Advances in machine learning for directed evolution

Design and Evolution of an Enzyme for the Asymmetric Michael Addition of Cyclic Ketones to Nitroolefins by Enamine Catalysis

Consistent introduction of novel enzymes is required for developing efficient biocatalysts for challenging biotransformations. Absorbing catalytic modes from organocatalysis may be fruitful for designing new-to-nature enzymes with novel functions. Herein we report a newly designed artificial enzyme harboring a catalytic pyrrolidine residue that catalyzes the asymmetric Michael addition of cyclic ketones to nitroolefins through enamine activation with high efficiency. Diverse chiral γ-nitro cyclic ketones with two stereocenters were efficiently prepared with excellent stereoselectivity (up to 97 % e.e., >20 : 1 d.r.) and good yield (up to 86 %). This work provides an efficient biocatalytic strategy for cyclic ketone functionalization, and highlights the usefulness of artificial enzymes for extending biocatalysis to further non-natural reactions.

A combinatorially complete epistatic fitness landscape in an enzyme active site

Protein engineering often targets binding pockets or active sites which are enriched in epistasis-nonadditive interactions between amino acid substitutions-and where the combined effects of multiple single substitutions are difficult to predict. Few existing sequence-fitness datasets capture epistasis at large scale, especially for enzyme catalysis, limiting the development and assessment of model-guided enzyme engineering approaches. We present here a combinatorially complete, 160,000-variant fitness landscape across four residues in the active site of an enzyme. Assaying the native reaction of a thermostable β-subunit of tryptophan synthase (TrpB) in a nonnative environment yielded a landscape characterized by significant epistasis and many local optima. These effects prevent simulated directed evolution approaches from efficiently reaching the global optimum. There is nonetheless wide variability in the effectiveness of different directed evolution approaches, which together provide experimental benchmarks for computational and machine learning workflows. The most-fit TrpB variants contain a substitution that is nearly absent in natural TrpB sequences-a result that conservation-based predictions would not capture. Thus, although fitness prediction using evolutionary data can enrich in more-active variants, these approaches struggle to identify and differentiate among the most-active variants, even for this near-native function. Overall, this work presents a large-scale testing ground for model-guided enzyme engineering and suggests that efficient navigation of epistatic fitness landscapes can be improved by advances in both machine learning and physical modeling.

Deep-learning-based design of synthetic orthologs of SH3 signaling domains

Evolution-based deep generative models represent an exciting direction in understanding and designing proteins. An open question is whether such models can learn specialized functional constraints that control fitness in specific biological contexts. Here, we examine the ability of generative models to produce synthetic versions of Src-homology 3 (SH3) domains that mediate signaling in the Sho1 osmotic stress response pathway of yeast. We show that a variational autoencoder (VAE) model produces artificial sequences that experimentally recapitulate the function of natural SH3 domains. More generally, the model organizes all fungal SH3 domains such that locality in the model latent space (but not simply locality in sequence space) enriches the design of synthetic orthologs and exposes non-obvious amino acid constraints distributed near and far from the SH3 ligand-binding site. The ability of generative models to design ortholog-like functions in vivo opens new avenues for engineering protein function in specific cellular contexts and environments.

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.

Engineering the next-generation synthetic cell factory driven by protein engineering

Synthetic cell factory offers substantial advantages in economically efficient production of biofuels, chemicals, and pharmaceutical compounds. However, to create a high-performance synthetic cell factory, precise regulation of cellular material and energy flux is essential. In this context, protein components including enzymes, transcription factor-based biosensors and transporters play pivotal roles. Protein engineering aims to create novel protein variants with desired properties by modifying or designing protein sequences. This review focuses on summarizing the latest advancements of protein engineering in optimizing various aspects of synthetic cell factory, including: enhancing enzyme activity to eliminate production bottlenecks, altering enzyme selectivity to steer metabolic pathways towards desired products, modifying enzyme promiscuity to explore innovative routes, and improving the efficiency of transporters. Furthermore, the utilization of protein engineering to modify protein-based biosensors accelerates evolutionary process and optimizes the regulation of metabolic pathways. The remaining challenges and future opportunities in this field are also discussed.

Peptide-based drug discovery through artificial intelligence: towards an autonomous design of therapeutic peptides

With their diverse biological activities, peptides are promising candidates for therapeutic applications, showing antimicrobial, antitumour and hormonal signalling capabilities. Despite their advantages, therapeutic peptides face challenges such as short half-life, limited oral bioavailability and susceptibility to plasma degradation. The rise of computational tools and artificial intelligence (AI) in peptide research has spurred the development of advanced methodologies and databases that are pivotal in the exploration of these complex macromolecules. This perspective delves into integrating AI in peptide development, encompassing classifier methods, predictive systems and the avant-garde design facilitated by deep-generative models like generative adversarial networks and variational autoencoders. There are still challenges, such as the need for processing optimization and careful validation of predictive models. This work outlines traditional strategies for machine learning model construction and training techniques and proposes a comprehensive AI-assisted peptide design and validation pipeline. The evolving landscape of peptide design using AI is emphasized, showcasing the practicality of these methods in expediting the development and discovery of novel peptides within the context of peptide-based drug discovery.

