Signal Peptides Generated by Attention-Based Neural Networks

Protein subcellular localization prediction tools

Protein subcellular localization prediction is of great significance in bioinformatics and biological research. Most of the proteins do not have experimentally determined localization information, computational prediction methods and tools have been acting as an active research area for more than two decades now. Knowledge of the subcellular location of a protein provides valuable information about its functionalities, the functioning of the cell, and other possible interactions with proteins. Fast, reliable, and accurate predictors provides platforms to harness the abundance of sequence data to predict subcellular locations accordingly. During the last decade, there has been a considerable amount of research effort aimed at developing subcellular localization predictors. This paper reviews recent subcellular localization prediction tools in the Eukaryotic, Prokaryotic, and Virus-based categories followed by a detailed analysis. Each predictor is discussed based on its main features, strengths, weaknesses, algorithms used, prediction techniques, and analysis. This review is supported by prediction tools taxonomies that highlight their rele- vant area and examples for uncomplicated categorization and ease of understandability. These taxonomies help users find suitable tools according to their needs. Furthermore, recent research gaps and challenges are discussed to cover areas that need the utmost attention. This survey provides an in-depth analysis of the most recent prediction tools to facilitate readers and can be considered a quick guide for researchers to identify and explore the recent literature advancements.

Efficient Production of an Alginate Lyase in Bacillus subtilis with Combined Strategy: Vector and Host Selection, Promoter and Signal Peptide Screening, and Modification of a Translation Initiation Region

Alginate lyases (ALys) whose degrading products, alginate oligosaccharides, exhibit various outstanding biochemical activities have aroused increasing interest of researchers in the marine bioresource field. However, their predominant sourcing from marine bacteria, with limited yields and unclear genetic backgrounds, presents a challenge for industrial production. In this study, ALys (Aly01) from Vibrio natriegens SK 42.001 was expressed in Bacillus subtilis (B. subtilis), a nonpathogenic microorganism recognized as generally safe (GRAS). This accomplishment was realized through a comprehensive strategy involving vector and host selection, promoter and signal peptide screening, and engineering of the ribosome binding site (RBS) and the N-terminal coding sequence (NCS). The optimal combination was identified as the pP43NMK and B. subtilis WB600. Among the 19 reported strong promoters, PnprE exhibited the best performance, showing intracellular enzyme activities of 4.47 U/mL. Despite expectations, dual promoter construction did not yield a significant increase. Further, SPydhT demonstrated the highest extracellular activity (1.33 U/mL), which was further improved by RBS/NCS engineering, reaching 4.58 U/mL. Finally, after fed-batch fermentation, the extracellular activity reached 18.01 U/mL, which was the highest of ALys with a high molecular weight expressed in B. subtilis. These findings are expected to offer valuable insights into the heterologous expression of ALys in B. subtilis.

Addressing epistasis in the design of protein function

Mutations in protein active sites can dramatically improve function. The active site, however, is densely packed and extremely sensitive to mutations. Therefore, some mutations may only be tolerated in combination with others in a phenomenon known as epistasis. Epistasis reduces the likelihood of obtaining improved functional variants and dramatically slows natural and lab evolutionary processes. Research has shed light on the molecular origins of epistasis and its role in shaping evolutionary trajectories and outcomes. In addition, sequence- and AI-based strategies that infer epistatic relationships from mutational patterns in natural or experimental evolution data have been used to design functional protein variants. In recent years, combinations of such approaches and atomistic design calculations have successfully predicted highly functional combinatorial mutations in active sites. These were used to design thousands of functional active-site variants, demonstrating that, while our understanding of epistasis remains incomplete, some of the determinants that are critical for accurate design are now sufficiently understood. We conclude that the space of active-site variants that has been explored by evolution may be expanded dramatically to enhance natural activities or discover new ones. Furthermore, design opens the way to systematically exploring sequence and structure space and mutational impacts on function, deepening our understanding and control over protein activity.

A bacteriocin expression platform for targeting pathogenic bacterial species

Bacteriocins are antimicrobial peptides that are naturally produced by many bacteria. They hold great potential in the fight against antibiotic resistant bacteria, including ESKAPE pathogens. Engineered live biotherapeutic products (eLBPs) that secrete bacteriocins can be created to deliver targeted bacteriocin production. Here we develop a modular bacteriocin secretion platform that can be used to express and secrete multiple bacteriocins from non-pathogenic Escherichia coli host strains. As a proof of concept we create Enterocin A (EntA) and Enterocin B (EntB) secreting strains that show strong antimicrobial activity against Enterococcus faecalis and Enterococcus faecium in vitro, and characterise this activity in both solid culture and liquid co-culture. We then develop a Lotka-Volterra model that can be used to capture the interactions of these competitor strains. We show that simultaneous exposure to EntA and EntB can delay Enterococcus growth. Our system has the potential to be used as an eLBP to secrete additional bacteriocins for the targeted killing of pathogenic bacteria.

