Machine learning applications in systems metabolic engineering

Promoting lignocellulosic biorefinery by machine learning: progress, perspectives and challenges

The lignocellulosic biorefinery involves pretreatment, enzymatic hydrolysis, mixed sugar fermentation, and optional anaerobic digestion. This pipeline could be effectively implemented through machine learning (ML)-guided process optimization and strain modification rather than experimental or experience-based ones. This review takes a holistic perspective on the entire pipeline, discussing how ML could aid lignocellulosic, while other published work has focused on individual modules within the pipeline. This review also explores the model construction and evaluation strategies and highlights the emerging potential of transfer learning and hybrid ML models to address data insufficiency and improve model interpretability. Furthermore, challenges and future prospects of ML in lignocellulosic biorefinery will be elaborated in this review. Integrating ML into lignocellulosic biorefinery offers a promising pathway towards sustainable and competitive biorefinery systems.

Intelligent biomanufacturing of water-soluble vitamins

Given the crucial role of water-soluble vitamins in the human body and the rising demand for natural sources, their biosynthesis has gained the attention of researchers. This review offers a comprehensive look at recent progress in water-soluble vitamin biosynthesis, emphasizing synthetic biotechnology for green biomanufacturing. Specifically, it encompasses the optimization of biological components, pathways, and systems, as well as energy metabolism regulation, stress-tolerance enhancement, high-throughput screening, and the upscaling of production processes. It also envisages intelligent biomanufacturing platforms, highlighting the role of systems biology and artificial intelligence (AI), and proposes future research directions, such as integrating AI-driven metabolic models, enzyme engineering, and cell-free systems, to address limitations in the efficiency, toxicity, and scalability of water-soluble vitamin production.

Technological advancement spurs Komagataella phaffii as a next-generation platform for sustainable biomanufacturing

Biomanufacturing stands as a cornerstone of sustainable industrial development, necessitating a shift toward non-food carbon feedstocks to alleviate agricultural resource competition and advance a circular bioeconomy. Methanol, a renewable one‑carbon substrate, has emerged as a pivotal candidate due to its abundance, cost-effectiveness, and high reduction potential, further bolstered by breakthroughs in CO₂ hydrogenation-based synthesis. Capitalizing on this momentum, the methylotrophic yeast Komagataella phaffii has undergone transformative technological upgrades, evolving from a conventional protein expression workhorse into an intelligent bioproduction chassis. This paradigm shift is fundamentally driven by converging innovations across CRISPR-empowered advancement in genome editing and AI-powered metabolic pathway design in K. phaffii. The integration of CRISPR systems with droplet microfluidics high-throughput screening has redefined strain engineering efficiency, achieving much higher editing precision than traditional homologous recombination while compressing the "design-build-test-learn" cycle. Concurrently, machine learning-enhanced genome-scale metabolic models facilitate dynamic flux balancing, enabling simultaneous improvements in product titers, carbon yields, and volumetric productivity. Finally, technological advancement promotes the application of K. phaffii, including directing more efficiently metabolic flux toward nutrient products, and strengthening efficient synthesis of excreted proteins. As DNA synthesis automation and robotic experimentation platforms mature, next-generation breakthroughs in genome modification, cofactor engineering, and AI-guided autonomous evolution will further cement K. phaffii as a next-generation platform for decarbonizing global manufacturing paradigms. This technological trajectory positions methanol-based biomanufacturing as a cornerstone of the low-carbon circular economy.

Recent Advances in Multiple Strategies for the Biosynthesis of Sesquiterpenols

Sesquiterpenols, a class of natural compounds composed of three isoprene units that form a 15-carbon skeleton with hydroxyl (-OH) group, are characterized by their volatility and potent aromatic properties. These compounds exhibit a wide range of biological activities, including antitumor, antibacterial, anti-inflammatory, anti-neurotoxic, antiviral, immunosuppressive, hepatoprotective, and cardiotonic effects. Due to their diverse physiological functionalities, sesquiterpenols serve as critical raw materials in the pharmaceutical, food, and cosmetic industries. In recent years, research on the heterologous synthesis of sesquiterpenol compounds using microbial systems has surged, attracting significant scientific interest. However, challenges such as low yields and high production costs have impeded their industrial-scale application. The rapid development of synthetic biology has introduced innovative methodologies for the microbial production of sesquiterpenol compounds. Herein, we examine the latest synthetic biology strategies and progress in microbial sesquiterpenol production, focusing on adaptive sesquiterpenol synthase screening and expression, synthesis pathway regulation, intracellular compartmentalized expression strategies, and tolerance to terpenoid-related toxicity. Critical challenges and future directions are also discussed to advance research in sesquiterpenol biosynthesis.

Unlocking the potential of Shewanella in metabolic engineering: Current status, challenges, and opportunities

Shewanella species are facultative anaerobes with distinctive electrochemical properties, making them valuable for applications in energy conversion and environmental bioremediation. Due to their well-characterized electron transfer mechanisms and ease of genetic manipulation, Shewanella spp. have emerged as a promising chassis for metabolic engineering. In this review, we provide a comprehensive overview of the advancements in Shewanella-based metabolic engineering. We begin by discussing the physiological characteristics of Shewanella, with a particular focus on its extracellular electron transfer (EET) capability. Next, we outline the use of Shewanella as a metabolic engineering chassis, presenting a general framework for strain construction based on the Design-Build-Test-Learn (DBTL) cycle and summarizing key advancements in the engineering of Shewanella's metabolic modules. Finally, we offer a perspective on the future development of Shewanella chassis, highlighting the need for deeper mechanistic insights, rational strain design, and interdisciplinary collaboration to drive further progress.

Machine Learning-Powered Optimization of a CHO Cell Cultivation Process

Chinese Hamster Ovary (CHO) cells are the most widely used cell lines to produce recombinant therapeutic proteins such as monoclonal antibodies (mAbs). However, the optimization of the CHO cell culture process is very complex and influenced by various factors. This study investigates the use of machine learning (ML) algorithms to optimize an established industrial CHO cell cultivation process. A ML algorithm in the form of an artificial neural network (ANN) was used and trained on datasets from historical and newly generated CHO cell cultivation runs. The algorithm was then used to find better cultivation conditions and improve cell productivity. The selected artificial intelligence (AI) tool was able to suggest optimized cultivation settings and new condition combinations, which promised both increased cell growth and increased mAb titers. After performing the validation experiments, it was shown that the ML algorithm was able to successfully optimize the cultivation process and significantly improve the antibody production. The best results showed an increase in final mAb titer up to 48%, demonstrating that the use of ML algorithms is a promising approach to optimize the productivity of bioprocesses like CHO cell cultivation processes clearly.