Protein Engineering with Lightweight Graph Denoising Neural Networks

Protein engineering faces challenges in finding optimal mutants from a massive pool of candidate mutants. In this study, we introduce a deep-learning-based data-efficient fitness prediction tool to steer protein engineering. Our methodology establishes a lightweight graph neural network scheme for protein structures, which efficiently analyzes the microenvironment of amino acids in wild-type proteins and reconstructs the distribution of the amino acid sequences that are more likely to pass natural selection. This distribution serves as a general guidance for scoring proteins toward arbitrary properties on any order of mutations. Our proposed solution undergoes extensive wet-lab experimental validation spanning diverse physicochemical properties of various proteins, including fluorescence intensity, antigen-antibody affinity, thermostability, and DNA cleavage activity. More than 40% of ProtLGN-designed single-site mutants outperform their wild-type counterparts across all studied proteins and targeted properties. More importantly, our model can bypass the negative epistatic effect to combine single mutation sites and form deep mutants with up to seven mutation sites in a single round, whose physicochemical properties are significantly improved. This observation provides compelling evidence of the structure-based model's potential to guide deep mutations in protein engineering. Overall, our approach emerges as a versatile tool for protein engineering, benefiting both the computational and bioengineering communities.

A systematic analysis of regression models for protein engineering

To optimize proteins for particular traits holds great promise for industrial and pharmaceutical purposes. Machine Learning is increasingly applied in this field to predict properties of proteins, thereby guiding the experimental optimization process. A natural question is: How much progress are we making with such predictions, and how important is the choice of regressor and representation? In this paper, we demonstrate that different assessment criteria for regressor performance can lead to dramatically different conclusions, depending on the choice of metric, and how one defines generalization. We highlight the fundamental issues of sample bias in typical regression scenarios and how this can lead to misleading conclusions about regressor performance. Finally, we make the case for the importance of calibrated uncertainty in this domain.

Importance of Inter-residue Contacts for Understanding Protein Folding and Unfolding Rates, Remote Homology, and Drug Design

Inter-residue interactions in protein structures provide valuable insights into protein folding and stability. Understanding these interactions can be helpful in many crucial applications, including rational design of therapeutic small molecules and biologics, locating functional protein sites, and predicting protein-protein and protein-ligand interactions. The process of developing machine learning models incorporating inter-residue interactions has been improved recently. This review highlights the theoretical models incorporating inter-residue interactions in predicting folding and unfolding rates of proteins. Utilizing contact maps to depict inter-residue interactions aids researchers in developing computer models for detecting remote homologs and interface residues within protein-protein complexes which, in turn, enhances our knowledge of the relationship between sequence and structure of proteins. Further, the application of contact maps derived from inter-residue interactions is highlighted in the field of drug discovery. Overall, this review presents an extensive assessment of the significant models that use inter-residue interactions to investigate folding rates, unfolding rates, remote homology, and drug development, providing potential future advancements in constructing efficient computational models in structural biology.

Effective engineering of a ketoreductase for the biocatalytic synthesis of an ipatasertib precursor

Semi-rational enzyme engineering is a powerful method to develop industrial biocatalysts. Profiting from advances in molecular biology and bioinformatics, semi-rational approaches can effectively accelerate enzyme engineering campaigns. Here, we present the optimization of a ketoreductase from Sporidiobolus salmonicolor for the chemo-enzymatic synthesis of ipatasertib, a potent protein kinase B inhibitor. Harnessing the power of mutational scanning and structure-guided rational design, we created a 10-amino acid substituted variant exhibiting a 64-fold higher apparent kcat and improved robustness under process conditions compared to the wild-type enzyme. In addition, the benefit of algorithm-aided enzyme engineering was studied to derive correlations in protein sequence-function data, and it was found that the applied Gaussian processes allowed us to reduce enzyme library size. The final scalable and high performing biocatalytic process yielded the alcohol intermediate with ≥ 98% conversion and a diastereomeric excess of 99.7% (R,R-trans) from 100 g L-1 ketone after 30 h. Modelling and kinetic studies shed light on the mechanistic factors governing the improved reaction outcome, with mutations T134V, A238K, M242W and Q245S exerting the most beneficial effect on reduction activity towards the target ketone.

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

Enzymes can be engineered at the level of their amino acid sequences to optimize key properties such as expression, stability, substrate range, and catalytic efficiency-or even to unlock new catalytic activities not found in nature. Because the search space of possible proteins is vast, enzyme engineering usually involves discovering an enzyme starting point that has some level of the desired activity followed by directed evolution to improve its "fitness" for a desired application. Recently, machine learning (ML) has emerged as a powerful tool to complement this empirical process. ML models can contribute to (1) starting point discovery by functional annotation of known protein sequences or generating novel protein sequences with desired functions and (2) navigating protein fitness landscapes for fitness optimization by learning mappings between protein sequences and their associated fitness values. In this Outlook, we explain how ML complements enzyme engineering and discuss its future potential to unlock improved engineering outcomes.

Perspectives on Computational Enzyme Modeling: From Mechanisms to Design and Drug Development

Understanding enzyme mechanisms is essential for unraveling the complex molecular machinery of life. In this review, we survey the field of computational enzymology, highlighting key principles governing enzyme mechanisms and discussing ongoing challenges and promising advances. Over the years, computer simulations have become indispensable in the study of enzyme mechanisms, with the integration of experimental and computational exploration now established as a holistic approach to gain deep insights into enzymatic catalysis. Numerous studies have demonstrated the power of computer simulations in characterizing reaction pathways, transition states, substrate selectivity, product distribution, and dynamic conformational changes for various enzymes. Nevertheless, significant challenges remain in investigating the mechanisms of complex multistep reactions, large-scale conformational changes, and allosteric regulation. Beyond mechanistic studies, computational enzyme modeling has emerged as an essential tool for computer-aided enzyme design and the rational discovery of covalent drugs for targeted therapies. Overall, enzyme design/engineering and covalent drug development can greatly benefit from our understanding of the detailed mechanisms of enzymes, such as protein dynamics, entropy contributions, and allostery, as revealed by computational studies. Such a convergence of different research approaches is expected to continue, creating synergies in enzyme research. This review, by outlining the ever-expanding field of enzyme research, aims to provide guidance for future research directions and facilitate new developments in this important and evolving field.