TULIP: A transformer-based unsupervised language model for interacting peptides and T cell receptors that generalizes to unseen epitopes

The accurate prediction of binding between T cell receptors (TCR) and their cognate epitopes is key to understanding the adaptive immune response and developing immunotherapies. Current methods face two significant limitations: the shortage of comprehensive high-quality data and the bias introduced by the selection of the negative training data commonly used in the supervised learning approaches. We propose a method, Transformer-based Unsupervised Language model for Interacting Peptides and T cell receptors (TULIP), that addresses both limitations by leveraging incomplete data and unsupervised learning and using the transformer architecture of language models. Our model is flexible and integrates all possible data sources, regardless of their quality or completeness. We demonstrate the existence of a bias introduced by the sampling procedure used in previous supervised approaches, emphasizing the need for an unsupervised approach. TULIP recognizes the specific TCRs binding an epitope, performing well on unseen epitopes. Our model outperforms state-of-the-art models and offers a promising direction for the development of more accurate TCR epitope recognition models.

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.

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.

Enhanced bovine genome annotation through integration of transcriptomics and epi-transcriptomics datasets facilitates genomic biology

BACKGROUND

The accurate identification of the functional elements in the bovine genome is a fundamental requirement for high-quality analysis of data informing both genome biology and genomic selection. Functional annotation of the bovine genome was performed to identify a more complete catalog of transcript isoforms across bovine tissues.

RESULTS

A total of 160,820 unique transcripts (50% protein coding) representing 34,882 unique genes (60% protein coding) were identified across tissues. Among them, 118,563 transcripts (73% of the total) were structurally validated by independent datasets (PacBio isoform sequencing data, Oxford Nanopore Technologies sequencing data, de novo assembled transcripts from RNA sequencing data) and comparison with Ensembl and NCBI gene sets. In addition, all transcripts were supported by extensive data from different technologies such as whole transcriptome termini site sequencing, RNA Annotation and Mapping of Promoters for the Analysis of Gene Expression, chromatin immunoprecipitation sequencing, and assay for transposase-accessible chromatin using sequencing. A large proportion of identified transcripts (69%) were unannotated, of which 86% were produced by annotated genes and 14% by unannotated genes. A median of two 5' untranslated regions were expressed per gene. Around 50% of protein-coding genes in each tissue were bifunctional and transcribed both coding and noncoding isoforms. Furthermore, we identified 3,744 genes that functioned as noncoding genes in fetal tissues but as protein-coding genes in adult tissues. Our new bovine genome annotation extended more than 11,000 annotated gene borders compared to Ensembl or NCBI annotations. The resulting bovine transcriptome was integrated with publicly available quantitative trait loci data to study tissue-tissue interconnection involved in different traits and construct the first bovine trait similarity network.

CONCLUSIONS

These validated results show significant improvement over current bovine genome annotations.

Secretome Analysis: Reading Cellular Sign Language to Understand Intercellular Communication

A significant portion of mammalian proteomes is secreted to the extracellular space to fulfill crucial roles in cell-to-cell communication. To best recapitulate the intricate and multi-faceted crosstalk between cells in a live organism, there is an ever-increasing need for methods to study protein secretion in model systems that include multiple cell types. In addition, posttranslational modifications further expand the complexity and versatility of cellular communication. This review aims to summarize recent strategies and model systems that employ cellular coculture, chemical biology tools, protein enrichment, and proteomic methods to characterize the composition and function of cellular secretomes. This is all geared towards gaining better understanding of organismal biology in vivo mediated by secretory signaling.

Efficient Computational Framework for Target-Specific Active Peptide Discovery: A Case Study on IL-17C Targeting Cyclic Peptides

The development of potentially active peptides for specific targets is critical for the modern pharmaceutical industry's growth. In this study, we present an efficient computational framework for the discovery of active peptides targeting a specific pharmacological target, which combines a conditional variational autoencoder (CVAE) and a classifier named TCPP based on the Transformer and convolutional neural network. In our example scenario, we constructed an active cyclic peptide library targeting interleukin-17C (IL-17C) through a library-based in vitro selection strategy. The CVAE model is trained on the preprocessed peptide data sets to generate potentially active peptides and the TCPP further screens the generated peptides. Ultimately, six candidate peptides predicted by the model were synthesized and assayed for their activity, and four of them exhibited promising binding affinity to IL-17C. Our study provides a one-stop-shop for target-specific active peptide discovery, which is expected to boost up the process of peptide drug development.

Let There Be Light: Genome Reduction Enables Bacillus subtilis to Produce Disulfide-Bonded Gaussia Luciferase

Bacillus subtilis is a major workhorse for enzyme production in industrially relevant quantities. Compared to mammalian-based expression systems, B. subtilis presents intrinsic advantages, such as high growth rates, high space-time yield, unique protein secretion capabilities, and low maintenance costs. However, B. subtilis shows clear limitations in the production of biopharmaceuticals, especially proteins from eukaryotic origin that contain multiple disulfide bonds. In the present study, we deployed genome minimization, signal peptide screening, and coexpression of recombinant thiol oxidases as strategies to improve the ability of B. subtilis to secrete proteins with multiple disulfide bonds. Different genome-reduced strains served as the chassis for expressing the model protein Gaussia Luciferase (GLuc), which contains five disulfide bonds. These chassis lack extracellular proteases, prophages, and key sporulation genes. Importantly, compared to the reference strain with a full-size genome, the best-performing genome-minimized strain achieved over 3000-fold increased secretion of active GLuc while growing to lower cell densities. Our results show that high-level GLuc secretion relates, at least in part, to the absence of major extracellular proteases. In addition, we show that the thiol-disulfide oxidoreductase requirements for disulfide bonding have changed upon genome reduction. Altogether, our results highlight genome-engineered Bacillus strains as promising expression platforms for proteins with multiple disulfide bonds.

