Deep diversification of an AAV capsid protein by machine learning

Enhanced transgene expression from single-stranded AAV vectors in human cells in vitro and in murine hepatocytes in vivo

We identified that distal 10 nucleotides in the D-sequence in AAV2 inverted terminal repeat (ITR) share partial sequence homology to 1/2 binding site of glucocorticoid receptor-binding element (GRE). Here, we describe that (1) purified GR binds to AAV2 D-sequence, and the D-sequence competes with GR binding to its cognate binding site; (2) dexamethasone-mediated activation of GR pathway significantly increases the transduction efficiency of AAV2 vectors in human cells; (3) human osteosarcoma cells, U2OS, which lack expression of GR, are poorly transduced by AAV2 vectors, but stable transfection with a GR expression plasmid restores vector-mediated transgene expression; (4) replacement of the distal 10 nucleotides in the D-sequence of the AAV2 ITR with a full-length GRE consensus sequence significantly enhances transgene expression in human cells in vitro and in murine hepatocytes in vivo; and (5) none of the ITRs in AAV1, AAV3, AAV4, AAV5, and AAV6 genomes contains the GRE 1/2 binding site, and insertion of a full-length GRE consensus sequence in the AAV6-ITR also significantly enhances transgene expression from AAV6 vectors, both in vitro and in vivo. These novel vectors, termed generation Y AAV vectors, which are serotype, transgene, or promoter agnostic, should be useful in human gene therapy.

Strategies for enhanced gene delivery to the central nervous system

The delivery of genes to the central nervous system (CNS) has been a persistent challenge due to various biological barriers. The blood-brain barrier (BBB), in particular, hampers the access of systemically injected drugs to parenchymal cells, allowing only a minimal percentage (<1%) to pass through. Recent scientific insights highlight the crucial role of the extracellular space (ECS) in governing drug diffusion. Taking into account advancements in vectors, techniques, and knowledge, the discussion will center on the most notable vectors utilized for gene delivery to the CNS. This review will explore the influence of the ECS - a dynamically regulated barrier-on drug diffusion. Furthermore, we will underscore the significance of employing remote-control technologies to facilitate BBB traversal and modulate the ECS. Given the rapid progress in gene editing, our discussion will also encompass the latest advances focused on delivering therapeutic editing in vivo to the CNS tissue. In the end, a brief summary on the impact of Artificial Intelligence (AI)/Machine Learning (ML), ultrasmall, soft endovascular robots, and high-resolution endovascular cameras on improving the gene delivery to the CNS will be provided.

Advances and opportunities in process analytical technologies for viral vector manufacturing

Viral vectors are an emerging, exciting class of biologics whose application in vaccines, oncology, and gene therapy has grown exponentially in recent years. Following first regulatory approval, this class of therapeutics has been vigorously pursued to treat monogenic disorders including orphan diseases, entering hundreds of new products into pipelines. Viral vector manufacturing supporting clinical efforts has spurred the introduction of a broad swath of analytical techniques dedicated to assessing the diverse and evolving panel of Critical Quality Attributes (CQAs) of these products. Herein, we provide an overview of the current state of analytics enabling measurement of CQAs such as capsid and vector identities, product titer, transduction efficiency, impurity clearance etc. We highlight orthogonal methods and discuss the advantages and limitations of these techniques while evaluating their adaptation as process analytical technologies. Finally, we identify gaps and propose opportunities in enabling existing technologies for real-time monitoring from hardware, software, and data analysis viewpoints for technology development within viral vector biomanufacturing.

Novel Drug Delivery Systems: An Important Direction for Drug Innovation Research and Development

The escalating demand for enhanced therapeutic efficacy and reduced adverse effects in the pharmaceutical domain has catalyzed a new frontier of innovation and research in the field of pharmacy: novel drug delivery systems. These systems are designed to address the limitations of conventional drug administration, such as abbreviated half-life, inadequate targeting, low solubility, and bioavailability. As the disciplines of pharmacy, materials science, and biomedicine continue to advance and converge, the development of efficient and safe drug delivery systems, including biopharmaceutical formulations, has garnered significant attention both domestically and internationally. This article presents an overview of the latest advancements in drug delivery systems, categorized into four primary areas: carrier-based and coupling-based targeted drug delivery systems, intelligent drug delivery systems, and drug delivery devices, based on their main objectives and methodologies. Additionally, it critically analyzes the technological bottlenecks, current research challenges, and future trends in the application of novel drug delivery systems.

An adeno-associated virus variant enabling efficient ocular-directed gene delivery across species

Recombinant adeno-associated viruses (rAAVs) have emerged as promising gene therapy vectors due to their proven efficacy and safety in clinical applications. In non-human primates (NHPs), rAAVs are administered via suprachoroidal injection at a higher dose. However, high doses of rAAVs tend to increase additional safety risks. Here, we present a novel AAV capsid (AAVv128), which exhibits significantly enhanced transduction efficiency for photoreceptors and retinal pigment epithelial (RPE) cells, along with a broader distribution across the layers of retinal tissues in different animal models (mice, rabbits, and NHPs) following intraocular injection. Notably, the suprachoroidal delivery of AAVv128-anti-VEGF vector completely suppresses the Grade IV lesions in a laser-induced choroidal neovascularization (CNV) NHP model for neovascular age-related macular degeneration (nAMD). Furthermore, cryo-EM analysis at 2.1 Å resolution reveals that the critical residues of AAVv128 exhibit a more robust advantage in AAV binding, the nuclear uptake and endosome escaping. Collectively, our findings highlight the potential of AAVv128 as a next generation ocular gene therapy vector, particularly using the suprachoroidal delivery route.

