A deep learning approach to programmable RNA switches

Mechanism-based and data-driven modeling in cell-free synthetic biology

Cell-free systems have emerged as a versatile platform in synthetic biology, finding applications in various areas such as prototyping synthetic circuits, biosensor development, and biomanufacturing. To streamline the prototyping process, cell-free systems often incorporate a modeling step that predicts the outcomes of various experimental scenarios, providing a deeper insight into the underlying mechanisms and functions. There are two recognized approaches for modeling these systems: mechanism-based modeling, which models the underlying reaction mechanisms; and data-driven modeling, which makes predictions based on data without preconceived interactions between system components. In this highlight, we focus on the latest advancements in both modeling approaches for cell-free systems, exploring their potential for the design and optimization of synthetic genetic circuits.

Transfer learning for cross-context prediction of protein expression from 5'UTR sequence

Model-guided DNA sequence design can accelerate the reprogramming of living cells. It allows us to engineer more complex biological systems by removing the need to physically assemble and test each potential design. While mechanistic models of gene expression have seen some success in supporting this goal, data-centric, deep learning-based approaches often provide more accurate predictions. This accuracy, however, comes at a cost - a lack of generalization across genetic and experimental contexts that has limited their wider use outside the context in which they were trained. Here, we address this issue by demonstrating how a simple transfer learning procedure can effectively tune a pre-trained deep learning model to predict protein translation rate from 5' untranslated region (5'UTR) sequence for diverse contexts in Escherichia coli using a small number of new measurements. This allows for important model features learnt from expensive massively parallel reporter assays to be easily transferred to new settings. By releasing our trained deep learning model and complementary calibration procedure, this study acts as a starting point for continually refined model-based sequence design that builds on previous knowledge and future experimental efforts.

A single-vector intersectional AAV strategy for interrogating cellular diversity and brain function

As discovery of cellular diversity in the brain accelerates, so does the need for tools that target cells based on multiple features. Here we developed Conditional Viral Expression by Ribozyme Guided Degradation (ConVERGD), an adeno-associated virus-based, single-construct, intersectional targeting strategy that combines a self-cleaving ribozyme with traditional FLEx switches to deliver molecular cargo to specific neuronal subtypes. ConVERGD offers benefits over existing intersectional expression platforms, such as expanded intersectional targeting with up to five recombinase-based features, accommodation of larger and more complex payloads and a vector that is easy to modify for rapid toolkit expansion. In the present report we employed ConVERGD to characterize an unexplored subpopulation of norepinephrine (NE)-producing neurons within the rodent locus coeruleus that co-express the endogenous opioid gene prodynorphin (Pdyn). These studies showcase ConVERGD as a versatile tool for targeting diverse cell types and reveal Pdyn-expressing NE+ locus coeruleus neurons as a small neuronal subpopulation capable of driving anxiogenic behavioral responses in rodents.

RNA-Based Sensor Systems for Affordable Diagnostics in the Age of Pandemics

In the era of the COVID-19 pandemic, the significance of point-of-care (POC) diagnostic tools has become increasingly vital, driven by the need for quick and precise virus identification. RNA-based sensors, particularly toehold sensors, have emerged as promising candidates for POC detection systems due to their selectivity and sensitivity. Toehold sensors operate by employing an RNA switch that changes the conformation when it binds to a target RNA molecule, resulting in a detectable signal. This review focuses on the development and deployment of RNA-based sensors for POC viral RNA detection with a particular emphasis on toehold sensors. The benefits and limits of toehold sensors are explored, and obstacles and future directions for improving their performance within POC detection systems are presented. The use of RNA-based sensors as a technology for rapid and sensitive detection of viral RNA holds great potential for effectively managing (dealing/coping) with present and future pandemics in resource-constrained settings.

Cas9-assisted biological containment of a genetically engineered human commensal bacterium and genetic elements

Sophisticated gene circuits built by synthetic biology can enable bacteria to sense their environment and respond predictably. Engineered biosensing bacteria outfitted with such circuits can potentially probe the human gut microbiome to prevent, diagnose, or treat disease. To provide robust biocontainment for engineered bacteria, we devised a Cas9-assisted auxotrophic biocontainment system combining thymidine auxotrophy, an Engineered Riboregulator (ER) for controlled gene expression, and a CRISPR Device (CD). The CD prevents the engineered bacteria from acquiring thyA via horizontal gene transfer, which would disrupt the biocontainment system, and inhibits the spread of genetic elements by killing bacteria harboring the gene cassette. This system tunably controlled gene expression in the human gut commensal bacterium Bacteroides thetaiotaomicron, prevented escape from thymidine auxotrophy, and blocked transgene dissemination. These capabilities were validated in vitro and in vivo. This biocontainment system exemplifies a powerful strategy for bringing genetically engineered microorganisms safely into biomedicine.

Machine Learning and Deep Learning in Synthetic Biology: Key Architectures, Applications, and Challenges

Machine learning (ML), particularly deep learning (DL), has made rapid and substantial progress in synthetic biology in recent years. Biotechnological applications of biosystems, including pathways, enzymes, and whole cells, are being probed frequently with time. The intricacy and interconnectedness of biosystems make it challenging to design them with the desired properties. ML and DL have a synergy with synthetic biology. Synthetic biology can be employed to produce large data sets for training models (for instance, by utilizing DNA synthesis), and ML/DL models can be employed to inform design (for example, by generating new parts or advising unrivaled experiments to perform). This potential has recently been brought to light by research at the intersection of engineering biology and ML/DL through achievements like the design of novel biological components, best experimental design, automated analysis of microscopy data, protein structure prediction, and biomolecular implementations of ANNs (Artificial Neural Networks). I have divided this review into three sections. In the first section, I describe predictive potential and basics of ML along with myriad applications in synthetic biology, especially in engineering cells, activity of proteins, and metabolic pathways. In the second section, I describe fundamental DL architectures and their applications in synthetic biology. Finally, I describe different challenges causing hurdles in the progress of ML/DL and synthetic biology along with their solutions.