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.

Improving plastic degrading enzymes via directed evolution

Plastic degrading enzymes have immense potential for use in industrial applications. Protein engineering efforts over the last decade have resulted in considerable enhancement of many properties of these enzymes. Directed evolution, a protein engineering approach that mimics the natural process of evolution in a laboratory, has been particularly useful in overcoming some of the challenges of structure-based protein engineering. For example, directed evolution has been used to improve the catalytic activity and thermostability of polyethylene terephthalate (PET)-degrading enzymes, although its use for the improvement of other desirable properties, such as solvent tolerance, has been less studied. In this review, we aim to identify some of the knowledge gaps and current challenges, and highlight recent studies related to the directed evolution of plastic-degrading enzymes.

Machine Learning-Supported Enzyme Engineering toward Improved CO2-Fixation of Glycolyl-CoA Carboxylase

Glycolyl-CoA carboxylase (GCC) is a new-to-nature enzyme that catalyzes the key reaction in the tartronyl-CoA (TaCo) pathway, a synthetic photorespiration bypass that was recently designed to improve photosynthetic CO2 fixation. GCC was created from propionyl-CoA carboxylase (PCC) through five mutations. However, despite reaching activities of naturally evolved biotin-dependent carboxylases, the quintuple substitution variant GCC M5 still lags behind 4-fold in catalytic efficiency compared to its template PCC and suffers from futile ATP hydrolysis during CO2 fixation. To further improve upon GCC M5, we developed a machine learning-supported workflow that reduces screening efforts for identifying improved enzymes. Using this workflow, we present two novel GCC variants with 2-fold increased carboxylation rate and 60% reduced energy demand, respectively, which are able to address kinetic and thermodynamic limitations of the TaCo pathway. Our work highlights the potential of combining machine learning and directed evolution strategies to reduce screening efforts in enzyme engineering.

A comprehensive framework for advanced protein classification and function prediction using synergistic approaches: Integrating bispectral analysis, machine learning, and deep learning

Proteins are fundamental components of diverse cellular systems and play crucial roles in a variety of disease processes. Consequently, it is crucial to comprehend their structure, function, and intricate interconnections. Classifying proteins into families or groups with comparable structural and functional characteristics is a crucial aspect of this comprehension. This classification is crucial for evolutionary research, predicting protein function, and identifying potential therapeutic targets. Sequence alignment and structure-based alignment are frequently ineffective techniques for identifying protein families.This study addresses the need for a more efficient and accurate technique for feature extraction and protein classification. The research proposes a novel method that integrates bispectrum characteristics, deep learning techniques, and machine learning algorithms to overcome the limitations of conventional methods. The proposed method uses numbers to represent protein sequences, utilizes bispectrum analysis, uses different topologies for convolutional neural networks to pull out features, and chooses robust features to classify protein families. The goal is to outperform existing methods for identifying protein families, thereby enhancing classification metrics. The materials consist of numerous protein datasets, whereas the methods incorporate bispectrum characteristics and deep learning strategies. The results of this study demonstrate that the proposed method for identifying protein families is superior to conventional approaches. Significantly enhanced quality metrics demonstrated the efficacy of the combined bispectrum and deep learning approaches. These findings have the potential to advance the field of protein biology and facilitate pharmaceutical innovation. In conclusion, this study presents a novel method that employs bispectrum characteristics and deep learning techniques to improve the precision and efficiency of protein family identification. The demonstrated advancements in classification metrics demonstrate this method's applicability to numerous scientific disciplines. This furthers our understanding of protein function and its implications for disease and treatment.

Artificial intelligence-driven systems engineering for next-generation plant-derived biopharmaceuticals

Recombinant biopharmaceuticals including antigens, antibodies, hormones, cytokines, single-chain variable fragments, and peptides have been used as vaccines, diagnostics and therapeutics. Plant molecular pharming is a robust platform that uses plants as an expression system to produce simple and complex recombinant biopharmaceuticals on a large scale. Plant system has several advantages over other host systems such as humanized expression, glycosylation, scalability, reduced risk of human or animal pathogenic contaminants, rapid and cost-effective production. Despite many advantages, the expression of recombinant proteins in plant system is hindered by some factors such as non-human post-translational modifications, protein misfolding, conformation changes and instability. Artificial intelligence (AI) plays a vital role in various fields of biotechnology and in the aspect of plant molecular pharming, a significant increase in yield and stability can be achieved with the intervention of AI-based multi-approach to overcome the hindrance factors. Current limitations of plant-based recombinant biopharmaceutical production can be circumvented with the aid of synthetic biology tools and AI algorithms in plant-based glycan engineering for protein folding, stability, viability, catalytic activity and organelle targeting. The AI models, including but not limited to, neural network, support vector machines, linear regression, Gaussian process and regressor ensemble, work by predicting the training and experimental data sets to design and validate the protein structures thereby optimizing properties such as thermostability, catalytic activity, antibody affinity, and protein folding. This review focuses on, integrating systems engineering approaches and AI-based machine learning and deep learning algorithms in protein engineering and host engineering to augment protein production in plant systems to meet the ever-expanding therapeutics market.