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.

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.

Novel alleles in the era of next-generation sequencing-based HLA typing calls for standardization and policy

Next-Generation Sequencing (NGS) has transformed clinical histocompatibility laboratories through its capacity to provide accurate, high-throughput, high-resolution typing of Human Leukocyte Antigen (HLA) genes, which is critical for transplant safety and success. As this technology becomes widely used for clinical genotyping, histocompatibility laboratories now have an increased capability to identify novel HLA alleles that previously would not be detected using traditional genotyping methods. Standard guidelines for the clinical verification and reporting of novelties in the era of NGS are greatly needed. Here, we describe the experience of a clinical histocompatibility laboratory's use of NGS for HLA genotyping and its management of novel alleles detected in an ethnically-diverse population of British Columbia, Canada. Over a period of 18 months, 3,450 clinical samples collected for the purpose of solid organ or hematopoietic stem cell transplantation were sequenced using NGS. Overall, 29 unique novel alleles were identified at a rate of ∼1.6 per month. The majority of novelties (52%) were detected in the alpha chains of class II (HLA-DQA1 and -DPA1). Novelties were found in all 11 HLA classical genes except for HLA-DRB3, -DRB4, and -DQB1. All novelties were single nucleotide polymorphisms, where more than half led to an amino acid change, and one resulted in a premature stop codon. Missense mutations were evaluated for changes in their amino acid properties to assess the potential effect on the novel HLA protein. All novelties identified were confirmed independently at another accredited HLA laboratory using a different NGS assay and platform to ensure validity in the reporting of novelties. The novel alleles were submitted to the Immuno Polymorphism Database-Immunogenetics/HLA (IPD-IMGT/HLA) for official allele name designation and inclusion in future database releases. A nationwide survey involving all Canadian HLA laboratories confirmed the common occurrence of novel allele detection but identified a wide variability in the assessment and reporting of novelties. In summary, a considerable proportion of novel alleles were identified in routine clinical testing. We propose a framework for the standardization of policies on the clinical management of novel alleles and inclusion in proficiency testing programs in the era of NGS-based HLA genotyping.

Generative models for protein sequence modeling: recent advances and future directions

The widespread adoption of high-throughput omics technologies has exponentially increased the amount of protein sequence data involved in many salient disease pathways and their respective therapeutics and diagnostics. Despite the availability of large-scale sequence data, the lack of experimental fitness annotations underpins the need for self-supervised and unsupervised machine learning (ML) methods. These techniques leverage the meaningful features encoded in abundant unlabeled sequences to accomplish complex protein engineering tasks. Proficiency in the rapidly evolving fields of protein engineering and generative AI is required to realize the full potential of ML models as a tool for protein fitness landscape navigation. Here, we support this work by (i) providing an overview of the architecture and mathematical details of the most successful ML models applicable to sequence data (e.g. variational autoencoders, autoregressive models, generative adversarial neural networks, and diffusion models), (ii) guiding how to effectively implement these models on protein sequence data to predict fitness or generate high-fitness sequences and (iii) highlighting several successful studies that implement these techniques in protein engineering (from paratope regions and subcellular localization prediction to high-fitness sequences and protein design rules generation). By providing a comprehensive survey of model details, novel architecture developments, comparisons of model applications, and current challenges, this study intends to provide structured guidance and robust framework for delivering a prospective outlook in the ML-driven protein engineering field.

Protein-Specific Signal Peptides for Mammalian Vector Engineering

Expression of recombinant proteins in mammalian cell factories relies on synthetic assemblies of genetic parts to optimally control flux through the product biosynthetic pathway. In comparison to other genetic part-types, there is a relative paucity of characterized signal peptide components, particularly for mammalian cell contexts. In this study, we describe a toolkit of signal peptide elements, created using bioinformatics-led and synthetic design approaches, that can be utilized to enhance production of biopharmaceutical proteins in Chinese hamster ovary cell factories. We demonstrate, for the first time in a mammalian cell context, that machine learning can be used to predict how discrete signal peptide elements will perform when utilized to drive endoplasmic reticulum (ER) translocation of specific single chain protein products. For more complex molecular formats, such as multichain monoclonal antibodies, we describe how a combination of in silico and targeted design rule-based in vitro testing can be employed to rapidly identify product-specific signal peptide solutions from minimal screening spaces. The utility of this technology is validated by deriving vector designs that increase product titers ≥1.8×, compared to standard industry systems, for a range of products, including a difficult-to-express monoclonal antibody. The availability of a vastly expanded toolbox of characterized signal peptide parts, combined with streamlined in silico/in vitro testing processes, will permit efficient expression vector re-design to maximize titers of both simple and complex protein products.

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.

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.