RNA-based controllers for engineering gene and cell therapies

Engineered RNA-based genetic controllers provide compact, tunable, post-transcriptional gene regulation. As RNA devices are generally small, these devices are portable to DNA and RNA viral vectors. RNA tools have recently expanded to allow reading and editing of endogenous RNAs for profiling and programming of transcriptional states. With their expanded capabilities and highly compact, modular, and programmable nature, RNA-based controllers will support greater safety, efficacy, and performance in gene and cell-based therapies. In this review, we highlight RNA-based controllers and their potential as user-guided and autonomous systems for control of gene and cell-based therapies.

Development of a cell-free toehold switch for hepatitis A virus type I on-site detection

Picornavirus hepatitis A virus (HAV) is a common cause of hepatitis worldwide. It is spread primarily through contaminated food and water or person-to-person contact. HAV I has been identified as the most common type of human HAV infection. Here, we have developed a cell-free toehold switch sensor for HAV I detection. We screened 10 suitable toehold switch sequences using NUPACK software, and the VP1 gene was used as the target gene. The optimal toehold switch sequence was selected by in vivo expression. The best toehold switch concentration was further found to be 20 nM in a cell-free system. 5 nM trigger RNA activated the toehold switch to generate visible green fluorescence. The minimum detection concentration decreased to 1 pM once combined with NASBA. HAV I trigger RNA could be detected accurately with excellent specificity. In addition, the cell-free toehold switch sensor was verified in HAV I entities. The successful construction of the cell-free toehold switch sensor provided a convenient, rapid, and accurate method for HAV I on-site detection, especially in developing countries, without the involvement of expensive facilities and additional professional operators.

Applications of artificial intelligence and machine learning in dynamic pathway engineering

Dynamic pathway engineering aims to build metabolic production systems embedded with intracellular control mechanisms for improved performance. These control systems enable host cells to self-regulate the temporal activity of a production pathway in response to perturbations, using a combination of biosensors and feedback circuits for controlling expression of heterologous enzymes. Pathway design, however, requires assembling together multiple biological parts into suitable circuit architectures, as well as careful calibration of the function of each component. This results in a large design space that is costly to navigate through experimentation alone. Methods from artificial intelligence (AI) and machine learning are gaining increasing attention as tools to accelerate the design cycle, owing to their ability to identify hidden patterns in data and rapidly screen through large collections of designs. In this review, we discuss recent developments in the application of machine learning methods to the design of dynamic pathways and their components. We cover recent successes and offer perspectives for future developments in the field. The integration of AI into metabolic engineering pipelines offers great opportunities to streamline design and discover control systems for improved production of high-value chemicals.

Modularized synthetic biology enabled intelligent biosensors

Biosensors that sense the concentration of a specified target and produce a specific signal output have become important technology for biological analysis. Recently, intelligent biosensors have received great interest due to their adaptability to meet sophisticated demands. Advances in developing standard modules and carriers in synthetic biology have shed light on intelligent biosensors that can implement advanced analytical processing to better accommodate practical applications. This review focuses on intelligent synthetic biology-enabled biosensors (SBBs). First, we illustrate recent progress in intelligent SBBs with the capability of computation, memory storage, and self-calibration. Then, we discuss emerging applications of SBBs in point-of-care testing (POCT) and wearable monitoring. Finally, future perspectives on intelligent SBBs are proposed.

Transformer Architecture and Attention Mechanisms in Genome Data Analysis: A Comprehensive Review

The emergence and rapid development of deep learning, specifically transformer-based architectures and attention mechanisms, have had transformative implications across several domains, including bioinformatics and genome data analysis. The analogous nature of genome sequences to language texts has enabled the application of techniques that have exhibited success in fields ranging from natural language processing to genomic data. This review provides a comprehensive analysis of the most recent advancements in the application of transformer architectures and attention mechanisms to genome and transcriptome data. The focus of this review is on the critical evaluation of these techniques, discussing their advantages and limitations in the context of genome data analysis. With the swift pace of development in deep learning methodologies, it becomes vital to continually assess and reflect on the current standing and future direction of the research. Therefore, this review aims to serve as a timely resource for both seasoned researchers and newcomers, offering a panoramic view of the recent advancements and elucidating the state-of-the-art applications in the field. Furthermore, this review paper serves to highlight potential areas of future investigation by critically evaluating studies from 2019 to 2023, thereby acting as a stepping-stone for further research endeavors.

Leveraging artificial intelligence in the fight against infectious diseases

Despite advances in molecular biology, genetics, computation, and medicinal chemistry, infectious disease remains an ominous threat to public health. Addressing the challenges posed by pathogen outbreaks, pandemics, and antimicrobial resistance will require concerted interdisciplinary efforts. In conjunction with systems and synthetic biology, artificial intelligence (AI) is now leading to rapid progress, expanding anti-infective drug discovery, enhancing our understanding of infection biology, and accelerating the development of diagnostics. In this Review, we discuss approaches for detecting, treating, and understanding infectious diseases, underscoring the progress supported by AI in each case. We suggest future applications of AI and how it might be harnessed to help control infectious disease outbreaks and pandemics.