Heterologous production of caffeic acid in microbial hosts: current status and perspectives

Caffeic acid, a plant-derived phenolic compound, has attracted much attention in the fields of medicines and cosmetics due to its remarkable physiological activities including antioxidant, anti-inflammation, antibacteria, antivirus and hemostasis. However, traditional plant extraction and chemical synthesis methods exist some problems such as high production costs, low extraction efficiency and environmental pollution. In recent years, the construction of microbial cell factories for the biosynthesis of caffeic acid has attracted much attention due to its potential to offer an efficient and environmentally-friendly alternative for caffeic acid production. This review introduces the caffeic acid biosynthesis pathway first, after which the characteristics of microbial hosts for caffeic acid production are analyzed. Then, the main strategies for caffeic acid production in microbial hosts, including selection and optimization of heterologous enzymes, enhancement of the metabolic flux to caffeic acid, supply and recycling of cofactor, and optimization of the production process, are summarized and discussed. Finally, the future prospects and perspectives of microbial caffeic acid production are discussed. Recent breakthroughs have achieved caffeic acid titers of up to 6.17 g/L, demonstrating the potential of microbial biosynthesis. Future research can focus on the enhancement of metabolic flux to caffeic acid biosynthesis pathway, the development of robust microbial hosts with improved tolerance to caffeic acid and its precursors, and the establishment of cost-effective industrial production processes.

Chromatic symphony of fleshy fruits: functions, biosynthesis and metabolic engineering of bioactive compounds

Fleshy fruits are popular among consumers due to their significant nutritional value, which includes essential bioactive compounds such as pigments, vitamins, and minerals. Notably, plant-derived pigments are generally considered safe and reliable, helping to protect humans against various inflammatory diseases. Although the phytochemical diversity and their biological activities have been extensively reviewed and summarized, the status of bioactive nutrients in fleshy fruits, particularly with a focusing on different colors, has received less attention. Therefore, this review introduces five common types of fleshy fruits based on coloration and summarizes their major bioactive compounds. It also provides the latest advancements on the function, biosynthesis, and metabolic engineering of plant-derived pigments. In this review, we emphasize that promoting the consumption of a diverse array of colorful fruits can contribute to a balanced diet; however, optimal intake levels still require further clinical validation. This review may serve as a useful guide for decisions that enhance the understanding of natural pigments and accelerate their application in agriculture and medicine.

Clarification of the biosynthetic gene cluster involved in the antifungal prodrug echinocandin B and its robust production in engineered Aspergillus pachycristatus

Echinocandin antifungals exhibit high efficacy against drug-resistant strains due to their unique mechanism of action. The production of their semi-synthetic precursors relies solely on microbial metabolism, leading to elevated production costs. Anidulafungin, an excellent echinocandin drug, is derived from echinocandin B (ECB), which is industrially produced by Aspergillus pachycristatus. However, the genes involved in the actual ECB biosynthesis remain unclear, which hinders yield improvements through engineered strains. This study systematically investigated the putative ECB biosynthetic gene cluster using genomic and transcriptomic profiling combined with gene editing. Among the 18 putative genes previously reported, only a 13-gene cluster (ecdA, ecdG-J, htyA-F) was found to be actively involved in ECB biosynthesis, while the remaining 5 genes (ecdB-F) were non-essential and functioned independently. Notably, we identified that htyC and htyD were involved in L-homotyrosine biosynthesis, while HtyF catalyzed the C4 hydroxylation of 3S-hydroxyl-L-homotyrosine. Most importantly, EcdJ was identified as a crucial global transcriptional activator regulating the ECB gene cluster. Deletion of ecdJ silenced all related genes and abolished ECB production. Accordingly, overexpressing ecdJ alone or combining ecdA and htyF together significantly enhanced ECB yield. Under optimized liquid fermentation conditions, ECB production in the OEecdJ strain achieved 841 ± 23.11 mg/L. Solid-state fermentation further enhanced the ECB yield to 1.5 g/L, which is 7.7-fold higher than that of the wild-type strain under initial liquid fermentation conditions. This study has thoroughly elucidated the functions of key genes involved in the ECB biosynthesis and provided effective strategies for enhancing antifungal prodrug-ECB production, achieving the highest ECB production in an engineering A. pachycristatus strain.

Deep learning prioritizes cancer mutations that alter protein nucleocytoplasmic shuttling to drive tumorigenesis

Genetic variants can affect protein function by driving aberrant subcellular localization. However, comprehensive analysis of how mutations promote tumor progression by influencing nuclear localization is currently lacking. Here, we systematically characterize potential shuttling-attacking mutations (SAMs) across cancers through developing the deep learning model pSAM for the ab initio decoding of the sequence determinants of nucleocytoplasmic shuttling. Leveraging cancer mutations across 11 cancer types, we find that SAMs enrich functional genetic variations and critical genes in cancer. We experimentally validate a dozen SAMs, among which R14M in PTEN, P255L in CHFR, etc. are identified to disrupt the nuclear localization signals through interfering their interactions with importins. Further studies confirm that the nucleocytoplasmic shuttling altered by SAMs in PTEN and CHFR rewire the downstream signaling and eliminate their function of tumor suppression. Thus, this study will help to understand the molecular traits of nucleocytoplasmic shuttling and their dysfunctions mediated by genetic variants.

De novo biosynthesis of 4,6-dihydroxycoumarin in Escherichia coli

Coumarins and their derivatives possess crucial biochemical and pharmaceutical properties. However, the exploration of the coumarin biosynthesis pathways remains limited, restricting their microbial biosynthesis, especially for hydroxycoumarins. In this work, we designed and verified novel artificial pathways to produce a valuable compound 4,6-dihydroxycoumarin (4,6-DHC) in Escherichia coli. Based on the retrosynthesis analysis, multiple routes were designed and verified by extending the shikimate pathway, screening the potential enzymes, and characterizing the enzymes involved. Rare codon optimization and protein engineering strategies were applied to optimize the rate-limiting steps. De novo biosynthesis of 4,6-DHC was achieved using the cheap carbon source glycerol, and the titer can reach 18.3 ± 0.7 mg L-1. Ultimately, inducible regulation of critical pathway genes with a tetracycline-inducible controller yielded a significant boost in 4,6-DHC production, achieving a titer of 56.7 ± 2.1 mg L-1. This research successfully created a microbial platform for 4,6-dihydroxycoumarin production and demonstrated a generalizable strategy for synthesizing valuable compounds.

