Massive computational acceleration by using neural networks to emulate mechanism-based biological models

Machine Learning Surrogate Models for Mechanistic Kinetics: Embedding Atom Balance and Positivity

Multiscale simulations of reactive flows are critical in many fields. However, their application is often hindered by the high computational cost of solving detailed chemical kinetics. Recent advances in surrogate models for reactive chemistry offer promising speedups, but ensuring physical consistency remains challenging. In particular, machine learning models for chemical kinetics must enforce atom balance and guarantee the positivity of predicted concentrations. Here, we introduce a positivity preserving projection and a correction by linear interpolation backtracking which simultaneously guarantee both constraints. We demonstrate this using two practical examples from atmospheric chemistry and heterogeneous catalysis, as well as for a large number of random, synthetically generated reaction systems. In all cases, our approach yields exclusively positive model predictions conforming to the atom balance, without reducing the overall accuracy of the model.

Assessing the Effectiveness of Neural Networks and Molecular Dynamics Simulations in Predicting Viscosity of Small Organic Molecules

Viscosity is a crucial material property that influences a wide range of applications, including three-dimensional (3D) printing, lubricants, and solvents. However, experimental approaches to measuring viscosity face challenges such as handling multiple samples, high costs, and limited compound availability. To address these limitations, we have developed computational models for viscosity prediction of small organic molecules, utilizing machine learning (ML) and nonequilibrium molecular dynamics (NEMD) simulations. Our ML framework, which includes feed-forward neural networks (FNN) and physics-informed neural networks (PINN), is based on the largest data set of small molecule viscosities compiled from the literature. The PINN model, in particular, incorporates temperature dependence through a four-parameter model, allowing for the direct prediction of continuous temperature-dependent viscosity curves. The ML models demonstrate exceptional prediction accuracy for the viscosity of various organic compounds across a wide range of temperatures. External validation of our models further confirms that the ML prediction models outperform the NEMD approach in predicting viscosity across a diverse range of organic molecules and temperatures. This highlights the potential of ML models to overcome limitations in traditional MD simulations, which often struggle with accuracy for specific molecules or temperature ranges. Our further feature importance analysis revealed a strong correlation between molecular structure and viscosity. We emphasize the key role of substructures in determining viscosity, offering deeper molecular insights for material design with tailored viscosity.

ML-enhanced peroxisome capacity enables compartmentalization of multienzyme pathway

Repurposing an organelle for specialized metabolism provides an avenue for fermentable, unicellular organisms such as Saccharomyces cerevisiae to mimic compartmentalization of metabolic pathways within different plant tissues. Peroxisomes are attractive organelles for repurposing as they are not required for yeast viability when grown on glucose and can efficiently compartmentalize heterologous enzymes to enable physical separation of cytosolic native metabolism and peroxisomal engineered metabolism. However, when not required, peroxisomes are repressed, leading to low functional capacities for heterologous proteins. Here we engineer peroxisomes with enhanced functional capacities, with the goal of compartmentalizing up to eight metabolic enzymes to enhance titers. We implement a machine learning pipeline that allows the identification of factors to overexpress, culminating in a 137% increase in peroxisome functional capacity compared to a wild-type strain. Improved pathway compartmentalization enables an 80% increase in the biosynthesis titers of the monoterpene geraniol, up to 9.5 g L-1.

Evaluation of water environmental capacity in a northern river-reservoir continuum using environmental fluid dynamics code

The hydrodynamics, water temperature, and water quality model for the Dan River and Renzhuang Reservoir continuum were developed using field monitoring data and the Environmental Fluid Dynamics Code (EFDC). An in-situ water discharge experiment enabled the calculation of water propagation time using a simulated flood progression method and the hydrodynamics module of EFDC. Based on these model results, degradation coefficients for chemical oxygen demand, biochemical oxygen demand, nitrogen (N), phosphorus (P), fluoride, arsenic were determined, revealing significantly higher values when the wetland barrage was opening. According to field monitoring and EFDC model outputs, the water environment capacity (WEC) of total nitrogen (TN) in Dan River was even negative, with the WEC of TN and total phosphorus (TP) in Renzhuang Reservoir being very low at 147 kg/d and 1490 kg/d, respective. Furthermore, the water environment carrying capacity (WECC) was found to be extremely low (~0.2 %) and limited by TN and TP. The results demonstrated that the spatiotemporal variations of water quality presented by the EFDC model facilitate an intuitive and comprehensive observation of water quality compliance rates over time and space, providing valuable references for decision-makers. The WEC and WECC of the study area underscored the urgency of N and P coordination control.

Exploring the spatial effects influencing the EGFR/ERK pathway dynamics with machine learning surrogate models

The fate of cells is regulated by biochemical reactions taking place inside of them, known as intracellular pathways. Cells display a variety of characteristics related to their shape, structure and contained fluid, which influences the diffusion of proteins and their interactions. To gain insights into the spatial effects shaping intracellular regulation, we apply machine learning (ML) to explore a previously developed spatial model of the epidermal growth factor receptor (EGFR) signaling. The model describes the reactions between molecular species inside of cells following the transient activation of EGF receptors. To train our ML models, we conduct 10,000 numerical simulations in parallel where we calculate the cumulative activation of molecules and transcription factors under various conditions such as different diffusion speeds, inactivation rates, and cell structures. We take advantage of the low computational cost of ML algorithms to investigate the effects of cell and nucleus sizes, the diffusion speed of proteins, and the inactivation rate of the Ras molecules on the activation strength of transcription factors. Our results suggest that the predictions by both neural networks and random forests yielded minimal mean square error (MSEs), while linear generalized models displayed a significantly larger MSE. The exploration of the surrogate models has shown that smaller cell and nucleus radii as well, larger diffusion coefficients, and reduced inactivation rates increase the activation of transcription factors. These results are confirmed by numerical simulations. Our ML algorithms can be readily incorporated within multiscale models of tumor growth to embed the spatial effects regulating intracellular pathways, enabling the use of complex cell models within multiscale models while reducing the computational cost.

