochemical properties, physiological functions, applications, and biosynthesis. It also briefly discusses the crucial role of understanding aldoketose isomerase structure and optimizing its performance in D-allose synthesis. Furthermore, it delves into the challenges and future perspectives in D-allose bioproduction. Early efforts focused on identifying and characterizing enzymes responsible for D-allose production, followed by detailed crystal structure analysis to improve performance through molecular modification. Strategies such as enzyme immobilization and implementing multi-enzyme cascade reactions, utilizing more cost-effective feedstocks, were explored. Despite progress, challenges remain, including the lack of efficient high-throughput screening methods for enzyme modification, the need for food-grade expression systems, the establishment of ordered substrate channels in multi-enzyme cascade reactions, and the development of downstream separation and purification processes.

De novo design of cavity-containing proteins with a backbone-centered neural network energy function

The design of small-molecule-binding proteins requires protein backbones that contain cavities. Previous design efforts were based on naturally occurring cavity-containing backbone architectures. Here, we designed diverse cavity-containing backbones without predefined architectures by introducing tailored restraints into the backbone sampling driven by SCUBA (Side Chain-Unknown Backbone Arrangement), a neural network statistical energy function. For 521 out of 5816 designs, the root-mean-square deviations (RMSDs) of the Cα atoms for the AlphaFold2-predicted structures and our designed structures are within 2.0 Å. We experimentally tested 10 designed proteins and determined the crystal structures of two of them. One closely agrees with the designed model, while the other forms a domain-swapped dimer, where the partial structures are in agreement with the designed structures. Our results indicate that data-driven methods such as SCUBA hold great potential for designing de novo proteins with tailored small-molecule-binding function.

Sparks of function by de novo protein design

Information in proteins flows from sequence to structure to function, with each step causally driven by the preceding one. Protein design is founded on inverting this process: specify a desired function, design a structure executing this function, and find a sequence that folds into this structure. This 'central dogma' underlies nearly all de novo protein-design efforts. Our ability to accomplish these tasks depends on our understanding of protein folding and function and our ability to capture this understanding in computational methods. In recent years, deep learning-derived approaches for efficient and accurate structure modeling and enrichment of successful designs have enabled progression beyond the design of protein structures and towards the design of functional proteins. We examine these advances in the broader context of classical de novo protein design and consider implications for future challenges to come, including fundamental capabilities such as sequence and structure co-design and conformational control considering flexibility, and functional objectives such as antibody and enzyme design.

Machine learning for functional protein design

Recent breakthroughs in AI coupled with the rapid accumulation of protein sequence and structure data have radically transformed computational protein design. New methods promise to escape the constraints of natural and laboratory evolution, accelerating the generation of proteins for applications in biotechnology and medicine. To make sense of the exploding diversity of machine learning approaches, we introduce a unifying framework that classifies models on the basis of their use of three core data modalities: sequences, structures and functional labels. We discuss the new capabilities and outstanding challenges for the practical design of enzymes, antibodies, vaccines, nanomachines and more. We then highlight trends shaping the future of this field, from large-scale assays to more robust benchmarks, multimodal foundation models, enhanced sampling strategies and laboratory automation.

Design of complicated all-α protein structures

A wide range of de novo protein structure designs have been achieved, but the complexity of naturally occurring protein structures is still far beyond these designs. Here, to expand the diversity and complexity of de novo designed protein structures, we sought to develop a method for designing 'difficult-to-describe' α-helical protein structures composed of irregularly aligned α-helices like globins. Backbone structure libraries consisting of a myriad of α-helical structures with five or six helices were generated by combining 18 helix-loop-helix motifs and canonical α-helices, and five distinct topologies were selected for de novo design. The designs were found to be monomeric with high thermal stability in solution and fold into the target topologies with atomic accuracy. This study demonstrated that complicated α-helical proteins are created using typical building blocks. The method we developed will enable us to explore the universe of protein structures for designing novel functional proteins.

De novo protein design-From new structures to programmable functions

Methods from artificial intelligence (AI) trained on large datasets of sequences and structures can now "write" proteins with new shapes and molecular functions de novo, without starting from proteins found in nature. In this Perspective, I will discuss the state of the field of de novo protein design at the juncture of physics-based modeling approaches and AI. New protein folds and higher-order assemblies can be designed with considerable experimental success rates, and difficult problems requiring tunable control over protein conformations and precise shape complementarity for molecular recognition are coming into reach. Emerging approaches incorporate engineering principles-tunability, controllability, and modularity-into the design process from the beginning. Exciting frontiers lie in deconstructing cellular functions with de novo proteins and, conversely, constructing synthetic cellular signaling from the ground up. As methods improve, many more challenges are unsolved.

Small-molecule binding and sensing with a designed protein family

Despite transformative advances in protein design with deep learning, the design of small-molecule-binding proteins and sensors for arbitrary ligands remains a grand challenge. Here we combine deep learning and physics-based methods to generate a family of proteins with diverse and designable pocket geometries, which we employ to computationally design binders for six chemically and structurally distinct small-molecule targets. Biophysical characterization of the designed binders revealed nanomolar to low micromolar binding affinities and atomic-level design accuracy. The bound ligands are exposed at one edge of the binding pocket, enabling the de novo design of chemically induced dimerization (CID) systems; we take advantage of this to create a biosensor with nanomolar sensitivity for cortisol. Our approach provides a general method to design proteins that bind and sense small molecules for a wide range of analytical, environmental, and biomedical applications.