Synthetic biology strategies for sustainable bioplastic production by yeasts

The increasing environmental concerns regarding conventional plastics have led to a growing demand for sustainable alternatives, such as biodegradable plastics. Yeast cell factories, specifically Saccharomyces cerevisiae and Yarrowia lipolytica, have emerged as promising platforms for bioplastic production due to their scalability, robustness, and ease of manipulation. This review highlights synthetic biology approaches aimed at developing yeast cell factories to produce key biodegradable plastics, including polylactic acid (PLA), polyhydroxyalkanoates (PHAs), and poly (butylene adipate-co-terephthalate) (PBAT). We explore recent advancements in engineered yeast strains that utilize various synthetic biology strategies, such as the incorporation of new genetic elements at the gene, pathway, and cellular system levels. The combined efforts of metabolic engineering, protein engineering, and adaptive evolution have enhanced strain efficiency and maximized product yields. Additionally, this review addresses the importance of integrating computational tools and machine learning into the Design-Build-Test-Learn cycle for strain development. This integration aims to facilitate strain development while minimizing effort and maximizing performance. However, challenges remain in improving strain robustness and scaling up industrial production processes. By combining advanced synthetic biology techniques with computational approaches, yeast cell factories hold significant potential for the sustainable and scalable production of bioplastics, thus contributing to a greener bioeconomy.

Overcoming Bacteriophage Contamination in Bioprocessing: Strategies and Applications

Bacteriophage contamination has a devastating impact on the viability of bacterial hosts and can significantly reduce the productivity of bioprocesses in biotechnological industries. The consequences range from widespread fermentation failure to substantial economic losses, highlighting the urgent need for effective countermeasures. Conventional prevention methods, which focus primarily on the physical removal of bacteriophages from equipment, bioprocess units, and the environment, have proven ineffective in preventing phage entry and contamination. The coevolutionary dynamics between phages and their bacterial hosts have spurred the development of a diverse repertoire of antiviral defense mechanisms within microbial communities. These naturally occurring defense strategies can be harnessed through genetic engineering to convert phage-sensitive hosts into robust, phage-resistant cell factories, providing a strategic approach to mitigate the threats posed by bacteriophages to industrial bacterial processes. In this review, an overview of the various defense strategies and immune systems that curb the propagation of bacteriophages and highlight their applications in fermentation bioprocesses to combat phage contamination is provided. Additionally, the tactics employed by phages to circumvent these defense strategies are also discussed, as preventing the emergence of phage escape mutants is a key component of effective contamination management.

Microbial Technologies Enhanced by Artificial Intelligence for Healthcare Applications

The combination of artificial intelligence (AI) with microbial technology marks the start of a major transformation, improving applications throughout biotechnology, especially in healthcare. With the capability of AI to process vast amounts of biological big data, advanced microbial technology allows for a comprehensive understanding of complex biological systems, advancing disease diagnosis, treatment and the development of microbial therapeutics. This mini review explores the impact of AI-integrated microbial technologies in healthcare, highlighting advancements in microbial biomarker-based diagnosis, the development of microbial therapeutics and the microbial production of therapeutic compounds. This exploration promises significant improvements in the design and implementation of health-related solutions, steering a new era in biotechnological applications.

The transformative potential of AI-driven CRISPR-Cas9 genome editing to enhance CAR T-cell therapy

This narrative review examines the promising potential of integrating artificial intelligence (AI) with CRISPR-Cas9 genome editing to advance CAR T-cell therapy. AI algorithms offer unparalleled precision in identifying genetic targets, essential for enhancing the therapeutic efficacy of CAR T-cell treatments. This precision is critical for eliminating negative regulatory elements that undermine therapy effectiveness. Additionally, AI streamlines the manufacturing process, significantly reducing costs and increasing accessibility, thereby encouraging further research and development investment. A key benefit of AI integration is improved safety; by predicting and minimizing off-target effects, AI enhances the specificity of CRISPR-Cas9 edits, contributing to safer CAR T-cell therapy. This advancement is crucial for patient safety and broader clinical adoption. The convergence of AI and CRISPR-Cas9 has transformative potential, poised to revolutionize personalized immunotherapy. These innovations could expand the application of CAR T-cell therapy beyond hematologic malignancies to various solid tumors and other non-hematologic conditions, heralding a new era in cancer treatment that substantially improves patient outcomes.

Relieving metabolic burden to improve robustness and bioproduction by industrial microorganisms

Metabolic burden is defined by the influence of genetic manipulation and environmental perturbations on the distribution of cellular resources. The rewiring of microbial metabolism for bio-based chemical production often leads to a metabolic burden, followed by adverse physiological effects, such as impaired cell growth and low product yields. Alleviating the burden imposed by undesirable metabolic changes has become an increasingly attractive approach for constructing robust microbial cell factories. In this review, we provide a brief overview of metabolic burden engineering, focusing specifically on recent developments and strategies for diminishing the burden while improving robustness and yield. A variety of examples are presented to showcase the promise of metabolic burden engineering in facilitating the design and construction of robust microbial cell factories. Finally, challenges and limitations encountered in metabolic burden engineering are discussed.

Artificial Intelligence in Metabolomics: A Current Review

Metabolomics and artificial intelligence (AI) form a synergistic partnership. Metabolomics generates large datasets comprising hundreds to thousands of metabolites with complex relationships. AI, aiming to mimic human intelligence through computational modeling, possesses extraordinary capabilities for big data analysis. In this review, we provide a recent overview of the methodologies and applications of AI in metabolomics studies in the context of systems biology and human health. We first introduce the AI concept, history, and key algorithms for machine learning and deep learning, summarizing their strengths and weaknesses. We then discuss studies that have successfully used AI across different aspects of metabolomic analysis, including analytical detection, data preprocessing, biomarker discovery, predictive modeling, and multi-omics data integration. Lastly, we discuss the existing challenges and future perspectives in this rapidly evolving field. Despite limitations and challenges, the combination of metabolomics and AI holds great promises for revolutionary advancements in enhancing human health.

Enhancing Withanolide Production in the Withania Species: Advances in In Vitro Culture and Synthetic Biology Approaches

Withanolides are naturally occurring steroidal lactones found in certain species of the Withania genus, especially Withania somnifera (commonly known as Ashwagandha). These compounds have gained considerable attention due to their wide range of therapeutic properties and potential applications in modern medicine. To meet the rapidly growing demand for withanolides, innovative approaches such as in vitro culture techniques and synthetic biology offer promising solutions. In recent years, synthetic biology has enabled the production of engineered withanolides using heterologous systems, such as yeast and bacteria. Additionally, in vitro methods like cell suspension culture and hairy root culture have been employed to enhance withanolide production. Nevertheless, one of the primary obstacles to increasing the production of withanolides using these techniques has been the intricacy of the biosynthetic pathways for withanolides. The present article examines new developments in withanolide production through in vitro culture. A comprehensive summary of viable traditional methods for producing withanolide is also provided. The development of withanolide production in heterologous systems is examined and emphasized. The use of machine learning as a potent tool to model and improve the bioprocesses involved in the generation of withanolide is then discussed. In addition, the control and modification of the withanolide biosynthesis pathway by metabolic engineering mediated by CRISPR are discussed.