AI-powered simulation-based inference of a genuinely spatial-stochastic gene regulation model of early mouse embryogenesis

Understanding how multicellular organisms reliably orchestrate cell-fate decisions is a central challenge in developmental biology, particularly in early mammalian development, where tissue-level differentiation arises from seemingly cell-autonomous mechanisms. In this study, we present a multi-scale, spatial-stochastic simulation framework for mouse embryogenesis, focusing on inner cell mass (ICM) differentiation into epiblast (EPI) and primitive endoderm (PRE) at the blastocyst stage. Our framework models key regulatory and tissue-scale interactions in a biophysically realistic fashion, capturing the inherent stochasticity of intracellular gene expression and intercellular signaling, while efficiently simulating these processes by advancing event-driven simulation techniques. Leveraging the power of Simulation-Based Inference (SBI) through the AI-driven Sequential Neural Posterior Estimation (SNPE) algorithm, we conduct a large-scale Bayesian inferential analysis to identify parameter sets that faithfully reproduce experimentally observed features of ICM specification. Our results reveal mechanistic insights into how the combined action of autocrine and paracrine FGF4 signaling coordinates stochastic gene expression at the cellular scale to achieve robust and reproducible ICM patterning at the tissue scale. We further demonstrate that the ICM exhibits a specific time window of sensitivity to exogenous FGF4, enabling lineage proportions to be adjusted based on timing and dosage, thereby extending current experimental findings and providing quantitative predictions for both mutant and wild-type ICM systems. Notably, FGF4 signaling not only ensures correct EPI-PRE lineage proportions but also enhances ICM resilience to perturbations, reducing fate-proportioning errors by 10-20% compared to a purely cell-autonomous system. Additionally, we uncover a surprising role for variability in intracellular initial conditions, showing that high gene-expression heterogeneity can improve both the accuracy and precision of cell-fate proportioning, which remains robust when fewer than 25% of the ICM population experiences perturbed initial conditions. Our work offers a comprehensive, spatial-stochastic description of the biochemical processes driving ICM differentiation and identifies the necessary conditions for its robust unfolding. It also provides a framework for future exploration of similar spatial-stochastic systems in developmental biology.

DyMMM-LEAPS: An ML-based framework for modulating evenness and stability in synthetic microbial communities

There have been a growing number of computational strategies to aid in the design of synthetic microbial consortia. A framework to identify regions in parametric space to maximize two essential properties, evenness and stability, is critical. In this study, we introduce DyMMM-LEAPS (dynamic multispecies metabolic modeling-locating evenness and stability in large parametric space), an extension of the DyMMM framework. Our method explores the large parametric space of genetic circuits in synthetic microbial communities to identify regions of evenness and stability. Due to the high computational costs of exhaustive sampling, we utilize adaptive sampling and surrogate modeling to reduce the number of simulations required to map the vast space. Our framework predicts engineering targets and computes their operating ranges to maximize the probability of the engineered community to have high evenness and stability. We demonstrate our approach by simulating five cocultures and one three-strain culture with different social interactions (cooperation, competition, and predation) employing quorum-sensing-based genetic circuits. In addition to guiding circuit tuning, our pipeline gives an opportunity for a detailed analysis of pockets of evenness and stability for the circuit under investigation, which can further help dissect the relationship between the two properties. DyMMM-LEAPS is easily customizable and can be expanded to a larger community with more complex interactions.

AI in cellular engineering and reprogramming

During the last decade, artificial intelligence (AI) has increasingly been applied in biophysics and related fields, including cellular engineering and reprogramming, offering novel approaches to understand, manipulate, and control cellular function. The potential of AI lies in its ability to analyze complex datasets and generate predictive models. AI algorithms can process large amounts of data from single-cell genomics and multiomic technologies, allowing researchers to gain mechanistic insights into the control of cell identity and function. By integrating and interpreting these complex datasets, AI can help identify key molecular events and regulatory pathways involved in cellular reprogramming. This knowledge can inform the design of precision engineering strategies, such as the development of new transcription factor and signaling molecule cocktails, to manipulate cell identity and drive authentic cell fate across lineage boundaries. Furthermore, when used in combination with computational methods, AI can accelerate and improve the analysis and understanding of the intricate relationships between genes, proteins, and cellular processes. In this review article, we explore the current state of AI applications in biophysics with a specific focus on cellular engineering and reprogramming. Then, we showcase a couple of recent applications where we combined machine learning with experimental and computational techniques. Finally, we briefly discuss the challenges and prospects of AI in cellular engineering and reprogramming, emphasizing the potential of these technologies to revolutionize our ability to engineer cells for a variety of applications, from disease modeling and drug discovery to regenerative medicine and biomanufacturing.