Computationally guided AAV engineering for enhanced gene delivery

Gene delivery vehicles based on adeno-associated viruses (AAVs) are enabling increasing success in human clinical trials, and they offer the promise of treating a broad spectrum of both genetic and non-genetic disorders. However, delivery efficiency and targeting must be improved to enable safe and effective therapies. In recent years, considerable effort has been invested in creating AAV variants with improved delivery, and computational approaches have been increasingly harnessed for AAV engineering. In this review, we discuss how computationally designed AAV libraries are enabling directed evolution. Specifically, we highlight approaches that harness sequences outputted by next-generation sequencing (NGS) coupled with machine learning (ML) to generate new functional AAV capsids and related regulatory elements, pushing the frontier of what vector engineering and gene therapy may achieve.

Adeno-associated virus as a delivery vector for gene therapy of human diseases

Adeno-associated virus (AAV) has emerged as a pivotal delivery tool in clinical gene therapy owing to its minimal pathogenicity and ability to establish long-term gene expression in different tissues. Recombinant AAV (rAAV) has been engineered for enhanced specificity and developed as a tool for treating various diseases. However, as rAAV is being more widely used as a therapy, the increased demand has created challenges for the existing manufacturing methods. Seven rAAV-based gene therapy products have received regulatory approval, but there continue to be concerns about safely using high-dose viral therapies in humans, including immune responses and adverse effects such as genotoxicity, hepatotoxicity, thrombotic microangiopathy, and neurotoxicity. In this review, we explore AAV biology with an emphasis on current vector engineering strategies and manufacturing technologies. We discuss how rAAVs are being employed in ongoing clinical trials for ocular, neurological, metabolic, hematological, neuromuscular, and cardiovascular diseases as well as cancers. We outline immune responses triggered by rAAV, address associated side effects, and discuss strategies to mitigate these reactions. We hope that discussing recent advancements and current challenges in the field will be a helpful guide for researchers and clinicians navigating the ever-evolving landscape of rAAV-based gene therapy.

Convolutions are competitive with transformers for protein sequence pretraining

Pretrained protein sequence language models have been shown to improve the performance of many prediction tasks and are now routinely integrated into bioinformatics tools. However, these models largely rely on the transformer architecture, which scales quadratically with sequence length in both run-time and memory. Therefore, state-of-the-art models have limitations on sequence length. To address this limitation, we investigated whether convolutional neural network (CNN) architectures, which scale linearly with sequence length, could be as effective as transformers in protein language models. With masked language model pretraining, CNNs are competitive with, and occasionally superior to, transformers across downstream applications while maintaining strong performance on sequences longer than those allowed in the current state-of-the-art transformer models. Our work suggests that computational efficiency can be improved without sacrificing performance, simply by using a CNN architecture instead of a transformer, and emphasizes the importance of disentangling pretraining task and model architecture. A record of this paper's transparent peer review process is included in the supplemental information.

Rationally engineered novel AAV capsids for intra-articular gene delivery

Intra-articular adeno-associated virus (AAV) gene therapy has been explored as a potential strategy for joint diseases. However, concerns of low transduction efficacy, off-target expression, and neutralizing antibodies (Nabs) still need to be addressed. In this study, we demonstrated that AAV6 was the best serotype to transduce joints after screening serotypes 1 to 9. To develop a more effective AAV vector, a set of novel AAV capsids were rationally engineered. The mutant AAV62 created by swapping variable region I (VRI) of AAV2 into AAV6 induced a higher transduction efficiency per AAV genome copy number. To further investigate the roles of specific amino acids in the transduction of AAV62 and AAV6, we found out that AAV6D with the deletion of threonine at residue 265 induced a 2-fold higher transduction than AAV6, while the transduction efficiency from AAV6M with the mutation of alanine to glutamine at residue 263 was 10-fold lower. AAV6D efficiently transduced both synoviocytes and chondrocytes with low AAV genome copy numbers in other tissues and less Nab formation. This study demonstrates that novel AAV mutants with rational engineering may enhance joint transduction after intra-articular administration in mice, with the potential to evade AAV Nabs and minimize off-target effects in the liver.

In vivo imaging of astrocytes in the whole brain with engineered AAVs and diffusion-weighted magnetic resonance imaging

Astrocytes constitute a major part of the central nervous system and the delineation of their activity patterns is conducive to a better understanding of brain network dynamics. This study aimed to develop a magnetic resonance imaging (MRI)-based method in order to monitor the brain-wide or region-specific astrocytes in live animals. Adeno-associated virus (AAVs) vectors carrying the human glial fibrillary acidic protein (GFAP) promoter driving the EGFP-AQP1 (Aquaporin-1, an MRI reporter) fusion gene were employed. The following steps were included: constructing recombinant AAV vectors for astrocyte-specific expression, detecting MRI reporters in cell culture, brain regions, or whole brain following cell transduction, stereotactic injection, or tail vein injection. The astrocytes were detected by both fluorescent imaging and Diffusion-weighted MRI. The novel AAV mutation (Site-directed mutagenesis of surface-exposed tyrosine (Y) residues on the AAV5 capsid) significantly increased fluorescence intensity (p < 0.01) compared with the AAV5 wild type. Transduction of the rAAV2/5 carrying AQP1 induced the titer-dependent changes in MRI contrast in cell cultures (p < 0.05) and caudate-putamen (CPu) in the brain (p < 0.05). Furthermore, the MRI revealed a good brain-wide alignment between AQP1 levels and ADC signals, which increased over time in most of the transduced brain regions. In addition, the rAAV2/PHP.eB serotype efficiently introduced AOP1 expression in the whole brain via tail vein injection. This study provides an MRI-based approach to detect dynamic changes in astrocytes in live animals. The novel in vivo tool could help us to understand the complexity of neuronal and glial networks in different pathophysiological conditions.