Emerging trends in production of plant natural products and new-to-nature biopharmaceuticals in yeast

Natural products represent an inestimable source of valuable compounds for human health. Notably, those produced by plants remain challenging to access due to their low production. Potential shortages of plant-derived biopharmaceuticals caused by climate change or pandemics also regularly tense the market trends. Thus, biotechnological alternatives of supply based on synthetic biology have emerged. These innovative strategies mostly rely on the use of engineered microbial systems for compound synthesis. In this regard, yeasts remain the easiest-tractable eukaryotic models and a convenient chassis for reconstructing whole biosynthetic routes for the heterologous production of plant-derived metabolites. Here, we highlight the recent discoveries dedicated to the bioproduction of new-to-nature compounds in yeasts and provide an overview of emerging strategies for optimising bioproduction.

Can digital twin efforts shape microorganism-based alternative food?

With the continuous increment in global population growth, compounded by post-pandemic food security challenges due to labor shortages, effects of climate change, political conflicts, limited land for agriculture, and carbon emissions control, addressing food production in a sustainable manner for future generations is critical. Microorganisms are potential alternative food sources that can help close the gap in food production. For the development of more efficient and yield-enhancing products, it is necessary to have a better understanding on the underlying regulatory molecular pathways of microbial growth. Nevertheless, as microbes are regulated at multiomics scales, current research focusing on single omics (genomics, proteomics, or metabolomics) independently is inadequate for optimizing growth and product output. Here, we discuss digital twin (DT) approaches that integrate systems biology and artificial intelligence in analyzing multiomics datasets to yield a microbial replica model for in silico testing before production. DT models can thus provide a holistic understanding of microbial growth, metabolite biosynthesis mechanisms, as well as identifying crucial production bottlenecks. Our argument, therefore, is to support the development of novel DT models that can potentially revolutionize microorganism-based alternative food production efficiency.

Changes in Rhizosphere Soil Microorganisms and Metabolites during the Cultivation of Fritillaria cirrhosa

Fritillaria cirrhosa is an important cash crop, and its industrial development is being hampered by continuous cropping obstacles, but the composition and changes of rhizosphere soil microorganisms and metabolites in the cultivation process of Fritillaria cirrhosa have not been revealed. We used metagenomics sequencing to analyze the changes of the microbiome in rhizosphere soil during a three-year cultivation process, and combined it with LC-MS/MS to detect the changes of metabolites. Results indicate that during the cultivation of Fritillaria cirrhosa, the composition and structure of the rhizosphere soil microbial community changed significantly, especially regarding the relative abundance of some beneficial bacteria. The abundance of Bradyrhizobium decreased from 7.04% in the first year to about 5% in the second and third years; the relative abundance of Pseudomonas also decreased from 6.20% in the first year to 2.22% in the third year; and the relative abundance of Lysobacter decreased significantly from more than 4% in the first two years of cultivation to 1.01% in the third year of cultivation. However, the relative abundance of some harmful fungi has significantly increased, such as Botrytis, which increased significantly from less than 3% in the first two years to 7.93% in the third year, and Talaromyces fungi, which were almost non-existent in the first two years of cultivation, significantly increased to 3.43% in the third year of cultivation. The composition and structure of Fritillaria cirrhosa rhizosphere metabolites also changed significantly, the most important of which were carbohydrates represented by sucrose (48.00-9.36-10.07%) and some amino acid compounds related to continuous cropping obstacles. Co-occurrence analysis showed that there was a significant correlation between differential microorganisms and differential metabolites, but Procrustes analysis showed that the relationship between bacteria and metabolites was closer than that between fungi and metabolites. In general, in the process of Fritillaria cirrhosa cultivation, the beneficial bacteria in the rhizosphere decreased, the harmful bacteria increased, and the relative abundance of carbohydrate and amino acid compounds related to continuous cropping obstacles changed significantly. There is a significant correlation between microorganisms and metabolites, and the shaping of the Fritillaria cirrhosa rhizosphere's microecology by bacteria is more relevant.

Bioproduction of porphyrins, phycobilins, and their proteins using microbial cell factories: engineering, metabolic regulations, challenges, and perspectives

Porphyrins, phycobilins, and their proteins have abundant π-electrons and strongly absorb visible light, some of which bind a metal ion in the center. Because of the structural and optical properties, they not only play critical roles as an essential component in natural systems but also have attracted much attention as a high value specialty chemical in various fields, including renewable energy, cosmetics, medicines, and foods. However, their commercial application seems to be still limited because the market price of porphyrins and phycobilins is generally expensive to apply them easily. Furthermore, their petroleum-based chemical synthesis is energy-intensive and emits a pollutant. Recently, to replace petroleum-based production, many studies on the bioproduction of metalloporphyrins, including Zn-porphyrin, Co-porphyrin, and heme, porphyrin derivatives including chlorophyll, biliverdin, and phycobilins, and their proteins including hemoproteins, phycobiliproteins, and phytochromes from renewable carbon sources using microbial cell factories have been reported. This review outlines recent advances in the bioproduction of porphyrins, phycobilins, and their proteins using microbial cell factories developed by various microbial biotechnology techniques, provides well-organized information on metabolic regulations of the porphyrin metabolism, and then critically discusses challenges and future perspectives. Through these, it is expected to be able to achieve possible solutions and insights and to develop an outstanding platform to be applied to the industry in future research.

Automated in vivo enzyme engineering accelerates biocatalyst optimization

Achieving cost-competitive bio-based processes requires development of stable and selective biocatalysts. Their realization through in vitro enzyme characterization and engineering is mostly low throughput and labor-intensive. Therefore, strategies for increasing throughput while diminishing manual labor are gaining momentum, such as in vivo screening and evolution campaigns. Computational tools like machine learning further support enzyme engineering efforts by widening the explorable design space. Here, we propose an integrated solution to enzyme engineering challenges whereby ML-guided, automated workflows (including library generation, implementation of hypermutation systems, adapted laboratory evolution, and in vivo growth-coupled selection) could be realized to accelerate pipelines towards superior biocatalysts.

Predicting pathways for old and new metabolites through clustering

The diverse metabolic pathways are fundamental to all living organisms, as they harvest energy, synthesize biomass components, produce molecules to interact with the microenvironment, and neutralize toxins. While the discovery of new metabolites and pathways continues, the prediction of pathways for new metabolites can be challenging. It can take vast amounts of time to elucidate pathways for new metabolites; thus, according to HMDB (Human Metabolome Database), only 60% of metabolites get assigned to pathways. Here, we present an approach to identify pathways based on metabolite structure. We extracted 201 features from SMILES annotations and identified new metabolites from PubMed abstracts and HMDB. After applying clustering algorithms to both groups of features, we quantified correlations between metabolites, and found the clusters accurately linked 92% of known metabolites to their respective pathways. Thus, this approach could be valuable for predicting metabolic pathways for new metabolites.