Hallucination of closed repeat proteins containing central pockets

In pseudocyclic proteins, such as TIM barrels, β barrels, and some helical transmembrane channels, a single subunit is repeated in a cyclic pattern, giving rise to a central cavity that can serve as a pocket for ligand binding or enzymatic activity. Inspired by these proteins, we devised a deep-learning-based approach to broadly exploring the space of closed repeat proteins starting from only a specification of the repeat number and length. Biophysical data for 38 structurally diverse pseudocyclic designs produced in Escherichia coli are consistent with the design models, and the three crystal structures we were able to obtain are very close to the designed structures. Docking studies suggest the diversity of folds and central pockets provide effective starting points for designing small-molecule binders and enzymes.

De novo design of protein structure and function with RFdiffusion

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.

Mega-scale experimental analysis of protein folding stability in biology and design

Advances in DNA sequencing and machine learning are providing insights into protein sequences and structures on an enormous scale1. However, the energetics driving folding are invisible in these structures and remain largely unknown2. The hidden thermodynamics of folding can drive disease3,4, shape protein evolution5-7 and guide protein engineering8-10, and new approaches are needed to reveal these thermodynamics for every sequence and structure. Here we present cDNA display proteolysis, a method for measuring thermodynamic folding stability for up to 900,000 protein domains in a one-week experiment. From 1.8 million measurements in total, we curated a set of around 776,000 high-quality folding stabilities covering all single amino acid variants and selected double mutants of 331 natural and 148 de novo designed protein domains 40-72 amino acids in length. Using this extensive dataset, we quantified (1) environmental factors influencing amino acid fitness, (2) thermodynamic couplings (including unexpected interactions) between protein sites, and (3) the global divergence between evolutionary amino acid usage and protein folding stability. We also examined how our approach could identify stability determinants in designed proteins and evaluate design methods. The cDNA display proteolysis method is fast, accurate and uniquely scalable, and promises to reveal the quantitative rules for how amino acid sequences encode folding stability.

Peptides of a Feather: How Computation Is Taking Peptide Therapeutics under Its Wing

Leveraging computation in the development of peptide therapeutics has garnered increasing recognition as a valuable tool to generate novel therapeutics for disease-related targets. To this end, computation has transformed the field of peptide design through identifying novel therapeutics that exhibit enhanced pharmacokinetic properties and reduced toxicity. The process of in-silico peptide design involves the application of molecular docking, molecular dynamics simulations, and machine learning algorithms. Three primary approaches for peptide therapeutic design including structural-based, protein mimicry, and short motif design have been predominantly adopted. Despite the ongoing progress made in this field, there are still significant challenges pertaining to peptide design including: enhancing the accuracy of computational methods; improving the success rate of preclinical and clinical trials; and developing better strategies to predict pharmacokinetics and toxicity. In this review, we discuss past and present research pertaining to the design and development of in-silico peptide therapeutics in addition to highlighting the potential of computation and artificial intelligence in the future of disease therapeutics.

De novo design of small beta barrel proteins

Small beta barrel proteins are attractive targets for computational design because of their considerable functional diversity despite their very small size (<70 amino acids). However, there are considerable challenges to designing such structures, and there has been little success thus far. Because of the small size, the hydrophobic core stabilizing the fold is necessarily very small, and the conformational strain of barrel closure can oppose folding; also intermolecular aggregation through free beta strand edges can compete with proper monomer folding. Here, we explore the de novo design of small beta barrel topologies using both Rosetta energy-based methods and deep learning approaches to design four small beta barrel folds: Src homology 3 (SH3) and oligonucleotide/oligosaccharide-binding (OB) topologies found in nature and five and six up-and-down-stranded barrels rarely if ever seen in nature. Both approaches yielded successful designs with high thermal stability and experimentally determined structures with less than 2.4 Å rmsd from the designed models. Using deep learning for backbone generation and Rosetta for sequence design yielded higher design success rates and increased structural diversity than Rosetta alone. The ability to design a large and structurally diverse set of small beta barrel proteins greatly increases the protein shape space available for designing binders to protein targets of interest.

Possibilities of Using De Novo Design for Generating Diverse Functional Food Enzymes

Food enzymes have an important role in the improvement of certain food characteristics, such as texture improvement, elimination of toxins and allergens, production of carbohydrates, enhancing flavor/appearance characteristics. Recently, along with the development of artificial meats, food enzymes have been employed to achieve more diverse functions, especially in converting non-edible biomass to delicious foods. Reported food enzyme modifications for specific applications have highlighted the significance of enzyme engineering. However, using direct evolution or rational design showed inherent limitations due to the mutation rates, which made it difficult to satisfy the stability or specific activity needs for certain applications. Generating functional enzymes using de novo design, which highly assembles naturally existing enzymes, provides potential solutions for screening desired enzymes. Here, we describe the functions and applications of food enzymes to introduce the need for food enzymes engineering. To illustrate the possibilities of using de novo design for generating diverse functional proteins, we reviewed protein modelling and de novo design methods and their implementations. The future directions for adding structural data for de novo design model training, acquiring diversified training data, and investigating the relationship between enzyme-substrate binding and activity were highlighted as challenges to overcome for the de novo design of food enzymes.