A new age in protein design empowered by deep learning

The rapid progress in the field of deep learning has had a significant impact on protein design. Deep learning methods have recently produced a breakthrough in protein structure prediction, leading to the availability of high-quality models for millions of proteins. Along with novel architectures for generative modeling and sequence analysis, they have revolutionized the protein design field in the past few years remarkably by improving the accuracy and ability to identify novel protein sequences and structures. Deep neural networks can now learn and extract the fundamental features of protein structures, predict how they interact with other biomolecules, and have the potential to create new effective drugs for treating disease. As their applicability in protein design is rapidly growing, we review the recent developments and technology in deep learning methods and provide examples of their performance to generate novel functional proteins.

Engineering of Therapeutic and Detoxifying Enzymes

Therapeutic enzymes present excellent opportunities for the treatment of human disease, modulation of metabolic pathways and system detoxification. However, current use of enzyme therapy in the clinic is limited as naturally occurring enzymes are seldom optimal for such applications and require substantial improvement by protein engineering. Engineering strategies such as design and directed evolution that have been successfully implemented for industrial biocatalysis can significantly advance the field of therapeutic enzymes, leading to biocatalysts with new-to-nature therapeutic activities, high selectivity, and suitability for medical applications. This minireview highlights case studies of how state-of-the-art and emerging methods in protein engineering are explored for the generation of therapeutic enzymes and discusses gaps and future opportunities in the field of enzyme therapy.

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. Our findings highlight how each architecture's inductive biases prime them to learn different aspects of the protein fitness landscape.

Exploring the Protein Sequence Space with Global Generative Models

Recent advancements in specialized large-scale architectures for training images and language have profoundly impacted the field of computer vision and natural language processing (NLP). Language models, such as the recent ChatGPT and GPT-4, have demonstrated exceptional capabilities in processing, translating, and generating human language. These breakthroughs have also been reflected in protein research, leading to the rapid development of numerous new methods in a short time, with unprecedented performance. Several of these models have been developed with the goal of generating sequences in novel regions of the protein space. In this work, we provide an overview of the use of protein generative models, reviewing (1) language models for the design of novel artificial proteins, (2) works that use non-transformer architectures, and (3) applications in directed evolution approaches.

Statistical Analysis and Tokenization of Epitopes to Construct Artificial Neoepitope Libraries

Epitopes are specific regions on an antigen's surface that the immune system recognizes. Epitopes are usually protein regions on foreign immune-stimulating entities such as viruses and bacteria, and in some cases, endogenous proteins may act as antigens. Identifying epitopes is crucial for accelerating the development of vaccines and immunotherapies. However, mapping epitopes in pathogen proteomes is challenging using conventional methods. Screening artificial neoepitope libraries against antibodies can overcome this issue. Here, we applied conventional sequence analysis and methods inspired in natural language processing to reveal specific sequence patterns in the linear epitopes deposited in the Immune Epitope Database (www.iedb.org) that can serve as building blocks for the design of universal epitope libraries. Our results reveal that amino acid frequency in annotated linear epitopes differs from that in the human proteome. Aromatic residues are overrepresented, while the presence of cysteines is practically null in epitopes. Byte pair encoding tokenization shows high frequencies of tryptophan in tokens of 5, 6, and 7 amino acids, corroborating the findings of the conventional sequence analysis. These results can be applied to reduce the diversity of linear epitope libraries by orders of magnitude.

Fishing for Catalysis: Experimental Approaches to Narrowing Search Space in Directed Evolution of Enzymes

Directed evolution has transformed protein engineering offering a path to rapid improvement of protein properties. Yet, in practice it is limited by the hyper-astronomic protein sequence search space, and approaches to identify mutagenic hot spots, i.e., locations where mutations are most likely to have a productive impact, are needed. In this perspective, we categorize and discuss recent progress in the experimental approaches (broadly defined as structural, bioinformatic, and dynamic) to hot spot identification. Recent successes in harnessing protein dynamics and machine learning approaches provide new opportunities for the field and will undoubtedly help directed evolution reach its full potential.

Artificial intelligence-aided protein engineering: from topological data analysis to deep protein language models

Protein engineering is an emerging field in biotechnology that has the potential to revolutionize various areas, such as antibody design, drug discovery, food security, ecology, and more. However, the mutational space involved is too vast to be handled through experimental means alone. Leveraging accumulative protein databases, machine learning (ML) models, particularly those based on natural language processing (NLP), have considerably expedited protein engineering. Moreover, advances in topological data analysis (TDA) and artificial intelligence-based protein structure prediction, such as AlphaFold2, have made more powerful structure-based ML-assisted protein engineering strategies possible. This review aims to offer a comprehensive, systematic, and indispensable set of methodological components, including TDA and NLP, for protein engineering and to facilitate their future development.

A Customized Bayesian Algorithm to Optimize Enzyme-Catalyzed Reactions

Design of experiments (DoE) plays an important role in optimizing the catalytic performance of chemical reactions. The most commonly used DoE relies on the response surface methodology (RSM) to model the variable space of experimental conditions with the fewest number of experiments. However, the RSM leads to an exponential increase in the number of required experiments as the number of variables increases. Herein we describe a Bayesian optimization algorithm (BOA) to optimize the continuous parameters (e.g., temperature, reaction time, reactant and enzyme concentrations, etc.) of enzyme-catalyzed reactions with the aim of maximizing performance. Compared to existing Bayesian optimization methods, we propose an improved algorithm that leads to better results under limited resources and time for experiments. To validate the versatility of the BOA, we benchmarked its performance with biocatalytic C-C bond formation and amination for the optimization of the turnover number. Gratifyingly, up to 80% improvement compared to RSM and up to 360% improvement vs previous Bayesian optimization algorithms were obtained. Importantly, this strategy enabled simultaneous optimization of both the enzyme's activity and selectivity for cross-benzoin condensation.