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.

Protein design using structure-based residue preferences

Recent developments in protein design rely on large neural networks with up to 100s of millions of parameters, yet it is unclear which residue dependencies are critical for determining protein function. Here, we show that amino acid preferences at individual residues-without accounting for mutation interactions-explain much and sometimes virtually all of the combinatorial mutation effects across 8 datasets (R2 ~ 78-98%). Hence, few observations (~100 times the number of mutated residues) enable accurate prediction of held-out variant effects (Pearson r > 0.80). We hypothesized that the local structural contexts around a residue could be sufficient to predict mutation preferences, and develop an unsupervised approach termed CoVES (Combinatorial Variant Effects from Structure). Our results suggest that CoVES outperforms not just model-free methods but also similarly to complex models for creating functional and diverse protein variants. CoVES offers an effective alternative to complicated models for identifying functional protein mutations.

Accurate top protein variant discovery via low-N pick-and-validate machine learning

A strategy to obtain the greatest number of best-performing variants with least amount of experimental effort over the vast combinatorial mutational landscape would have enormous utility in boosting resource producibility for protein engineering. Toward this goal, we present a simple and effective machine learning-based strategy that outperforms other state-of-the-art methods. Our strategy integrates zero-shot prediction and multi-round sampling to direct active learning via experimenting with only a few predicted top variants. We find that four rounds of low-N pick-and-validate sampling of 12 variants for machine learning yielded the best accuracy of up to 92.6% in selecting the true top 1% variants in combinatorial mutant libraries, whereas two rounds of 24 variants can also be used. We demonstrate our strategy in successfully discovering high-performance protein variants from diverse families including the CRISPR-based genome editors, supporting its generalizable application for solving protein engineering tasks. A record of this paper's transparent peer review process is included in the supplemental information.

In vivo functional phenotypes from a computational epistatic model of evolution

Computational models of evolution are valuable for understanding the dynamics of sequence variation, to infer phylogenetic relationships or potential evolutionary pathways and for biomedical and industrial applications. Despite these benefits, few have validated their propensities to generate outputs with in vivo functionality, which would enhance their value as accurate and interpretable evolutionary algorithms. We demonstrate the power of epistasis inferred from natural protein families to evolve sequence variants in an algorithm we developed called sequence evolution with epistatic contributions (SEEC). Utilizing the Hamiltonian of the joint probability of sequences in the family as fitness metric, we sampled and experimentally tested for in vivo [Formula: see text]-lactamase activity in Escherichia coli TEM-1 variants. These evolved proteins can have dozens of mutations dispersed across the structure while preserving sites essential for both catalysis and interactions. Remarkably, these variants retain family-like functionality while being more active than their wild-type predecessor. We found that depending on the inference method used to generate the epistatic constraints, different parameters simulate diverse selection strengths. Under weaker selection, local Hamiltonian fluctuations reliably predict relative changes to variant fitness, recapitulating neutral evolution. SEEC has the potential to explore the dynamics of neofunctionalization, characterize viral fitness landscapes, and facilitate vaccine development.

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.

Is Novelty Predictable?

Machine learning-based design has gained traction in the sciences, most notably in the design of small molecules, materials, and proteins, with societal applications ranging from drug development and plastic degradation to carbon sequestration. When designing objects to achieve novel property values with machine learning, one faces a fundamental challenge: how to push past the frontier of current knowledge, distilled from the training data into the model, in a manner that rationally controls the risk of failure. If one trusts learned models too much in extrapolation, one is likely to design rubbish. In contrast, if one does not extrapolate, one cannot find novelty. Herein, we ponder how one might strike a useful balance between these two extremes. We focus in particular on designing proteins with novel property values, although much of our discussion is relevant to machine learning-based design more broadly.

Protein language models meet reduced amino acid alphabets

MOTIVATION

Protein language models (PLMs), which borrowed ideas for modelling and inference from natural language processing, have demonstrated the ability to extract meaningful representations in an unsupervised way. This led to significant performance improvement in several downstream tasks. Clustering amino acids based on their physical-chemical properties to achieve reduced alphabets has been of interest in past research, but their application to PLMs or folding models is unexplored.

RESULTS

Here, we investigate the efficacy of PLMs trained on reduced amino acid alphabets in capturing evolutionary information, and we explore how the loss of protein sequence information impacts learned representations and downstream task performance. Our empirical work shows that PLMs trained on the full alphabet and a large number of sequences capture fine details that are lost in alphabet reduction methods. We further show the ability of a structure prediction model(ESMFold) to fold CASP14 protein sequences translated using a reduced alphabet. For 10 proteins out of the 50 targets, reduced alphabets improve structural predictions with LDDT-Cα differences of up to 19%.

AVAILABILITY AND IMPLEMENTATION

Trained models and code are available at github.com/Ieremie/reduced-alph-PLM.

Optimal trade-off control in machine learning-based library design, with application to adeno-associated virus (AAV) for gene therapy

Adeno-associated viruses (AAVs) hold tremendous promise as delivery vectors for gene therapies. AAVs have been successfully engineered-for instance, for more efficient and/or cell-specific delivery to numerous tissues-by creating large, diverse starting libraries and selecting for desired properties. However, these starting libraries often contain a high proportion of variants unable to assemble or package their genomes, a prerequisite for any gene delivery goal. Here, we present and showcase a machine learning (ML) method for designing AAV peptide insertion libraries that achieve fivefold higher packaging fitness than the standard NNK library with negligible reduction in diversity. To demonstrate our ML-designed library's utility for downstream engineering goals, we show that it yields approximately 10-fold more successful variants than the NNK library after selection for infection of human brain tissue, leading to a promising glial-specific variant. Moreover, our design approach can be applied to other types of libraries for AAV and beyond.