De novo design of luciferases using deep learning

De novo enzyme design has sought to introduce active sites and substrate-binding pockets that are predicted to catalyse a reaction of interest into geometrically compatible native scaffolds1,2, but has been limited by a lack of suitable protein structures and the complexity of native protein sequence-structure relationships. Here we describe a deep-learning-based 'family-wide hallucination' approach that generates large numbers of idealized protein structures containing diverse pocket shapes and designed sequences that encode them. We use these scaffolds to design artificial luciferases that selectively catalyse the oxidative chemiluminescence of the synthetic luciferin substrates diphenylterazine3 and 2-deoxycoelenterazine. The designed active sites position an arginine guanidinium group adjacent to an anion that develops during the reaction in a binding pocket with high shape complementarity. For both luciferin substrates, we obtain designed luciferases with high selectivity; the most active of these is a small (13.9 kDa) and thermostable (with a melting temperature higher than 95 °C) enzyme that has a catalytic efficiency on diphenylterazine (kcat/Km = 106 M-1 s-1) comparable to that of native luciferases, but a much higher substrate specificity. The creation of highly active and specific biocatalysts from scratch with broad applications in biomedicine is a key milestone for computational enzyme design, and our approach should enable generation of a wide range of luciferases and other enzymes.

Design of Diverse Asymmetric Pockets in De Novo Homo-oligomeric Proteins

A challenge for design of protein-small-molecule recognition is that incorporation of cavities with size, shape, and composition suitable for specific recognition can considerably destabilize protein monomers. This challenge can be overcome through binding pockets formed at homo-oligomeric interfaces between folded monomers. Interfaces surrounding the central homo-oligomer symmetry axes necessarily have the same symmetry and so may not be well suited to binding asymmetric molecules. To enable general recognition of arbitrary asymmetric substrates and small molecules, we developed an approach to designing asymmetric interfaces at off-axis sites on homo-oligomers, analogous to those found in native homo-oligomeric proteins such as glutamine synthetase. We symmetrically dock curved helical repeat proteins such that they form pockets at the asymmetric interface of the oligomer with sizes ranging from several angstroms, appropriate for binding a single ion, to up to more than 20 Å across. Of the 133 proteins tested, 84 had soluble expression in E. coli, 47 had correct oligomeric states in solution, 35 had small-angle X-ray scattering (SAXS) data largely consistent with design models, and 8 had negative-stain electron microscopy (nsEM) 2D class averages showing the structures coming together as designed. Both an X-ray crystal structure and a cryogenic electron microscopy (cryoEM) structure are close to the computational design models. The nature of these proteins as homo-oligomers allows them to be readily built into higher-order structures such as nanocages, and the asymmetric pockets of these structures open rich possibilities for small-molecule binder design free from the constraints associated with monomer destabilization.

MlaC belongs to a unique class of non-canonical substrate-binding proteins and follows a novel phospholipid-binding mechanism

The outer membrane (OM) of Gram-negative bacteria acts as a formidable barrier against a plethora of detrimental compounds owing to its asymmetric nature. This is because the OM possesses lipopolysaccharides (LPSs) in the outer leaflet and phospholipids (PLs) in the inner leaflet. The maintenance of lipid asymmetry (Mla) system is involved in preserving the distribution of PLs in OM. The periplasmic component of the system MlaC serves as the substrate-binding protein (SBP) that shuttles PLs between the inner and outer membranes. However, an in-depth report highlighting its mechanism of ligand binding is still lacking. This study reports the crystal structure of MlaC from Escherichia coli (EcMlaC) at a resolution of 2.5 Å in a quasi-open state, complexed with PL. The structural analysis reveals that EcMlaC and orthologs comprise two major domains, viz. nuclear transport factor 2-like (NTF2-like) and phospholipid-binding protein (PBP). Each domain can be further divided into two subdomains arranged in a discontinuous fashion. This study further reveals that EcMlaC is polyspecific in nature and follows a reverse mechanism of the opening of the substrate-binding site during the ligand binding. Furthermore, MlaC can bind two PLs by forming subsites in the binding pocket. These findings, altogether, have led to the proposition of the unique "segmented domain movement" mechanism of PL binding, not reported for any known SBP to date. Further, unlike typical SBPs, MlaC has originated from a cystatin-like fold. Overall, this study establishes MlaC to be a non-canonical SBP with a unique ligand-binding mechanism.