Improving therapeutic protein secretion in the probiotic yeast Saccharomyces boulardii using a multifactorial engineering approach

The probiotic yeast Saccharomyces boulardii (Sb) is a promising chassis to deliver therapeutic proteins to the gut due to Sb's innate therapeutic properties, resistance to phage and antibiotics, and high protein secretion capacity. To maintain therapeutic efficacy in the context of challenges such as washout, low rates of diffusion, weak target binding, and/or high rates of proteolysis, it is desirable to engineer Sb strains with enhanced levels of protein secretion. In this work, we explored genetic modifications in both cis- (i.e. to the expression cassette of the secreted protein) and trans- (i.e. to the Sb genome) that enhance Sb's ability to secrete proteins, taking a Clostridioides difficile Toxin A neutralizing peptide (NPA) as our model therapeutic. First, by modulating the copy number of the NPA expression cassette, we found NPA concentrations in the supernatant could be varied by sixfold (76-458 mg/L) in microbioreactor fermentations. In the context of high NPA copy number, we found a previously-developed collection of native and synthetic secretion signals could further tune NPA secretion between 121 and 463 mg/L. Then, guided by prior knowledge of S. cerevisiae's secretion mechanisms, we generated a library of homozygous single gene deletion strains, the most productive of which achieved 2297 mg/L secretory production of NPA. We then expanded on this library by performing combinatorial gene deletions, supplemented by proteomics experiments. We ultimately constructed a quadruple protease-deficient Sb strain that produces 5045 mg/L secretory NPA, an improvement of > tenfold over wild-type Sb. Overall, this work systematically explores a broad collection of engineering strategies to improve protein secretion in Sb and highlights the ability of proteomics to highlight under-explored mediators of this process. In doing so, we created a set of probiotic strains that are capable of delivering a wide range of protein titers and therefore furthers the ability of Sb to deliver therapeutics to the gut and other settings to which it is adapted.

Signal Peptide Efficiency: From High-Throughput Data to Prediction and Explanation

The passage of proteins across biological membranes via the general secretory (Sec) pathway is a universally conserved process with critical functions in cell physiology and important industrial applications. Proteins are directed into the Sec pathway by a signal peptide at their N-terminus. Estimating the impact of physicochemical signal peptide features on protein secretion levels has not been achieved so far, partially due to the extreme sequence variability of signal peptides. To elucidate relevant features of the signal peptide sequence that influence secretion efficiency, an evaluation of ∼12,000 different designed signal peptides was performed using a novel miniaturized high-throughput assay. The results were used to train a machine learning model, and a post-hoc explanation of the model is provided. By describing each signal peptide with a selection of 156 physicochemical features, it is now possible to both quantify feature importance and predict the protein secretion levels directed by each signal peptide. Our analyses allow the detection and explanation of the relevant signal peptide features influencing the efficiency of protein secretion, generating a versatile tool for the de novo design and in silico evaluation of signal peptides.

Funneling modulatory peptide design with generative models: Discovery and characterization of disruptors of calcineurin protein-protein interactions

Design of peptide binders is an attractive strategy for targeting "undruggable" protein-protein interfaces. Current design protocols rely on the extraction of an initial sequence from one known protein interactor of the target protein, followed by in-silico or in-vitro mutagenesis-based optimization of its binding affinity. Wet lab protocols can explore only a minor portion of the vast sequence space and cannot efficiently screen for other desirable properties such as high specificity and low toxicity, while in-silico design requires intensive computational resources and often relies on simplified binding models. Yet, for a multivalent protein target, dozens to hundreds of natural protein partners already exist in the cellular environment. Here, we describe a peptide design protocol that harnesses this diversity via a machine learning generative model. After identifying putative natural binding fragments by literature and homology search, a compositional Restricted Boltzmann Machine is trained and sampled to yield hundreds of diverse candidate peptides. The latter are further filtered via flexible molecular docking and an in-vitro microchip-based binding assay. We validate and test our protocol on calcineurin, a calcium-dependent protein phosphatase involved in various cellular pathways in health and disease. In a single screening round, we identified multiple 16-length peptides with up to six mutations from their closest natural sequence that successfully interfere with the binding of calcineurin to its substrates. In summary, integrating protein interaction and sequence databases, generative modeling, molecular docking and interaction assays enables the discovery of novel protein-protein interaction modulators.

Conformal prediction under feedback covariate shift for biomolecular design

Many applications of machine-learning methods involve an iterative protocol in which data are collected, a model is trained, and then outputs of that model are used to choose what data to consider next. For example, a data-driven approach for designing proteins is to train a regression model to predict the fitness of protein sequences and then use it to propose new sequences believed to exhibit greater fitness than observed in the training data. Since validating designed sequences in the wet laboratory is typically costly, it is important to quantify the uncertainty in the model's predictions. This is challenging because of a characteristic type of distribution shift between the training and test data that arises in the design setting-one in which the training and test data are statistically dependent, as the latter is chosen based on the former. Consequently, the model's error on the test data-that is, the designed sequences-has an unknown and possibly complex relationship with its error on the training data. We introduce a method to construct confidence sets for predictions in such settings, which account for the dependence between the training and test data. The confidence sets we construct have finite-sample guarantees that hold for any regression model, even when it is used to choose the test-time input distribution. As a motivating use case, we use real datasets to demonstrate how our method quantifies uncertainty for the predicted fitness of designed proteins and can therefore be used to select design algorithms that achieve acceptable tradeoffs between high predicted fitness and low predictive uncertainty.