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.

Tpgen: a language model for stable protein design with a specific topology structure

BACKGROUND

Natural proteins occupy a small portion of the protein sequence space, whereas artificial proteins can explore a wider range of possibilities within the sequence space. However, specific requirements may not be met when generating sequences blindly. Research indicates that small proteins have notable advantages, including high stability, accurate resolution prediction, and facile specificity modification.

RESULTS

This study involves the construction of a neural network model named TopoProGenerator(TPGen) using a transformer decoder. The model is trained with sequences consisting of a maximum of 65 amino acids. The training process of TopoProGenerator incorporates reinforcement learning and adversarial learning, for fine-tuning. Additionally, it encompasses a stability predictive model trained with a dataset comprising over 200,000 sequences. The results demonstrate that TopoProGenerator is capable of designing stable small protein sequences with specified topology structures.

CONCLUSION

TPGen has the ability to generate protein sequences that fold into the specified topology, and the pretraining and fine-tuning methods proposed in this study can serve as a framework for designing various types of proteins.

Meta learning addresses noisy and under-labeled data in machine learning-guided antibody engineering

Machine learning-guided protein engineering is rapidly progressing; however, collecting high-quality, large datasets remains a bottleneck. Directed evolution and protein engineering studies often require extensive experimental processes to eliminate noise and label protein sequence-function data. Meta learning has proven effective in other fields in learning from noisy data via bi-level optimization given the availability of a small dataset with trusted labels. Here, we leverage meta learning approaches to overcome noisy and under-labeled data and expedite workflows in antibody engineering. We generate yeast display antibody mutagenesis libraries and screen them for target antigen binding followed by deep sequencing. We then create representative learning tasks, including learning from noisy training data, positive and unlabeled learning, and learning out of distribution properties. We demonstrate that meta learning has the potential to reduce experimental screening time and improve the robustness of machine learning models by training with noisy and under-labeled training data.

Machine-learning-guided Directed Evolution for AAV Capsid Engineering

Target gene delivery is crucial to gene therapy. Adeno-associated virus (AAV) has emerged as a primary gene therapy vector due to its broad host range, long-term expression, and low pathogenicity. However, AAV vectors have some limitations, such as immunogenicity and insufficient targeting. Designing or modifying capsids is a potential method of improving the efficacy of gene delivery, but hindered by weak biological basis of AAV, complexity of the capsids, and limitations of current screening methods. Artificial intelligence (AI), especially machine learning (ML), has great potential to accelerate and improve the optimization of capsid properties as well as decrease their development time and manufacturing costs. This review introduces the traditional methods of designing AAV capsids and the general steps of building a sequence-function ML model, highlights the applications of ML in the development workflow, and summarizes its advantages and challenges.

Self-driving laboratories to autonomously navigate the protein fitness landscape

Protein engineering has nearly limitless applications across chemistry, energy and medicine, but creating new proteins with improved or novel functions remains slow, labor-intensive and inefficient. Here we present the Self-driving Autonomous Machines for Protein Landscape Exploration (SAMPLE) platform for fully autonomous protein engineering. SAMPLE is driven by an intelligent agent that learns protein sequence-function relationships, designs new proteins and sends designs to a fully automated robotic system that experimentally tests the designed proteins and provides feedback to improve the agent's understanding of the system. We deploy four SAMPLE agents with the goal of engineering glycoside hydrolase enzymes with enhanced thermal tolerance. Despite showing individual differences in their search behavior, all four agents quickly converge on thermostable enzymes. Self-driving laboratories automate and accelerate the scientific discovery process and hold great potential for the fields of protein engineering and synthetic biology.

ACIDES: on-line monitoring of forward genetic screens for protein engineering

Forward genetic screens of mutated variants are a versatile strategy for protein engineering and investigation, which has been successfully applied to various studies like directed evolution (DE) and deep mutational scanning (DMS). While next-generation sequencing can track millions of variants during the screening rounds, the vast and noisy nature of the sequencing data impedes the estimation of the performance of individual variants. Here, we propose ACIDES that combines statistical inference and in-silico simulations to improve performance estimation in the library selection process by attributing accurate statistical scores to individual variants. We tested ACIDES first on a random-peptide-insertion experiment and then on multiple public datasets from DE and DMS studies. ACIDES allows experimentalists to reliably estimate variant performance on the fly and can aid protein engineering and research pipelines in a range of applications, including gene therapy.

ProtWave-VAE: Integrating Autoregressive Sampling with Latent-Based Inference for Data-Driven Protein Design

Deep generative models (DGMs) have shown great success in the understanding and data-driven design of proteins. Variational autoencoders (VAEs) are a popular DGM approach that can learn the correlated patterns of amino acid mutations within a multiple sequence alignment (MSA) of protein sequences and distill this information into a low-dimensional latent space to expose phylogenetic and functional relationships and guide generative protein design. Autoregressive (AR) models are another popular DGM approach that typically lacks a low-dimensional latent embedding but does not require training sequences to be aligned into an MSA and enable the design of variable length proteins. In this work, we propose ProtWave-VAE as a novel and lightweight DGM, employing an information maximizing VAE with a dilated convolution encoder and an autoregressive WaveNet decoder. This architecture blends the strengths of the VAE and AR paradigms in enabling training over unaligned sequence data and the conditional generative design of variable length sequences from an interpretable, low-dimensional learned latent space. We evaluated the model's ability to infer patterns and design rules within alignment-free homologous protein family sequences and to design novel synthetic proteins in four diverse protein families. We show that our model can infer meaningful functional and phylogenetic embeddings within latent spaces and make highly accurate predictions within semisupervised downstream fitness prediction tasks. In an application to the C-terminal SH3 domain in the Sho1 transmembrane osmosensing receptor in baker's yeast, we subject ProtWave-VAE-designed sequences to experimental gene synthesis and select-seq assays for the osmosensing function to show that the model enables synthetic protein design, conditional C-terminus diversification, and engineering of the osmosensing function into SH3 paralogues.