Generative and predictive neural networks for the design of functional RNA molecules

RNA is a remarkably versatile molecule that has been engineered for applications in therapeutics, diagnostics, and in vivo information-processing systems. However, the complex relationship between the sequence and structural properties of an RNA molecule and its ability to perform specific functions often necessitates extensive experimental screening of candidate sequences. Here we present a generalized neural network architecture that utilizes the sequence and structure of RNA molecules (SANDSTORM) to inform functional predictions. We demonstrate that this approach achieves state-of-the-art performance across several distinct RNA prediction tasks, while learning interpretable abstractions of RNA secondary structure. We paired these predictive models with generative adversarial RNA design networks (GARDN), allowing the generative modelling of novel mRNA 5' untranslated regions and toehold switch riboregulators exhibiting a predetermined fitness. This approach enabled the design of novel toehold switches with a 43-fold increase in experimentally characterized dynamic range compared to those designed using classic thermodynamic algorithms. SANDSTORM and GARDN thus represent powerful new predictive and generative tools for the development of diagnostic and therapeutic RNA molecules with improved function.

Harnessing synthetic biology to enhance ocean health

Ocean health is faltering, its capability for regeneration and renewal being eroded by a steady pulse of anthropomorphic impacts. Plastic waste has infiltrated all ocean biomes, climate change threatens coral reefs with extinction, and eutrophication has unleashed vast algal blooms. In the face of these challenges, synthetic biology approaches may hold untapped solutions to mitigate adverse effects, repair ecosystems, and put us on a path towards sustainable stewardship of our planet. Leveraging synthetic biology tools would enable innovative engineering approaches to augment the natural adaptive capacity of ocean biological systems to cope with the swiftness of human-induced change. Here, we present a framework for developing synthetic biology solutions for the challenges of plastic pollution, coral bleaching, and harmful algal blooms.

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.

A blueprint for a synthetic genetic feedback optimizer

Biomolecular control enables leveraging cells as biomanufacturing factories. Despite recent advancements, we currently lack genetically encoded modules that can be deployed to dynamically fine-tune and optimize cellular performance. Here, we address this shortcoming by presenting the blueprint of a genetic feedback module to optimize a broadly defined performance metric by adjusting the production and decay rate of a (set of) regulator species. We demonstrate that the optimizer can be implemented by combining available synthetic biology parts and components, and that it can be readily integrated with existing pathways and genetically encoded biosensors to ensure its successful deployment in a variety of settings. We further illustrate that the optimizer successfully locates and tracks the optimum in diverse contexts when relying on mass action kinetics-based dynamics and parameter values typical in Escherichia coli.

Deep learning for optimization of protein expression

Recent progress in high-throughput DNA synthesis and sequencing has enabled the development of massively parallel reporter assays for strain characterization. These datasets map a large number of DNA sequences to protein expression levels, sparking increased interest in data-driven methods for sequence-to-expression modeling. Here, we highlight advances in deep learning models of protein expression and their potential for optimizing strains engineered to produce recombinant proteins. We review recent works that built highly accurate models and discuss challenges that hinder adoption by end users. There is a need to better align this technology with the constraints encountered in strain engineering, particularly the cost of acquiring large amounts of data and the requirement for interpretable models that generalize beyond the training data. Overcoming these barriers will help to incentivize academic and industrial laboratories to tap into a new era of data-centric strain engineering.

"Toehold Switches; a foothold for Synthetic Biology"

Toehold switches are de novo designed riboregulators that contain two RNA components interacting through linear-linear RNA interactions, regulating the gene expression. These are highly versatile, exhibit excellent orthogonality, wide dynamic range, and are highly programmable, so can be used for various applications in synthetic biology. In this review, we summarized and discussed the design characteristics and benefits of toehold switch riboregulators over conventional riboregulators. We also discussed applications and recent advancements of toehold switch riboregulators in various fields like gene editing, DNA nanotechnology, translational repression, and diagnostics (detection of microRNAs and some pathogens). Toehold switches, therefore, furnished advancement in synthetic biology applications in various fields with their prominent features.

Ultra-high throughput mapping of genetic design space

Massively parallel genetic screens have been used to map sequence-to-function relationships for a variety of genetic elements. However, because these approaches only interrogate short sequences, it remains challenging to perform high throughput (HT) assays on constructs containing combinations of sequence elements arranged across multi-kb length scales. Overcoming this barrier could accelerate synthetic biology; by screening diverse gene circuit designs, "composition-to-function" mappings could be created that reveal genetic part composability rules and enable rapid identification of behavior-optimized variants. Here, we introduce CLASSIC, a generalizable genetic screening platform that combines long- and short-read next-generation sequencing (NGS) modalities to quantitatively assess pooled libraries of DNA constructs of arbitrary length. We show that CLASSIC can measure expression profiles of >10 ⁵ drug-inducible gene circuit designs (ranging from 6-9 kb) in a single experiment in human cells. Using statistical inference and machine learning (ML) approaches, we demonstrate that data obtained with CLASSIC enables predictive modeling of an entire circuit design landscape, offering critical insight into underlying design principles. Our work shows that by expanding the throughput and understanding gained with each design-build-test-learn (DBTL) cycle, CLASSIC dramatically augments the pace and scale of synthetic biology and establishes an experimental basis for data-driven design of complex genetic systems.

Autocatalytic base editing for RNA-responsive translational control

Genetic circuits that control transgene expression in response to pre-defined transcriptional cues would enable the development of smart therapeutics. To this end, here we engineer programmable single-transcript RNA sensors in which adenosine deaminases acting on RNA (ADARs) autocatalytically convert target hybridization into a translational output. Dubbed DART VADAR (Detection and Amplification of RNA Triggers via ADAR), our system amplifies the signal from editing by endogenous ADAR through a positive feedback loop. Amplification is mediated by the expression of a hyperactive, minimal ADAR variant and its recruitment to the edit site via an orthogonal RNA targeting mechanism. This topology confers high dynamic range, low background, minimal off-target effects, and a small genetic footprint. We leverage DART VADAR to detect single nucleotide polymorphisms and modulate translation in response to endogenous transcript levels in mammalian cells.