Highly efficient modeling and optimization of neural fiber responses to electrical stimulation

Peripheral neuromodulation has emerged as a powerful modality for controlling physiological functions and treating a variety of medical conditions including chronic pain and organ dysfunction. The underlying complexity of the nonlinear responses to electrical stimulation make it challenging to design precise and effective neuromodulation protocols. Computational models have thus become indispensable in advancing our understanding and control of neural responses to electrical stimulation. However, existing approaches suffer from computational bottlenecks, rendering them unsuitable for real-time applications, large-scale parameter sweeps, or sophisticated optimization. In this work, we introduce an approach for massively parallel estimation and optimization of neural fiber responses to electrical stimulation using machine learning techniques. By leveraging advances in high-performance computing and parallel programming, we present a surrogate fiber model that generates spatiotemporal responses to a wide variety of cuff-based electrical peripheral nerve stimulation protocols. We used our surrogate fiber model to design stimulation parameters for selective stimulation of pig and human vagus nerves. Our approach yields a several-orders-of-magnitude improvement in computational efficiency while retaining generality and high predictive accuracy, demonstrating its robustness and potential to enhance the design and optimization of peripheral neuromodulation therapies.

A perspective-driven and technical evaluation of machine learning in bioreactor scale-up: A case-study for potential model developments

Bioreactor scale-up and scale-down have always been a topical issue for the biopharmaceutical industry and despite considerable effort, the identification of a fail-safe strategy for bioprocess development across scales remains a challenge. With the ubiquitous growth of digital transformation technologies, new scaling methods based on computer models may enable more effective scaling. This study aimed to evaluate the potential application of machine learning (ML) algorithms for bioreactor scale-up, with a specific focus on the prediction of scaling parameters. Factors critical to the development of such models were identified and data for bioreactor scale-up studies involving CHO cell-generated mAb products collated from the literature and public sources for the development of unsupervised and supervised ML models. Comparison of bioreactor performance across scales identified similarities between the different processes and primary differences between small- and large-scale bioreactors. A series of three case studies were developed to assess the relationship between cell growth and scale-sensitive bioreactor features. An embedding layer improved the capability of artificial neural network models to predict cell growth at a large-scale, as this approach captured similarities between the processes. Further models constructed to predict scaling parameters demonstrated how ML models may be applied to assist the scaling process. The development of data sets that include more characterization data with greater variability under different gassing and agitation regimes will also assist the future development of ML tools for bioreactor scaling.

Mechanism-based organization of neural networks to emulate systems biology and pharmacology models

Deep learning neural networks are often described as black boxes, as it is difficult to trace model outputs back to model inputs due to a lack of clarity over the internal mechanisms. This is even true for those neural networks designed to emulate mechanistic models, which simply learn a mapping between the inputs and outputs of mechanistic models, ignoring the underlying processes. Using a mechanistic model studying the pharmacological interaction between opioids and naloxone as a proof-of-concept example, we demonstrated that by reorganizing the neural networks' layers to mimic the structure of the mechanistic model, it is possible to achieve better training rates and prediction accuracy relative to the previously proposed black-box neural networks, while maintaining the interpretability of the mechanistic simulations. Our framework can be used to emulate mechanistic models in a large parameter space and offers an example on the utility of increasing the interpretability of deep learning networks.

Incorporating temporal information during feature engineering bolsters emulation of spatio-temporal emergence

MOTIVATION

Emergent biological dynamics derive from the evolution of lower-level spatial and temporal processes. A long-standing challenge for scientists and engineers is identifying simple low-level rules that give rise to complex higher-level dynamics. High-resolution biological data acquisition enables this identification and has evolved at a rapid pace for both experimental and computational approaches. Simultaneously harnessing the resolution and managing the expense of emerging technologies-e.g. live cell imaging, scRNAseq, agent-based models-requires a deeper understanding of how spatial and temporal axes impact biological systems. Effective emulation is a promising solution to manage the expense of increasingly complex high-resolution computational models. In this research, we focus on the emulation of a tumor microenvironment agent-based model to examine the relationship between spatial and temporal environment features, and emergent tumor properties.

RESULTS

Despite significant feature engineering, we find limited predictive capacity of tumor properties from initial system representations. However, incorporating temporal information derived from intermediate simulation states dramatically improves the predictive performance of machine learning models. We train a deep-learning emulator on intermediate simulation states and observe promising enhancements over emulators trained solely on initial conditions. Our results underscore the importance of incorporating temporal information in the evaluation of spatio-temporal emergent behavior. Nevertheless, the emulators exhibit inconsistent performance, suggesting that the underlying model characterizes unique cell populations dynamics that are not easily replaced.

AVAILABILITY AND IMPLEMENTATION

All source codes for the agent-based model, emulation, and analyses are publicly available at the corresponding DOIs: 10.5281/zenodo.10622155, 10.5281/zenodo.10611675, 10.5281/zenodo.10621244, respectively.

Systems Biology Approaches to Understanding COVID-19 Spread in the Population

In essence, the COVID-19 pandemic can be regarded as a systems biology problem, with the entire world as the system, and the human population as the element transitioning from one state to another with certain transition rates. While capturing all the relevant features of such a complex system is hardly possible, compartmental epidemiological models can be used as an appropriate simplification to model the system's dynamics and infer its important characteristics, such as basic and effective reproductive numbers of the virus. These measures can later be used as response variables in feature selection methods to uncover the main factors contributing to disease transmissibility. We here demonstrate that a combination of dynamic modeling and machine learning approaches can represent a powerful tool in understanding the spread, not only of COVID-19, but of any infectious disease of epidemiological proportions.