Dissecting the stability determinants of a challenging de novo protein fold using massively parallel design and experimentation

Designing entirely new protein structures remains challenging because we do not fully understand the biophysical determinants of folding stability. Yet, some protein folds are easier to design than others. Previous work identified the 43-residue ɑββɑ fold as especially challenging: The best designs had only a 2% success rate, compared to 39 to 87% success for other simple folds [G. J. Rocklin et al., Science 357, 168-175 (2017)]. This suggested the ɑββɑ fold would be a useful model system for gaining a deeper understanding of folding stability determinants and for testing new protein design methods. Here, we designed over 10,000 new ɑββɑ proteins and found over 3,000 of them to fold into stable structures using a high-throughput protease-based assay. NMR, hydrogen-deuterium exchange, circular dichroism, deep mutational scanning, and scrambled sequence control experiments indicated that our stable designs fold into their designed ɑββɑ structures with exceptional stability for their small size. Our large dataset enabled us to quantify the influence of universal stability determinants including nonpolar burial, helix capping, and buried unsatisfied polar atoms, as well as stability determinants unique to the ɑββɑ topology. Our work demonstrates how large-scale design and test cycles can solve challenging design problems while illuminating the biophysical determinants of folding.

Identifying well-folded de novo proteins in the new era of accurate structure prediction

Computational de novo protein design tailors proteins for target structures and oligomerisation states with high stability, which allows overcoming many limitations of natural proteins when redesigned for new functions. Despite significant advances in the field over the past decade, it remains challenging to predict sequences that will fold as stable monomers in solution or binders to a particular protein target; thereby requiring substantial experimental resources to identify proteins with the desired properties. To overcome this, here we leveraged the large amount of design data accumulated in the last decade, and the breakthrough in protein structure prediction from last year to investigate on improved ways of selecting promising designs before experimental testing. We collected de novo proteins from previous studies, 518 designed as monomers of different folds and 2112 as binders against the Botulinum neurotoxin, and analysed their structures with AlphaFold2, RoseTTAFold and fragment quality descriptors in combination with other properties related to surface interactions. These features showed high complementarity in rationalizing the experimental results, which allowed us to generate quite accurate machine learning models for predicting well-folded monomers and binders with a small set of descriptors. Cross-validating designs with varied orthogonal computational techniques should guide us for identifying design imperfections, rescuing designs and making more robust design selections before experimental testing.

Redesigning Enzymes for Biocatalysis: Exploiting Structural Understanding for Improved Selectivity

The discovery of new enzymes, alongside the push to make chemical processes more sustainable, has resulted in increased industrial interest in the use of biocatalytic processes to produce high-value and chiral precursor chemicals. Huge strides in protein engineering methodology and in silico tools have facilitated significant progress in the discovery and production of enzymes for biocatalytic processes. However, there are significant gaps in our knowledge of the relationship between enzyme structure and function. This has demonstrated the need for improved computational methods to model mechanisms and understand structure dynamics. Here, we explore efforts to rationally modify enzymes toward changing aspects of their catalyzed chemistry. We highlight examples of enzymes where links between enzyme function and structure have been made, thus enabling rational changes to the enzyme structure to give predictable chemical outcomes. We look at future directions the field could take and the technologies that will enable it.

Oligomeric interactions maintain active-site structure in a noncooperative enzyme family

The evolutionary benefit accounting for widespread conservation of oligomeric structures in proteins lacking evidence of intersubunit cooperativity remains unclear. Here, crystal and cryo-EM structures, and enzymological data, demonstrate that a conserved tetramer interface maintains the active-site structure in one such class of proteins, the short-chain dehydrogenase/reductase (SDR) superfamily. Phylogenetic comparisons support a significantly longer polypeptide being required to maintain an equivalent active-site structure in the context of a single subunit. Oligomerization therefore enhances evolutionary fitness by reducing the metabolic cost of enzyme biosynthesis. The large surface area of the structure-stabilizing oligomeric interface yields a synergistic gain in fitness by increasing tolerance to activity-enhancing yet destabilizing mutations. We demonstrate that two paralogous SDR superfamily enzymes with different specificities can form mixed heterotetramers that combine their individual enzymological properties. This suggests that oligomerization can also diversify the functions generated by a given metabolic investment, enhancing the fitness advantage provided by this architectural strategy.

Harnessing the Structural and Functional Diversity of Protein Filaments as Biomaterial Scaffolds

The natural ability of many proteins to polymerize into highly structured filaments has been harnessed as scaffolds to align functional molecules in a diverse range of biomaterials. Protein-engineering methodologies also enable the structural and physical properties of filaments to be tailored for specific biomaterial applications through genetic engineering or filaments built from the ground up using advances in the computational prediction of protein folding and assembly. Using these approaches, protein filament-based biomaterials have been engineered to accelerate enzymatic catalysis, provide routes for the biomineralization of inorganic materials, facilitate energy production and transfer, and provide support for mammalian cells for tissue engineering. In this review, we describe how the unique structural and functional diversity in natural and computationally designed protein filaments can be harnessed in biomaterials. In addition, we detail applications of these protein assemblies as material scaffolds with a particular emphasis on applications that exploit unique properties of specific filaments. Through the diversity of protein filaments, the biomaterial engineer's toolbox contains many modular protein filaments that will likely be incorporated as the main structural component of future biomaterials.