Engineered Living Materials For Sustainability

Recent advances in synthetic biology and materials science have given rise to a new form of materials, namely engineered living materials (ELMs), which are composed of living matter or cell communities embedded in self-regenerating matrices of their own or artificial scaffolds. Like natural materials such as bone, wood, and skin, ELMs, which possess the functional capabilities of living organisms, can grow, self-organize, and self-repair when needed. They also spontaneously perform programmed biological functions upon sensing external cues. Currently, ELMs show promise for green energy production, bioremediation, disease treatment, and fabricating advanced smart materials. This review first introduces the dynamic features of natural living systems and their potential for developing novel materials. We then summarize the recent research progress on living materials and emerging design strategies from both synthetic biology and materials science perspectives. Finally, we discuss the positive impacts of living materials on promoting sustainability and key future research directions.

Predicting DNA kinetics with a truncated continuous-time Markov chain method

Predicting the kinetics of reactions involving nucleic acid strands is a fundamental task in biology and biotechnology. Reaction kinetics can be modeled as an elementary step continuous-time Markov chain, where states correspond to secondary structures and transitions correspond to base pair formation and breakage. Since the number of states in the Markov chain could be large, rates are determined by estimating the mean first passage time from sampled trajectories. As a result, the cost of kinetic predictions becomes prohibitively expensive for rare events with extremely long trajectories. Also problematic are scenarios where multiple predictions are needed for the same reaction, e.g., under different environmental conditions, or when calibrating model parameters, because a new set of trajectories is needed multiple times. We propose a new method, called pathway elaboration, to handle these scenarios. Pathway elaboration builds a truncated continuous-time Markov chain through both biased and unbiased sampling. The resulting Markov chain has moderate state space size, so matrix methods can efficiently compute reaction rates, even for rare events. Also the transition rates of the truncated Markov chain can easily be adapted when model or environmental parameters are perturbed, making model calibration feasible. We illustrate the utility of pathway elaboration on toehold-mediated strand displacement reactions, show that it well-approximates trajectory-based predictions of unbiased elementary step models on a wide range of reaction types for which such predictions are feasible, and demonstrate that it performs better than alternative truncation-based approaches that are applicable for mean first passage time estimation. Finally, in a small study, we use pathway elaboration to optimize the Metropolis kinetic model of Multistrand, an elementary step simulator, showing that the optimized parameters greatly improve reaction rate predictions. Our framework and dataset are available at https://github.com/DNA-and-Natural-Algorithms-Group/PathwayElaboration.

A Novel Single Vector Intersectional AAV Strategy for Interrogating Cellular Diversity and Brain Function

As the discovery of cellular diversity in the brain accelerates, so does the need for functional tools that target cells based on multiple features, such as gene expression and projection target. By selectively driving recombinase expression in a feature-specific manner, one can utilize intersectional strategies to conditionally promote payload expression only where multiple features overlap. We developed Conditional Viral Expression by Ribozyme Guided Degradation (ConVERGD), a single-construct intersectional targeting strategy that combines a self-cleaving ribozyme with traditional FLEx switches. ConVERGD offers benefits over existing platforms, such as expanded intersectionality, the ability to accommodate larger and more complex payloads, and a vector design that is easily modified to better facilitate rapid toolkit expansion. To demonstrate its utility for interrogating neural circuitry, we employed ConVERGD to target an unexplored subpopulation of norepinephrine (NE)-producing neurons within the rodent locus coeruleus (LC) identified via single-cell transcriptomic profiling to co-express the stress-related endogenous opioid gene prodynorphin (Pdyn). These studies showcase ConVERGD as a versatile tool for targeting diverse cell types and reveal Pdyn-expressing NE+ LC neurons as a small neuronal subpopulation capable of driving anxiogenic behavioral responses in rodents.

Genetic switches based on nucleic acid strand displacement

Toehold-mediated strand displacement (TMSD) is an isothermal switching process that enables the sequence-programmable and reversible conversion of DNA or RNA strands between single- and double-stranded conformations or other secondary structures. TMSD processes have already found widespread application in DNA nanotechnology, where they are used to drive DNA-based molecular devices or for the realization of synthetic biochemical computing circuits. Recently, researchers have started to employ TMSD also for the control of RNA-based gene regulatory processes in vivo, in particular in the context of synthetic riboregulators and conditional guide RNAs for CRISPR/Cas. Here, we provide a review over recent developments in this emerging field and discuss the opportunities and challenges for such systems in in vivo applications.

Precision engineering of biological function with large-scale measurements and machine learning

As synthetic biology expands and accelerates into real-world applications, methods for quantitatively and precisely engineering biological function become increasingly relevant. This is particularly true for applications that require programmed sensing to dynamically regulate gene expression in response to stimuli. However, few methods have been described that can engineer biological sensing with any level of quantitative precision. Here, we present two complementary methods for precision engineering of genetic sensors: in silico selection and machine-learning-enabled forward engineering. Both methods use a large-scale genotype-phenotype dataset to identify DNA sequences that encode sensors with quantitatively specified dose response. First, we show that in silico selection can be used to engineer sensors with a wide range of dose-response curves. To demonstrate in silico selection for precise, multi-objective engineering, we simultaneously tune a genetic sensor's sensitivity (EC50) and saturating output to meet quantitative specifications. In addition, we engineer sensors with inverted dose-response and specified EC50. Second, we demonstrate a machine-learning-enabled approach to predictively engineer genetic sensors with mutation combinations that are not present in the large-scale dataset. We show that the interpretable machine learning results can be combined with a biophysical model to engineer sensors with improved inverted dose-response curves.