A newcomer's guide to deep learning for inverse design in nano-photonics

Nanophotonic devices manipulate light at sub-wavelength scales, enabling tasks such as light concentration, routing, and filtering. Designing these devices to achieve precise light-matter interactions using structural parameters and materials is a challenging task. Traditionally, solving this problem has relied on computationally expensive, iterative methods. In recent years, deep learning techniques have emerged as promising tools for tackling the inverse design of nanophotonic devices. While several review articles have provided an overview of the progress in this rapidly evolving field, there is a need for a comprehensive tutorial that specifically targets newcomers without prior experience in deep learning. Our goal is to address this gap and provide practical guidance for applying deep learning to individual scientific problems. We introduce the fundamental concepts of deep learning and critically discuss the potential benefits it offers for various inverse design problems in nanophotonics. We present a suggested workflow and detailed, practical design guidelines to help newcomers navigate the challenges they may encounter. By following our guide, newcomers can avoid frustrating roadblocks commonly experienced when venturing into deep learning for the first time. In a second part, we explore different iterative and direct deep learning-based techniques for inverse design, and evaluate their respective advantages and limitations. To enhance understanding and facilitate implementation, we supplement the manuscript with detailed Python notebook examples, illustrating each step of the discussed processes. While our tutorial primarily focuses on researchers in (nano-)photonics, it is also relevant for those working with deep learning in other research domains. We aim at providing a solid starting point to empower researchers to leverage the potential of deep learning in their scientific pursuits.

Studying stochastic systems biology of the cell with single-cell genomics data

Recent experimental developments in genome-wide RNA quantification hold considerable promise for systems biology. However, rigorously probing the biology of living cells requires a unified mathematical framework that accounts for single-molecule biological stochasticity in the context of technical variation associated with genomics assays. We review models for a variety of RNA transcription processes, as well as the encapsulation and library construction steps of microfluidics-based single-cell RNA sequencing, and present a framework to integrate these phenomena by the manipulation of generating functions. Finally, we use simulated scenarios and biological data to illustrate the implications and applications of the approach.

Deep Neural Networks for Predicting Single-Cell Responses and Probability Landscapes

Engineering biology relies on the accurate prediction of cell responses. However, making these predictions is challenging for a variety of reasons, including the stochasticity of biochemical reactions, variability between cells, and incomplete information about underlying biological processes. Machine learning methods, which can model diverse input-output relationships without requiring a priori mechanistic knowledge, are an ideal tool for this task. For example, such approaches can be used to predict gene expression dynamics given time-series data of past expression history. To explore this application, we computationally simulated single-cell responses, incorporating different sources of noise and alternative genetic circuit designs. We showed that deep neural networks trained on these simulated data were able to correctly infer the underlying dynamics of a cell response even in the presence of measurement noise and stochasticity in the biochemical reactions. The training set size and the amount of past data provided as inputs both affected prediction quality, with cascaded genetic circuits that introduce delays requiring more past data. We also tested prediction performance on a bistable auto-activation circuit, finding that our initial method for predicting a single trajectory was fundamentally ill-suited for multimodal dynamics. To address this, we updated the network architecture to predict the entire distribution of future states, showing it could accurately predict bimodal expression distributions. Overall, these methods can be readily applied to the diverse prediction tasks necessary to predict and control a variety of biological circuits, a key aspect of many synthetic biology applications.

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.

Studying stochastic systems biology of the cell with single-cell genomics data

Recent experimental developments in genome-wide RNA quantification hold considerable promise for systems biology. However, rigorously probing the biology of living cells requires a unified mathematical framework that accounts for single-molecule biological stochasticity in the context of technical variation associated with genomics assays. We review models for a variety of RNA transcription processes, as well as the encapsulation and library construction steps of microfluidics-based single-cell RNA sequencing, and present a framework to integrate these phenomena by the manipulation of generating functions. Finally, we use simulated scenarios and biological data to illustrate the implications and applications of the approach.

A novel framework for the evaluation of coastal protection schemes through integration of numerical modelling and artificial intelligence into the Sand Engine App

There is growing interest in the adoption of Engineering with Nature or Nature Based Solutions for coastal protection including large mega-nourishment interventions. However, there are still many unknowns on the variables and design features influencing their functionalities. There are also challenges in the optimization of coastal modelling outputs or information usage in support of decision-making. In this study, more than five hundred numerical simulations with different sandengine designs and different locations along Morecambe Bay (UK) were conducted in Delft3D. Twelve Artificial Neural Networking ensemble models structures were trained on the simulated data to predict the influence of different sand engines on water depth, wave height and sediment transports with good performance. The ensemble models were then packed into a Sand Engine App developed in MATLAB and designed to calculate the impact of different sand engine features on the above variables based on users' inputs of sandengine designs.