SignalP 6.0 predicts all five types of signal peptides using protein language models

Signal peptides (SPs) are short amino acid sequences that control protein secretion and translocation in all living organisms. SPs can be predicted from sequence data, but existing algorithms are unable to detect all known types of SPs. We introduce SignalP 6.0, a machine learning model that detects all five SP types and is applicable to metagenomic data.

Deep generative models for peptide design

Computers can already be programmed for superhuman pattern recognition of images and text. For machines to discover novel molecules, they must first be trained to sort through the many characteristics of molecules and determine which properties should be retained, suppressed, or enhanced to optimize functions of interest. Machines need to be able to understand, read, write, and eventually create new molecules. Today, this creative process relies on deep generative models, which have gained popularity since powerful deep neural networks were introduced to generative model frameworks. In recent years, they have demonstrated excellent ability to model complex distribution of real-word data (e.g., images, audio, text, molecules, and biological sequences). Deep generative models can generate data beyond those provided in training samples, thus yielding an efficient and rapid tool for exploring the massive search space of high-dimensional data such as DNA/protein sequences and facilitating the design of biomolecules with desired functions. Here, we review the emerging field of deep generative models applied to peptide science. In particular, we discuss several popular deep generative model frameworks as well as their applications to generate peptides with various kinds of properties (e.g., antimicrobial, anticancer, cell penetration, etc). We conclude our review with a discussion of current limitations and future perspectives in this emerging field.

Proteome-Scale Screening to Identify High-Expression Signal Peptides with Minimal N-Terminus Biases via Yeast Display

Signal peptides are critical for the efficient expression and routing of extracellular and secreted proteins. Most protein production and screening technologies rely upon a relatively small set of signal peptides. Despite their central role in biotechnology, there are limited studies comprehensively examining the interplay between signal peptides and expressed protein sequences. Here, we describe a high-throughput method to screen novel signal peptides that maintain a high degree of surface expression across a range of protein scaffolds with highly variable N-termini. We find that the canonical signal peptide used in yeast surface display, derived from Aga2p, fails to achieve high surface expression for 42.5% of constructs containing diverse N-termini. To circumvent this, we have identified two novel signal peptides derived from endogenous yeast proteins, SRL1 and KISH, which are highly tolerant to diverse N-terminal sequences. This pipeline can be used to expand our understanding of signal peptide function, identify improved signal peptides for protein expression, and refine the computational tools used for signal peptide prediction.

Machine learning to navigate fitness landscapes for protein engineering

Machine learning (ML) is revolutionizing our ability to understand and predict the complex relationships between protein sequence, structure, and function. Predictive sequence-function models are enabling protein engineers to efficiently search the sequence space for useful proteins with broad applications in biotechnology. In this review, we highlight the recent advances in applying ML to protein engineering. We discuss supervised learning methods that infer the sequence-function mapping from experimental data and new sequence representation strategies for data-efficient modeling. We then describe the various ways in which ML can be incorporated into protein engineering workflows, including purely in silico searches, ML-assisted directed evolution, and generative models that can learn the underlying distribution of the protein function in a sequence space. ML-driven protein engineering will become increasingly powerful with continued advances in high-throughput data generation, data science, and deep learning.

Recent Advances in the Prediction of Subcellular Localization of Proteins and Related Topics

Prediction of subcellular localization of proteins from their amino acid sequences has a long history in bioinformatics and is still actively developing, incorporating the latest advances in machine learning and proteomics. Notably, deep learning-based methods for natural language processing have made great contributions. Here, we review recent advances in the field as well as its related fields, such as subcellular proteomics and the prediction/recognition of subcellular localization from image data.

Protein design via deep learning

Proteins with desired functions and properties are important in fields like nanotechnology and biomedicine. De novo protein design enables the production of previously unseen proteins from the ground up and is believed as a key point for handling real social challenges. Recent introduction of deep learning into design methods exhibits a transformative influence and is expected to represent a promising and exciting future direction. In this review, we retrospect the major aspects of current advances in deep-learning-based design procedures and illustrate their novelty in comparison with conventional knowledge-based approaches through noticeable cases. We not only describe deep learning developments in structure-based protein design and direct sequence design, but also highlight recent applications of deep reinforcement learning in protein design. The future perspectives on design goals, challenges and opportunities are also comprehensively discussed.

SignalP 6.0 predicts all five types of signal peptides using protein language models

Signal peptides (SPs) are short amino acid sequences that control protein secretion and translocation in all living organisms. SPs can be predicted from sequence data, but existing algorithms are unable to detect all known types of SPs. We introduce SignalP 6.0, a machine learning model that detects all five SP types and is applicable to metagenomic data.

A new version of SignalP predicts all types of signal peptides.

Progress and challenges for the machine learning-based design of fit-for-purpose monoclonal antibodies

Although the therapeutic efficacy and commercial success of monoclonal antibodies (mAbs) are tremendous, the design and discovery of new candidates remain a time and cost-intensive endeavor. In this regard, progress in the generation of data describing antigen binding and developability, computational methodology, and artificial intelligence may pave the way for a new era of in silico on-demand immunotherapeutics design and discovery. Here, we argue that the main necessary machine learning (ML) components for an in silico mAb sequence generator are: understanding of the rules of mAb-antigen binding, capacity to modularly combine mAb design parameters, and algorithms for unconstrained parameter-driven in silico mAb sequence synthesis. We review the current progress toward the realization of these necessary components and discuss the challenges that must be overcome to allow the on-demand ML-based discovery and design of fit-for-purpose mAb therapeutic candidates.