Nucleic acid strand displacement - from DNA nanotechnology to translational regulation

Nucleic acid strand displacement reactions involve the competition of two or more DNA or RNA strands of similar sequence for binding to a complementary strand, and facilitate the isothermal replacement of an incumbent strand by an invader. The process can be biased by augmenting the duplex comprising the incumbent with a single-stranded extension, which can act as a toehold for a complementary invader. The toehold gives the invader a thermodynamic advantage over the incumbent, and can be programmed as a unique label to activate a specific strand displacement process. Toehold-mediated strand displacement processes have been extensively utilized for the operation of DNA-based molecular machines and devices as well as for the design of DNA-based chemical reaction networks. More recently, principles developed initially in the context of DNA nanotechnology have been applied for the de novo design of gene regulatory switches that can operate inside living cells. The article specifically focuses on the design of RNA-based translational regulators termed toehold switches. Toehold switches utilize toehold-mediated strand invasion to either activate or repress translation of an mRNA in response to the binding of a trigger RNA molecule. The basic operation principles of toehold switches will be discussed as well as their applications in sensing and biocomputing. Finally, strategies for their optimization will be described as well as challenges for their operation in vivo.

Design and Analysis of Massively Parallel Reporter Assays Using FORECAST

Machine learning is revolutionizing molecular biology and bioengineering by providing powerful insights and predictions. Massively parallel reporter assays (MPRAs) have emerged as a particularly valuable class of high-throughput technique to support such algorithms. MPRAs enable the simultaneous characterization of thousands or even millions of genetic constructs and provide the large amounts of data needed to train models. However, while the scale of this approach is impressive, the design of effective MPRA experiments is challenging due to the many factors that can be varied and the difficulty in predicting how these will impact the quality and quantity of data obtained. Here, we present a computational tool called FORECAST, which can simulate MPRA experiments based on fluorescence-activated cell sorting and subsequent sequencing (commonly referred to as Flow-seq or Sort-seq experiments), as well as carry out rigorous statistical estimation of construct performance from this type of experimental data. FORECAST can be used to develop workflows to aid the design of MPRA experiments and reanalyze existing MPRA data sets.

Genetically encoded biosensors for microbial synthetic biology: From conceptual frameworks to practical applications

Genetically encoded biosensors are the vital components of synthetic biology and metabolic engineering, as they are regarded as powerful devices for the dynamic control of genotype metabolism and evolution/screening of desirable phenotypes. This review summarized the recent advances in the construction and applications of different genetically encoded biosensors, including fluorescent protein-based biosensors, nucleic acid-based biosensors, allosteric transcription factor-based biosensors and two-component system-based biosensors. First, the construction frameworks of these biosensors were outlined. Then, the recent progress of biosensor applications in creating versatile microbial cell factories for the bioproduction of high-value chemicals was summarized. Finally, the challenges and prospects for constructing robust and sophisticated biosensors were discussed. This review provided theoretical guidance for constructing genetically encoded biosensors to create desirable microbial cell factories for sustainable bioproduction.

Synthetic Biology Meets Machine Learning

This chapter outlines the myriad applications of machine learning (ML) in synthetic biology, specifically in engineering cell and protein activity, and metabolic pathways. Though by no means comprehensive, the chapter highlights several prominent computational tools applied in the field and their potential use cases. The examples detailed reinforce how ML algorithms can enhance synthetic biology research by providing data-driven insights into the behavior of living systems, even without detailed knowledge of their underlying mechanisms. By doing so, ML promises to increase the efficiency of research projects by modeling hypotheses in silico that can then be tested through experiments. While challenges related to training dataset generation and computational costs remain, ongoing improvements in ML tools are paving the way for smarter and more streamlined synthetic biology workflows that can be readily employed to address grand challenges across manufacturing, medicine, engineering, agriculture, and beyond.

Accuracy and data efficiency in deep learning models of protein expression

Synthetic biology often involves engineering microbial strains to express high-value proteins. Thanks to progress in rapid DNA synthesis and sequencing, deep learning has emerged as a promising approach to build sequence-to-expression models for strain optimization. But such models need large and costly training data that create steep entry barriers for many laboratories. Here we study the relation between accuracy and data efficiency in an atlas of machine learning models trained on datasets of varied size and sequence diversity. We show that deep learning can achieve good prediction accuracy with much smaller datasets than previously thought. We demonstrate that controlled sequence diversity leads to substantial gains in data efficiency and employed Explainable AI to show that convolutional neural networks can finely discriminate between input DNA sequences. Our results provide guidelines for designing genotype-phenotype screens that balance cost and quality of training data, thus helping promote the wider adoption of deep learning in the biotechnology sector.

Oligonucleotide separation techniques for purification and analysis: What can we learn for today's tasks?

Nucleic acids are the blueprint of life. They are not only the construction plan of the single cell or higher associations of them, but also necessary for function, communication and regulation. Due to the pandemic, the attention shifted in particular to their therapeutic potential as a vaccine. As pharmaceutical oligonucleotides are unique in terms of their stability and application, special delivery systems were also considered. Oligonucleotide production systems can vary and depend on the feasibility, availability, price and intended application. To achieve good purity, reliable results and match the strict specifications in the pharmaceutical industry, the separation of oligonucleotides is always essential. Besides the separation required for production, additional and specifically different separation techniques are needed for analysis to determine if the product complies with the designated specifications. After a short introduction to ribonucleic acids (RNAs), messenger RNA vaccines, and their production and delivery systems, an overview regarding separation techniques will be provided. This not only emphasises electrophoretic separations but also includes spin columns, extractions, precipitations, magnetic nanoparticles and several chromatographic separation principles, such as ion exchange chromatography, ion-pair reversed-phase, size exclusion and affinity.