Decoding CAR T cell phenotype using combinatorial signaling motif libraries and machine learning

Chimeric antigen receptor (CAR) costimulatory domains derived from native immune receptors steer the phenotypic output of therapeutic T cells. We constructed a library of CARs containing ~2300 synthetic costimulatory domains, built from combinations of 13 signaling motifs. These CARs promoted diverse human T cell fates, which were sensitive to motif combinations and configurations. Neural networks trained to decode the combinatorial grammar of CAR signaling motifs allowed extraction of key design rules. For example, non-native combinations of motifs that bind tumor necrosis factor receptor-associated factors (TRAFs) and phospholipase C gamma 1 (PLCγ1) enhanced cytotoxicity and stemness associated with effective tumor killing. Thus, libraries built from minimal building blocks of signaling, combined with machine learning, can efficiently guide engineering of receptors with desired phenotypes.

Application of a Machine Learning Algorithm for Evaluation of Stiff Fractional Modeling of Polytropic Gas Spheres and Electric Circuits

Fractional polytropic gas sphere problems and electrical engineering models typically simulated with interconnected circuits have numerous applications in physical, astrophysical phenomena, and thermionic currents. Generally, most of these models are singular-nonlinear, symmetric, and include time delay, which has increased attention to them among researchers. In this work, we explored deep neural networks (DNNs) with an optimization algorithm to calculate the approximate solutions for nonlinear fractional differential equations (NFDEs). The target data-driven design of the DNN-LM algorithm was further implemented on the fractional models to study the rigorous impact and symmetry of different parameters on RL, RC circuits, and polytropic gas spheres. The targeted data generated from the analytical and numerical approaches in the literature for different cases were utilized by the deep neural networks to predict the numerical solutions by minimizing the differences in mean square error using the Levenberg–Marquardt algorithm. The numerical solutions obtained by the designed technique were contrasted with the multi-step reproducing kernel Hilbert space method (MS-RKM), Laplace transformation method (LTM), and Padé approximations. The results demonstrate the accuracy of the design technique as the DNN-LM algorithm overlaps with the actual results with minimum percentage absolute errors that lie between 10−8 and 10−12. The extensive graphical and statistical analysis of the designed technique showed that the DNN-LM algorithm is dependable and facilitates the examination of higher-order nonlinear complex problems due to the flexibility of the DNN architecture and the effectiveness of the optimization procedure.

Toward predictive engineering of gene circuits

Many synthetic biology applications rely on programming living cells using gene circuits - the assembly and wiring of genetic elements to control cellular behaviors. Extensive progress has been made in constructing gene circuits with diverse functions and applications. For many circuit functions, however, it remains challenging to ensure that the circuits operate in a predictable manner. Although the notion of predictability may appear intuitive, close inspection suggests that it is not always clear what constitutes predictability. We dissect this concept and how it can be confounded by the complexity of a circuit, the complexity of the context, and the interplay between the two. We discuss circuit engineering strategies, in both computation and experiment, that have been used to improve the predictability of gene circuits.

Calibrating spatiotemporal models of microbial communities to microscopy data: A review

Spatiotemporal models that account for heterogeneity within microbial communities rely on single-cell data for calibration and validation. Such data, commonly collected via microscopy and flow cytometry, have been made more accessible by recent advances in microfluidics platforms and data processing pipelines. However, validating models against such data poses significant challenges. Validation practices vary widely between modelling studies; systematic and rigorous methods have not been widely adopted. Similar challenges are faced by the (macrobial) ecology community, in which systematic calibration approaches are often employed to improve quantitative predictions from computational models. Here, we review single-cell observation techniques that are being applied to study microbial communities and the calibration strategies that are being employed for accompanying spatiotemporal models. To facilitate future calibration efforts, we have compiled a list of summary statistics relevant for quantifying spatiotemporal patterns in microbial communities. Finally, we highlight some recently developed techniques that hold promise for improved model calibration, including algorithmic guidance of summary statistic selection and machine learning approaches for efficient model simulation.

Approximating solutions of the Chemical Master equation using neural networks

The Chemical Master Equation (CME) provides an accurate description of stochastic biochemical reaction networks in well-mixed conditions, but it cannot be solved analytically for most systems of practical interest. Although Monte Carlo methods provide a principled means to probe system dynamics, the large number of simulations typically required can render the estimation of molecule number distributions and other quantities infeasible. In this article, we aim to leverage the representational power of neural networks to approximate the solutions of the CME and propose a framework for the Neural Estimation of Stochastic Simulations for Inference and Exploration (Nessie). Our approach is based on training neural networks to learn the distributions predicted by the CME from relatively few stochastic simulations. We show on biologically relevant examples that simple neural networks with one hidden layer can capture highly complex distributions across parameter space, thereby accelerating computationally intensive tasks such as parameter exploration and inference.

Transition State Theory-Inspired Neural Network for Estimating the Viscosity of Deep Eutectic Solvents

The lack of accurate methods for predicting the viscosity of solvent materials, especially those with complex interactions, remains unresolved. Deep eutectic solvents (DESs), an emerging class of green solvents, have a severe lack of viscosity data, resulting in their application still staying at the stage of random trial and error, and it is difficult for them to be implemented on an industrial scale. In this work, we demonstrate the successful prediction of the viscosity of DESs based on the transition state theory-inspired neural network (TSTiNet). The TSTiNet adopts multilayer perceptron (MLP) for the transition state theory-inspired equation (TSTiEq) parameters calculation and verification using the most comprehensive DESs viscosity data set to date. For the energy parameters of the TSTiEq, the constant assumption and the fast iteration with the help of MLP can allow TSTiNet to achieve the best performance (the average absolute relative deviation on the test set of 6.84% and R ² of 0.9805). Compared with the traditional machine learning methods, the TSTiNet has better generalization ability and dramatically reduces the maximum relative deviation of prediction under the constraints of the thermodynamic formulation. It requires only the structural information on DESs and is the most accurate and reliable model available for DESs viscosity prediction.

