Principles for designing ideal protein structures

Opportunities and challenges in design and optimization of protein function

The field of protein design has made remarkable progress over the past decade. Historically, the low reliability of purely structure-based design methods limited their application, but recent strategies that combine structure-based and sequence-based calculations, as well as machine learning tools, have dramatically improved protein engineering and design. In this Review, we discuss how these methods have enabled the design of increasingly complex structures and therapeutically relevant activities. Additionally, protein optimization methods have improved the stability and activity of complex eukaryotic proteins. Thanks to their increased reliability, computational design methods have been applied to improve therapeutics and enzymes for green chemistry and have generated vaccine antigens, antivirals and drug-delivery nano-vehicles. Moreover, the high success of design methods reflects an increased understanding of basic rules that govern the relationships among protein sequence, structure and function. However, de novo design is still limited mostly to α-helix bundles, restricting its potential to generate sophisticated enzymes and diverse protein and small-molecule binders. Designing complex protein structures is a challenging but necessary next step if we are to realize our objective of generating new-to-nature activities.

Design of amyloidogenic peptide traps

Segments of proteins with high β-strand propensity can self-associate to form amyloid fibrils implicated in many diseases. We describe a general approach to bind such segments in β-strand and β-hairpin conformations using de novo designed scaffolds that contain deep peptide-binding clefts. The designs bind their cognate peptides in vitro with nanomolar affinities. The crystal structure of a designed protein-peptide complex is close to the design model, and NMR characterization reveals how the peptide-binding cleft is protected in the apo state. We use the approach to design binders to the amyloid-forming proteins transthyretin, tau, serum amyloid A1 and amyloid β1-42 (Aβ42). The Aβ binders block the assembly of Aβ fibrils as effectively as the most potent of the clinically tested antibodies to date and protect cells from toxic Aβ42 species.

Sculpting conducting nanopore size and shape through de novo protein design

Transmembrane β-barrels have considerable potential for a broad range of sensing applications. Current engineering approaches for nanopore sensors are limited to naturally occurring channels, which provide suboptimal starting points. By contrast, de novo protein design can in principle create an unlimited number of new nanopores with any desired properties. Here we describe a general approach to designing transmembrane β-barrel pores with different diameters and pore geometries. Nuclear magnetic resonance and crystallographic characterization show that the designs are stably folded with structures resembling those of the design models. The designs have distinct conductances that correlate with their pore diameter, ranging from 110 picosiemens (~0.5 nanometer pore diameter) to 430 picosiemens (~1.1 nanometer pore diameter). Our approach opens the door to the custom design of transmembrane nanopores for sensing and sequencing applications.

Validation of de novo designed water-soluble and transmembrane β-barrels by in silico folding and melting

In silico validation of de novo designed proteins with deep learning (DL)-based structure prediction algorithms has become mainstream. However, formal evidence of the relationship between a high-quality predicted model and the chance of experimental success is lacking. We used experimentally characterized de novo water-soluble and transmembrane β-barrel designs to show that AlphaFold2 and ESMFold excel at different tasks. ESMFold can efficiently identify designs generated based on high-quality (designable) backbones. However, only AlphaFold2 can predict which sequences have the best chance of experimentally folding among similar designs. We show that ESMFold can generate high-quality structures from just a few predicted contacts and introduce a new approach based on incremental perturbation of the prediction ("in silico melting"), which can reveal differences in the presence of favorable contacts between designs. This study provides a new insight on DL-based structure prediction models explainability and on how they could be leveraged for the design of increasingly complex proteins; in particular membrane proteins which have historically lacked basic in silico validation tools.

De novo design of alpha-beta repeat proteins

Proteins composed of a single structural unit tandemly repeated multiple times carry out a wide range of functions in biology. There has hence been considerable interest in designing such repeat proteins; previous approaches have employed strict constraints on secondary structure types and relative geometries, and most characterized designs either mimic a known natural topology, adhere closely to a parametric helical bundle architecture, or exploit very short repetitive sequences. Here, we describe Rosetta-based and deep learning hallucination methods for generating novel repeat protein architectures featuring mixed alpha-helix and beta-strand topologies, and 25 new highly stable alpha-beta proteins designed using these methods. We find that incorporation of terminal caps which prevent beta strand mediated intermolecular interactions increases the solubility and monomericity of individual designs as well as overall design success rate.

Diversifying de novo TIM barrels by hallucination

De novo protein design expands the protein universe by creating new sequences to accomplish tailor-made enzymes in the future. A promising topology to implement diverse enzyme functions is the ubiquitous TIM-barrel fold. Since the initial de novo design of an idealized four-fold symmetric TIM barrel, the family of de novo TIM barrels is expanding rapidly. Despite this and in contrast to natural TIM barrels, these novel proteins lack cavities and structural elements essential for the incorporation of binding sites or enzymatic functions. In this work, we diversified a de novo TIM barrel by extending multiple βα-loops using constrained hallucination. Experimentally tested designs were found to be soluble upon expression in Escherichia coli and well-behaved. Biochemical characterization and crystal structures revealed successful extensions with defined α-helical structures. These diversified de novo TIM barrels provide a framework to explore a broad spectrum of functions based on the potential of natural TIM barrels.