Artificial Immune Cell, AI-cell, a New Tool to Predict Interferon Production by Peripheral Blood Monocytes in Response to Nucleic Acid Nanoparticles

Nucleic acid nanoparticles, or NANPs, rationally designed to communicate with the human immune system, can offer innovative therapeutic strategies to overcome the limitations of traditional nucleic acid therapies. Each set of NANPs is unique in their architectural parameters and physicochemical properties, which together with the type of delivery vehicles determine the kind and the magnitude of their immune response. Currently, there are no predictive tools that would reliably guide the design of NANPs to the desired immunological outcome, a step crucial for the success of personalized therapies. Through a systematic approach investigating physicochemical and immunological profiles of a comprehensive panel of various NANPs, the research team developes and experimentally validates a computational model based on the transformer architecture able to predict the immune activities of NANPs. It is anticipated that the freely accessible computational tool that is called an "artificial immune cell," or AI-cell, will aid in addressing the current critical public health challenges related to safety criteria of nucleic acid therapies in a timely manner and promote the development of novel biomedical tools.

Multi-modality artificial intelligence in digital pathology

In common medical procedures, the time-consuming and expensive nature of obtaining test results plagues doctors and patients. Digital pathology research allows using computational technologies to manage data, presenting an opportunity to improve the efficiency of diagnosis and treatment. Artificial intelligence (AI) has a great advantage in the data analytics phase. Extensive research has shown that AI algorithms can produce more up-to-date and standardized conclusions for whole slide images. In conjunction with the development of high-throughput sequencing technologies, algorithms can integrate and analyze data from multiple modalities to explore the correspondence between morphological features and gene expression. This review investigates using the most popular image data, hematoxylin-eosin stained tissue slide images, to find a strategic solution for the imbalance of healthcare resources. The article focuses on the role that the development of deep learning technology has in assisting doctors' work and discusses the opportunities and challenges of AI.

Engineered living materials (ELMs) design: From function allocation to dynamic behavior modulation

Natural materials possess many distinctive "living" attributes, such as self-growth, self-healing, environmental responsiveness, and evolvability, that are beyond the reach of many existing synthetic materials. The emerging field of engineered living materials (ELMs) takes inspiration from nature and harnesses engineered living systems to produce dynamic and responsive materials with genetically programmable functionalities. Here, we identify and review two main directions for the rational design of ELMs: first, engineering of living materials with enhanced performances by incorporating functional material modules, including engineered biological building blocks (proteins, polysaccharides, and nucleic acids) or well-defined artificial materials; second, engineering of smart ELMs that can sense and respond to their surroundings by programming dynamic cellular behaviors regulated via cell-cell or cell-environment interactions. We next discuss the strengths and challenges of current ELMs and conclude by providing a perspective of future directions in this promising area.

Deep Learning Concepts and Applications for Synthetic Biology

Synthetic biology has a natural synergy with deep learning. It can be used to generate large data sets to train models, for example by using DNA synthesis, and deep learning models can be used to inform design, such as by generating novel parts or suggesting optimal experiments to conduct. Recently, research at the interface of engineering biology and deep learning has highlighted this potential through successes including the design of novel biological parts, protein structure prediction, automated analysis of microscopy data, optimal experimental design, and biomolecular implementations of artificial neural networks. In this review, we present an overview of synthetic biology-relevant classes of data and deep learning architectures. We also highlight emerging studies in synthetic biology that capitalize on deep learning to enable novel understanding and design, and discuss challenges and future opportunities in this space.

Artificial Intelligence (AI) and Internet of Medical Things (IoMT) Assisted Biomedical Systems for Intelligent Healthcare

Artificial intelligence (AI) is a modern approach based on computer science that develops programs and algorithms to make devices intelligent and efficient for performing tasks that usually require skilled human intelligence. AI involves various subsets, including machine learning (ML), deep learning (DL), conventional neural networks, fuzzy logic, and speech recognition, with unique capabilities and functionalities that can improve the performances of modern medical sciences. Such intelligent systems simplify human intervention in clinical diagnosis, medical imaging, and decision-making ability. In the same era, the Internet of Medical Things (IoMT) emerges as a next-generation bio-analytical tool that combines network-linked biomedical devices with a software application for advancing human health. In this review, we discuss the importance of AI in improving the capabilities of IoMT and point-of-care (POC) devices used in advanced healthcare sectors such as cardiac measurement, cancer diagnosis, and diabetes management. The role of AI in supporting advanced robotic surgeries developed for advanced biomedical applications is also discussed in this article. The position and importance of AI in improving the functionality, detection accuracy, decision-making ability of IoMT devices, and evaluation of associated risks assessment is discussed carefully and critically in this review. This review also encompasses the technological and engineering challenges and prospects for AI-based cloud-integrated personalized IoMT devices for designing efficient POC biomedical systems suitable for next-generation intelligent healthcare.

Interpretable modeling of genotype-phenotype landscapes with state-of-the-art predictive power

Large-scale measurements linking genetic background to biological function have driven a need for models that can incorporate these data for reliable predictions and insight into the underlying biophysical system. Recent modeling efforts, however, prioritize predictive accuracy at the expense of model interpretability. Here, we present LANTERN (landscape interpretable nonparametric model, https://github.com/usnistgov/lantern), a hierarchical Bayesian model that distills genotype-phenotype landscape (GPL) measurements into a low-dimensional feature space that represents the fundamental biological mechanisms of the system while also enabling straightforward, explainable predictions. Across a benchmark of large-scale datasets, LANTERN equals or outperforms all alternative approaches, including deep neural networks. LANTERN furthermore extracts useful insights of the landscape, including its inherent dimensionality, a latent space of additive mutational effects, and metrics of landscape structure. LANTERN facilitates straightforward discovery of fundamental mechanisms in GPLs, while also reliably extrapolating to unexplored regions of genotypic space.