Engineering whole-cell microbial biosensors: Design principles and applications in monitoring and treatment of heavy metals and organic pollutants

Biosensors have been widely used as cost-effective, rapid, in situ, and real-time analytical tools for monitoring environments. The development of synthetic biology has enabled emergence of genetically engineered whole-cell microbial biosensors. This review updates the design and optimization principles for a diverse array of whole-cell biosensors based on transcription factors (TF) including activators or repressors derived from heavy metal resistance systems, alkanes, and aromatics metabolic pathways of bacteria. By designing genetic circuits, the whole-cell biosensors could be engineered to intelligently sense heavy metals (Hg²⁺, Zn²⁺, Pb²⁺, Au³⁺, Cd²⁺, As³⁺, Ni²⁺, Cu²⁺, and UO₂²⁺) or organic compounds (alcohols, alkanes, phenols, and benzenes) through one-component or two-component system-based TFs, transduce signals through genetic amplifiers, and response as various outputs such as cell fluorescence and bioelectricity for monitoring heavy metals and organic pollutants in real conditions, synthetic curli and surface metal-binding peptides for in situ bio-sorption of heavy metals. We further review strategies that have been implemented to optimize the selectivity and correlation between ligand concentration and output signal of the TF-based biosensors, so as to meet requirements of practical applications. The optimization strategies include protein engineering to change specificities, promoter engineering to improve sensitivities, and genetic circuit-based amplification to enhance dynamic ranges via designing transcriptional amplifiers, logic gates, and feedback loops. At last, we outlook future trends in developing novel forms of biosensors.

Recurrent neural networks enable design of multifunctional synthetic human gut microbiome dynamics

Predicting the dynamics and functions of microbiomes constructed from the bottom-up is a key challenge in exploiting them to our benefit. Current models based on ecological theory fail to capture complex community behaviors due to higher order interactions, do not scale well with increasing complexity and in considering multiple functions. We develop and apply a long short-term memory (LSTM) framework to advance our understanding of community assembly and health-relevant metabolite production using a synthetic human gut community. A mainstay of recurrent neural networks, the LSTM learns a high dimensional data-driven non-linear dynamical system model. We show that the LSTM model can outperform the widely used generalized Lotka-Volterra model based on ecological theory. We build methods to decipher microbe-microbe and microbe-metabolite interactions from an otherwise black-box model. These methods highlight that Actinobacteria, Firmicutes and Proteobacteria are significant drivers of metabolite production whereas Bacteroides shape community dynamics. We use the LSTM model to navigate a large multidimensional functional landscape to design communities with unique health-relevant metabolite profiles and temporal behaviors. In sum, the accuracy of the LSTM model can be exploited for experimental planning and to guide the design of synthetic microbiomes with target dynamic functions.

A Bayesian finite-element trained machine learning approach for predicting post-burn contraction

Burn injuries can decrease the quality of life of a patient tremendously, because of esthetic reasons and because of contractions that result from them. In severe case, skin contraction takes place at such a large extent that joint mobility of a patient is significantly inhibited. In these cases, one refers to a contracture. In order to predict the evolution of post-wounding skin, several mathematical model frameworks have been set up. These frameworks are based on complicated systems of partial differential equations that need finite element-like discretizations for the approximation of the solution. Since these computational frameworks can be expensive in terms of computation time and resources, we study the applicability of neural networks to reproduce the finite element results. Our neural network is able to simulate the evolution of skin in terms of contraction for over one year. The simulations are based on 25 input parameters that are characteristic for the patient and the injury. One of such input parameters is the stiffness of the skin. The neural network results have yielded an average goodness of fit (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^2$$\end{document}R2) of 0.9928 (± 0.0013). Further, a tremendous speed-up of 19354X was obtained with the neural network. We illustrate the applicability by an online medical App that takes into account the age of the patient and the length of the burn.

A Computational Platform Integrating a Mechanistic Model of Crohn's Disease for Predicting Temporal Progression of Mucosal Damage and Healing

INTRODUCTION

Physicians are often required to make treatment decisions for patients with Crohn's disease on the basis of limited objective information about the state of the patient's gastrointestinal tissue while aiming to achieve mucosal healing. Tools to predict changes in mucosal health with treatment are needed. We evaluated a computational approach integrating a mechanistic model of Crohn's disease with a responder classifier to predict temporal changes in mucosal health.

METHODS

A hybrid mechanistic-statistical platform was developed to predict biomarker and tissue health time courses in patients with Crohn's disease. Eligible patients from the VERSIFY study (n = 69) were classified into archetypical response cohorts using a decision tree based on early treatment data and baseline characteristics. A virtual patient matching algorithm assigned a digital twin to each patient from their corresponding response cohort. The digital twin was used to forecast response to treatment using the mechanistic model.

RESULTS

The responder classifier predicted endoscopic remission and mucosal healing for treatment with vedolizumab over 26 weeks, with overall sensitivities of 80% and 75% and overall specificities of 69% and 70%, respectively. Predictions for changes in tissue damage over time in the validation set (n = 31), a measure of the overall performance of the platform, were considered good (at least 70% of data points matched), fair (at least 50%), and poor (less than 50%) for 71%, 23%, and 6% of patients, respectively.