Mimetic Neural Networks: A Unified Framework for Protein Design and Folding

Recent advancements in machine learning techniques for protein structure prediction motivate better results in its inverse problem–protein design. In this work we introduce a new graph mimetic neural network, MimNet, and show that it is possible to build a reversible architecture that solves the structure and design problems in tandem, allowing to improve protein backbone design when the structure is better estimated. We use the ProteinNet data set and show that the state of the art results in protein design can be met and even improved, given recent architectures for protein folding.

Molecular and experimental tools to design synthetic enhancers

Understanding the grammar of enhancers and how they regulate gene expression is key for both basic research and for the pharma and biotech industries. The design and characterization of synthetic enhancers can expand the known regulatory space. This is achieved by the utilization of DNA Oligo Libraries (OLs), which facilitates screening of as many as millions of synthetic enhancer variants simultaneously. This review includes the latest commercial DNA OL synthesis technology and its capabilities, and a general 'know-how' guide for the design, construction, and analysis of OL-based synthetic enhancer characterization experiments. Specifically, we focus on synthetic-enhancer-based massively parallel reporter assay, Sort-seq methodologies (e.g. flow cytometry, deep sequencing), and a brief description of machine learning-based attempts for OL-analysis and follow-up validation experiments.

Advances in the Computational Design of Small-Molecule-Controlled Protein-Based Circuits for Synthetic Biology

Synthetic biology approaches living systems with an engineering perspective and promises to deliver solutions to global challenges in healthcare and sustainability. A critical component is the design of biomolecular circuits with programmable input-output behaviors. Such circuits typically rely on a sensor module that recognizes molecular inputs, which is coupled to a functional output via protein-level circuits or regulating the expression of a target gene. While gene expression outputs can be customized relatively easily by exchanging the target genes, sensing new inputs is a major limitation. There is a limited repertoire of sensors found in nature, and there are often difficulties with interfacing them with engineered circuits. Computational protein design could be a key enabling technology to address these challenges, as it allows for the engineering of modular and tunable sensors that can be tailored to the circuit's application. In this article, we review recent computational approaches to design protein-based sensors for small-molecule inputs with particular focus on those based on the widely used Rosetta software suite. Furthermore, we review mechanisms that have been harnessed to couple ligand inputs to functional outputs. Based on recent literature, we illustrate how the combination of protein design and synthetic biology enables new sensors for diverse applications ranging from biomedicine to metabolic engineering. We conclude with a perspective on how strategies to address frontiers in protein design and cellular circuit design may enable the next generation of sense-response networks, which may increasingly be assembled from de novo components to display diverse and engineerable input-output behaviors.

Recent Advances in Biocatalysis for Drug Synthesis

Biocatalysis is constantly providing novel options for the synthesis of active pharmaceutical ingredients (APIs). In addition to drug development and manufacturing, biocatalysis also plays a role in drug discovery and can support many active ingredient syntheses at an early stage to build up entire scaffolds in a targeted and preparative manner. Recent progress in recruiting new enzymes by genome mining and screening or adapting their substrate, as well as product scope, by protein engineering has made biocatalysts a competitive tool applied in academic and industrial spheres. This is especially true for the advances in the field of nonribosomal peptide synthesis and enzyme cascades that are expanding the capabilities for the discovery and synthesis of new bioactive compounds via biotransformation. Here we highlight some of the most recent developments to add to the portfolio of biocatalysis with special relevance for the synthesis and late-stage functionalization of APIs, in order to bypass pure chemical processes.

Large-scale design and refinement of stable proteins using sequence-only models

Engineered proteins generally must possess a stable structure in order to achieve their designed function. Stable designs, however, are astronomically rare within the space of all possible amino acid sequences. As a consequence, many designs must be tested computationally and experimentally in order to find stable ones, which is expensive in terms of time and resources. Here we use a high-throughput, low-fidelity assay to experimentally evaluate the stability of approximately 200,000 novel proteins. These include a wide range of sequence perturbations, providing a baseline for future work in the field. We build a neural network model that predicts protein stability given only sequences of amino acids, and compare its performance to the assayed values. We also report another network model that is able to generate the amino acid sequences of novel stable proteins given requested secondary sequences. Finally, we show that the predictive model-despite weaknesses including a noisy data set-can be used to substantially increase the stability of both expert-designed and model-generated proteins.

The stability and dynamics of computationally designed proteins

Protein stability, dynamics and function are intricately linked. Accordingly, protein designers leverage dynamics in their designs and gain insight to their successes and failures by analyzing their proteins' dynamics. Molecular dynamics (MD) simulations are a powerful computational tool for quantifying both local and global protein dynamics. This review highlights studies where MD simulations were applied to characterize the stability and dynamics of designed proteins and where dynamics were incorporated into computational protein design. First, we discuss the structural basis underlying the extreme stability and thermostability frequently observed in computationally designed proteins. Next, we discuss examples of designed proteins, where dynamics were not explicitly accounted for in the design process, whose coordinated motions or active site dynamics, as observed by MD simulation, enhanced or detracted from their function. Many protein functions depend on sizeable or subtle conformational changes, so we finally discuss the computational design of proteins to perform a specific function that requires consideration of motion by multi-state design.