Engineering Toehold-Mediated Switches for Native RNA Detection and Regulation in Bacteria

RNA switches are versatile tools in synthetic biology for sensing and regulation applications. The discoveries of RNA-mediated translational and transcriptional control have facilitated the development of complexde novodesigns of RNA switches. Specifically, RNA toehold-mediated switches, in which binding to the toehold sensing domain controls the transition between switch states via strand displacement, have been extensively adapted for coupling systems responses to specifictrans-RNA inputs. This review highlights some of the challenges associated with applying these switches for native RNA detectionin vivo, including transferability between organisms. The applicability and design considerations of toehold-mediated switches are discussed by highlighting twelve recently developed switch designs. This review finishes with future perspectives to address current gaps in the field, particularly regarding the power of structural prediction algorithms for improved in vivo functionality of RNA switches.

Toehold switch based biosensors for sensing the highly trafficked rosewood Dalbergia maritima

Nucleic acid sensing is a 3 decades old but still challenging area of application for different biological sub-domains, from pathogen detection to single cell transcriptomics analysis. The many applications of nucleic acid detection and identification are mostly carried out by PCR techniques, sequencing, and their derivatives used at large scale. However, these methods’ limitations on speed, cost, complexity and specificity have motivated the development of innovative detection methods among which nucleic acid biosensing technologies seem promising. Toehold switches are a particular class of RNA sensing devices relying on a conformational switch of secondary structure induced by the pairing of the detected trigger RNA with a de novo designed synthetic sensing mRNA molecule. Here we describe a streamlined methodology enabling the development of such a sensor for the RNA-mediated detection of an endangered plant species in a cell-free reaction system. We applied this methodology to help identify the rosewood Dalbergia maritima, a highly trafficked wood, whose protection is limited by the capacity of the authorities to distinguish protected logs from other unprotected but related species. The streamlined pipeline presented in this work is a versatile framework enabling cheap and rapid development of new sensors for custom RNA detection.

End-to-end computational approach to the design of RNA biosensors for detecting miRNA biomarkers of cervical cancer

Cervical cancer is a global public health subject as it affects women in the reproductive ages, and accounts for the second largest burden among cancer patients worldwide with an unforgiving 50% mortality rate. Relatively scant awareness and limited access to effective diagnosis have led to this enormous disease burden, calling for point-of-care, minimally invasive diagnosis methods. Here, an end-to-end quantitative unified pipeline for diagnosis has been developed, beginning with identification of optimal biomarkers, concurrent design of toehold switch sensors, and finally simulation of the designed diagnostic circuits to assess performance. Using miRNA expression data in the public domain, we identified miR-21–5p and miR-20a-5p as blood-based miRNA biomarkers specific to early-stage cervical cancer employing a multi-tier algorithmic screening. Synthetic riboregulators called toehold switches specific to the biomarker panel were then designed. To predict the dynamic range of toehold switches for use in genetic circuits as biosensors, we used a generic grammar of these switches, and built a neural network model of dynamic range using thermodynamic features derived from mRNA secondary structure and interaction. Second-generation toehold switches were used to overcome the design challenges associated with miRNA biomarkers. The resultant model yielded an adj. R² ∼0.71, outperforming earlier models of toehold-switch dynamic range. Reaction kinetics modelling was performed to predict the sensitivity of the second-generation toehold switches to the miRNA biomarkers. Simulations showed a linear response between 10 nM and 100 nM before saturation. Our study demonstrates an end-to-end computational workflow for the efficient design of genetic circuits geared towards the effective detection of unique genomic/nucleic-acid signatures. The approach has the potential to replace iterative experimental trial and error, and focus time, money, and efforts. All software including the toehold grammar parser, neural network model and reaction kinetics simulation are available as open-source software (https://github.com/SASTRA-iGEM2019) under GNU GPLv3 licence.

Deep Learning in RNA Structure Studies

Deep learning, or artificial neural networks, is a type of machine learning algorithm that can decipher underlying relationships from large volumes of data and has been successfully applied to solve structural biology questions, such as RNA structure. RNA can fold into complex RNA structures by forming hydrogen bonds, thereby playing an essential role in biological processes. While experimental effort has enabled resolving RNA structure at the genome-wide scale, deep learning has been more recently introduced for studying RNA structure and its functionality. Here, we discuss successful applications of deep learning to solve RNA problems, including predictions of RNA structures, non-canonical G-quadruplex, RNA-protein interactions and RNA switches. Following these cases, we give a general guide to deep learning for solving RNA structure problems.

Predicting target-ligand interactions with graph convolutional networks for interpretable pharmaceutical discovery

Drug Discovery is an active research area that demands great investments and generates low returns due to its inherent complexity and great costs. To identify potential therapeutic candidates more effectively, we propose protein–ligand with adversarial augmentations network (PLA-Net), a deep learning-based approach to predict target–ligand interactions. PLA-Net consists of a two-module deep graph convolutional network that considers ligands’ and targets’ most relevant chemical information, successfully combining them to find their binding capability. Moreover, we generate adversarial data augmentations that preserve relevant biological backgrounds and improve the interpretability of our model, highlighting the relevant substructures of the ligands reported to interact with the protein targets. Our experiments demonstrate that the joint ligand–target information and the adversarial augmentations significantly increase the interaction prediction performance. PLA-Net achieves 86.52% in mean average precision for 102 target proteins with perfect performance for 30 of them, in a curated version of actives as decoys dataset. Lastly, we accurately predict pharmacologically-relevant molecules when screening the ligands of ChEMBL and drug repurposing Hub datasets with the perfect-scoring targets.

