Deep learning predicts microbial interactions from self-organized spatiotemporal patterns

Disentangling the feedback loops driving spatial patterning in microbial communities

The properties of multispecies biofilms are determined by how species are arranged in space. How these patterns emerge is a complex and largely unsolved problem. Here, we synthesize the known factors affecting pattern formation, identify the interdependencies and feedback loops coupling them, and discuss approaches to disentangle their effects. Finally, we propose an interdisciplinary research program that could create a predictive understanding of pattern formation in microbial communities.

Deciphering and designing microbial communities by genome-scale metabolic modelling

Microbial communities are shaped by the complex interactions among organisms and the environment. Genome-scale metabolic models (GEMs) can provide deeper insights into the complexity and ecological properties of various microbial communities, revealing their intricate interactions. Many researchers have modified GEMs for the microbial communities based on specific needs. Thus, GEMs need to be comprehensively summarized to better understand the trends in their development. In this review, we summarized the key developments in deciphering and designing microbial communities using different GEMs. A timeline of selected highlights in GEMs indicated that this area is evolving from the single-strain level to the microbial community level. Then, we outlined a framework for constructing GEMs of microbial communities. We also summarized the models and resources of static and dynamic community-level GEMs. We focused on the role of external environmental and intracellular resources in shaping the assembly of microbial communities. Finally, we discussed the key challenges and future directions of GEMs, focusing on the integration of GEMs with quorum sensing mechanisms, microbial ecology interactions, machine learning algorithms, and automatic modeling, all of which contribute to consortia-based applications in different fields.

SPLANG-a synthetic poisson-lognormal-based abundance and network generative model for microbial interaction inference algorithms

Microbes are pervasive and their interaction with each other and the environment can impact fields as diverse as health and agriculture. While network inference and related algorithms that use abundance data from pyrosequencing can infer microbial interaction networks, the ambiguity surrounding the actual underlying networks hampers the validation of these algorithms. This study introduces a generative model to synthesize both the underlying interactive network and observable abundance data, serving as a test bed for the existing and future network inference algorithms. We tested our generative model with four typical network inference algorithms; our results suggest that none of these algorithms demonstrate adequate accuracy for inferring ecologies of non-commensalistic species, either mutualistic or competitive. We further explored the potential for predictability by combining existing algorithms with an oracle algorithm built by fusing the results of several existing algorithms. The oracle algorithm reveals promising improvements in predictability, although it falls short when applied to networks characterized by dense interspecies taxa interactions. Our work underscores the need for the continued development and validation of algorithms to unravel the intricacies of microbial interaction networks.

Engineering agricultural soil microbiomes and predicting plant phenotypes

Plant growth-promoting rhizobacteria (PGPR) can improve crop yields, nutrient use efficiency, plant tolerance to stressors, and confer benefits to future generations of crops grown in the same soil. Unlocking the potential of microbial communities in the rhizosphere and endosphere is therefore of great interest for sustainable agriculture advancements. Before plant microbiomes can be engineered to confer desirable phenotypic effects on their plant hosts, a deeper understanding of the interacting factors influencing rhizosphere community structure and function is needed. Dealing with this complexity is becoming more feasible using computational approaches. In this review, we discuss recent advances at the intersection of experimental and computational strategies for the investigation of plant-microbiome interactions and the engineering of desirable soil microbiomes.

Drivers and human impacts on topsoil bacterial and fungal community biogeography across Australia

Soil microbial diversity mediates a wide range of key processes and ecosystem services influencing planetary health. Our knowledge of microbial biogeography patterns, spatial drivers and human impacts at the continental scale remains limited. Here, we reveal the drivers of bacterial and fungal community distribution in Australian topsoils using 1384 soil samples from diverse bioregions. Our findings highlight that climate factors, particularly precipitation and temperature, along with soil properties, are the primary drivers of topsoil microbial biogeography. Using random forest machine-learning models, we generated high-resolution maps of soil bacteria and fungi across continental Australia. The maps revealed microbial hotspots, for example, the eastern coast, southeastern coast, and west coast were dominated by Proteobacteria and Acidobacteria. Fungal distribution is strongly influenced by precipitation, with Ascomycota dominating the central region. This study also demonstrated the impact of human modification on the underground microbial community at the continental scale, which significantly increased the relative abundance of Proteobacteria and Ascomycota, but decreased Chloroflexi and Basidiomycota. The variations in microbial phyla could be attributed to distinct responses to altered environmental factors after human modifications. This study provides insights into the biogeography of soil microbiota, valuable for regional soil biodiversity assessments and monitoring microbial responses to global changes.