Design principles of protein switches

Protein switches perform essential roles in many biological processes and are exciting targets for de novo protein design, which aims to produce proteins of arbitrary shape and functionality. However, the biophysical requirements for switch function - multiple conformational states, fine-tuned energetics, and stimuli-responsiveness - pose a formidable challenge for design by computation (or intuition). A variety of methods have been developed toward tackling this challenge, usually taking inspiration from the wealth of sequence and structural information available for naturally occurring protein switches. More recently, modular switches have been designed computationally, and new methods have emerged for sampling unexplored structure space, providing promising new avenues toward the generation of purpose-built switches and de novo signaling systems for cellular engineering.

Thermostabilization mechanisms in thermophilic versus mesophilic three-helix bundle proteins

The engineered three‐helix bundle, UVF, is thermostabilized entropically due to heightened, native‐state dynamics. However, it is unclear whether this thermostabilization strategy is observed in natural proteins from thermophiles. We performed all‐atom, explicit solvent molecular dynamics simulations of two three‐helix bundles from thermophilic H. butylicus (2lvsN and 2lvsC) and compared their dynamics to a mesophilic three‐helix bundle, the Engrailed homeodomain (EnHD). Like UVF, 2lvsC had heightened native dynamics, which it maintained without unfolding at 100°C. Shortening and rigidification of loops in 2lvsN and 2lvsC and increased surface hydrogen bonds in 2lvsN were observed, as is common in thermophilic proteins. A buried disulfide and salt bridge in 2lvsN and 2lvsC, respectively, provided some stabilization, and addition of a homologous disulfide bond in EnHD slowed unfolding. The transferability and commonality of stabilization strategies among members of the three‐helix bundle fold suggest that these strategies may be general and deployable in designing thermostable proteins.

Prediction of Residue-specific Contributions to Binding and Thermal Stability Using Yeast Surface Display

Accurate prediction of residue burial as well as quantitative prediction of residue-specific contributions to protein stability and activity is challenging, especially in the absence of experimental structural information. This is important for prediction and understanding of disease causing mutations, and for protein stabilization and design. Using yeast surface display of a saturation mutagenesis library of the bacterial toxin CcdB, we probe the relationship between ligand binding and expression level of displayed protein, with in vivo solubility in E. coli and in vitro thermal stability. We find that both the stability and solubility correlate well with the total amount of active protein on the yeast cell surface but not with total amount of expressed protein. We coupled FACS and deep sequencing to reconstruct the binding and expression mean fluorescent intensity of each mutant. The reconstructed mean fluorescence intensity (MFIseq) was used to differentiate between buried site, exposed non active-site and exposed active-site positions with high accuracy. The MFIseq was also used as a criterion to identify destabilized as well as stabilized mutants in the library, and to predict the melting temperatures of destabilized mutants. These predictions were experimentally validated and were more accurate than those of various computational predictors. The approach was extended to successfully identify buried and active-site residues in the receptor binding domain of the spike protein of SARS-CoV-2, suggesting it has general applicability.

Oligo Pools as an Affordable Source of Synthetic DNA for Cost-Effective Library Construction in Protein- and Metabolic Pathway Engineering

The construction of custom libraries is critical for rational protein engineering and directed evolution. Array‐synthesized oligo pools of thousands of user‐defined sequences (up to ∼350 bases in length) have emerged as a low‐cost commercially available source of DNA. These pools cost ≤10 % (depending on error rate and length) of other commercial sources of custom DNA, and this significant cost difference can determine whether an enzyme engineering project can be realized on a given research budget. However, while being cheap, oligo pools do suffer from a low concentration of individual oligos and relatively high error rates. Several powerful techniques that specifically make use of oligo pools have been developed and proven valuable or even essential for next‐generation protein and pathway engineering strategies, such as sequence‐function mapping, enzyme minimization, or de‐novo design. Here we consolidate the knowledge on these techniques and their applications to facilitate the use of oligo pools within the protein engineering community.

De novo protein fold families expand the designable ligand binding site space

A major challenge in designing proteins de novo to bind user-defined ligands with high affinity is finding backbones structures into which a new binding site geometry can be engineered with high precision. Recent advances in methods to generate protein fold families de novo have expanded the space of accessible protein structures, but it is not clear to what extend de novo proteins with diverse geometries also expand the space of designable ligand binding functions. We constructed a library of 25,806 high-quality ligand binding sites and developed a fast protocol to place (“match”) these binding sites into both naturally occurring and de novo protein families with two fold topologies: Rossman and NTF2. Each matching step involves engineering new binding site residues into each protein “scaffold”, which is distinct from the problem of comparing already existing binding pockets. 5,896 and 7,475 binding sites could be matched to the Rossmann and NTF2 fold families, respectively. De novo designed Rossman and NTF2 protein families can support 1,791 and 678 binding sites that cannot be matched to naturally existing structures with the same topologies, respectively. While the number of protein residues in ligand binding sites is the major determinant of matching success, ligand size and primary sequence separation of binding site residues also play important roles. The number of matched binding sites are power law functions of the number of members in a fold family. Our results suggest that de novo sampling of geometric variations on diverse fold topologies can significantly expand the space of designable ligand binding sites for a wealth of possible new protein functions.