DeCOIL: Optimization of Degenerate Codon Libraries for Machine Learning-Assisted Protein Engineering

With advances in machine learning (ML)-assisted protein engineering, models based on data, biophysics, and natural evolution are being used to propose informed libraries of protein variants to explore. Synthesizing these libraries for experimental screens is a major bottleneck, as the cost of obtaining large numbers of exact gene sequences is often prohibitive. Degenerate codon (DC) libraries are a cost-effective alternative for generating combinatorial mutagenesis libraries where mutations are targeted to a handful of amino acid sites. However, existing computational methods to optimize DC libraries to include desired protein variants are not well suited to design libraries for ML-assisted protein engineering. To address these drawbacks, we present DEgenerate Codon Optimization for Informed Libraries (DeCOIL), a generalized method that directly optimizes DC libraries to be useful for protein engineering: to sample protein variants that are likely to have both high fitness and high diversity in the sequence search space. Using computational simulations and wet-lab experiments, we demonstrate that DeCOIL is effective across two specific case studies, with the potential to be applied to many other use cases. DeCOIL offers several advantages over existing methods, as it is direct, easy to use, generalizable, and scalable. With accompanying software (https://github.com/jsunn-y/DeCOIL), DeCOIL can be readily implemented to generate desired informed libraries.

In vitro continuous protein evolution empowered by machine learning and automation

Directed evolution has become one of the most successful and powerful tools for protein engineering. However, the efforts required for designing, constructing, and screening a large library of variants can be laborious, time-consuming, and costly. With the recent advent of machine learning (ML) in the directed evolution of proteins, researchers can now evaluate variants in silico and guide a more efficient directed evolution campaign. Furthermore, recent advancements in laboratory automation have enabled the rapid execution of long, complex experiments for high-throughput data acquisition in both industrial and academic settings, thus providing the means to collect a large quantity of data required to develop ML models for protein engineering. In this perspective, we propose a closed-loop in vitro continuous protein evolution framework that leverages the best of both worlds, ML and automation, and provide a brief overview of the recent developments in the field.

Protein engineering via sequence-performance mapping

Discovery and evolution of new and improved proteins has empowered molecular therapeutics, diagnostics, and industrial biotechnology. Discovery and evolution both require efficient screens and effective libraries, although they differ in their challenges because of the absence or presence, respectively, of an initial protein variant with the desired function. A host of high-throughput technologies-experimental and computational-enable efficient screens to identify performant protein variants. In partnership, an informed search of sequence space is needed to overcome the immensity, sparsity, and complexity of the sequence-performance landscape. Early in the historical trajectory of protein engineering, these elements aligned with distinct approaches to identify the most performant sequence: selection from large, randomized combinatorial libraries versus rational computational design. Substantial advances have now emerged from the synergy of these perspectives. Rational design of combinatorial libraries aids the experimental search of sequence space, and high-throughput, high-integrity experimental data inform computational design. At the core of the collaborative interface, efficient protein characterization (rather than mere selection of optimal variants) maps sequence-performance landscapes. Such quantitative maps elucidate the complex relationships between protein sequence and performance-e.g., binding, catalytic efficiency, biological activity, and developability-thereby advancing fundamental protein science and facilitating protein discovery and evolution.

Using Data Science for Mechanistic Insights and Selectivity Predictions in a Non-Natural Biocatalytic Reaction

The study of non-natural biocatalytic transformations relies heavily on empirical methods, such as directed evolution, for identifying improved variants. Although exceptionally effective, this approach provides limited insight into the molecular mechanisms behind the transformations and necessitates multiple protein engineering campaigns for new reactants. To address this limitation, we disclose a strategy to explore the biocatalytic reaction space and garner insight into the molecular mechanisms driving enzymatic transformations. Specifically, we explored the selectivity of an "ene"-reductase, GluER-T36A, to create a data-driven toolset that explores reaction space and rationalizes the observed and predicted selectivities of substrate/mutant combinations. The resultant statistical models related structural features of the enzyme and substrate to selectivity and were used to effectively predict selectivity in reactions with out-of-sample substrates and mutants. Our approach provided a deeper understanding of enantioinduction by GluER-T36A and holds the potential to enhance the virtual screening of enzyme mutants.

Cross-protein transfer learning substantially improves disease variant prediction

BACKGROUND

Genetic variation in the human genome is a major determinant of individual disease risk, but the vast majority of missense variants have unknown etiological effects. Here, we present a robust learning framework for leveraging saturation mutagenesis experiments to construct accurate computational predictors of proteome-wide missense variant pathogenicity.

RESULTS

We train cross-protein transfer (CPT) models using deep mutational scanning (DMS) data from only five proteins and achieve state-of-the-art performance on clinical variant interpretation for unseen proteins across the human proteome. We also improve predictive accuracy on DMS data from held-out proteins. High sensitivity is crucial for clinical applications and our model CPT-1 particularly excels in this regime. For instance, at 95% sensitivity of detecting human disease variants annotated in ClinVar, CPT-1 improves specificity to 68%, from 27% for ESM-1v and 55% for EVE. Furthermore, for genes not used to train REVEL, a supervised method widely used by clinicians, we show that CPT-1 compares favorably with REVEL. Our framework combines predictive features derived from general protein sequence models, vertebrate sequence alignments, and AlphaFold structures, and it is adaptable to the future inclusion of other sources of information. We find that vertebrate alignments, albeit rather shallow with only 100 genomes, provide a strong signal for variant pathogenicity prediction that is complementary to recent deep learning-based models trained on massive amounts of protein sequence data. We release predictions for all possible missense variants in 90% of human genes.