A computational toolbox for the assembly yield of complex and heterogeneous structures

The self-assembly of complex structures from a set of non-identical building blocks is a hallmark of soft matter and biological systems, including protein complexes, colloidal clusters, and DNA-based assemblies. Predicting the dependence of the equilibrium assembly yield on the concentrations and interaction energies of building blocks is highly challenging, owing to the difficulty of computing the entropic contributions to the free energy of the many structures that compete with the ground state configuration. While these calculations yield well known results for spherically symmetric building blocks, they do not hold when the building blocks have internal rotational degrees of freedom. Here we present an approach for solving this problem that works with arbitrary building blocks, including proteins with known structure and complex colloidal building blocks. Our algorithm combines classical statistical mechanics with recently developed computational tools for automatic differentiation. Automatic differentiation allows efficient evaluation of equilibrium averages over configurations that would otherwise be intractable. We demonstrate the validity of our framework by comparison to molecular dynamics simulations of simple examples, and apply it to calculate the yield curves for known protein complexes and for the assembly of colloidal shells.

AAV Immunotoxicity: Implications in Anti-HBV Gene Therapy

Hepatitis B virus (HBV) has afflicted humankind for decades and there is still no treatment that can clear the infection. The development of recombinant adeno-associated virus (rAAV)-based gene therapy for HBV infection has become important in recent years and research has made exciting leaps. Initial studies, mainly using mouse models, showed that rAAVs are non-toxic and induce minimal immune responses. However, several later studies demonstrated rAAV toxicity, which is inextricably associated with immunogenicity. This is a major setback for the progression of rAAV-based therapies toward clinical application. Research aimed at understanding the mechanisms behind rAAV immunity and toxicity has contributed significantly to the inception of approaches to overcoming these challenges. The target tissue, the features of the vector, and the vector dose are some of the determinants of AAV toxicity, with the latter being associated with the most severe adverse events. This review discusses our current understanding of rAAV immunogenicity, toxicity, and approaches to overcoming these hurdles. How this information and current knowledge about HBV biology and immunity can be harnessed in the efforts to design safe and effective anti-HBV rAAVs is discussed.

Multiparametric domain insertional profiling of adeno-associated virus VP1

Several evolved properties of adeno-associated virus (AAV), such as broad tropism and immunogenicity in humans, are barriers to AAV-based gene therapy. Most efforts to re-engineer these properties have focused on variable regions near AAV's 3-fold protrusions and capsid protein termini. To comprehensively survey AAV capsids for engineerable hotspots, we determined multiple AAV fitness phenotypes upon insertion of six structured protein domains into the entire AAV-DJ capsid protein VP1. This is the largest and most comprehensive AAV domain insertion dataset to date. Our data revealed a surprising robustness of AAV capsids to accommodate large domain insertions. Insertion permissibility depended strongly on insertion position, domain type, and measured fitness phenotype, which clustered into contiguous structural units that we could link to distinct roles in AAV assembly, stability, and infectivity. We also identified engineerable hotspots of AAV that facilitate the covalent attachment of binding scaffolds, which may represent an alternative approach to re-direct AAV tropism.

Intravitreal injection of a rationally designed AAV capsid library in non-human primate identifies variants with enhanced retinal transduction and neutralizing antibody evasion

Adeno-associated virus (AAV) continues to be the gold standard vector for therapeutic gene delivery and has proven especially useful for treating ocular disease. Intravitreal injection (IVtI) is a promising delivery route because it increases accessibility of gene therapies to larger patient populations. However, data from clinical and non-human primate (NHP) studies utilizing currently available capsids indicate that anatomical barriers to AAV and pre-existing neutralizing antibodies can restrict gene expression to levels that are "sub-therapeutic" in a substantial proportion of patients. Here, we performed a combination of directed evolution in NHPs of an AAV2-based capsid library with simultaneous mutations across six surface-exposed variable regions and rational design to identify novel capsid variants with improved retinal transduction following IVtI. Following two rounds of screening in NHP, enriched variants were characterized in intravitreally injected mice and NHPs and shown to have increased transduction relative to AAV2. Lead capsid variant, P2-V1, demonstrated an increased ability to evade neutralizing antibodies in human vitreous samples relative to AAV2 and AAV2.7m8. Taken together, this study further contributed to our understanding of the selective pressures associated with retinal transduction via the vitreous and identified promising novel AAV capsid variants for clinical consideration.

Advances in generative modeling methods and datasets to design novel enzymes for renewable chemicals and fuels

Biotechnology has revolutionized the development of sustainable energy sources by harnessing biomass as a feedstock for energy production. However, challenges such as recalcitrant feedstocks and inefficient metabolic pathways hinder the large-scale integration of renewable energy systems. Enzyme engineering has emerged as a powerful tool to address these challenges by enhancing enzyme activity, specificity, and stability. Generative machine learning (ML) models have shown great promise in accelerating protein design, allowing for the generation of novel protein sequences with desired properties by navigating vast spaces. This review paper aims to summarize the state of the art in generative models for protein design and how they can be applied to bioenergy applications, including the underlying architectures and training strategies. Additionally, it highlights the importance of high-quality datasets for training and evaluating generative models, organizes available datasets for generative protein design, and discusses the potential of applying generative models to strain design for bioenergy production.