CONCLUSION

Hybrid computational tools including mechanistic components represent a promising form of decision support that can predict outcomes and patient progress in Crohn's disease.

Machine Learning-Based Continuous Intracranial Pressure Prediction for Traumatic Injury Patients

Structured Abstract—Objective: Abnormal elevation of intracranial pressure (ICP) can cause dangerous or even fatal outcomes. The early detection of high intracranial pressure events can be crucial in saving lives in an intensive care unit (ICU). Despite many applications of machine learning (ML) techniques related to clinical diagnosis, ML applications for continuous ICP detection or short-term predictions have been rarely reported. This study proposes an efficient method of applying an artificial recurrent neural network on the early prediction of ICP evaluation continuously for TBI patients. Methods: After ICP data preprocessing, the learning model is generated for thirteen patients to continuously predict the ICP signal occurrence and classify events for the upcoming 10 minutes by inputting the previous 20-minutes of the ICP signal. Results: As the overall model performance, the average accuracy is 94.62%, the average sensitivity is 74.91%, the average specificity is 94.83%, and the average root mean square error is approximately 2.18 mmHg. Conclusion: This research addresses a significant clinical problem with the management of traumatic brain injury patients. The machine learning model data enables early prediction of ICP continuously in a real-time fashion, which is crucial for appropriate clinical interventions. The results show that our machine learning-based model has high adaptive performance, accuracy, and efficiency.

The Use of Artificial Neural Networks to Forecast the Behavior of Agent-Based Models of Pathophysiology: An Example Utilizing an Agent-Based Model of Sepsis

Introduction: Disease states are being characterized at finer and finer levels of resolution via biomarker or gene expression profiles, while at the same time. Machine learning (ML) is increasingly used to analyze and potentially classify or predict the behavior of biological systems based on such characterization. As ML applications are extremely data-intensive, given the relative sparsity of biomedical data sets ML training of artificial neural networks (ANNs) often require the use of synthetic training data. Agent-based models (ABMs) that incorporate known biological mechanisms and their associated stochastic properties are a potential means of generating synthetic data. Herein we present an example of ML used to train an artificial neural network (ANN) as a surrogate system used to predict the time evolution of an ABM focusing on the clinical condition of sepsis. Methods: The disease trajectories for clinical sepsis, in terms of temporal cytokine and phenotypic dynamics, can be interpreted as a random dynamical system. The Innate Immune Response Agent-based Model (IIRABM) is a well-established model that utilizes known cellular and molecular rules to simulate disease trajectories corresponding to clinical sepsis. We have utilized two distinct neural network architectures, Long Short-Term Memory and Multi-Layer Perceptron, to take a time sequence of five measurements of eleven IIRABM simulated serum cytokine concentrations as input and to return both the future cytokine trajectories as well as an aggregate metric representing the patient's state of health. Results: The ANNs predicted model trajectories with the expected amount of error, due to stochasticity in the simulation, and recognizing that the mapping from a specific cytokine profile to a state-of-health is not unique. The Multi-Layer Perceptron neural network, generated predictions with a more accurate forecasted trajectory cone. Discussion: This work serves as a proof-of-concept for the use of ANNs to predict disease progression in sepsis as represented by an ABM. The findings demonstrate that multicellular systems with intrinsic stochasticity can be approximated with an ANN, but that forecasting a specific trajectory of the system requires sequential updating of the system state to provide a rolling forecast horizon.

AI-driven prediction of SARS-CoV-2 variant binding trends from atomistic simulations

We present a novel technique to predict binding affinity trends between two molecules from atomistic molecular dynamics simulations. The technique uses a neural network algorithm applied to a series of images encoding the distance between two molecules in time. We demonstrate that our algorithm is capable of separating with high accuracy non-hydrophobic mutations with low binding affinity from those with high binding affinity. Moreover, we show high accuracy in prediction using a small subset of the simulation, therefore requiring a much shorter simulation time. We apply our algorithm to the binding between several variants of the SARS-CoV-2 spike protein and the human receptor ACE2.

Synthetic biology as driver for the biologization of materials sciences

Materials in nature have fascinating properties that serve as a continuous source of inspiration for materials scientists. Accordingly, bio-mimetic and bio-inspired approaches have yielded remarkable structural and functional materials for a plethora of applications. Despite these advances, many properties of natural materials remain challenging or yet impossible to incorporate into synthetic materials. Natural materials are produced by living cells, which sense and process environmental cues and conditions by means of signaling and genetic programs, thereby controlling the biosynthesis, remodeling, functionalization, or degradation of the natural material. In this context, synthetic biology offers unique opportunities in materials sciences by providing direct access to the rational engineering of how a cell senses and processes environmental information and translates them into the properties and functions of materials. Here, we identify and review two main directions by which synthetic biology can be harnessed to provide new impulses for the biologization of the materials sciences: first, the engineering of cells to produce precursors for the subsequent synthesis of materials. This includes materials that are otherwise produced from petrochemical resources, but also materials where the bio-produced substances contribute unique properties and functions not existing in traditional materials. Second, engineered living materials that are formed or assembled by cells or in which cells contribute specific functions while remaining an integral part of the living composite material. We finally provide a perspective of future scientific directions of this promising area of research and discuss science policy that would be required to support research and development in this field.