CONCLUSIONS

Our results demonstrate the utility of mutational scanning data for learning properties of variants that transfer to unseen proteins.

The Surprising Ease of Finding Optimal Solutions for Controlling Nonlinear Phenomena in Quantum and Classical Complex Systems

This Perspective addresses the often observed surprising ease of achieving optimal control of nonlinear phenomena in quantum and classical complex systems. The circumstances involved are wide-ranging, with scenarios including manipulation of atomic scale processes, maximization of chemical and material properties or synthesis yields, Nature's optimization of species' populations by natural selection, and directed evolution. Natural evolution will mainly be discussed in terms of laboratory experiments with microorganisms, and the field is also distinct from the other domains where a scientist specifies the goal(s) and oversees the control process. We use the word "control" in reference to all of the available variables, regardless of the circumstance. The empirical observations on the ease of achieving at least good, if not excellent, control in diverse domains of science raise the question of why this occurs despite the generally inherent complexity of the systems in each scenario. The key to addressing the question lies in examining the associated control landscape, which is defined as the optimization objective as a function of the control variables that can be as diverse as the phenomena under consideration. Controls may range from laser pulses, chemical reagents, chemical processing conditions, out to nucleic acids in the genome and more. This Perspective presents a conjecture, based on present findings, that the systematics of readily finding good outcomes from controlled phenomena may be unified through consideration of control landscapes with the same common set of three underlying assumptions─the existence of an optimal solution, the ability for local movement on the landscape, and the availability of sufficient control resources─whose validity needs assessment in each scenario. In practice, many cases permit using myopic gradient-like algorithms while other circumstances utilize algorithms having some elements of stochasticity or introduced noise, depending on whether the landscape is locally smooth or rough. The overarching observation is that only relatively short searches are required despite the common high dimensionality of the available controls in typical scenarios.

Ultrahigh-Throughput Enzyme Engineering and Discovery in In Vitro Compartments

Novel and improved biocatalysts are increasingly sourced from libraries via experimental screening. The success of such campaigns is crucially dependent on the number of candidates tested. Water-in-oil emulsion droplets can replace the classical test tube, to provide in vitro compartments as an alternative screening format, containing genotype and phenotype and enabling a readout of function. The scale-down to micrometer droplet diameters and picoliter volumes brings about a >10⁷-fold volume reduction compared to 96-well-plate screening. Droplets made in automated microfluidic devices can be integrated into modular workflows to set up multistep screening protocols involving various detection modes to sort >10⁷ variants a day with kHz frequencies. The repertoire of assays available for droplet screening covers all seven enzyme commission (EC) number classes, setting the stage for widespread use of droplet microfluidics in everyday biochemical experiments. We review the practicalities of adapting droplet screening for enzyme discovery and for detailed kinetic characterization. These new ways of working will not just accelerate discovery experiments currently limited by screening capacity but profoundly change the paradigms we can probe. By interfacing the results of ultrahigh-throughput droplet screening with next-generation sequencing and deep learning, strategies for directed evolution can be implemented, examined, and evaluated.

Persistent spectral theory-guided protein engineering

While protein engineering, which iteratively optimizes protein fitness by screening the gigantic mutational space, is constrained by experimental capacity, various machine learning models have substantially expedited protein engineering. Three-dimensional protein structures promise further advantages, but their intricate geometric complexity hinders their applications in deep mutational screening. Persistent homology, an established algebraic topology tool for protein structural complexity reduction, fails to capture the homotopic shape evolution during the filtration of a given data. This work introduces a Topology-offered protein Fitness (TopFit) framework to complement protein sequence and structure embeddings. Equipped with an ensemble regression strategy, TopFit integrates the persistent spectral theory, a new topological Laplacian, and two auxiliary sequence embeddings to capture mutation-induced topological invariant, shape evolution, and sequence disparity in the protein fitness landscape. The performance of TopFit is assessed by 34 benchmark datasets with 128,634 variants, involving a vast variety of protein structure acquisition modalities and training set size variations.

Data-driven enzyme engineering to identify function-enhancing enzymes

Identifying function-enhancing enzyme variants is a 'holy grail' challenge in protein science because it will allow researchers to expand the biocatalytic toolbox for late-stage functionalization of drug-like molecules, environmental degradation of plastics and other pollutants, and medical treatment of food allergies. Data-driven strategies, including statistical modeling, machine learning, and deep learning, have largely advanced the understanding of the sequence-structure-function relationships for enzymes. They have also enhanced the capability of predicting and designing new enzymes and enzyme variants for catalyzing the transformation of new-to-nature reactions. Here, we reviewed the recent progresses of data-driven models that were applied in identifying efficiency-enhancing mutants for catalytic reactions. We also discussed existing challenges and obstacles faced by the community. Although the review is by no means comprehensive, we hope that the discussion can inform the readers about the state-of-the-art in data-driven enzyme engineering, inspiring more joint experimental-computational efforts to develop and apply data-driven modeling to innovate biocatalysts for synthetic and pharmaceutical applications.