Learning Strategies in Protein Directed Evolution

Synthetic biology is a fast-evolving research field that combines biology and engineering principles to develop new biological systems for medical, pharmacological, and industrial applications. Synthetic biologists use iterative "design, build, test, and learn" cycles to efficiently engineer genetic systems that are reliable, reproducible, and predictable. Protein engineering by directed evolution can benefit from such a systematic engineering approach for various reasons. Learning can be carried out before starting, throughout or after finalizing a directed evolution project. Computational tools, bioinformatics, and scanning mutagenesis methods can be excellent starting points, while molecular dynamics simulations and other strategies can guide engineering efforts. Similarly, studying protein intermediates along evolutionary pathways offers fascinating insights into the molecular mechanisms shaped by evolution. The learning step of the cycle is not only crucial for proteins or enzymes that are not suitable for high-throughput screening or selection systems, but it is also valuable for any platform that can generate a large amount of data that can be aided by machine learning algorithms. The main challenge in protein engineering is to predict the effect of a single mutation on one functional parameter-to say nothing of several mutations on multiple parameters. This is largely due to nonadditive mutational interactions, known as epistatic effects-beneficial mutations present in a genetic background may not be beneficial in another genetic background. In this work, we provide an overview of experimental and computational strategies that can guide the user to learn protein function at different stages in a directed evolution project. We also discuss how epistatic effects can influence the success of directed evolution projects. Since machine learning is gaining momentum in protein engineering and the field is becoming more interdisciplinary thanks to collaboration between mathematicians, computational scientists, engineers, molecular biologists, and chemists, we provide a general workflow that familiarizes nonexperts with the basic concepts, dataset requirements, learning approaches, model capabilities and performance metrics of this intriguing area. Finally, we also provide some practical recommendations on how machine learning can harness epistatic effects for engineering proteins in an "outside-the-box" way.

A roadmap for metagenomic enzyme discovery

Covering: up to 2021Metagenomics has yielded massive amounts of sequencing data offering a glimpse into the biosynthetic potential of the uncultivated microbial majority. While genome-resolved information about microbial communities from nearly every environment on earth is now available, the ability to accurately predict biocatalytic functions directly from sequencing data remains challenging. Compared to primary metabolic pathways, enzymes involved in secondary metabolism often catalyze specialized reactions with diverse substrates, making these pathways rich resources for the discovery of new enzymology. To date, functional insights gained from studies on environmental DNA (eDNA) have largely relied on PCR- or activity-based screening of eDNA fragments cloned in fosmid or cosmid libraries. As an alternative, shotgun metagenomics holds underexplored potential for the discovery of new enzymes directly from eDNA by avoiding common biases introduced through PCR- or activity-guided functional metagenomics workflows. However, inferring new enzyme functions directly from eDNA is similar to searching for a 'needle in a haystack' without direct links between genotype and phenotype. The goal of this review is to provide a roadmap to navigate shotgun metagenomic sequencing data and identify new candidate biosynthetic enzymes. We cover both computational and experimental strategies to mine metagenomes and explore protein sequence space with a spotlight on natural product biosynthesis. Specifically, we compare in silico methods for enzyme discovery including phylogenetics, sequence similarity networks, genomic context, 3D structure-based approaches, and machine learning techniques. We also discuss various experimental strategies to test computational predictions including heterologous expression and screening. Finally, we provide an outlook for future directions in the field with an emphasis on meta-omics, single-cell genomics, cell-free expression systems, and sequence-independent methods.

Intelligent host engineering for metabolic flux optimisation in biotechnology

Optimising the function of a protein of length N amino acids by directed evolution involves navigating a 'search space' of possible sequences of some 20N. Optimising the expression levels of P proteins that materially affect host performance, each of which might also take 20 (logarithmically spaced) values, implies a similar search space of 20P. In this combinatorial sense, then, the problems of directed protein evolution and of host engineering are broadly equivalent. In practice, however, they have different means for avoiding the inevitable difficulties of implementation. The spare capacity exhibited in metabolic networks implies that host engineering may admit substantial increases in flux to targets of interest. Thus, we rehearse the relevant issues for those wishing to understand and exploit those modern genome-wide host engineering tools and thinking that have been designed and developed to optimise fluxes towards desirable products in biotechnological processes, with a focus on microbial systems. The aim throughput is 'making such biology predictable'. Strategies have been aimed at both transcription and translation, especially for regulatory processes that can affect multiple targets. However, because there is a limit on how much protein a cell can produce, increasing kcat in selected targets may be a better strategy than increasing protein expression levels for optimal host engineering.

Protein Design with Deep Learning

Computational Protein Design (CPD) has produced impressive results for engineering new proteins, resulting in a wide variety of applications. In the past few years, various efforts have aimed at replacing or improving existing design methods using Deep Learning technology to leverage the amount of publicly available protein data. Deep Learning (DL) is a very powerful tool to extract patterns from raw data, provided that data are formatted as mathematical objects and the architecture processing them is well suited to the targeted problem. In the case of protein data, specific representations are needed for both the amino acid sequence and the protein structure in order to capture respectively 1D and 3D information. As no consensus has been reached about the most suitable representations, this review describes the representations used so far, discusses their strengths and weaknesses, and details their associated DL architecture for design and related tasks.