Heat-induced structural and chemical changes to a computationally designed miniprotein

The de novo design of miniprotein inhibitors has recently emerged as a new technology to create proteins that bind with high affinity to specific therapeutic targets. Their size, ease of expression, and apparent high stability makes them excellent candidates for a new class of protein drugs. However, beyond circular dichroism melts and hydrogen/deuterium exchange experiments, little is known about their dynamics, especially at the elevated temperatures they seemingly tolerate quite well. To address that and gain insight for future designs, we have focused on identifying unintended and previously overlooked heat-induced structural and chemical changes in a particularly stable model miniprotein, EHEE_rd2_0005. Nuclear magnetic resonance (NMR) studies suggest the presence of dynamics on multiple time and temperature scales. Transiently elevating the temperature results in spontaneous chemical deamidation visible in the NMR spectra, which we validate using both capillary electrophoresis and mass spectrometry (MS) experiments. High temperatures also result in greatly accelerated intrinsic rates of hydrogen exchange and signal loss in NMR heteronuclear single quantum coherence spectra from local unfolding. These losses are in excellent agreement with both room temperature hydrogen exchange experiments and hydrogen bond disruption in replica exchange molecular dynamics simulations. Our analysis reveals important principles for future miniprotein designs and the potential for high stability to result in long-lived alternate conformational states.

Structure-Based De Novo Design for the Discovery of Miniprotein Inhibitors Targeting Oncogenic Mutant BRAF

Being a component of the Ras/Raf/MEK/ERK signaling pathway crucial for cellular responses, the VRAF murine sarcoma viral oncogene homologue B1 (BRAF) kinase has emerged as a promising target for anticancer drug discovery due to oncogenic mutations that lead to pathway hyperactivation. Despite the discovery of several small-molecule BRAF kinase inhibitors targeting oncogenic mutants, their clinical utility has been limited by challenges such as off-target effects and suboptimal pharmacological properties. This study focuses on identifying miniprotein inhibitors for the oncogenic V600E mutant BRAF, leveraging their potential as versatile drug candidates. Using a structure-based de novo design approach based on binding affinity to V600E mutant BRAF and hydration energy, 39 candidate miniprotein inhibitors comprising three helices and 69 amino acids were generated from the substructure of the endogenous ligand protein (14-3-3). Through in vitro binding and kinase inhibition assays, two miniproteins (63 and 76) were discovered as novel inhibitors of V600E mutant BRAF with low-micromolar activity, with miniprotein 76 demonstrating a specific impediment to MEK1 phosphorylation in mammalian cells. These findings highlight miniprotein 76 as a potential lead compound for developing new cancer therapeutics, and the structural features contributing to its biochemical potency against V600E mutant BRAF are discussed in detail.

LoCoHD: a metric for comparing local environments of proteins

Protein folds and the local environments they create can be compared using a variety of differently designed measures, such as the root mean squared deviation, the global distance test, the template modeling score or the local distance difference test. Although these measures have proven to be useful for a variety of tasks, each fails to fully incorporate the valuable chemical information inherent to atoms and residues, and considers these only partially and indirectly. Here, we develop the highly flexible local composition Hellinger distance (LoCoHD) metric, which is based on the chemical composition of local residue environments. Using LoCoHD, we analyze the chemical heterogeneity of amino acid environments and identify valines having the most conserved-, and arginines having the most variable chemical environments. We use LoCoHD to investigate structural ensembles, to evaluate critical assessment of structure prediction (CASP) competitors, to compare the results with the local distance difference test (lDDT) scoring system, and to evaluate a molecular dynamics simulation. We show that LoCoHD measurements provide unique information about protein structures that is distinct from, for example, those derived using the alignment-based RMSD metric, or the similarly distance matrix-based but alignment-free lDDT metric.

In silico design of misfolding resistant proteins: the role of structural similarity of a competing conformational ensemble in the optimization of frustration

Most state-of-the-art in silico design methods fail due to misfolding of designed sequences to a conformation other than the target. Thus, a method to design misfolding resistant proteins will provide a better understanding of the misfolding phenomenon and will also increase the success rate of in silico design methods. In this work, we optimize the conformational ensemble to be selected for negative design purposes based on the similarity of the conformational ensemble to the target. Five ensembles with different degrees of similarity to the target are created and destabilized and the target is stabilized while designing sequences using mean field theory and Monte Carlo simulation methods. The results suggest that the degree of similarity of the non-native conformations to the target plays a prominent role in designing misfolding resistant protein sequences. The design procedures that destabilize the conformational ensemble with moderate similarity to the target have proven to be more promising. Incorporation of either highly similar or highly dissimilar conformations to the target conformation into the non-native ensemble to be destabilized may lead to sequences with a higher misfolding propensity. This will significantly reduce the conformational space to be considered in any protein design procedure. Interestingly, the results suggest that a sequence with higher frustration in the target structure does not necessarily lead to a misfold prone sequence. A successful design method may purposefully choose a frustrated sequence in the target conformation if that sequence is even more frustrated in the competing non-native conformations.

Regenerated Fiber's Ideal Target: Comparable to Natural Fiber

The toughness of silk naturally obtained from spiders and silkworms exceeds that of all other natural and man-made fibers. These insects transform aqueous protein feedstocks into mechanically specialized materials, which represents an engineering phenomenon that has developed over millions of years of natural evolution. Silkworms have become a new research hotspot due to the difficulties in collecting spider silk and other challenges. According to continuous research on the natural spinning process of the silkworm, it is possible to divide the main aspects of bionic spinning into two main segments: the solvent and behavior. This work focuses on the various methods currently used for the spinning of artificial silk fibers to replicate natural silk fibers, providing new insights based on changes in the fiber properties and production processes over time.

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.