A Machine Learning Approach for Predicting Essentiality of Metabolic Genes

The identification of essential genes is a key challenge in systems and synthetic biology, particularly for engineering metabolic pathways that convert feedstocks into valuable products. Assessment of gene essentiality at a genome scale requires large and costly growth assays of knockout strains. Here we describe a strategy to predict the essentiality of metabolic genes using binary classification algorithms. The approach combines elements from genome-scale metabolic models, directed graphs, and machine learning into a predictive model that can be trained on small knockout data. We demonstrate the efficacy of this approach using the most complete metabolic model of Escherichia coli and various machine learning algorithms for binary classification.

Prediction of Thermostability of Enzymes Based on the Amino Acid Index (AAindex) Database and Machine Learning

The combination of wet-lab experimental data on multi-site combinatorial mutations and machine learning is an innovative method in protein engineering. In this study, we used an innovative sequence-activity relationship (innov'SAR) methodology based on novel descriptors and digital signal processing (DSP) to construct a predictive model. In this paper, 21 experimental (R)-selective amine transaminases from Aspergillus terreus (AT-ATA) were used as an input to predict higher thermostability mutants than those predicted using the existing data. We successfully improved the coefficient of determination (R2) of the model from 0.66 to 0.92. In addition, root-mean-squared deviation (RMSD), root-mean-squared fluctuation (RMSF), solvent accessible surface area (SASA), hydrogen bonds, and the radius of gyration were estimated based on molecular dynamics simulations, and the differences between the predicted mutants and the wild-type (WT) were analyzed. The successful application of the innov'SAR algorithm in improving the thermostability of AT-ATA may help in directed evolutionary screening and open up new avenues for protein engineering.

A journey with psychedelic mushrooms: From historical relevance to biology, cultivation, medicinal uses, biotechnology, and beyond

Psychedelic mushrooms containing psilocybin and related tryptamines have long been used for ethnomycological purposes, but emerging evidence points to the potential therapeutic value of these mushrooms to address modern neurological, psychiatric health, and related disorders. As a result, psilocybin containing mushrooms represent a re-emerging frontier for mycological, biochemical, neuroscience, and pharmacology research. This work presents crucial information related to traditional use of psychedelic mushrooms, as well as research trends and knowledge gaps related to their diversity and distribution, technologies for quantification of tryptamines and other tryptophan-derived metabolites, as well as biosynthetic mechanisms for their production within mushrooms. In addition, we explore the current state of knowledge for how psilocybin and related tryptamines are metabolized in humans and their pharmacological effects, including beneficial and hazardous human health implications. Finally, we describe opportunities and challenges for investigating the production of psychedelic mushrooms and metabolic engineering approaches to alter secondary metabolite profiles using biotechnology integrated with machine learning. Ultimately, this critical review of all aspects related to psychedelic mushrooms represents a roadmap for future research efforts that will pave the way to new applications and refined protocols.

Recent advances in systems metabolic engineering

Systems metabolic engineering, which integrates metabolic engineering with systems biology, synthetic biology, and evolutionary engineering, has revolutionized the sustainable production of fuels and materials through the creation of efficient microbial cell factories. Recent advancements in systems metabolic engineering targeting different biological components of the host cell have enabled the creation of highly productive microbial cell factories. This article provides a review of the recent tools and strategies used for enzyme-, genetic module-, pathway-, flux-, genome-, and cell-level engineering, supported by illustrative examples. Furthermore, we highlight recent trends in systems metabolic engineering, which involve the application of multiple tools discussed in this review. Finally, the paper addresses the challenges and perspectives of transitioning academic-level metabolic engineering studies to commercial-scale production.

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.

Biotechnological production of omega-3 fatty acids: current status and future perspectives

Omega-3 fatty acids, including alpha-linolenic acids (ALA), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), have shown major health benefits, but the human body's inability to synthesize them has led to the necessity of dietary intake of the products. The omega-3 fatty acid market has grown significantly, with a global market from an estimated USD 2.10 billion in 2020 to a predicted nearly USD 3.61 billion in 2028. However, obtaining a sufficient supply of high-quality and stable omega-3 fatty acids can be challenging. Currently, fish oil serves as the primary source of omega-3 fatty acids in the market, but it has several drawbacks, including high cost, inconsistent product quality, and major uncertainties in its sustainability and ecological impact. Other significant sources of omega-3 fatty acids include plants and microalgae fermentation, but they face similar challenges in reducing manufacturing costs and improving product quality and sustainability. With the advances in synthetic biology, biotechnological production of omega-3 fatty acids via engineered microbial cell factories still offers the best solution to provide a more stable, sustainable, and affordable source of omega-3 fatty acids by overcoming the major issues associated with conventional sources. This review summarizes the current status, key challenges, and future perspectives for the biotechnological production of major omega-3 fatty acids.

Deep learning for metabolic pathway design

The establishment of a bio-based circular economy is imperative in tackling the climate crisis and advancing sustainable development. In this realm, the creation of microbial cell factories is central to generating a variety of chemicals and materials. The design of metabolic pathways is crucial in shaping these microbial cell factories, especially when it comes to producing chemicals with yet-to-be-discovered biosynthetic routes. To aid in navigating the complexities of chemical and metabolic domains, computer-supported tools for metabolic pathway design have emerged. In this paper, we evaluate how digital strategies can be employed for pathway prediction and enzyme discovery. Additionally, we touch upon the recent strides made in using deep learning techniques for metabolic pathway prediction. These computational tools and strategies streamline the design of metabolic pathways, facilitating the development of microbial cell factories. Leveraging the capabilities of deep learning in metabolic pathway design is profoundly promising, potentially hastening the advent of a bio-based circular economy.

Progress and Prospects of Natural Glycoside Sweetener Biosynthesis: A Review

To achieve an adequate sense of sweetness with a healthy low-sugar diet, it is necessary to explore and produce sugar alternatives. Recently, glycoside sweeteners and their biosynthetic approaches have attracted the attention of researchers. In this review, we first outlined the synthetic pathways of glycoside sweeteners, including the key enzymes and rate-limiting steps. Next, we reviewed the progress in engineered microorganisms producing glycoside sweeteners, including de novo synthesis, whole-cell catalysis synthesis, and in vitro synthesis. The applications of metabolic engineering strategies, such as cofactor engineering and enzyme modification, in the optimization of glycoside sweetener biosynthesis were summarized. Finally, the prospects of combining enzyme engineering and machine learning strategies to enhance the production of glycoside sweeteners were discussed. This review provides a perspective on synthesizing glycoside sweeteners in microbial cells, theoretically guiding the bioproduction of glycoside sweeteners.