Improved Bst DNA Polymerase Variants Derived via a Machine Learning Approach

The DNA polymerase I from Geobacillus stearothermophilus (also known as Bst DNAP) is widely used in isothermal amplification reactions, where its strand displacement ability is prized. More robust versions of this enzyme should be enabled for diagnostic applications, especially for carrying out higher temperature reactions that might proceed more quickly. To this end, we appended a short fusion domain from the actin-binding protein villin that improved both stability and purification of the enzyme. In parallel, we have developed a machine learning algorithm that assesses the relative fit of individual amino acids to their chemical microenvironments at any position in a protein and applied this algorithm to predict sequence substitutions in Bst DNAP. The top predicted variants had greatly improved thermotolerance (heating prior to assay), and upon combination, the mutations showed additive thermostability, with denaturation temperatures up to 2.5 °C higher than the parental enzyme. The increased thermostability of the enzyme allowed faster loop-mediated isothermal amplification assays to be carried out at 73 °C, where both Bst DNAP and its improved commercial counterpart Bst 2.0 are inactivated. Overall, this is one of the first examples of the application of machine learning approaches to the thermostabilization of an enzyme.

Insertions and Deletions (Indels): A Missing Piece of the Protein Engineering Jigsaw

Over the years, protein engineers have studied nature and borrowed its tricks to accelerate protein evolution in the test tube. While there have been considerable advances, our ability to generate new proteins in the laboratory is seemingly limited. One explanation for these shortcomings may be that insertions and deletions (indels), which frequently arise in nature, are largely overlooked during protein engineering campaigns. The profound effect of indels on protein structures, by way of drastic backbone alterations, could be perceived as "saltation" events that bring about significant phenotypic changes in a single mutational step. Should we leverage these effects to accelerate protein engineering and gain access to unexplored regions of adaptive landscapes? In this Perspective, we describe the role played by indels in the functional diversification of proteins in nature and discuss their untapped potential for protein engineering, despite their often-destabilizing nature. We hope to spark a renewed interest in indels, emphasizing that their wider study and use may prove insightful and shape the future of protein engineering by unlocking unique functional changes that substitutions alone could never achieve.

Efficient Exploration of Sequence Space by Sequence-Guided Protein Engineering and Design

The rapid growth of sequence databases over the past two decades means that protein engineers faced with optimizing a protein for any given task will often have immediate access to a vast number of related protein sequences. These sequences encode information about the evolutionary history of the protein and the underlying sequence requirements to produce folded, stable, and functional protein variants. Methods that can take advantage of this information are an increasingly important part of the protein engineering tool kit. In this Perspective, we discuss the utility of sequence data in protein engineering and design, focusing on recent advances in three main areas: the use of ancestral sequence reconstruction as an engineering tool to generate thermostable and multifunctional proteins, the use of sequence data to guide engineering of multipoint mutants by structure-based computational protein design, and the use of unlabeled sequence data for unsupervised and semisupervised machine learning, allowing the generation of diverse and functional protein sequences in unexplored regions of sequence space. Altogether, these methods enable the rapid exploration of sequence space within regions enriched with functional proteins and therefore have great potential for accelerating the engineering of stable, functional, and diverse proteins for industrial and biomedical applications.

High-throughput biochemistry in RNA sequence space: predicting structure and function

RNAs are central to fundamental biological processes in all known organisms. The set of possible intramolecular interactions of RNA nucleotides defines the range of alternative structural conformations of a specific RNA that can coexist, and these structures enable functional catalytic properties of RNAs and/or their productive intermolecular interactions with other RNAs or proteins. However, the immense combinatorial space of potential RNA sequences has precluded predictive mapping between RNA sequence and molecular structure and function. Recent advances in high-throughput approaches in vitro have enabled quantitative thermodynamic and kinetic measurements of RNA-RNA and RNA-protein interactions, across hundreds of thousands of sequence variations. In this Review, we explore these techniques, how they can be used to understand RNA function and how they might form the foundations of an accurate model to predict the structure and function of an RNA directly from its nucleotide sequence. The experimental techniques and modelling frameworks discussed here are also highly relevant for the sampling of sequence-structure-function space of DNAs and proteins.

Predicting plant Rubisco kinetics from RbcL sequence data using machine learning

Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is responsible for the conversion of atmospheric CO2 to organic carbon during photosynthesis, and often acts as a rate limiting step in the later process. Screening the natural diversity of Rubisco kinetics is the main strategy used to find better Rubisco enzymes for crop engineering efforts. Here, we demonstrate the use of Gaussian processes (GPs), a family of Bayesian models, coupled with protein encoding schemes, for predicting Rubisco kinetics from Rubisco large subunit (RbcL) sequence data. GPs trained on published experimentally obtained Rubisco kinetic datasets were applied to over 9000 sequences encoding RbcL to predict Rubisco kinetic parameters. Notably, our predicted kinetic values were in agreement with known trends, e.g. higher carboxylation turnover rates (Kcat) for Rubisco enzymes from C4 or crassulacean acid metabolism (CAM) species, compared with those found in C3 species. This is the first study demonstrating machine learning approaches as a tool for screening and predicting Rubisco kinetics, which could be applied to other enzymes.

Biomanufacturing by In Vitro Biotransformation (ivBT) Using Purified Cascade Multi-enzymes

In vitro biotransformation (ivBT) refers to the use of an artificial biological reaction system that employs purified enzymes for the one-pot conversion of low-cost materials into biocommodities such as ethanol, organic acids, and amino acids. Unshackled from cell growth and metabolism, ivBT exhibits distinct advantages compared with metabolic engineering, including but not limited to high engineering flexibility, ease of operation, fast reaction rate, high product yields, and good scalability. These characteristics position ivBT as a promising next-generation biomanufacturing platform. Nevertheless, challenges persist in the enhancement of bulk enzyme preparation methods, the acquisition of enzymes with superior catalytic properties, and the development of sophisticated approaches for pathway design and system optimization. In alignment with the workflow of ivBT development, this chapter presents a systematic introduction to pathway design, enzyme mining and engineering, system construction, and system optimization. The chapter also proffers perspectives on ivBT development.