Microbial community composition predicts bacterial production across ocean ecosystems

Microbial ecological functions are an emergent property of community composition. For some ecological functions, this link is strong enough that community composition can be used to estimate the quantity of an ecological function. Here, we apply random forest regression models to compare the predictive performance of community composition and environmental data for bacterial production (BP). Using data from two independent long-term ecological research sites-Palmer LTER in Antarctica and Station SPOT in California-we found that community composition was a strong predictor of BP. The top performing model achieved an R2 of 0.84 and RMSE of 20.2 pmol L-1 hr-1 on independent validation data, outperforming a model based solely on environmental data (R2 = 0.32, RMSE = 51.4 pmol L-1 hr-1). We then operationalized our top performing model, estimating BP for 346 Antarctic samples from 2015 to 2020 for which only community composition data were available. Our predictions resolved spatial trends in BP with significance in the Antarctic (P value = 1 × 10-4) and highlighted important taxa for BP across ocean basins. Our results demonstrate a strong link between microbial community composition and microbial ecosystem function and begin to leverage long-term datasets to construct models of BP based on microbial community composition.

Machine learning in fermentative biohydrogen production: Advantages, challenges, and applications

Hydrogen can be produced in an environmentally friendly manner through biological processes using a variety of organic waste and biomass as feedstock. However, the complexity of biological processes limits their predictability and reliability, which hinders the scale-up and dissemination. This article reviews contemporary research and perspectives on the application of machine learning in biohydrogen production technology. Several machine learning algorithems have recently been implemented for modeling the nonlinear and complex relationships among operational and performance parameters in biohydrogen production as well as predicting the process performance and microbial population dynamics. Reinforced machine learning methods exhibited precise state prediction and retrieved the underlying kinetics effectively. Machine-learning based prediction was also improved by using microbial sequencing data as input parameters. Further research on machine learning could be instrumental in designing a process control tool to maintain reliable hydrogen production performance and identify connection between the process performance and the microbial population.

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.

Machine Learning Advances in Microbiology: A Review of Methods and Applications

Microorganisms play an important role in natural material and elemental cycles. Many common and general biology research techniques rely on microorganisms. Machine learning has been gradually integrated with multiple fields of study. Machine learning, including deep learning, aims to use mathematical insights to optimize variational functions to aid microbiology using various types of available data to help humans organize and apply collective knowledge of various research objects in a systematic and scaled manner. Classification and prediction have become the main achievements in the development of microbial community research in the direction of computational biology. This review summarizes the application and development of machine learning and deep learning in the field of microbiology and shows and compares the advantages and disadvantages of different algorithm tools in four fields: microbiome and taxonomy, microbial ecology, pathogen and epidemiology, and drug discovery.

Synthetic Soil Aggregates: Bioprinted Habitats for High-Throughput Microbial Metaphenomics

The dynamics of microbial processes are difficult to study in natural soil, owing to the small spatial scales on which microorganisms operate and to the opacity and chemical complexity of the soil habitat. To circumvent these challenges, we have created a 3D-bioprinted habitat that mimics aspects of natural soil aggregates while providing a chemically defined and translucent alternative culturing method for soil microorganisms. Our Synthetic Soil Aggregates (SSAs) retain the porosity, permeability, and patchy resource distribution of natural soil aggregates—parameters that are expected to influence emergent microbial community interactions. We demonstrate the printability and viability of several different microorganisms within SSAs and show how the SSAs can be integrated into a multi-omics workflow for single SSA resolution genomics, metabolomics, proteomics, lipidomics, and biogeochemical assays. We study the impact of the structured habitat on the distribution of a model co-culture microbial community and find that it is significantly different from the spatial organization of the same community in liquid culture, indicating a potential for SSAs to reproduce naturally occurring emergent community phenotypes. The SSAs have the potential as a tool to help researchers quantify microbial scale processes in situ and achieve high-resolution data from the interplay between environmental properties and microbial ecology.