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.

Optimizing Ethanol Production in Saccharomyces cerevisiae at Ambient and Elevated Temperatures through Machine Learning-Guided Combinatorial Promoter Modifications

Bioethanol has gained popularity in recent decades as an ecofriendly alternative to fossil fuels due to increasing concerns about global climate change. However, economically viable ethanol fermentation remains a challenge. High-temperature fermentation can reduce production costs, but Saccharomyces cerevisiae yeast strains normally ferment poorly under high temperatures. In this study, we present a machine learning (ML) approach to optimize bioethanol production in S. cerevisiae by fine-tuning the promoter activities of three endogenous genes. We created 216 combinatorial strains of S. cerevisiae by replacing native promoters with five promoters of varying strengths to regulate ethanol production. Promoter replacement resulted in a 63% improvement in ethanol production at 30 °C. We created an ML-guided workflow by utilizing XGBoost to train high-performance models based on promoter strengths and cellular metabolite concentrations obtained from ethanol production of 216 combinatorial strains at 30 °C. This strategy was then applied to optimize ethanol production at 40 °C, where we selected 31 strains for experimental fermentation. This reduced experimental load led to a 7.4% increase in ethanol production in the second round of the ML-guided workflow. Our study offers a comprehensive library of promoter strength modifications for key ethanol production enzymes, showcasing how machine learning can guide yeast strain optimization and make bioethanol production more cost-effective and efficient. Furthermore, we demonstrate that metabolic engineering processes can be accelerated and optimized through this approach.

Simulated Design-Build-Test-Learn Cycles for Consistent Comparison of Machine Learning Methods in Metabolic Engineering

Combinatorial pathway optimization is an important tool in metabolic flux optimization. Simultaneous optimization of a large number of pathway genes often leads to combinatorial explosions. Strain optimization is therefore often performed using iterative design-build-test-learn (DBTL) cycles. The aim of these cycles is to develop a product strain iteratively, every time incorporating learning from the previous cycle. Machine learning methods provide a potentially powerful tool to learn from data and propose new designs for the next DBTL cycle. However, due to the lack of a framework for consistently testing the performance of machine learning methods over multiple DBTL cycles, evaluating the effectiveness of these methods remains a challenge. In this work, we propose a mechanistic kinetic model-based framework to test and optimize machine learning for iterative combinatorial pathway optimization. Using this framework, we show that gradient boosting and random forest models outperform the other tested methods in the low-data regime. We demonstrate that these methods are robust for training set biases and experimental noise. Finally, we introduce an algorithm for recommending new designs using machine learning model predictions. We show that when the number of strains to be built is limited, starting with a large initial DBTL cycle is favorable over building the same number of strains for every cycle.

Efforts to Minimise the Bacterial Genome as a Free-Living Growing System

Exploring the minimal genetic requirements for cells to maintain free living is an exciting topic in biology. Multiple approaches are employed to address the question of the minimal genome. In addition to constructing the synthetic genome in the test tube, reducing the size of the wild-type genome is a practical approach for obtaining the essential genomic sequence for living cells. The well-studied Escherichia coli has been used as a model organism for genome reduction owing to its fast growth and easy manipulation. Extensive studies have reported how to reduce the bacterial genome and the collections of genomic disturbed strains acquired, which were sufficiently reviewed previously. However, the common issue of growth decrease caused by genetic disturbance remains largely unaddressed. This mini-review discusses the considerable efforts made to improve growth fitness, which was decreased due to genome reduction. The proposal and perspective are clarified for further accumulated genetic deletion to minimise the Escherichia coli genome in terms of genome reduction, experimental evolution, medium optimization, and machine learning.

DeepRT: Predicting compounds presence in pathway modules and classifying into module classes using deep neural networks based on molecular properties

Metabolic pathways play a crucial role in understanding the biochemistry of organisms. In metabolic pathways, modules refer to clusters of interconnected reactions or sub-networks representing specific functional units or biological processes within the overall pathway. In pathway modules, compounds are major elements and refer to the various molecules that participate in the biochemical reactions within the pathway modules. These molecules can include substrates, intermediates and final products. Determining the presence relation of compounds and pathway modules is essential for synthesizing new molecules and predicting hidden reactions. To date, several computational methods have been proposed to address this problem. However, all methods only predict the metabolic pathways and their types, not the pathway modules. To address this issue, we proposed a novel deep learning model, DeepRT that integrates message passing neural networks (MPNNs) and transformer encoder. This combination allows DeepRT to effectively extract global and local structure information from the molecular graph. The model is designed to perform two tasks: first, determining the present relation of the compound with the pathway module, and second, predicting the relation of query compound and module classes. The proposed DeepRT model evaluated on a dataset comprising compounds and pathway modules, and it outperforms existing approaches.

RespectM revealed metabolic heterogeneity powers deep learning for reshaping the DBTL cycle

Synthetic biology, relying on Design-Build-Test-Learn (DBTL) cycle, aims to solve medicine, manufacturing, and agriculture problems. However, the DBTL cycle's Learn (L) step lacks predictive power for the behavior of biological systems, resulting from the incompatibility between sparse testing data and chaotic metabolic networks. Herein, we develop a method, "RespectM," based on mass spectrometry imaging, which is able to detect metabolites at a rate of 500 cells per hour with high efficiency. In this study, 4,321 single cell level metabolomics data were acquired, representing metabolic heterogeneity. An optimizable deep neural network was applied to learn from metabolic heterogeneity and a "heterogeneity-powered learning (HPL)" based model was trained as well. By testing the HPL based model, we suggest minimal operations to achieve high triglyceride production for engineering. The HPL strategy could revolutionize rational design and reshape the DBTL cycle.

Developing high-dimensional machine learning models to improve generalization ability and overcome data insufficiency for mixed sugar fermentation simulation

Biorefinery can be promoted by building accurate machine learning models. This work proposed a strategy to enhance model's generalization ability and overcome insufficient data conditions for mixed sugar fermentation simulation. Multiple inputs single output models, using initial glucose, initial xylose, and time together as inputs, have higher generalization ability than single input single output models with time as sole input in predicting glucose, xylose, ethanol, or biomass separately. Multiple inputs multiple outputs models, integrating outputs, enhanced model accuracy and resulted in an average R² at 0.99. To overcome data insufficiency conditions, consensus yeast (CY) model, through consolidating data from 4 yeasts, obtained R² at 0.90. By adjusting the pretrained CY model, the model can save more than 50% data and get R² at 0.95 and 0.93 for yeast and bacterial fermentation simulation. The strategy can expand the application range and save costs of data curation for ANN models.