AI in cellular engineering and reprogramming

During the last decade, artificial intelligence (AI) has increasingly been applied in biophysics and related fields, including cellular engineering and reprogramming, offering novel approaches to understand, manipulate, and control cellular function. The potential of AI lies in its ability to analyze complex datasets and generate predictive models. AI algorithms can process large amounts of data from single-cell genomics and multiomic technologies, allowing researchers to gain mechanistic insights into the control of cell identity and function. By integrating and interpreting these complex datasets, AI can help identify key molecular events and regulatory pathways involved in cellular reprogramming. This knowledge can inform the design of precision engineering strategies, such as the development of new transcription factor and signaling molecule cocktails, to manipulate cell identity and drive authentic cell fate across lineage boundaries. Furthermore, when used in combination with computational methods, AI can accelerate and improve the analysis and understanding of the intricate relationships between genes, proteins, and cellular processes. In this review article, we explore the current state of AI applications in biophysics with a specific focus on cellular engineering and reprogramming. Then, we showcase a couple of recent applications where we combined machine learning with experimental and computational techniques. Finally, we briefly discuss the challenges and prospects of AI in cellular engineering and reprogramming, emphasizing the potential of these technologies to revolutionize our ability to engineer cells for a variety of applications, from disease modeling and drug discovery to regenerative medicine and biomanufacturing.

The structural landscape of the immunoglobulin fold by large-scale de novo design

De novo designing immunoglobulin-like frameworks that allow for functional loop diversification shows great potential for crafting antibody-like scaffolds with fully customizable structures and functions. In this work, we combined de novo parametric design with deep-learning methods for protein structure prediction and design to explore the structural landscape of 7-stranded immunoglobulin domains. After screening folding of nearly 4 million designs, we have assembled a structurally diverse library of ~50,000 immunoglobulin domains with high-confidence AlphaFold2 predictions and structures diverging from naturally occurring ones. The designed dataset enabled us to identify structural requirements for the correct folding of immunoglobulin domains, shed light on β-sheet-β-sheet rotational preferences and how these are linked to functional properties. Our approach eliminates the need for preset loop conformations and opens the route to large-scale de novo design of immunoglobulin-like frameworks.

HighFold: accurately predicting structures of cyclic peptides and complexes with head-to-tail and disulfide bridge constraints

In recent years, cyclic peptides have emerged as a promising therapeutic modality due to their diverse biological activities. Understanding the structures of these cyclic peptides and their complexes is crucial for unlocking invaluable insights about protein target-cyclic peptide interaction, which can facilitate the development of novel-related drugs. However, conducting experimental observations is time-consuming and expensive. Computer-aided drug design methods are not practical enough in real-world applications. To tackles this challenge, we introduce HighFold, an AlphaFold-derived model in this study. By integrating specific details about the head-to-tail circle and disulfide bridge structures, the HighFold model can accurately predict the structures of cyclic peptides and their complexes. Our model demonstrates superior predictive performance compared to other existing approaches, representing a significant advancement in structure-activity research. The HighFold model is openly accessible at https://github.com/hongliangduan/HighFold.

Lipid vesicle-based molecular robots

A molecular robot, which is a system comprised of one or more molecular machines and computers, can execute sophisticated tasks in many fields that span from nanomedicine to green nanotechnology. The core parts of molecular robots are fairly consistent from system to system and always include (i) a body to encapsulate molecular machines, (ii) sensors to capture signals, (iii) computers to make decisions, and (iv) actuators to perform tasks. This review aims to provide an overview of approaches and considerations to develop molecular robots. We first introduce the basic technologies required for constructing the core parts of molecular robots, describe the recent progress towards achieving higher functionality, and subsequently discuss the current challenges and outlook. We also highlight the applications of molecular robots in sensing biomarkers, signal communications with living cells, and conversion of energy. Although molecular robots are still in their infancy, they will unquestionably initiate massive change in biomedical and environmental technology in the not too distant future.

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.

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.

Fusion proteins development strategies and their role as cancer therapeutic agents

Cancer continues to be leading cause of morbidity and mortality despite decades of research and advancement in chemotherapy. Most tumors can be reduced via standard oncology treatments, such as chemotherapy, radiotherapy, and surgical resection, and they frequently recur. Significant progress has been made since targeted cancer therapy inception in creation of medications that exhibit improved tumor-selective action. Particularly in preclinical and clinical investigations, fusion proteins have shown strong activity and improved treatment outcomes for a number of human cancers. Synergistically combining many proteins into one complex allows the creation of synthetic fusion proteins with enhanced characteristics or new capabilities. Signal transduction pathways are important for onset, development, and spread of cancer. As result, signaling molecules are desirable targets for cancer therapies, and significant effort has been made into developing fusion proteins that would act as inhibitors of these pathways. A wide range of biotechnological and medicinal applications are made possible by fusion of protein domains that improves bioactivities or creates new functional combinations. Such proteins may function as immune effectors cell recruiters to tumors or as decoy receptors for various ligands. In this review article, we have outlined the standard methods for creating fusion proteins and covered the applications of fusion proteins in treatment of cancer. This article also highlights the role of fusion proteins in targeting the signaling pathways involved in cancer for effective treatment.

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.

Computational Design of Peptide Assemblies

With the ongoing development of peptide self-assembling materials, there is growing interest in exploring novel functional peptide sequences. From short peptides to long polypeptides, as the functionality increases, the sequence space is also expanding exponentially. Consequently, attempting to explore all functional sequences comprehensively through experience and experiments alone has become impractical. By utilizing computational methods, especially artificial intelligence enhanced molecular dynamics (MD) simulation and de novo peptide design, there has been a significant expansion in the exploration of sequence space. Through these methods, a variety of supramolecular functional materials, including fibers, two-dimensional arrays, nanocages, etc., have been designed by meticulously controlling the inter- and intramolecular interactions. In this review, we first provide a brief overview of the current main computational methods and then focus on the computational design methods for various self-assembled peptide materials. Additionally, we introduce some representative protein self-assemblies to offer guidance for the design of self-assembling peptides.