Collective colony growth is optimized by branching pattern formation in Pseudomonas aeruginosa

Branching pattern formation is common in many microbes. Extensive studies have focused on addressing how such patterns emerge from local cell-cell and cell-environment interactions. However, little is known about whether and to what extent these patterns play a physiological role. Here, we consider the colonization of bacteria as an optimization problem to find the colony patterns that maximize colony growth efficiency under different environmental conditions. We demonstrate that Pseudomonas aeruginosa colonies develop branching patterns with characteristics comparable to the prediction of modeling; for example, colonies form thin branches in a nutrient-poor environment. Hence, the formation of branching patterns represents an optimal strategy for the growth of Pseudomonas aeruginosa colonies. The quantitative relationship between colony patterns and growth conditions enables us to develop a coarse-grained model to predict diverse colony patterns under more complex conditions, which we validated experimentally. Our results offer new insights into branching pattern formation as a problem-solving social behavior in microbes and enable fast and accurate predictions of complex spatial patterns in branching colonies.

Acceleration of PDE-Based Biological Simulation Through the Development of Neural Network Metamodels

PURPOSE OF REVIEW

Partial differential equation (PDE) mathematical models of biological systems and the simulation approaches used to solve them are widely used to test hypotheses and infer regulatory interactions based on optimization of the PDE model against the observed data. In this review, we discuss the ability of powerful machine learning methods to accelerate the parametric screening of biophysical informed- PDE systems.

RECENT FINDINGS

A major shortcoming in more broad adaptation of PDE-based models is the high computational complexity required to solve and optimize the models and it requires many simulations to traverse the very high-dimensional parameter spaces during model calibration and inference tasks. For instance, when scaling up to tens of millions of simulations for optimization and sensitivity analysis of the PDE models, compute times quickly extend from months to years for sufficient coverage to solve the problems. For many systems, this brute-force approach is simply not feasible. Recently, neural network metamodels have been shown to be an efficient way to accelerate PDE model calibration and here we look at the benefits and limitations in extending the PDE acceleration methods to improve optimization and sensitivity analysis.

SUMMARY

We use an example simulation to quantitatively and qualitatively show how neural network metamodels can be accurate and fast and demonstrate their potential for optimization of complex spatiotemporal problems in biology. We expect these approaches will be broadly applied to speed up scientific research and discovery in biology and other systems that can be described by complex PDE systems.

Engineering spatiotemporal organization and dynamics in synthetic cells

Constructing synthetic cells has recently become an appealing area of research. Decades of research in biochemistry and cell biology have amassed detailed part lists of components involved in various cellular processes. Nevertheless, recreating any cellular process in vitro in cell-sized compartments remains ambitious and challenging. Two broad features or principles are key to the development of synthetic cells-compartmentalization and self-organization/spatiotemporal dynamics. In this review article, we discuss the current state of the art and research trends in the engineering of synthetic cell membranes, development of internal compartmentalization, reconstitution of self-organizing dynamics, and integration of activities across scales of space and time. We also identify some research areas that could play a major role in advancing the impact and utility of engineered synthetic cells. This article is categorized under: Biology-Inspired Nanomaterials > Lipid-Based Structures Biology-Inspired Nanomaterials > Protein and Virus-Based Structures.

Application of artificial intelligence to the in silico assessment of antimicrobial resistance and risks to human and animal health presented by priority enteric bacterial pathogens

Each year, approximately one in eight Canadians are affected by foodborne illness, either through outbreaks or sporadic illness, with animals being the major reservoir for the pathogens. Whole genome sequence analyses are now routinely implemented by public and animal health laboratories to define epidemiological disease clusters and to identify potential sources of infection. Similarly, a number of bioinformatics tools can be used to identify virulence and antimicrobial resistance (AMR) determinants in the genomes of pathogenic strains. Many important clinical and phenotypic characteristics of these pathogens can now be predicted using machine learning algorithms applied to whole genome sequence data. In this overview, we compare the ability of support vector machines, gradient-boosted decision trees and artificial neural networks to predict the levels of AMR within Salmonella enterica and extended-spectrum β-lactamase (ESBL) producing Escherichia coli. We show that minimum inhibitory concentrations (MIC) for each of 13 antimicrobials for S. enterica strains can be accurately determined, and that ESBL-producing E. coli strains can be accurately classified as susceptible, intermediate or resistant for each of seven antimicrobials. In addition to AMR and bacterial populations of greatest risk to human health, artificial intelligence algorithms hold promise as tools to predict other clinically and epidemiologically important phenotypes of enteric pathogens.

Deep Learning Meets Nanophotonics: A Generalized Accurate Predictor for Near Fields and Far Fields of Arbitrary 3D Nanostructures

Deep artificial neural networks are powerful tools with many possible applications in nanophotonics. Here, we demonstrate how a deep neural network can be used as a fast, general purpose predictor of the full near-field and far-field response of plasmonic and dielectric nanostructures. A trained neural network is shown to infer the internal fields of arbitrary three-dimensional nanostructures many orders of magnitude faster compared to conventional numerical simulations. Secondary physical quantities are derived from the deep learning predictions and faithfully reproduce a wide variety of physical effects without requiring specific training. We discuss the strengths and limitations of the neural network approach using a number of model studies of single particles and their near-field interactions. Our approach paves the way for fast, yet universal, methods for design and analysis of nanophotonic systems.