Continuous evolution of user-defined genes at 1-million-times the genomic mutation rate

When nature maintains or evolves a gene's function over millions of years at scale, it produces a diversity of homologous sequences whose patterns of conservation and change contain rich structural, functional, and historical information about the gene. However, natural gene diversity likely excludes vast regions of functional sequence space and includes phylogenetic and evolutionary eccentricities, limiting what information we can extract. We introduce an accessible experimental approach for compressing long-term gene evolution to laboratory timescales, allowing for the direct observation of extensive adaptation and divergence followed by inference of structural, functional, and environmental constraints for any selectable gene. To enable this approach, we developed a new orthogonal DNA replication (OrthoRep) system that durably hypermutates chosen genes at a rate of >10 -4 substitutions per base in vivo . When OrthoRep was used to evolve a conditionally essential maladapted enzyme, we obtained thousands of unique multi-mutation sequences with many pairs >60 amino acids apart (>15% divergence), revealing known and new factors influencing enzyme adaptation. The fitness of evolved sequences was not predictable by advanced machine learning models trained on natural variation. We suggest that OrthoRep supports the prospective and systematic discovery of constraints shaping gene evolution, uncovering of new regions in fitness landscapes, and general applications in biomolecular engineering.

Structure-guided AAV capsid evolution strategies for enhanced CNS gene delivery

Over the past 5 years, our laboratory has systematically developed a structure-guided library approach to evolve new adeno-associated virus (AAV) capsids with altered tissue tropism, higher transduction efficiency and the ability to evade pre-existing humoral immunity. Here, we provide a detailed protocol describing two distinct evolution strategies using structurally divergent AAV serotypes as templates, exemplified by improving CNS gene transfer efficiency in vivo. We outline four major components of our strategy: (i) structure-guided design of AAV capsid libraries, (ii) AAV library production, (iii) library cycling in single versus multiple animal models, followed by (iv) evaluation of lead AAV vector candidates in vivo. The protocol spans ~95 d, excluding gene expression analysis in vivo, and can vary depending on user experience, resources and experimental design. A distinguishing attribute of the current protocol is the focus on providing biomedical researchers with 3D structural information to guide evolution of precise 'hotspots' on AAV capsids. Furthermore, the protocol outlines two distinct methods for AAV library evolution consisting of adenovirus-enabled infectious cycling in a single species and noninfectious cycling in a cross-species manner. Notably, our workflow can be seamlessly merged with other RNA transcript-based library strategies and tailored for tissue-specific capsid selection. Overall, the procedures outlined herein can be adapted to expand the AAV vector toolkit for genetic manipulation of animal models and development of human gene therapies.

Technological advances in the use of viral and non-viral vectors for delivering genetic and non-genetic cargos for cancer therapy

The burden of cancer is increasing globally. Several challenges facing its mainstream treatment approaches have formed the basis for the development of targeted delivery systems to carry and distribute anti-cancer payloads to their defined targets. This site-specific delivery of drug molecules and gene payloads to selectively target druggable biomarkers aimed at inducing cell death while sparing normal cells is the principal goal for cancer therapy. An important advantage of a delivery vector either viral or non-viral is the cumulative ability to penetrate the haphazardly arranged and immunosuppressive tumour microenvironment of solid tumours and or withstand antibody-mediated immune response. Biotechnological approaches incorporating rational protein engineering for the development of targeted delivery systems which may serve as vehicles for packaging and distribution of anti-cancer agents to selectively target and kill cancer cells are highly desired. Over the years, these chemically and genetically modified delivery systems have aimed at distribution and selective accumulation of drug molecules at receptor sites resulting in constant maintenance of high drug bioavailability for effective anti-tumour activity. In this review, we highlighted the state-of-the art viral and non-viral drug and gene delivery systems and those under developments focusing on cancer therapy.

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.

Advances in Recombinant Adeno-Associated Virus Vectors for Neurodegenerative Diseases

Recombinant adeno-associated virus (rAAV) vectors are gene therapy delivery tools that offer a promising platform for the treatment of neurodegenerative diseases. Keeping up with developments in this fast-moving area of research is a challenge. This review was thus written with the intention to introduce this field of study to those who are new to it and direct others who are struggling to stay abreast of the literature towards notable recent studies. In ten sections, we briefly highlight early milestones within this field and its first clinical success stories. We showcase current clinical trials, which focus on gene replacement, gene augmentation, or gene suppression strategies. Next, we discuss ongoing efforts to improve the tropism of rAAV vectors for brain applications and introduce pre-clinical research directed toward harnessing rAAV vectors for gene editing applications. Subsequently, we present common genetic elements coded by the single-stranded DNA of rAAV vectors, their so-called payloads. Our focus is on recent advances that are bound to increase treatment efficacies. As needed, we included studies outside the neurodegenerative disease field that showcased improved pre-clinical designs of all-in-one rAAV vectors for gene editing applications. Finally, we discuss risks associated with off-target effects and inadvertent immunogenicity that these technologies harbor as well as the mitigation strategies available to date to make their application safer.