Bacteriophages: a potential game changer in food processing industry

In the food industry, despite the widespread use of interventions such as preservatives and thermal and non-thermal processing technologies to improve food safety, incidences of foodborne disease continue to happen worldwide, prompting the search for alternative strategies. Bacteriophages, commonly known as phages, have emerged as a promising alternative for controlling pathogenic bacteria in food. This review emphasizes the potential applications of phages in biological sciences, food processing, and preservation, with a particular focus on their role as biocontrol agents for improving food quality and preservation. By shedding light on recent developments and future possibilities, this review highlights the significance of phages in the food industry. Additionally, it addresses crucial aspects such as regulatory status and safety concerns surrounding the use of bacteriophages. The inclusion of up-to-date literature further underscores the relevance of phage-based strategies in reducing foodborne pathogenic bacteria's presence in both food and the production environment. As we look ahead, new phage products are likely to be targeted against emerging foodborne pathogens. This will further advance the efficacy of approaches that are based on phages in maintaining the safety and security of food.

β-Strand-mediated Domain-swapping in the Absence of Hydrophobic Core Repacking

Domain swapping is a process wherein a portion of a protein is exchanged with its counterpart in another copy of the molecule, resulting in the formation of homo-oligomers with concomitant repacking of a hydrophobic core. Here, we report domain swapping triggered upon modifying a β-hairpin sequence within a single-layer β-sheet (SLB) of a model protein, OspA that did not involve the formation of a reorganized hydrophobic core. The replacement of two β-hairpin sequences with a Gly-Gly and shorteing of a β-hairpin resulted in a protein that formed two distinct crystal structures under similar conditions: one was monomeric, similar to the parental molecule, whereas the other was a domain-swapped dimer, mediated by an intermolecular β-sheet in the SLB portion. Based on the dimer interface structure, we replaced the Gly-Gly sequence with three-residue sequences that enable the formation of a consecutive intermolecular β-sheet, including the Cys-Thr-Cys sequence that formed a stable disulfide-linked dimer. These results provide new insights into protein folding, evolution, and the designability of protein structure.

The 100-protein NMR spectra dataset: A resource for biomolecular NMR data analysis

Multidimensional NMR spectra are the basis for studying proteins by NMR spectroscopy and crucial for the development and evaluation of methods for biomolecular NMR data analysis. Nevertheless, in contrast to derived data such as chemical shift assignments in the BMRB and protein structures in the PDB databases, this primary data is in general not publicly archived. To change this unsatisfactory situation, we present a standardized set of solution NMR data comprising 1329 2-4-dimensional NMR spectra and associated reference (chemical shift assignments, structures) and derived (peak lists, restraints for structure calculation, etc.) annotations. With the 100-protein NMR spectra dataset that was originally compiled for the development of the ARTINA deep learning-based spectra analysis method, 100 protein structures can be reproduced from their original experimental data. The 100-protein NMR spectra dataset is expected to help the development of computational methods for NMR spectroscopy, in particular machine learning approaches, and enable consistent and objective comparisons of these methods.

Establishment and characterization of a novel indirect ELISA method based on ASFV antigenic epitope-associated recombinant protein

African Swine Fever (ASF) is an acute and highly lethal disease in pigs caused by African Swine Fever Virus (ASFV). Viral proteins have been commonly used as antigenic targets for the development of ASF diagnostic methods. However, the prokaryotic expression of viral proteins has deficiencies such as instability, insolubility, and high cost in eukaryotic situations. This study screened and verified ASFV-encoded p72, p54, and p30 protein antigenic epitopes. Subsequently, a novel antigenic epitope-associated recombinant protein was designed based on an ideal structural protein and expressed in Escherichia coli (E. coli). Western blot analysis indicated that the recombinant protein could specifically react with the monoclonal antibody (mAb) of p72 and polyclonal antibodies of p54 and p30, respectively. Next, an ASF indirect ELISA (iELISA) method was established based on the recombinant protein, which has no specific reaction with sera of other important pig viral diseases. Meanwhile, it shows a sensitivity to detecting dilutions of ASF-positive reference serum up to 1:6400. The clinical sample detection results showed a high coincidence rate of 98 % with a commercial competition ELISA kit. In conclusion, we established a novel specific, and sensitive ASF serologic detection method that opens new avenues for ASF serodiagnostic method development.

Sculpting conducting nanopore size and shape through de novo protein design

Transmembrane β-barrels (TMBs) are widely used for single molecule DNA and RNA sequencing and have considerable potential for a broad range of sensing and sequencing applications. Current engineering approaches for nanopore sensors are limited to naturally occurring channels such as CsgG, which have evolved to carry out functions very different from sensing, and hence provide sub-optimal starting points. In contrast, de novo protein design can in principle create an unlimited number of new nanopores with any desired properties. Here we describe a general approach to the design of transmembrane β-barrel pores with different diameter and pore geometry. NMR and crystallographic characterization shows that the designs are stably folded with structures close to the design models. We report the first examples of de novo designed TMBs with 10, 12 and 14 stranded β-barrels. The designs have distinct conductances that correlate with their pore diameter, ranging from 110 pS (~0.5 nm pore diameter) to 430 pS (~1.1 nm pore diameter), and can be converted into sensitive small-molecule sensors with high signal to noise ratio. The capability to generate on demand β-barrel pores of defined geometry opens up fundamentally new opportunities for custom engineering of sequencing and sensing technologies.

Illuminating protein space with a programmable generative model

Three billion years of evolution has produced a tremendous diversity of protein molecules1, but the full potential of proteins is likely to be much greater. Accessing this potential has been challenging for both computation and experiments because the space of possible protein molecules is much larger than the space of those likely to have functions. Here we introduce Chroma, a generative model for proteins and protein complexes that can directly sample novel protein structures and sequences, and that can be conditioned to steer the generative process towards desired properties and functions. To enable this, we introduce a diffusion process that respects the conformational statistics of polymer ensembles, an efficient neural architecture for molecular systems that enables long-range reasoning with sub-quadratic scaling, layers for efficiently synthesizing three-dimensional structures of proteins from predicted inter-residue geometries and a general low-temperature sampling algorithm for diffusion models. Chroma achieves protein design as Bayesian inference under external constraints, which can involve symmetries, substructure, shape, semantics and even natural-language prompts. The experimental characterization of 310 proteins shows that sampling from Chroma results in proteins that are highly expressed, fold and have favourable biophysical properties. The crystal structures of two designed proteins exhibit atomistic agreement with Chroma samples (a backbone root-mean-square deviation of around 1.0 Å). With this unified approach to protein design, we hope to accelerate the programming of protein matter to benefit human health, materials science and synthetic biology.