De novo design of proteins that can bind to novel and highly diverse user-defined small molecule ligands could have broad biomedical and synthetic biology applications. Because ligand binding site geometries need to be accommodated by protein backbone scaffolds at high accuracy, the diversity of scaffolds is a major limitation for designing new ligand binding functions. Advances in computational protein structure design methods have significantly increased the number of accessible stable scaffold structures. Understanding how many new ligand binding sites can be designed into the de novo scaffolds is important for engineering novel ligand binding proteins. To answer this question, we constructed a large library of ligand binding sites from the Protein Data Bank (PDB). We tested the number of ligand binding sites that can be designed into de novo scaffolds and naturally existing scaffolds with the same fold topologies. The results showed that de novo scaffolds significantly expanded the potential ligand binding space of their respective fold topologies. We also identified factors that affect difficulties of binding site accommodation, as well as the relationship between the number of scaffolds and the accessible ligand binding site space. We believe our findings will benefit future method development and applications of ligand binding protein design.

Enantioselective Construction of Modular and Asymmetric Baskets

The precise positioning of functional groups about the inner space of abiotic hosts is a challenging task and of interest for developing more effective receptors and catalysts akin to those found in nature. To address it, we herein report a synthetic methodology for preparing basket-like cavitands comprised of three different aromatics as side arms with orthogonal esters at the rim for further functionalization. First, enantioenriched A (borochloronorbornene), B (iodobromonorbornene), and C (boronorbornene) building blocks were obtained by stereoselective syntheses. Second, consecutive A-to-B and then AB-to-C Suzuki-Miyaura (SM) couplings were optimized to give enantioenriched ABC cavitand as the principal product. The robust synthetic protocol allowed us to prepare (a) an enantioenriched basket with three benzene sides and each holding either tBu, Et, or Me esters, (b) both enantiomers of a so-called "spiral staircase" basket with benzene, naphthalene, and anthracene groups surrounding the inner space, and (c) a photo-responsive basket bearing one anthracene and two benzene arms.

How much can physics do for protein design?

Physics and physical chemistry are an important thread in computational protein design, complementary to knowledge-based tools. They provide molecular mechanics scoring functions that need little or no ad hoc parameter readjustment, methods to thoroughly sample equilibrium ensembles, and different levels of approximation for conformational flexibility. They led recently to the successful redesign of a small protein using a physics-based folded state energy. Adaptive Monte Carlo or molecular dynamics schemes were discovered where protein variants are populated as per their ligand-binding free energy or catalytic efficiency. Molecular dynamics have been used for backbone flexibility. Implicit solvent models have been refined, polarizable force fields applied, and many physical insights obtained.

Peptide-drug conjugate-based novel molecular drug delivery system in cancer

Drug delivery systems are generally believed to comprise drugs and excipients. A peptide-drug conjugate is a single molecule that can simultaneously play multiple roles in a drug delivery system, such as in vivo drug distribution, targeted release, and bioactivity functions. This molecule can be regarded as an integrated drug delivery system, so it is called a molecular drug delivery system. In the context of cancer therapy, a peptide-drug conjugate comprises a tumor-targeting peptide, a payload, and a linker. Tumor-targeting peptides specifically identify membrane receptors on tumor cells, improve drug-targeted therapeutic effects, and reduce toxic and side effects. Payloads with bioactive functions connect to tumor-targeting peptides through linkers. In this review, we explored ongoing clinical work on peptide-drug conjugates targeting various receptors. We discuss the binding mechanisms of tumor-targeting peptides and related receptors, as well as the limiting factors for peptide-drug conjugate-based molecular drug delivery systems.

Principles and Methods in Computational Membrane Protein Design

After decades of progress in computational protein design, the design of proteins folding and functioning in lipid membranes appears today as the next frontier. Some notable successes in the de novo design of simplified model membrane protein systems have helped articulate fundamental principles of protein folding, architecture and interaction in the hydrophobic lipid environment. These principles are reviewed here, together with the computational methods and approaches that were used to identify them. We provide an overview of the methodological innovations in the generation of new protein structures and functions and in the development of membrane-specific energy functions. We highlight the opportunities offered by new machine learning approaches applied to protein design, and by new experimental characterization techniques applied to membrane proteins. Although membrane protein design is in its infancy, it appears more reachable than previously thought.

Computer-aided understanding and engineering of enzymatic selectivity

Enzymes offering chemo-, regio-, and stereoselectivity enable the asymmetric synthesis of high-value chiral molecules. Unfortunately, the drawback that naturally occurring enzymes are often inefficient or have undesired selectivity toward non-native substrates hinders the broadening of biocatalytic applications. To match the demands of specific selectivity in asymmetric synthesis, biochemists have implemented various computer-aided strategies in understanding and engineering enzymatic selectivity, diversifying the available repository of artificial enzymes. Here, given that the entire asymmetric catalytic cycle, involving precise interactions within the active pocket and substrate transport in the enzyme channel, could affect the enzymatic efficiency and selectivity, we presented a comprehensive overview of the computer-aided workflow for enzymatic selectivity. This review includes a mechanistic understanding of enzymatic selectivity based on quantum mechanical calculations, rational design of enzymatic selectivity guided by enzyme-substrate interactions, and enzymatic selectivity regulation via enzyme channel engineering. Finally, we discussed the computational paradigm for designing enzyme selectivity in silico to facilitate the advancement of asymmetric biosynthesis.