Prediction of transport proteins from sequence information with the deep learning approach

Transport proteins (TPs) are vital to the growth and life of all living things, especially in fields of microbial pathogenesis and drug resistance of tumor cells. Accurately identifying potential TPs remains an important challenge for the advancement of functional genomics. This study aimed to develop a tool for predicting TPs using the deep learning approach. Here, we proposed DeepTP, a convolutional neural network model that uses parallel subnetworks to extract features from protein sequences and uses fully connected layers for TP classification. To train and evaluate the performance of the developed model, datasets were collected from the UniProtKB/Swiss-Prot database. The test results revealed that the proposed model could successfully identify TPs with the AUCROC, accuracy, F-value, and Matthews correlation coefficient of 0.9719, 0.9513, 0.8982, and 0.8679, respectively. By further comparison, DeepTP achieved better performance than other commonly used methods. Analysis of the gradients of prediction score concerning input suggested that DeepTP makes predictions by recognizing the functional domains of TPs. We anticipate that DeepTP will serve as a useful tool for predicting TPs in large-scale genome projects, which will facilitate the discovery of novel TPs.

Predicting the effects of cultivation condition on gene regulation in Escherichia coli by using deep learning

Cell's physiology is affected by cultivation conditions at varying degrees, including carbon sources and inorganic nutrients in growth medium, and the presence or absence of aeration. When examining the effects of cultivation conditions on the cell, the cell's transcriptional response is often examined first among other phenotypes (e.g., proteome and metabolome). In this regard, we developed DeepMGR, a deep learning model that predicts the effects of culture media on gene regulation in Escherichia coli. DeepMGR specifically classifies the direction of gene regulation (i.e., upregulation, no regulation, or downregulation) for an input gene in comparison with M9 minimal medium with glucose as a control condition. For this classification task, DeepMGR uses a feedforward neural network to process: i) DNA sequence of a target gene, ii) presence or absence of aeration and trace elements, and iii) concentration and structural information (SMILES) of up to ten nutrients. The complete DeepMGR showed accuracy of 0.867 and F1 score of 0.703 for a test set from the gold standard dataset. DeepMGR was further subjected to simulation studies for validation where regulation directions for groups of homologous genes were predicted, and the DeepMGR results were compared with the literature with focus on carbon sources that upregulate specific genes. DeepMGR will be useful for designing experiments to understand gene regulations, especially in the context of metabolic engineering.

Machine learning for metabolic pathway optimization: A review

Optimizing the metabolic pathways of microbial cell factories is essential for establishing viable biotechnological production processes. However, due to the limited understanding of the complex setup of cellular machinery, building efficient microbial cell factories remains tedious and time-consuming. Machine learning (ML), a powerful tool capable of identifying patterns within large datasets, has been used to analyze biological datasets generated using various high-throughput technologies to build data-driven models for complex bioprocesses. In addition, ML can also be integrated with Design-Build-Test-Learn to accelerate development. This review focuses on recent ML applications in genome-scale metabolic model construction, multistep pathway optimization, rate-limiting enzyme engineering, and gene regulatory element designing. In addition, we have discussed some limitations of these methods as well as potential solutions.

Enzyme Substrate Prediction from Three-Dimensional Feature Representations Using Space-Filling Curves

Compact and interpretable structural feature representations are required for accurately predicting properties and function of proteins. In this work, we construct and evaluate three-dimensional feature representations of protein structures based on space-filling curves (SFCs). We focus on the problem of enzyme substrate prediction, using two ubiquitous enzyme families as case studies: the short-chain dehydrogenase/reductases (SDRs) and the S-adenosylmethionine-dependent methyltransferases (SAM-MTases). Space-filling curves such as the Hilbert curve and the Morton curve generate a reversible mapping from discretized three-dimensional to one-dimensional representations and thus help to encode three-dimensional molecular structures in a system-independent way and with only a few adjustable parameters. Using three-dimensional structures of SDRs and SAM-MTases generated using AlphaFold2, we assess the performance of the SFC-based feature representations in predictions on a new benchmark database of enzyme classification tasks including their cofactor and substrate selectivity. Gradient-boosted tree classifiers yield binary prediction accuracy of 0.77-0.91 and area under curve (AUC) characteristics of 0.83-0.92 for the classification tasks. We investigate the effects of amino acid encoding, spatial orientation, and (the few) parameters of SFC-based encodings on the accuracy of the predictions. Our results suggest that geometry-based approaches such as SFCs are promising for generating protein structural representations and are complementary to the existing protein feature representations such as evolutionary scale modeling (ESM) sequence embeddings.

Improving the production of the micafungin precursor FR901379 in an industrial production strain

BACKGROUND

Micafungin is an echinocandin-type antifungal agent used for the clinical treatment of invasive fungal infections. It is semisynthesized from the sulfonated lipohexapeptide FR901379, a nonribosomal peptide produced by the filamentous fungus Coleophoma empetri. However, the low fermentation efficiency of FR901379 increases the cost of micafungin production and hinders its widespread clinical application.

RESULTS

Here, a highly efficient FR901379-producing strain was constructed via systems metabolic engineering in C. empetri MEFC09. First, the biosynthesis pathway of FR901379 was optimized by overexpressing the rate-limiting enzymes cytochrome P450 McfF and McfH, which successfully eliminated the accumulation of unwanted byproducts and increased the production of FR901379. Then, the functions of putative self-resistance genes encoding β-1,3-glucan synthase were evaluated in vivo. The deletion of CEfks1 affected growth and resulted in more spherical cells. Additionally, the transcriptional activator McfJ for the regulation of FR901379 biosynthesis was identified and applied in metabolic engineering. Overexpressing mcfJ markedly increased the production of FR901379 from 0.3 g/L to 1.3 g/L. Finally, the engineered strain coexpressing mcfJ, mcfF, and mcfH was constructed for additive effects, and the FR901379 titer reached 4.0 g/L under fed-batch conditions in a 5 L bioreactor.

CONCLUSIONS

This study represents a significant improvement for the production of FR901379 and provides guidance for the establishment of efficient fungal cell factories for other echinocandins.

Metabolic engineering for sustainability and health

Bio-based production of chemicals and materials has attracted much attention due to the urgent need to establish sustainability and enhance human health. Metabolic engineering (ME) allows purposeful modification of cellular metabolic, regulatory, and signaling networks to achieve enhanced production of desired chemicals and degradation of environmentally harmful chemicals. ME has significantly progressed over the past 30 years through further integration of the strategies of synthetic biology, systems biology, evolutionary engineering, and data science aided by artificial intelligence. Here we review the field of ME from its emergence to the current state-of-the-art, highlighting its contribution to sustainable production of chemicals, health, and the environment through representative examples. Future challenges of ME and perspectives are also discussed.