Harnessing the power of artificial intelligence to advance cell therapy

Cell therapies are powerful technologies in which human cells are reprogrammed for therapeutic applications such as killing cancer cells or replacing defective cells. The technologies underlying cell therapies are increasing in effectiveness and complexity, making rational engineering of cell therapies more difficult. Creating the next generation of cell therapies will require improved experimental approaches and predictive models. Artificial intelligence (AI) and machine learning (ML) methods have revolutionized several fields in biology including genome annotation, protein structure prediction, and enzyme design. In this review, we discuss the potential of combining experimental library screens and AI to build predictive models for the development of modular cell therapy technologies. Advances in DNA synthesis and high-throughput screening techniques enable the construction and screening of libraries of modular cell therapy constructs. AI and ML models trained on this screening data can accelerate the development of cell therapies by generating predictive models, design rules, and improved designs.

De novo design of knotted tandem repeat proteins

De novo protein design methods can create proteins with folds not yet seen in nature. These methods largely focus on optimizing the compatibility between the designed sequence and the intended conformation, without explicit consideration of protein folding pathways. Deeply knotted proteins, whose topologies may introduce substantial barriers to folding, thus represent an interesting test case for protein design. Here we report our attempts to design proteins with trefoil (31) and pentafoil (51) knotted topologies. We extended previously described algorithms for tandem repeat protein design in order to construct deeply knotted backbones and matching designed repeat sequences (N = 3 repeats for the trefoil and N = 5 for the pentafoil). We confirmed the intended conformation for the trefoil design by X ray crystallography, and we report here on this protein's structure, stability, and folding behaviour. The pentafoil design misfolded into an asymmetric structure (despite a 5-fold symmetric sequence); two of the four repeat-repeat units matched the designed backbone while the other two diverged to form local contacts, leading to a trefoil rather than pentafoil knotted topology. Our results also provide insights into the folding of knotted proteins.

Time-resolved enantiomer-exchange probed by using the orbital angular momentum of X-ray light

Molecular chirality, a geometric property of utmost importance in biochemistry, is now being investigated in the time-domain. Ultrafast chiral techniques can probe the formation or disappearance of stereogenic centers in molecules. The element-sensitivity of X-rays adds the capability to probe chiral nuclear dynamics locally within the molecular system. However, the implementation of ultrafast techniques for measuring transient chirality remains a challenge because of the intrinsic weakness of chiral-sensitive signals based on circularly polarized light. We propose a novel approach for probing the enantiomeric dynamics by using the orbital angular momentum (OAM) of X-ray light, which can directly monitor the real-time chirality of molecules. Our simulations probe the oscillations in excited chiral formamide on different potential energy surfaces and demonstrate that using the X-ray OAM can increase the measured asymmetry ratio. Moreover, combining the OAM and SAM (spin angular momentum) provides stronger dichroic signals than linearly polarized light, and offers a powerful scheme for chiral discrimination.

Single-chain dimers from de novo immunoglobulins as robust scaffolds for multiple binding loops

Antibody derivatives have sought to recapitulate the antigen binding properties of antibodies, but with improved biophysical attributes convenient for therapeutic, diagnostic and research applications. However, their success has been limited by the naturally occurring structure of the immunoglobulin dimer displaying hypervariable binding loops, which is hard to modify by traditional engineering approaches. Here, we devise geometrical principles for de novo designing single-chain immunoglobulin dimers, as a tunable two-domain architecture that optimizes biophysical properties through more favorable dimer interfaces. Guided by these principles, we computationally designed protein scaffolds that were hyperstable, structurally accurate and robust for accommodating multiple functional loops, both individually and in combination, as confirmed through biochemical assays and X-ray crystallography. We showcase the modularity of this architecture by deep-learning-based diversification, opening up the possibility for tailoring the number, positioning, and relative orientation of ligand-binding loops targeting one or two distal epitopes. Our results provide a route to custom-design robust protein scaffolds for harboring multiple functional loops.

A probabilistic view of protein stability, conformational specificity, and design

Various approaches have used neural networks as probabilistic models for the design of protein sequences. These "inverse folding" models employ different objective functions, which come with trade-offs that have not been assessed in detail before. This study introduces probabilistic definitions of protein stability and conformational specificity and demonstrates the relationship between these chemical properties and the [Formula: see text] Boltzmann probability objective. This links the Boltzmann probability objective function to experimentally verifiable outcomes. We propose a novel sequence decoding algorithm, referred to as "BayesDesign", that leverages Bayes' Rule to maximize the [Formula: see text] objective instead of the [Formula: see text] objective common in inverse folding models. The efficacy of BayesDesign is evaluated in the context of two protein model systems, the NanoLuc enzyme and the WW structural motif. Both BayesDesign and the baseline ProteinMPNN algorithm increase the thermostability of NanoLuc and increase the conformational specificity of WW. The possible sources of error in the model are analyzed.