A primer to directed evolution: current methodologies and future directions

This review summarises the methods available for directed evolution, including mutagenesis and variant selection techniques. The advantages and disadvantages of each technique are presented, and future challenges in the field are discussed.

Protein Stability: Enhancement and Measurement

This chapter defines protein stability, emphasizes its importance, and surveys the field of protein stabilization, with summary reference to a selection of 2014-2021 publications. One can enhance stability, particularly by protein engineering strategies but also by chemical modification and by other means. General protocols are set out on how to measure a given protein's (i) kinetic thermal stability and (ii) oxidative stability and (iii) how to undertake chemical modification of a protein in solution.

Nucleotide augmentation for machine learning-guided protein engineering

Summary

Machine learning-guided protein engineering is a rapidly advancing field. Despite major experimental and computational advances, collecting protein genotype (sequence) and phenotype (function) data remains time- and resource-intensive. As a result, the quality and quantity of training data are often a limiting factor in developing machine learning models. Data augmentation techniques have been successfully applied to the fields of computer vision and natural language processing; however, there is a lack of such augmentation techniques for biological sequence data. Towards this end, we develop nucleotide augmentation (NTA), which leverages natural nucleotide codon degeneracy to augment protein sequence data via synonymous codon substitution. As a proof of concept for protein engineering, we test several online and offline augmentation implementations to train machine learning models with benchmark datasets of protein genotype and phenotype, revealing performance gains on par and surpassing benchmark models using a fraction of the training data. NTA also enables substantial improvements for classification tasks under heavy class imbalance.

Availability and implementation

The code used in this study is publicly available at https://github.com/minotm/NTA.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

CLADE 2.0: Evolution-Driven Cluster Learning-Assisted Directed Evolution

Directed evolution, a revolutionary biotechnology in protein engineering, optimizes protein fitness by searching an astronomical mutational space via expensive experiments. The cluster learning-assisted directed evolution (CLADE) efficiently explores the mutational space via a combination of unsupervised hierarchical clustering and supervised learning. However, the initial-stage sampling in CLADE treats all clusters equally despite many clusters containing a large portion of non-functional mutations. Recent statistical and deep learning tools enable evolutionary density modeling to access protein fitness in an unsupervised manner. In this work, we construct an ensemble of multiple evolutionary scores to guide the initial sampling in CLADE. The resulting evolutionary score-enhanced CLADE, called CLADE 2.0, efficiently selects a training set within a small informative space using the evolution-driven clustering sampling. CLADE 2.0 is validated by using two benchmark libraries both having 160,000 sequences from four-site mutational combinations. Extensive computational experiments and comparisons with existing cutting-edge methods indicate that CLADE 2.0 is a new state-of-art tool for machine learning-assisted directed evolution.

Generalized Property-Based Encoders and Digital Signal Processing Facilitate Predictive Tasks in Protein Engineering

Computational methods in protein engineering often require encoding amino acid sequences, i.e., converting them into numeric arrays. Physicochemical properties are a typical choice to define encoders, where we replace each amino acid by its value for a given property. However, what property (or group thereof) is best for a given predictive task remains an open problem. In this work, we generalize property-based encoding strategies to maximize the performance of predictive models in protein engineering. First, combining text mining and unsupervised learning, we partitioned the AAIndex database into eight semantically-consistent groups of properties. We then applied a non-linear PCA within each group to define a single encoder to represent it. Then, in several case studies, we assess the performance of predictive models for protein and peptide function, folding, and biological activity, trained using the proposed encoders and classical methods (One Hot Encoder and TAPE embeddings). Models trained on datasets encoded with our encoders and converted to signals through the Fast Fourier Transform (FFT) increased their precision and reduced their overfitting substantially, outperforming classical approaches in most cases. Finally, we propose a preliminary methodology to create de novo sequences with desired properties. All these results offer simple ways to increase the performance of general and complex predictive tasks in protein engineering without increasing their complexity.

A versatile active learning workflow for optimization of genetic and metabolic networks

Optimization of biological networks is often limited by wet lab labor and cost, and the lack of convenient computational tools. Here, we describe METIS, a versatile active machine learning workflow with a simple online interface for the data-driven optimization of biological targets with minimal experiments. We demonstrate our workflow for various applications, including cell-free transcription and translation, genetic circuits, and a 27-variable synthetic CO₂-fixation cycle (CETCH cycle), improving these systems between one and two orders of magnitude. For the CETCH cycle, we explore 10²⁵ conditions with only 1,000 experiments to yield the most efficient CO₂-fixation cascade described to date. Beyond optimization, our workflow also quantifies the relative importance of individual factors to the performance of a system identifying unknown interactions and bottlenecks. Overall, our workflow opens the way for convenient optimization and prototyping of genetic and metabolic networks with customizable adjustments according to user experience, experimental setup, and laboratory facilities.

Optimization of biological networks is often limited by wet lab labor and cost, and the lack of convenient computational tools. Here, aimed at democratization and standardization, the authors describe METIS, a modular and versatile active machine learning workflow with a simple online interface for the optimization of biological target functions with minimal experimental datasets.