The Transporter-Mediated Cellular Uptake and Efflux of Pharmaceutical Drugs and Biotechnology Products: How and Why Phospholipid Bilayer Transport Is Negligible in Real Biomembranes

Over the years, my colleagues and I have come to realise that the likelihood of pharmaceutical drugs being able to diffuse through whatever unhindered phospholipid bilayer may exist in intact biological membranes in vivo is vanishingly low. This is because (i) most real biomembranes are mostly protein, not lipid, (ii) unlike purely lipid bilayers that can form transient aqueous channels, the high concentrations of proteins serve to stop such activity, (iii) natural evolution long ago selected against transport methods that just let any undesirable products enter a cell, (iv) transporters have now been identified for all kinds of molecules (even water) that were once thought not to require them, (v) many experiments show a massive variation in the uptake of drugs between different cells, tissues, and organisms, that cannot be explained if lipid bilayer transport is significant or if efflux were the only differentiator, and (vi) many experiments that manipulate the expression level of individual transporters as an independent variable demonstrate their role in drug and nutrient uptake (including in cytotoxicity or adverse drug reactions). This makes such transporters valuable both as a means of targeting drugs (not least anti-infectives) to selected cells or tissues and also as drug targets. The same considerations apply to the exploitation of substrate uptake and product efflux transporters in biotechnology. We are also beginning to recognise that transporters are more promiscuous, and antiporter activity is much more widespread, than had been realised, and that such processes are adaptive (i.e., were selected by natural evolution). The purpose of the present review is to summarise the above, and to rehearse and update readers on recent developments. These developments lead us to retain and indeed to strengthen our contention that for transmembrane pharmaceutical drug transport "phospholipid bilayer transport is negligible".

Harnessing the yeast Saccharomyces cerevisiae for the production of fungal secondary metabolites

Fungal secondary metabolites (FSMs) represent a remarkable array of bioactive compounds, with potential applications as pharmaceuticals, nutraceuticals, and agrochemicals. However, these molecules are typically produced only in limited amounts by their native hosts. The native organisms may also be difficult to cultivate and genetically engineer, and some can produce undesirable toxic side-products. Alternatively, recombinant production of fungal bioactives can be engineered into industrial cell factories, such as aspergilli or yeasts, which are well amenable for large-scale manufacturing in submerged fermentations. In this review, we summarize the development of baker's yeast Saccharomyces cerevisiae to produce compounds derived from filamentous fungi and mushrooms. These compounds mainly include polyketides, terpenoids, and amino acid derivatives. We also describe how native biosynthetic pathways can be combined or expanded to produce novel derivatives and new-to-nature compounds. We describe some new approaches for cell factory engineering, such as genome-scale engineering, biosensor-based high-throughput screening, and machine learning, and how these tools have been applied for S. cerevisiae strain improvement. Finally, we prospect the challenges and solutions in further development of yeast cell factories to more efficiently produce FSMs.

Protein sequence design with deep generative models

Protein engineering seeks to identify protein sequences with optimized properties. When guided by machine learning, protein sequence generation methods can draw on prior knowledge and experimental efforts to improve this process. In this review, we highlight recent applications of machine learning to generate protein sequences, focusing on the emerging field of deep generative methods.

Revolutionizing enzyme engineering through artificial intelligence and machine learning

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

A modular two yeast species secretion system for the production and preparative application of unspecific peroxygenases

Fungal unspecific peroxygenases (UPOs) represent an enzyme class catalysing versatile oxyfunctionalisation reactions on a broad substrate scope. They are occurring as secreted, glycosylated proteins bearing a haem-thiolate active site and rely on hydrogen peroxide as the oxygen source. However, their heterologous production in a fast-growing organism suitable for high throughput screening has only succeeded once-enabled by an intensive directed evolution campaign. We developed and applied a modular Golden Gate-based secretion system, allowing the first production of four active UPOs in yeast, their one-step purification and application in an enantioselective conversion on a preparative scale. The Golden Gate setup was designed to be universally applicable and consists of the three module types: i) signal peptides for secretion, ii) UPO genes, and iii) protein tags for purification and split-GFP detection. The modular episomal system is suitable for use in Saccharomyces cerevisiae and was transferred to episomal and chromosomally integrated expression cassettes in Pichia pastoris. Shake flask productions in Pichia pastoris yielded up to 24 mg/L secreted UPO enzyme, which was employed for the preparative scale conversion of a phenethylamine derivative reaching 98.6 % ee. Our results demonstrate a rapid, modular yeast secretion workflow of UPOs yielding preparative scale enantioselective biotransformations.

Data-driven computational protein design

Computational protein design can generate proteins not found in nature that adopt desired structures and perform novel functions. Although proteins could, in theory, be designed with ab initio methods, practical success has come from using large amounts of data that describe the sequences, structures, and functions of existing proteins and their variants. We present recent creative uses of multiple-sequence alignments, protein structures, and high-throughput functional assays in computational protein design. Approaches range from enhancing structure-based design with experimental data to building regression models to training deep neural nets that generate novel sequences. Looking ahead, deep learning will be increasingly important for maximizing the value of data for protein design.