De novo design of highly selective miniprotein inhibitors of integrins αvβ6 and αvβ8

The RGD (Arg-Gly-Asp)-binding integrins αvβ6 and αvβ8 are clinically validated cancer and fibrosis targets of considerable therapeutic importance. Compounds that can discriminate between homologous αvβ6 and αvβ8 and other RGD integrins, stabilize specific conformational states, and have high thermal stability could have considerable therapeutic utility. Existing small molecule and antibody inhibitors do not have all these properties, and hence new approaches are needed. Here we describe a generalized method for computationally designing RGD-containing miniproteins selective for a single RGD integrin heterodimer and conformational state. We design hyperstable, selective αvβ6 and αvβ8 inhibitors that bind with picomolar affinity. CryoEM structures of the designed inhibitor-integrin complexes are very close to the computational design models, and show that the inhibitors stabilize specific conformational states of the αvβ6 and the αvβ8 integrins. In a lung fibrosis mouse model, the αvβ6 inhibitor potently reduced fibrotic burden and improved overall lung mechanics, demonstrating the therapeutic potential of de novo designed integrin binding proteins with high selectivity.

A brief review on the mechanisms and approaches of silk spinning-inspired biofabrication

Silk spinning, observed in spiders and insects, exhibits a remarkable biological source of inspiration for advanced polymer fabrications. Because of the systems design, silk spinning represents a holistic and circular approach to sustainable polymer fabrication, characterized by renewable resources, ambient and aqueous processing conditions, and fully recyclable "wastes." Also, silk spinning results in structures that are characterized by the combination of monolithic proteinaceous composition and mechanical strength, as well as demonstrate tunable degradation profiles and minimal immunogenicity, thus making it a viable alternative to most synthetic polymers for the development of advanced biomedical devices. However, the fundamental mechanisms of silk spinning remain incompletely understood, thus impeding the efforts to harness the advantageous properties of silk spinning. Here, we present a concise and timely review of several essential features of silk spinning, including the molecular designs of silk proteins and the solvent cues along the spinning apparatus. The solvent cues, including salt ions, pH, and water content, are suggested to direct the hierarchical assembly of silk proteins and thus play a central role in silk spinning. We also discuss several hypotheses on the roles of solvent cues to provide a relatively comprehensive analysis and to identify the current knowledge gap. We then review the state-of-the-art bioinspired fabrications with silk proteins, including fiber spinning and additive approaches/three-dimensional (3D) printing. An emphasis throughout the article is placed on the universal characteristics of silk spinning developed through millions of years of individual evolution pathways in spiders and silkworms. This review serves as a stepping stone for future research endeavors, facilitating the in vitro recapitulation of silk spinning and advancing the field of bioinspired polymer fabrication.

Design of stimulus-responsive two-state hinge proteins

In nature, proteins that switch between two conformations in response to environmental stimuli structurally transduce biochemical information in a manner analogous to how transistors control information flow in computing devices. Designing proteins with two distinct but fully structured conformations is a challenge for protein design as it requires sculpting an energy landscape with two distinct minima. Here we describe the design of "hinge" proteins that populate one designed state in the absence of ligand and a second designed state in the presence of ligand. X-ray crystallography, electron microscopy, double electron-electron resonance spectroscopy, and binding measurements demonstrate that despite the significant structural differences the two states are designed with atomic level accuracy and that the conformational and binding equilibria are closely coupled.

Exploration of novel αβ-protein folds through de novo design

A fundamental question in protein evolution is whether nature has exhaustively sampled nearly all possible protein folds throughout evolution, or whether a large fraction of the possible folds remains unexplored. To address this question, we defined a set of rules for β-sheet topology to predict novel αβ-folds and carried out a systematic de novo protein design exploration of the novel αβ-folds predicted by the rules. The designs for all eight of the predicted novel αβ-folds with a four-stranded β-sheet, including a knot-forming one, folded into structures close to the design models. Further, the rules predicted more than 10,000 novel αβ-folds with five- to eight-stranded β-sheets; this number far exceeds the number of αβ-folds observed in nature so far. This result suggests that a vast number of αβ-folds are possible, but have not emerged or have become extinct due to evolutionary bias.

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.

Large language models generate functional protein sequences across diverse families

Deep-learning language models have shown promise in various biotechnological applications, including protein design and engineering. Here we describe ProGen, a language model that can generate protein sequences with a predictable function across large protein families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. The model was trained on 280 million protein sequences from >19,000 families and is augmented with control tags specifying protein properties. ProGen can be further fine-tuned to curated sequences and tags to improve controllable generation performance of proteins from families with sufficient homologous samples. Artificial proteins fine-tuned to five distinct lysozyme families showed similar catalytic efficiencies as natural lysozymes, with sequence identity to natural proteins as low as 31.4%. ProGen is readily adapted to diverse protein families, as we demonstrate with chorismate mutase and malate dehydrogenase.

Transient water wires mediate selective proton transport in designed channel proteins

Selective proton transport through proteins is essential for forming and using proton gradients in cells. Protons are conducted along hydrogen-bonded 'wires' of water molecules and polar side chains, which, somewhat surprisingly, are often interrupted by dry apolar stretches in the conduction pathways, inferred from static protein structures. Here we hypothesize that protons are conducted through such dry spots by forming transient water wires, often highly correlated with the presence of the excess protons in the water wire. To test this hypothesis, we performed molecular dynamics simulations to design transmembrane channels with stable water pockets interspersed by apolar segments capable of forming flickering water wires. The minimalist designed channels conduct protons at rates similar to viral proton channels, and they are at least 106-fold more selective for H+ over Na+. These studies inform the mechanisms of biological proton conduction and the principles for engineering proton-conductive materials.