Riboswitch-inspired toehold riboregulators for gene regulation in Escherichia coli

Regulatory RNA molecules have been widely investigated as components for synthetic gene circuits, complementing the use of protein-based transcription factors. Among the potential advantages of RNA-based gene regulators are their comparatively simple design, sequence-programmability, orthogonality, and their relatively low metabolic burden. In this work, we developed a set of riboswitch-inspired riboregulators in Escherichia coli that combine the concept of toehold-mediated strand displacement (TMSD) with the switching principles of naturally occurring transcriptional and translational riboswitches. Specifically, for translational activation and repression, we sequestered anti-anti-RBS or anti-RBS sequences, respectively, inside the loop of a stable hairpin domain, which is equipped with a single-stranded toehold region at its 5' end and is followed by regulated sequences on its 3' side. A trigger RNA binding to the toehold region can invade the hairpin, inducing a structural rearrangement that results in translational activation or deactivation. We also demonstrate that TMSD can be applied in the context of transcriptional regulation by switching RNA secondary structure involved in Rho-dependent termination. Our designs expand the repertoire of available synthetic riboregulators by a set of RNA switches with no sequence limitation, which should prove useful for the development of robust genetic sensors and circuits.

Differentially Optimized Cell-Free Buffer Enables Robust Expression from Unprotected Linear DNA in Exonuclease-Deficient Extracts

The use of linear DNA templates in cell-free systems promises to accelerate the prototyping and engineering of synthetic gene circuits. A key challenge is that linear templates are rapidly degraded by exonucleases present in cell extracts. Current approaches tackle the problem by adding exonuclease inhibitors and DNA-binding proteins to protect the linear DNA, requiring additional time- and resource-intensive steps. Here, we delete the recBCD exonuclease gene cluster from the Escherichia coli BL21 genome. We show that the resulting cell-free systems, with buffers optimized specifically for linear DNA, enable near-plasmid levels of expression from σ70 promoters in linear DNA templates without employing additional protection strategies. When using linear or plasmid DNA templates at the buffer calibration step, the optimal potassium glutamate concentrations obtained when using linear DNA were consistently lower than those obtained when using plasmid DNA for the same extract. We demonstrate the robustness of the exonuclease deficient extracts across seven different batches and a wide range of experimental conditions across two different laboratories. Finally, we illustrate the use of the ΔrecBCD extracts for two applications: toehold switch characterization and enzyme screening. Our work provides a simple, efficient, and cost-effective solution for using linear DNA templates in cell-free systems and highlights the importance of specifically tailoring buffer composition for the final experimental setup. Our data also suggest that similar exonuclease deletion strategies can be applied to other species suitable for cell-free synthetic biology.

Detection of pks Island mRNAs Using Toehold Sensors in Escherichia coli

Synthetic biologists have applied biomolecular engineering approaches toward the goal of novel biological devices and have shown progress in diverse areas of medicine and biotechnology. Especially promising is the application of synthetic biological devices towards a novel class of molecular diagnostics. As an example, a de-novo-designed riboregulator called toehold switch, with its programmability and compatibility with field-deployable devices showed promising in vitro applications for viral RNA detection such as Zika and Corona viruses. However, the in vivo application of high-performance RNA sensors remains challenging due to the secondary structure of long mRNA species. Here, we introduced ‘Helper RNAs’ that can enhance the functionality of toehold switch sensors by mitigating the effect of secondary structures around a target site. By employing the helper RNAs, previously reported mCherry mRNA sensor showed improved fold-changes in vivo. To further generalize the Helper RNA approaches, we employed automatic design pipeline for toehold sensors that target the essential genes within the pks island, an important target of biomedical research in connection with colorectal cancer. The toehold switch sensors showed fold-changes upon the expression of full-length mRNAs that apparently depended sensitively on the identity of the gene as well as the predicted local structure within the target region of the mRNA. Still, the helper RNAs could improve the performance of toehold switch sensors in many instances, with up to 10-fold improvement over no helper cases. These results suggest that the helper RNA approaches can further assist the design of functional RNA devices in vivo with the aid of the streamlined automatic design software developed here. Further, our solutions for screening and stabilizing single-stranded region of mRNA may find use in other in vivo mRNA-sensing applications such as cas13 crRNA design, transcriptome engineering, and trans-cleaving ribozymes.

Prediction of Breast Cancer Recurrence Using a Deep Convolutional Neural Network Without Region-of-Interest Labeling

Purpose

The present study aimed to assign a risk score for breast cancer recurrence based on pathological whole slide images (WSIs) using a deep learning model.

Methods

A total of 233 WSIs from 138 breast cancer patients were assigned either a low-risk or a high-risk score based on a 70-gene signature. These images were processed into patches of 512x512 pixels by the PyHIST tool and underwent color normalization using the Macenko method. Afterward, out of focus and pixelated patches were removed using the Laplacian algorithm. Finally, the remaining patches (n=294,562) were split into 3 parts for model training (50%), validation (7%) and testing (43%). We used 6 pretrained models for transfer learning and evaluated their performance using accuracy, precision, recall, F1 score, confusion matrix, and AUC. Additionally, to demonstrate the robustness of the final model and its generalization capacity, the testing set was used for model evaluation. Finally, the GRAD-CAM algorithm was used for model visualization.

Results

Six models, namely VGG16, ResNet50, ResNet101, Inception_ResNet, EfficientB5, and Xception, achieved high performance in the validation set with an overall accuracy of 0.84, 0.85, 0.83, 0.84, 0.87, and 0.91, respectively. We selected Xception for assessment of the testing set, and this model achieved an overall accuracy of 0.87 with a patch-wise approach and 0.90 and 1.00 with a patient-wise approach for high-risk and low-risk groups, respectively.

Conclusions

Our study demonstrated the feasibility and high performance of artificial intelligence models trained without region-of-interest labeling for predicting cancer recurrence based on a 70-gene signature risk score.