Fold2Seq: A Joint Sequence(1D)-Fold(3D) Embedding-based Generative Model for Protein Design

Designing novel protein sequences for a desired 3D topological fold is a fundamental yet nontrivial task in protein engineering. Challenges exist due to the complex sequence-fold relationship, as well as the difficulties to capture the diversity of the sequences (therefore structures and functions) within a fold. To overcome these challenges, we propose Fold2Seq, a novel transformer-based generative framework for designing protein sequences conditioned on a specific target fold. To model the complex sequence-structure relationship, Fold2Seq jointly learns a sequence embedding using a transformer and a fold embedding from the density of secondary structural elements in 3D voxels. On test sets with single, high-resolution and complete structure inputs for individual folds, our experiments demonstrate improved or comparable performance of Fold2Seq in terms of speed, coverage, and reliability for sequence design, when compared to existing state-of-the-art methods that include data-driven deep generative models and physics-based RosettaDesign. The unique advantages of fold-based Fold2Seq, in comparison to a structure-based deep model and RosettaDesign, become more evident on three additional real-world challenges originating from low-quality, incomplete, or ambiguous input structures. Source code and data are available at https://github.com/IBM/fold2seq.

Limitations of the ABEGO-based backbone design: ambiguity between αα-corner and αα-hairpin

ABEGO is a coarse-grained representation for poly-peptide backbone dihedral angles. The Ramachandran map is divided into four segments denoted as A, B, E, and G to represent the local conformation of poly-peptide chains in the character strings. Although the ABEGO representation is widely used in backbone building simulation for de novo protein design, it cannot capture minor differences in backbone dihedral angles, which potentially leads to ambiguity between two structurally distinct fragments. Here, I show a nontrivial example of two local motifs that could not be distinguished by their ABEGO representations. I found that two well-known local motifs αα-hairpins and αα-corners are both represented as α-GBB-α and thus indistinguishable in the ABEGO representation, although they show distinct arrangements of the flanking α-helices. I also found that α-GBB-α motifs caused a loss of efficiency in the ABEGO-based fragment-assembly simulations for de novo protein backbone design. Nevertheless, I was able to design amino-acid sequences that were predicted to fold into the target topologies that contained these α-GBB-α motifs, which suggests such topologies that are difficult to build by ABEGO-based simulations are designable once the backbone structures are modeled by some means. The finding that certain local motifs bottleneck the ABEGO-based fragment-assembly simulations for construction of backbone structures suggests that finer representations of backbone torsion angles are required for efficiently generating diverse topologies containing such indistinguishable local motifs.

The Steady March toward Biomimetic Nanoelectronics

Steady progress is being made in unveiling nature's long-range charge transport mechanisms in redox proteins and in the development of versatile self-assembling scaffolds and de novo proteins by design-two separate fields that soon may intersect to yield the first artificial bioelectronic wires. Here, we summarize compelling developments in these areas that put a spotlight on the prospect of their convergence, featuring, in particular, work by Dai et al. in this issue of ACS Nano that illustrates success in intentional design with nuanced control, binding multiple c-type hemes into a specific ordered array bearing the essential hallmarks of heme chains in bacterial multiheme cytochromes.

Revolutionizing enzyme engineering through artificial intelligence and machine learning

The combinatorial space of an enzyme sequence has astronomical possibilities and exploring it with contemporary experimental techniques is arduous and often ineffective. Multi-target objectives such as concomitantly achieving improved selectivity, solubility and activity of an enzyme have narrow plausibility under approaches of restricted mutagenesis and combinatorial search. Traditional enzyme engineering approaches have a limited scope for complex optimization due to the requirement of a priori knowledge or experimental burden of screening huge protein libraries. The recent surge in high-throughput experimental methods including Next Generation Sequencing and automated screening has flooded the field of molecular biology with big-data, which requires us to re-think our concurrent approaches towards enzyme engineering. Artificial Intelligence (AI) and Machine Learning (ML) have great potential to revolutionize smart enzyme engineering without the explicit need for a complete understanding of the underlying molecular system. Here, we portray the role and position of AI techniques in the field of enzyme engineering along with their scope and limitations. In addition, we explain how the traditional approaches of directed evolution and rational design can be extended through AI tools. Recent successful examples of AI-assisted enzyme engineering projects and their deviation from traditional approaches are highlighted. A comprehensive picture of current challenges and future avenues for AI in enzyme engineering are also discussed.

Breakthroughs in computational design methods open up new frontiers for de novo protein engineering

Proteins catalyze the majority of chemical reactions in organisms, and harnessing this power has long been the focus of the protein engineering field. Computational protein design aims to create new proteins and functions in silico, and in doing so, accelerate the process, reduce costs and enable more sophisticated engineering goals to be accomplished. Challenges that very recently seemed impossible are now within reach thanks to several landmark advances in computational protein design methods. Here, we summarize these new methods, with a particular emphasis on de novo protein design advancements occurring within the past 5 years.