De novo design of highly selective miniprotein inhibitors of integrins αvβ6 and αvβ8

The RGD (Arg-Gly-Asp)-binding integrins αvβ6 and αvβ8 are clinically validated cancer and fibrosis targets of considerable therapeutic importance. Compounds that can discriminate between the two closely related integrin proteins and other RGD integrins, stabilize specific conformational states, and have sufficient stability enabling tissue restricted administration could have considerable therapeutic utility. Existing small molecules and antibody inhibitors do not have all of these properties, and hence there is a need for new approaches. Here we describe a method for computationally designing hyperstable RGD-containing miniproteins that are highly selective for a single RGD integrin heterodimer and conformational state, and use this strategy to design inhibitors of αvβ6 and αvβ8 with high selectivity. The αvβ6 and αvβ8 inhibitors have picomolar affinities for their targets, and >1000-fold selectivity over other RGD integrins. CryoEM structures are within 0.6-0.7Å root-mean-square deviation (RMSD) to the computational design models; the designed αvβ6 inhibitor and native ligand stabilize the open conformation in contrast to the therapeutic anti-αvβ6 antibody BG00011 that stabilizes the bent-closed conformation and caused on-target toxicity in patients with lung fibrosis, and the αvβ8 inhibitor maintains the constitutively fixed extended-closed αvβ8 conformation. In a mouse model of bleomycin-induced lung fibrosis, the αvβ6 inhibitor potently reduced fibrotic burden and improved overall lung mechanics when delivered via oropharyngeal administration mimicking inhalation, demonstrating the therapeutic potential of de novo designed integrin binding proteins with high selectivity.

Why we are made of proteins and nucleic acids: Structural biology views on extraterrestrial life

Is it a miracle that life exists on the Earth, or is it a common phenomenon in the universe? If extraterrestrial organisms exist, what are they like? To answer these questions, we must understand what kinds of molecules could evolve into life, or in other words, what properties are generally required to perform biological functions and store genetic information. This review summarizes recent findings on simple ancestral proteins, outlines the basic knowledge in textbooks, and discusses the generally required properties for biological molecules from structural biology viewpoints (e.g., restriction of shapes, and types of intra- and intermolecular interactions), leading to the conclusion that proteins and nucleic acids are at least one of the simplest (and perhaps very common) forms of catalytic and genetic biopolymers in the universe. This review article is an extended version of the Japanese article, On the Origin of Life: Coevolution between RNA and Peptide, published in SEIBUTSU BUTSURI Vol. 61, p. 232-235 (2021).

Computational modeling and prediction of deletion mutants

In-frame deletion mutations can result in disease. The impact of these mutations on protein structure and subsequent functional changes remain understudied, partially due to the lack of comprehensive datasets including a structural readout. In addition, the recent breakthrough in structure prediction through deep learning demands an update of computational deletion mutation prediction. In this study, we deleted individually every residue of a small α-helical sterile alpha motif domain and investigated the structural and thermodynamic changes using 2D NMR spectroscopy and differential scanning fluorimetry. Then, we tested computational protocols to model and classify observed deletion mutants. We show a method using AlphaFold2 followed by RosettaRelax performs the best overall. In addition, a metric containing pLDDT values and Rosetta ΔΔG is most reliable in classifying tolerated deletion mutations. We further test this method on other datasets and show they hold for proteins known to harbor disease-causing deletion mutations.

Cartilage-like protein hydrogels engineered via entanglement

Load-bearing tissues, such as muscle and cartilage, exhibit high elasticity, high toughness and fast recovery, but have different stiffness (with cartilage being significantly stiffer than muscle)^1-8. Muscle achieves its toughness through finely controlled forced domain unfolding-refolding in the muscle protein titin, whereas articular cartilage achieves its high stiffness and toughness through an entangled network comprising collagen and proteoglycans. Advancements in protein mechanics and engineering have made it possible to engineer titin-mimetic elastomeric proteins and soft protein biomaterials thereof to mimic the passive elasticity of muscle^9-11. However, it is more challenging to engineer highly stiff and tough protein biomaterials to mimic stiff tissues such as cartilage, or develop stiff synthetic matrices for cartilage stem and progenitor cell differentiation¹². Here we report the use of chain entanglements to significantly stiffen protein-based hydrogels without compromising their toughness. By introducing chain entanglements¹³ into the hydrogel network made of folded elastomeric proteins, we are able to engineer highly stiff and tough protein hydrogels, which seamlessly combine mutually incompatible mechanical properties, including high stiffness, high toughness, fast recovery and ultrahigh compressive strength, effectively converting soft protein biomaterials into stiff and tough materials exhibiting mechanical properties close to those of cartilage. Our study provides a general route towards engineering protein-based, stiff and tough biomaterials, which will find applications in biomedical engineering, such as osteochondral defect repair, and material sciences and engineering.

Sulfur-mediated chalcogen versus hydrogen bonds in proteins: a see-saw effect in the conformational space

Divalent sulfur (S) forms a chalcogen bond (Ch-bond) via its σ-holes and a hydrogen bond (H-bond) via its lone pairs. The relevance of these interactions and their interplay for protein structure and function is unclear. Based on the analyses of the crystal structures of small organic/organometallic molecules and proteins and their molecular electrostatic surface potential, we show that the reciprocity of the substituent-dependent strength of the σ-holes and lone pairs correlates with the formation of either Ch-bond or H-bond. In proteins, cystines preferentially form Ch-bonds, metal-chelated cysteines form H-bonds, while methionines form either of them with comparable frequencies. This has implications for the positioning of these residues and their role in protein structure and function. Computational analyses reveal that the S-mediated interactions stabilise protein secondary structures by mechanisms such as helix capping and protecting free β-sheet edges by negative design. The study highlights the importance of S-mediated Ch-bond and H-bond for understanding protein folding and function, the development of improved strategies for protein/peptide structure prediction and design and structure-based drug discovery.

