Machine Learning and Hybrid Methods for Metabolic Pathway Modeling

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.

Prospects for synthetic biology in 21st Century agriculture

Plant synthetic biology has emerged as a transformative field in agriculture, offering innovative solutions to enhance food security, provide resilience to climate change, and transition to sustainable farming practices. By integrating advanced genetic tools, computational modeling, and systems biology, researchers can precisely modify plant genomes to enhance traits such as yield, stress tolerance, and nutrient use efficiency. The ability to design plants with specific characteristics tailored to diverse environmental conditions and agricultural needs holds great potential to address global food security challenges. Here, we highlight recent advancements and applications of plant synthetic biology in agriculture, focusing on key areas such as photosynthetic efficiency, nitrogen fixation, drought tolerance, pathogen resistance, nutrient use efficiency, biofortification, climate resilience, microbiology engineering, synthetic plant genomes, and the integration of artificial intelligence (AI) with synthetic biology. These innovations aim to maximize resource use efficiency, reduce reliance on external inputs, and mitigate environmental impacts associated with conventional agricultural practices. Despite challenges related to regulatory approval and public acceptance, the integration of synthetic biology in agriculture holds immense promise for creating more resilient and sustainable agricultural systems, contributing to global food security and environmental sustainability. Rigorous multi-field testing of these approaches will undoubtedly be required to ensure reproducibility.

Enhanced diagnosing patients suspected of sarcoidosis using a hybrid support vector regression model with bald eagle and chimp optimizers

Searching for a reliable indicator of treatment response in sarcoidosis remains a challenge. The use of the soluble interleukin 2 receptor (sIL-2R) as a measure of disease activity has been proposed by researchers. A machine learning model was aimed to be developed in this study to predict sIL-2R levels based on a patient's serum angiotensin-converting enzyme (ACE) levels, potentially aiding in lung function evaluation. A novel forecasting model (SVR-BE-CO) for sIL-2R prediction is introduced, which combines support vector regression (SVR) with a hybrid optimization model (BES-CO); The hybrid optimization model composed of Bald Eagle Optimizer (BES) and Chimp Optimizer (CO) model. In this forecasting model, the hyper-parameters of the SVR model are optimized by the BES-CO hybrid optimization model, ultimately improving the accuracy of the predicted sIL-2R values. The hybrid forecasting model SVR-BE-CO model was evaluated against various forecasting methods, including Hybrid SVR with Firefly Algorithm (SVR-FFA), decision tree (DT), SVR with Gray Wolf Optimization (SVR-GWO) and random forest (RF). It was demonstrated that the hybrid SVR-BE-CO model surpasses all other methods in terms of accuracy.

Multivariate analyses and machine learning link sex and age with antibody responses to SARS-CoV-2 and vaccination

Prevention of negative COVID-19 infection outcomes is associated with the quality of antibody responses, whose variance by age and sex is poorly understood. Network approaches identified sex and age effects in antibody responses and neutralization potential of de novo infection and vaccination throughout the COVID-19 pandemic. Neutralization values followed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-specific receptor binding immunoglobulin G (RIgG), spike immunoglobulin G (SIgG) and spike and receptor immunoglobulin G (S, and RIgA) levels based on COVID-19 status. Serum immunoglobulin A (IgA) antibody titers correlated with neutralization only in females 40-60 years old (y.o.). Network analysis found males could improve IgA responses after vaccination dose 2. Complex correlation analyses found vaccination induced less antibody isotype switching and neutralization in older persons, especially in females. Sex-dependent antibody and neutralization decayed the fastest in older males. Shown sex and age characterization can direct studies integrating cell-mediated responses to define yet elusive correlates of protection and inform age and sex precision-focused vaccine design.

Expanding the structural diversity of terpenes by synthetic biology approaches

Terpenoids display chemical and structural diversities as well as important biological activities. Despite their extreme variability, the range of these structures is limited by the scope of natural products that canonically derive from interconvertible five-carbon (C5) isoprene units. New approaches have recently been developed to expand their structural diversity. This review systematically explores the combinatorial biosynthesis of noncanonical building blocks via the coexpression of the canonical mevalonate (MVA) pathway and C-methyltransferases (C-MTs), or by using the lepidopteran mevalonate (LMVA) pathway. Unnatural terpenoids can be created from farnesyl diphosphate (FPP) analogs by chemobiological synthesis and terpene cyclopropanation by artificial metalloenzymes (ArMs). Advanced technologies to accelerate terpene biosynthesis are discussed. This review provides a valuable reference for increasing the diversity of valuable terpenoids and their derivatives, as well as for expanding their potential applications.