The language of proteins: NLP, machine learning & protein sequences

Natural language processing (NLP) is a field of computer science concerned with automated text and language analysis. In recent years, following a series of breakthroughs in deep and machine learning, NLP methods have shown overwhelming progress. Here, we review the success, promise and pitfalls of applying NLP algorithms to the study of proteins. Proteins, which can be represented as strings of amino-acid letters, are a natural fit to many NLP methods. We explore the conceptual similarities and differences between proteins and language, and review a range of protein-related tasks amenable to machine learning. We present methods for encoding the information of proteins as text and analyzing it with NLP methods, reviewing classic concepts such as bag-of-words, k-mers/n-grams and text search, as well as modern techniques such as word embedding, contextualized embedding, deep learning and neural language models. In particular, we focus on recent innovations such as masked language modeling, self-supervised learning and attention-based models. Finally, we discuss trends and challenges in the intersection of NLP and protein research.

100th Anniversary of Macromolecular Science Viewpoint: Data-Driven Protein Design

The design of synthetic proteins with the desired function is a long-standing goal in biomolecular science, with broad applications in biochemical engineering, agriculture, medicine, and public health. Rational de novo design and experimental directed evolution have achieved remarkable successes but are challenged by the requirement to find functional "needles" in the vast "haystack" of protein sequence space. Data-driven models for fitness landscapes provide a predictive map between protein sequence and function and can prospectively identify functional candidates for experimental testing to greatly improve the efficiency of this search. This Viewpoint reviews the applications of machine learning and, in particular, deep learning as part of data-driven protein engineering platforms. We highlight recent successes, review promising computational methodologies, and provide an outlook on future challenges and opportunities. The article is written for a broad audience comprising both polymer and protein scientists and computer and data scientists interested in an up-to-date review of recent innovations and opportunities in this rapidly evolving field.

Machine Learning Enables Selection of Epistatic Enzyme Mutants for Stability Against Unfolding and Detrimental Aggregation

Machine learning (ML) has pervaded most areas of protein engineering, including stability and stereoselectivity. Using limonene epoxide hydrolase as the model enzyme and innov'SAR as the ML platform, comprising a digital signal process, we achieved high protein robustness that can resist unfolding with concomitant detrimental aggregation. Fourier transform (FT) allows us to take into account the order of the protein sequence and the nonlinear interactions between positions, and thus to grasp epistatic phenomena. The innov'SAR approach is interpolative, extrapolative and makes outside-the-box, predictions not found in other state-of-the-art ML or deep learning approaches. Equally significant is the finding that our approach to ML in the present context, flanked by advanced molecular dynamics simulations, uncovers the connection between epistatic mutational interactions and protein robustness.

Advances in machine learning for directed evolution

Machine learning (ML) can expedite directed evolution by allowing researchers to move expensive experimental screens in silico. Gathering sequence-function data for training ML models, however, can still be costly. In contrast, raw protein sequence data is widely available. Recent advances in ML approaches use protein sequences to augment limited sequence-function data for directed evolution. We highlight contributions in a growing effort to use sequences to reduce or eliminate the amount of sequence-function data needed for effective in silico screening. We also highlight approaches that use ML models trained on sequences to generate new functional sequence diversity, focusing on strategies that use these generative models to efficiently explore vast regions of protein space.

Living Therapeutics: The Next Frontier of Precision Medicine

Modern medicine has long studied the mechanism and impact of pathogenic microbes on human hosts, but has only recently shifted attention toward the complex and vital roles that commensal and probiotic microbes play in both health and dysbiosis. Fueled by an enhanced appreciation of the human-microbe holobiont, the past decade has yielded countless insights and established many new avenues of investigation in this area. In this review, we discuss advances, limitations, and emerging frontiers for microbes as agents of health maintenance, disease prevention, and cure. We highlight the flexibility of microbial therapeutics across disease states, with special consideration for the rational engineering of microbes toward precision medicine outcomes. As the field advances, we anticipate that tools of synthetic biology will be increasingly employed to engineer functional living therapeutics with the potential to address longstanding limitations of traditional drugs.

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

The number of 'small' molecules that may be of interest to chemical biologists - chemical space - is enormous, but the fraction that have ever been made is tiny. Most strategies are discriminative, i.e. have involved 'forward' problems (have molecule, establish properties). However, we normally wish to solve the much harder generative or inverse problem (describe desired properties, find molecule). 'Deep' (machine) learning based on large-scale neural networks underpins technologies such as computer vision, natural language processing, driverless cars, and world-leading performance in games such as Go; it can also be applied to the solution of inverse problems in chemical biology. In particular, recent developments in deep learning admit the in silico generation of candidate molecular structures and the prediction of their properties, thereby allowing one to navigate (bio)chemical space intelligently. These methods are revolutionary but require an understanding of both (bio)chemistry and computer science to be exploited to best advantage. We give a high-level (non-mathematical) background to the deep learning revolution, and set out the crucial issue for chemical biology and informatics as a two-way mapping from the discrete nature of individual molecules to the continuous but high-dimensional latent representation that may best reflect chemical space. A variety of architectures can do this; we focus on a particular type known as variational autoencoders. We then provide some examples of recent successes of these kinds of approach, and a look towards the future.