A humanized chemogenetic system inhibits murine pain-related behavior and hyperactivity in human sensory neurons

Hyperexcitability in sensory neurons is known to underlie many of the maladaptive changes associated with persistent pain. Chemogenetics has shown promise as a means to suppress such excitability, yet chemogenetic approaches suitable for human applications are needed. PSAM4-GlyR is a modular system based on the human α7 nicotinic acetylcholine and glycine receptors, which responds to inert chemical ligands and the clinically approved drug varenicline. Here, we demonstrated the efficacy of this channel in silencing both mouse and human sensory neurons by the activation of large shunting conductances after agonist administration. Virally mediated expression of PSAM4-GlyR in mouse sensory neurons produced behavioral hyposensitivity upon agonist administration, which was recovered upon agonist washout. Stable expression of the channel led to similar reversible suppression of pain-related behavior even after 10 months of viral delivery. Mechanical and spontaneous pain readouts were also ameliorated by PSAM4-GlyR activation in acute and joint pain inflammation mouse models. Furthermore, suppression of mechanical hypersensitivity generated by a spared nerve injury model of neuropathic pain was also observed upon activation of the channel. Effective silencing of behavioral hypersensitivity was reproduced in a human model of hyperexcitability and clinical pain: PSAM4-GlyR activation decreased the excitability of human-induced pluripotent stem cell-derived sensory neurons and spontaneous activity due to a gain-of-function NaV1.7 mutation causing inherited erythromelalgia. Our results demonstrate the contribution of sensory neuron hyperexcitability to neuropathic pain and the translational potential of an effective, stable, and reversible humanized chemogenetic system for the treatment of pain.

MBE: model-based enrichment estimation and prediction for differential sequencing data

Characterizing differences in sequences between two conditions, such as with and without drug exposure, using high-throughput sequencing data is a prevalent problem involving quantifying changes in sequence abundances, and predicting such differences for unobserved sequences. A key shortcoming of current approaches is their extremely limited ability to share information across related but non-identical reads. Consequently, they cannot use sequencing data effectively, nor be directly applied in many settings of interest. We introduce model-based enrichment (MBE) to overcome this shortcoming. We evaluate MBE using both simulated and real data. Overall, MBE improves accuracy compared to current differential analysis methods.

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.

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.

Linear-Scaling Kernels for Protein Sequences and Small Molecules Outperform Deep Learning While Providing Uncertainty Quantitation and Improved Interpretability

Gaussian process (GP) is a Bayesian model which provides several advantages for regression tasks in machine learning such as reliable quantitation of uncertainty and improved interpretability. Their adoption has been precluded by their excessive computational cost and by the difficulty in adapting them for analyzing sequences (e.g., amino acid sequences) and graphs (e.g., small molecules). In this study, we introduce a group of random feature-approximated kernels for sequences and graphs that exhibit linear scaling with both the size of the training set and the size of the sequences or graphs. We incorporate these new kernels into our new Python library for GP regression, xGPR, and develop an efficient and scalable algorithm for fitting GPs equipped with these kernels to large datasets. We compare the performance of xGPR on 17 different benchmarks with both standard and state-of-the-art deep learning models and find that GP regression achieves highly competitive accuracy for these tasks while providing with well-calibrated uncertainty quantitation and improved interpretability. Finally, in a simple experiment, we illustrate how xGPR may be used as part of an active learning strategy to engineer a protein with a desired property in an automated way without human intervention.

Functional biology and biotechnology of thermophilic viruses

Viruses have developed sophisticated biochemical and genetic mechanisms to manipulate and exploit their hosts. Enzymes derived from viruses have been essential research tools since the first days of molecular biology. However, most viral enzymes that have been commercialized are derived from a small number of cultivated viruses, which is remarkable considering the extraordinary diversity and abundance of viruses revealed by metagenomic analysis. Given the explosion of new enzymatic reagents derived from thermophilic prokaryotes over the past 40 years, those obtained from thermophilic viruses should be equally potent tools. This review discusses the still-limited state of the art regarding the functional biology and biotechnology of thermophilic viruses with a focus on DNA polymerases, ligases, endolysins, and coat proteins. Functional analysis of DNA polymerases and primase-polymerases from phages infecting Thermus, Aquificaceae, and Nitratiruptor has revealed new clades of enzymes with strong proofreading and reverse transcriptase capabilities. Thermophilic RNA ligase 1 homologs have been characterized from Rhodothermus and Thermus phages, with both commercialized for circularization of single-stranded templates. Endolysins from phages infecting Thermus, Meiothermus, and Geobacillus have shown high stability and unusually broad lytic activity against Gram-negative and Gram-positive bacteria, making them targets for commercialization as antimicrobials. Coat proteins from thermophilic viruses infecting Sulfolobales and Thermus strains have been characterized, with diverse potential applications as molecular shuttles. To gauge the scale of untapped resources for these proteins, we also document over 20,000 genes encoded by uncultivated viral genomes from high-temperature environments that encode DNA polymerase, ligase, endolysin, or coat protein domains.

Enabling technology and core theory of synthetic biology

Synthetic biology provides a new paradigm for life science research ("build to learn") and opens the future journey of biotechnology ("build to use"). Here, we discuss advances of various principles and technologies in the mainstream of the enabling technology of synthetic biology, including synthesis and assembly of a genome, DNA storage, gene editing, molecular evolution and de novo design of function proteins, cell and gene circuit engineering, cell-free synthetic biology, artificial intelligence (AI)-aided synthetic biology, as well as biofoundries. We also introduce the concept of quantitative synthetic biology, which is guiding synthetic biology towards increased accuracy and predictability or the real rational design. We conclude that synthetic biology will establish its disciplinary system with the iterative development of enabling technologies and the maturity of the core theory.

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.