Deep hybrid modeling of a HEK293 process: Combining long short-term memory networks with first principles equations

The combination of physical equations with deep learning is becoming a promising methodology for bioprocess digitalization. In this paper, we investigate for the first time the combination of long short-term memory (LSTM) networks with first principles equations in a hybrid workflow to describe human embryonic kidney 293 (HEK293) culture dynamics. Experimental data of 27 extracellular state variables in 20 fed-batch HEK293 cultures were collected in a parallel high throughput 250 mL cultivation system in an industrial process development setting. The adaptive moment estimation method with stochastic regularization and cross-validation were employed for deep learning. A total of 784 hybrid models with varying deep neural network architectures, depths, layers sizes and node activation functions were compared. In most scenarios, hybrid LSTM models outperformed classical hybrid Feedforward Neural Network (FFNN) models in terms of training and testing error. Hybrid LSTM models revealed to be less sensitive to data resampling than FFNN hybrid models. As disadvantages, Hybrid LSTM models are in general more complex (higher number of parameters) and have a higher computation cost than FFNN hybrid models. The hybrid model with the highest prediction accuracy consisted in a LSTM network with seven internal states connected in series with dynamic material balance equations. This hybrid model correctly predicted the dynamics of the 27 state variables (R2 = 0.93 in the test data set), including biomass, key substrates, amino acids and metabolic by-products for around 10 cultivation days.

From genotype to phenotype: computational approaches for inferring microbial traits relevant to the food industry

When selecting microbial strains for the production of fermented foods, various microbial phenotypes need to be taken into account to achieve target product characteristics, such as biosafety, flavor, texture, and health-promoting effects. Through continuous advances in sequencing technologies, microbial whole-genome sequences of increasing quality can now be obtained both cheaper and faster, which increases the relevance of genome-based characterization of microbial phenotypes. Prediction of microbial phenotypes from genome sequences makes it possible to quickly screen large strain collections in silico to identify candidates with desirable traits. Several microbial phenotypes relevant to the production of fermented foods can be predicted using knowledge-based approaches, leveraging our existing understanding of the genetic and molecular mechanisms underlying those phenotypes. In the absence of this knowledge, data-driven approaches can be applied to estimate genotype-phenotype relationships based on large experimental datasets. Here, we review computational methods that implement knowledge- and data-driven approaches for phenotype prediction, as well as methods that combine elements from both approaches. Furthermore, we provide examples of how these methods have been applied in industrial biotechnology, with special focus on the fermented food industry.

Ten quick tips for avoiding pitfalls in multi-omics data integration analyses

Data are the most important elements of bioinformatics: Computational analysis of bioinformatics data, in fact, can help researchers infer new knowledge about biology, chemistry, biophysics, and sometimes even medicine, influencing treatments and therapies for patients. Bioinformatics and high-throughput biological data coming from different sources can even be more helpful, because each of these different data chunks can provide alternative, complementary information about a specific biological phenomenon, similar to multiple photos of the same subject taken from different angles. In this context, the integration of bioinformatics and high-throughput biological data gets a pivotal role in running a successful bioinformatics study. In the last decades, data originating from proteomics, metabolomics, metagenomics, phenomics, transcriptomics, and epigenomics have been labelled -omics data, as a unique name to refer to them, and the integration of these omics data has gained importance in all biological areas. Even if this omics data integration is useful and relevant, due to its heterogeneity, it is not uncommon to make mistakes during the integration phases. We therefore decided to present these ten quick tips to perform an omics data integration correctly, avoiding common mistakes we experienced or noticed in published studies in the past. Even if we designed our ten guidelines for beginners, by using a simple language that (we hope) can be understood by anyone, we believe our ten recommendations should be taken into account by all the bioinformaticians performing omics data integration, including experts.

A General Hybrid Modeling Framework for Systems Biology Applications: Combining Mechanistic Knowledge with Deep Neural Networks under the SBML Standard

In this paper, a computational framework is proposed that merges mechanistic modeling with deep neural networks obeying the Systems Biology Markup Language (SBML) standard. Over the last 20 years, the systems biology community has developed a large number of mechanistic models that are currently stored in public databases in SBML. With the proposed framework, existing SBML models may be redesigned into hybrid systems through the incorporation of deep neural networks into the model core, using a freely available python tool. The so-formed hybrid mechanistic/neural network models are trained with a deep learning algorithm based on the adaptive moment estimation method (ADAM), stochastic regularization and semidirect sensitivity equations. The trained hybrid models are encoded in SBML and uploaded in model databases, where they may be further analyzed as regular SBML models. This approach is illustrated with three well-known case studies: the Escherichia coli threonine synthesis model, the P58IPK signal transduction model, and the Yeast glycolytic oscillations model. The proposed framework is expected to greatly facilitate the widespread use of hybrid modeling techniques for systems biology applications.

The Epidemiology of Infectious Diseases Meets AI: A Match Made in Heaven

Infectious diseases remain a major threat to public health [...].