Cytochrome P450s in plant terpenoid biosynthesis: discovery, characterization and metabolic engineering

As the largest family of natural products, terpenoids play valuable roles in medicine, agriculture, cosmetics and food. However, the traditional methods that rely on direct extraction from the original plants not only produce low yields, but also result in waste of resources, and are not applicable at all to endangered species. Modern heterologous biosynthesis is considered a promising, efficient, and sustainable production method, but it relies on the premise of a complete analysis of the biosynthetic pathway of terpenoids, especially the functionalization processes involving downstream cytochrome P450s. In this review, we systematically introduce the biotech approaches used to discover and characterize plant terpenoid-related P450s in recent years. In addition, we propose corresponding metabolic engineering approaches to increase the effective expression of P450 and improve the yield of terpenoids, and also elaborate on metabolic engineering strategies and examples of heterologous biosynthesis of terpenoids in Saccharomyces cerevisiae and plant hosts. Finally, we provide perspectives for the biotech approaches to be developed for future research on terpenoid-related P450.

Machine learning-guided prediction of potential engineering targets for microbial production of lycopene

The process of designing streamlined workflows for developing microbial strains using classical methods from vast amounts of biological big data has reached its limits. With the continuous increase in the amount of biological big data, data-driven machine learning approaches are being used to overcome the limits of classical approaches for strain development. Here, machine learning-guided engineering of Deinococcus radiodurans R1 for high-yield production of lycopene was demonstrated. The multilayer perceptron models were first trained using the mRNA expression levels of the key genes along with lycopene titers and yields obtained from 17 strains. Then, the potential overexpression targets from 2,047 possible combinations were predicted by the multilayer perceptron combined with a genetic algorithm. Through the machine learning-aided fine-tuning of the predicted genes, the final-engineered LY04 strain resulted in an 8-fold increase in the lycopene production, up to 1.25 g/L from glycerol, and a 6-fold increase in the lycopene yield.

Machine learning-assisted medium optimization revealed the discriminated strategies for improved production of the foreign and native metabolites

The composition of medium components is crucial for achieving the best performance of synthetic construction in genetically engineered cells. Which and how medium components determine the performance, e.g., productivity, remain poorly investigated. To address the questions, a comparative survey with two genetically engineered Escherichia coli strains was performed. As a case study, the strains carried the synthetic pathways for producing the aromatic compounds of 4-aminophenylalanine (4APhe) or tyrosine (Tyr), common in the upstream but differentiated in the downstream metabolism. Bacterial growth and compound production were examined in hundreds of medium combinations that comprised 48 pure chemicals. The resultant data sets linking the medium composition to bacterial growth and production were subjected to machine learning for improved production. Intriguingly, the primary medium components determining the production of 4PheA and Tyr were differentiated, which were the initial resource (glucose) of the synthetic pathway and the inducer (IPTG) of the synthetic construction, respectively. Fine-tuning of the primary component significantly increased the yields of 4APhe and Tyr, indicating that a single component could be crucial for the performance of synthetic construction. Transcriptome analysis observed the local and global changes in gene expression for improved production of 4APhe and Tyr, respectively, revealing divergent metabolic strategies for producing the foreign and native metabolites. The study demonstrated that ML-assisted medium optimization could provide a novel point of view on how to make the synthetic construction meet the designed working principle and achieve the expected biological function.

Current status and future prospects in cannabinoid production through in vitro culture and synthetic biology

For centuries, cannabis has been a rich source of fibrous, pharmaceutical, and recreational ingredients. Phytocannabinoids are the most important and well-known class of cannabis-derived secondary metabolites and display a broad range of health-promoting and psychoactive effects. The unique characteristics of phytocannabinoids (e.g., metabolite likeness, multi-target spectrum, and safety profile) have resulted in the development and approval of several cannabis-derived drugs. While most work has focused on the two main cannabinoids produced in the plant, over 150 unique cannabinoids have been identified. To meet the rapidly growing phytocannabinoid demand, particularly many of the minor cannabinoids found in low amounts in planta, biotechnology offers promising alternatives for biosynthesis through in vitro culture and heterologous systems. In recent years, the engineered production of phytocannabinoids has been obtained through synthetic biology both in vitro (cell suspension culture and hairy root culture) and heterologous systems. However, there are still several bottlenecks (e.g., the complexity of the cannabinoid biosynthetic pathway and optimizing the bioprocess), hampering biosynthesis and scaling up the biotechnological process. The current study reviews recent advances related to in vitro culture-mediated cannabinoid production. Additionally, an integrated overview of promising conventional approaches to cannabinoid production is presented. Progress toward cannabinoid production in heterologous systems and possible avenues for avoiding autotoxicity are also reviewed and highlighted. Machine learning is then introduced as a powerful tool to model, and optimize bioprocesses related to cannabinoid production. Finally, regulation and manipulation of the cannabinoid biosynthetic pathway using CRISPR- mediated metabolic engineering is discussed.

Recent advances in machine learning applications in metabolic engineering

Metabolic engineering encompasses several widely-used strategies, which currently hold a high seat in the field of biotechnology when its potential is manifesting through a plethora of research and commercial products with a strong societal impact. The genomic revolution that occurred almost three decades ago has initiated the generation of large omics-datasets which has helped in gaining a better understanding of cellular behavior. The itinerary of metabolic engineering that has occurred based on these large datasets has allowed researchers to gain detailed insights and a reasonable understanding of the intricacies of biosystems. However, the existing trail-and-error approaches for metabolic engineering are laborious and time-intensive when it comes to the production of target compounds with high yields through genetic manipulations in host organisms. Machine learning (ML) coupled with the available metabolic engineering test instances and omics data brings a comprehensive and multidisciplinary approach that enables scientists to evaluate various parameters for effective strain design. This vast amount of biological data should be standardized through knowledge engineering to train different ML models for providing accurate predictions in gene circuits designing, modification of proteins, optimization of bioprocess parameters for scaling up, and screening of hyper-producing robust cell factories. This review briefs on the premise of ML, followed by mentioning various ML methods and algorithms alongside the numerous omics datasets available to train ML models for predicting metabolic outcomes with high-accuracy. The combinative interplay between the ML algorithms and biological datasets through knowledge engineering have guided the recent advancements in applications such as CRISPR/Cas systems, gene circuits, protein engineering, metabolic pathway reconstruction, and bioprocess engineering. Finally, this review addresses the probable challenges of applying ML in metabolic engineering which will guide the researchers toward novel techniques to overcome the limitations.