Top-down design of protein architectures with reinforcement learning

As a result of evolutionary selection, the subunits of naturally occurring protein assemblies often fit together with substantial shape complementarity to generate architectures optimal for function in a manner not achievable by current design approaches. We describe a "top-down" reinforcement learning-based design approach that solves this problem using Monte Carlo tree search to sample protein conformers in the context of an overall architecture and specified functional constraints. Cryo-electron microscopy structures of the designed disk-shaped nanopores and ultracompact icosahedra are very close to the computational models. The icosohedra enable very-high-density display of immunogens and signaling molecules, which potentiates vaccine response and angiogenesis induction. Our approach enables the top-down design of complex protein nanomaterials with desired system properties and demonstrates the power of reinforcement learning in protein design.

Ab Initio Modelling of the Structure of ToxA-like and MAX Fungal Effector Proteins

Pathogenic fungal diseases in crops are mediated by the release of effector proteins that facilitate infection. Characterising the structure of these fungal effectors is vital to understanding their virulence mechanisms and interactions with their hosts, which is crucial in the breeding of plant cultivars for disease resistance. Several effectors have been identified and validated experimentally; however, their lack of sequence conservation often impedes the identification and prediction of their structure using sequence similarity approaches. Structural similarity has, nonetheless, been observed within fungal effector protein families, creating interest in validating the use of computational methods to predict their tertiary structure from their sequence. We used Rosetta ab initio modelling to predict the structures of members of the ToxA-like and MAX effector families for which experimental structures are known to validate this method. An optimised approach was then used to predict the structures of phenotypically validated effectors lacking known structures. Rosetta was found to successfully predict the structure of fungal effectors in the ToxA-like and MAX families, as well as phenotypically validated but structurally unconfirmed effector sequences. Interestingly, potential new effector structural families were identified on the basis of comparisons with structural homologues and the identification of associated protein domains.

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.

Alleviating the unwanted effects of oxidative stress on Aβ clearance: a review of related concepts and strategies for the development of computational modelling

Treatment for Alzheimer's disease (AD) can be more effective in the early stages. Although we do not completely understand the aetiology of the early stages of AD, potential pathological factors (amyloid beta [Aβ] and tau) and other co-factors have been identified as causes of AD, which may indicate some of the mechanism at work in the early stages of AD. Today, one of the primary techniques used to help delay or prevent AD in the early stages involves alleviating the unwanted effects of oxidative stress on Aβ clearance. 4-Hydroxynonenal (HNE), a product of lipid peroxidation caused by oxidative stress, plays a key role in the adduction of the degrading proteases. This HNE employs a mechanism which decreases catalytic activity. This process ultimately impairs Aβ clearance. The degradation of HNE-modified proteins helps to alleviate the unwanted effects of oxidative stress. Having a clear understanding of the mechanisms associated with the degradation of the HNE-modified proteins is essential for the development of strategies and for alleviating the unwanted effects of oxidative stress. The strategies which could be employed to decrease the effects of oxidative stress include enhancing antioxidant activity, as well as the use of nanozymes and/or specific inhibitors. One area which shows promise in reducing oxidative stress is protein design. However, more research is needed to improve the effectiveness and accuracy of this technique. This paper discusses the interplay of potential pathological factors and AD. In particular, it focuses on the effect of oxidative stress on the expression of the Aβ-degrading proteases through adduction of the degrading proteases caused by HNE. The paper also elucidates other strategies that can be used to alleviate the unwanted effects of oxidative stress on Aβ clearance. To improve the effectiveness and accuracy of protein design, we explain the application of quantum mechanical/molecular mechanical approach.

De novo protein fold design through sequence-independent fragment assembly simulations

De novo protein design generally consists of two steps, including structure and sequence design. Many protein design studies have focused on sequence design with scaffolds adapted from native structures in the PDB, which renders novel areas of protein structure and function space unexplored. We developed FoldDesign to create novel protein folds from specific secondary structure (SS) assignments through sequence-independent replica-exchange Monte Carlo (REMC) simulations. The method was tested on 354 non-redundant topologies, where FoldDesign consistently created stable structural folds, while recapitulating on average 87.7% of the SS elements. Meanwhile, the FoldDesign scaffolds had well-formed structures with buried residues and solvent-exposed areas closely matching their native counterparts. Despite the high fidelity to the input SS restraints and local structural characteristics of native proteins, a large portion of the designed scaffolds possessed global folds completely different from natural proteins in the PDB, highlighting the ability of FoldDesign to explore novel areas of protein fold space. Detailed data analyses revealed that the major contributions to the successful structure design lay in the optimal energy force field, which contains a balanced set of SS packing terms, and REMC simulations, which were coupled with multiple auxiliary movements to efficiently search the conformational space. Additionally, the ability to recognize and assemble uncommon super-SS geometries, rather than the unique arrangement of common SS motifs, was the key to generating novel folds. These results demonstrate a strong potential to explore both structural and functional spaces through computational design simulations that natural proteins have not reached through evolution.

Toward real-world automated antibody design with combinatorial Bayesian optimization

Antibodies are multimeric proteins capable of highly specific molecular recognition. The complementarity determining region 3 of the antibody variable heavy chain (CDRH3) often dominates antigen-binding specificity. Hence, it is a priority to design optimal antigen-specific CDRH3 to develop therapeutic antibodies. The combinatorial structure of CDRH3 sequences makes it impossible to query binding-affinity oracles exhaustively. Moreover, antibodies are expected to have high target specificity and developability. Here, we present AntBO, a combinatorial Bayesian optimization framework utilizing a CDRH3 trust region for an in silico design of antibodies with favorable developability scores. The in silico experiments on 159 antigens demonstrate that AntBO is a step toward practically viable in vitro antibody design. In under 200 calls to the oracle, AntBO suggests antibodies outperforming the best binding sequence from 6.9 million experimentally obtained CDRH3s. Additionally, AntBO finds very-high-affinity CDRH3 in only 38 protein designs while requiring no domain knowledge.