Targeting AAV vectors to the central nervous system by engineering capsid-receptor interactions that enable crossing of the blood-brain barrier

Viruses have evolved the ability to bind and enter cells through interactions with a wide variety of cell macromolecules. We engineered peptide-modified adeno-associated virus (AAV) capsids that transduce the brain through the introduction of de novo interactions with 2 proteins expressed on the mouse blood-brain barrier (BBB), LY6A or LY6C1. The in vivo tropisms of these capsids are predictable as they are dependent on the cell- and strain-specific expression of their target protein. This approach generated hundreds of capsids with dramatically enhanced central nervous system (CNS) tropisms within a single round of screening in vitro and secondary validation in vivo thereby reducing the use of animals in comparison to conventional multi-round in vivo selections. The reproducible and quantitative data derived via this method enabled both saturation mutagenesis and machine learning (ML)-guided exploration of the capsid sequence space. Notably, during our validation process, we determined that nearly all published AAV capsids that were selected for their ability to cross the BBB in mice leverage either the LY6A or LY6C1 protein, which are not present in primates. This work demonstrates that AAV capsids can be directly targeted to specific proteins to generate potent gene delivery vectors with known mechanisms of action and predictable tropisms.

Receptor-mediated transcytosis of macromolecules across the blood-brain barrier

INTRODUCTION

The blood-brain barrier (BBB) restricts brain access of virtually all macromolecules. Receptor-mediated transcytosis (RMT) is one strategy toward their brain delivery. In this strategy, targeting ligands conjugated to therapeutic payload or decorating particles containing the payload interact with targets on brain capillary endothelial cells (BCEC), triggering internalization, trafficking, and release from BCEC.

AREAS COVERED

RMT at the BBB has leveraged multiple formats of macromolecules and large particles. Interactions between those and BCEC have been studied primarily using antibodies, with findings applicable to the design of larger particles. BBB-penetrant constructs have also been identified in screening campaigns and directed evolution, and subsequently found to interact with RMT targets. In addition, BCEC targeted by constructs incorporating genomic payload can be made to produce therapeutic proteins.

EXPERT OPINION

While targeting may not be strictly necessary to reach a therapeutic effect for all macromolecules, it can improve a molecule's BBB transport, exposing it to the entire brain parenchyma and enhancing its effect. Constructs with better BCEC transcytosis may be designed rationally, leveraging knowledge about BCEC trafficking, and found in screening campaigns, where this knowledge can reduce the search space and improve iterative refinement. Identification of new targets may also help generate BBB-crossing constructs.

BioAutoMATED: An end-to-end automated machine learning tool for explanation and design of biological sequences

The design choices underlying machine-learning (ML) models present important barriers to entry for many biologists who aim to incorporate ML in their research. Automated machine-learning (AutoML) algorithms can address many challenges that come with applying ML to the life sciences. However, these algorithms are rarely used in systems and synthetic biology studies because they typically do not explicitly handle biological sequences (e.g., nucleotide, amino acid, or glycan sequences) and cannot be easily compared with other AutoML algorithms. Here, we present BioAutoMATED, an AutoML platform for biological sequence analysis that integrates multiple AutoML methods into a unified framework. Users are automatically provided with relevant techniques for analyzing, interpreting, and designing biological sequences. BioAutoMATED predicts gene regulation, peptide-drug interactions, and glycan annotation, and designs optimized synthetic biology components, revealing salient sequence characteristics. By automating sequence modeling, BioAutoMATED allows life scientists to incorporate ML more readily into their work.

Rationale and strategies for the development of safe and effective optimized AAV vectors for human gene therapy

Recombinant adeno-associated virus (AAV) vectors have been, or are currently in use, in 332 phase I/II/III clinical trials in a number of human diseases, and in some cases, remarkable clinical efficacy has also been achieved. There are now three US Food and Drug Administration (FDA)-approved AAV "drugs," but it has become increasingly clear that the first generation of AAV vectors are not optimal. In addition, relatively large vector doses are needed to achieve clinical efficacy, which has been shown to provoke host immune responses culminating in serious adverse events and, more recently, in the deaths of 10 patients to date. Thus, there is an urgent need for the development of the next generation of AAV vectors that are (1) safe, (2) effective, and (3) human tropic. This review describes the strategies to potentially overcome each of the limitations of the first generation of AAV vectors and the rationale and approaches for the development of the next generation of AAV serotype vectors. These vectors promise to be efficacious at significant reduced doses, likely to achieve clinical efficacy, thereby increasing the safety as well as reducing vector production costs, ensuring translation to the clinic with higher probability of success, without the need for the use of immune suppression, for gene therapy of a wide variety of diseases in humans.

Fast and efficient template-mediated synthesis of genetic variants

Efficient methods for the generation of specific mutations enable the study of functional variations in natural populations and lead to advances in genetic engineering applications. Here, we present a new approach, mutagenesis by template-guided amplicon assembly (MEGAA), for the rapid construction of kilobase-sized DNA variants. With this method, many mutations can be generated at a time to a DNA template at more than 90% efficiency per target in a predictable manner. We devised a robust and iterative protocol for an open-source laboratory automation robot that enables desktop production and long-read sequencing validation of variants. Using this system, we demonstrated the construction of 31 natural SARS-CoV2 spike gene variants and 10 recoded Escherichia coli genome fragments, with each 4 kb region containing up to 150 mutations. Furthermore, 125 defined combinatorial adeno-associated virus-2 cap gene variants were easily built using the system, which exhibited viral packaging enhancements of up to 10-fold compared with wild type. Thus, the MEGAA platform enables generation of multi-site sequence variants quickly, cheaply, and in a scalable manner for diverse applications in biotechnology.