Machine Learning Predicts Biogeochemistry from Microbial Community Structure in a Complex Model System

Microbial community structure is influenced by the environment and in turn exerts control on many environmental parameters. We applied this concept in a bioreactor study to test whether microbial community structure contains information sufficient to predict the concentration of H₂S as the product of sulfate reduction. Microbial sulfate reduction is a major source of H₂S in many industrial and environmental systems and is often influenced by the existing physicochemical conditions. Production of H₂S in industrial systems leads to occupational hazards and adversely affects the quality of products. A long-term (148 days) experiment was conducted in upflow bioreactors to mimic sulfidogenesis, followed by inhibition with nitrate salts and a resumption of H₂S generation when inhibition was released. We determined microbial community structure in 731 samples across 20 bioreactors using 16S rRNA gene sequencing and applied a random forest algorithm to successfully predict different phases of sulfidogenesis and mitigation (accuracy = 93.17%) and sessile and effluent microbial communities (accuracy = 100%). Similarly derived regression models that also included cell abundances were able to predict H₂S concentration with remarkably high fidelity (R² > 0.82). Metabolic profiles based on microbial community structure were also found to be reliable predictors for H₂S concentration (R² = 0.78). These results suggest that microbial community structure contains information sufficient to predict sulfidogenesis in a closed system, with anticipated applications to microbially driven processes in open environments.

IMPORTANCE Microbial communities control many biogeochemical processes. Many of these processes are impractical or expensive to measure directly. Because the taxonomic structure of the microbial community is indicative of its function, it encodes information that can be used to predict biogeochemistry. Here, we demonstrate how a machine learning technique can be used to predict sulfidogenesis, a key biogeochemical process in a model system. A distinction of this research was the ability to predict H₂S production in a bioreactor from the effluent bacterial community structure without direct observations of the sessile community or other environmental conditions. This study establishes the ability to use machine learning approaches in predicting sulfide concentrations in a closed system, which can be further developed as a valuable tool for predicting biogeochemical processes in open environments. As machine learning algorithms continue to improve, we anticipate increased applications of microbial community structure to predict key environmental and industrial processes.

Increasing a microscope's effective field of view via overlapped imaging and machine learning

This work demonstrates a multi-lens microscopic imaging system that overlaps multiple independent fields of view on a single sensor for high-efficiency automated specimen analysis. Automatic detection, classification and counting of various morphological features of interest is now a crucial component of both biomedical research and disease diagnosis. While convolutional neural networks (CNNs) have dramatically improved the accuracy of counting cells and sub-cellular features from acquired digital image data, the overall throughput is still typically hindered by the limited space-bandwidth product (SBP) of conventional microscopes. Here, we show both in simulation and experiment that overlapped imaging and co-designed analysis software can achieve accurate detection of diagnostically-relevant features for several applications, including counting of white blood cells and the malaria parasite, leading to multi-fold increase in detection and processing throughput with minimal reduction in accuracy.

Over-feeding the gut microbiome: A scoping review on health implications and therapeutic perspectives

The human gut microbiome has gained increasing attention over the past two decades. Several findings have shown that this complex and dynamic microbial ecosystem can contribute to the maintenance of host health or, when subject to imbalances, to the pathogenesis of various enteric and non-enteric diseases. This scoping review summarizes the current knowledge on how the gut microbiota and microbially-derived compounds affect host metabolism, especially in the context of obesity and related disorders. Examples of microbiome-based targeted intervention strategies that aim to restore and maintain an eubiotic layout are then discussed. Adjuvant therapeutic interventions to alleviate obesity and associated comorbidities are traditionally based on diet modulation and the supplementation of prebiotics, probiotics and synbiotics. However, these approaches have shown only moderate ability to induce sustained changes in the gut microbial ecosystem, making the development of innovative and tailored microbiome-based intervention strategies of utmost importance in clinical practice. In this regard, the administration of next-generation probiotics and engineered microbiomes has shown promising results, together with more radical intervention strategies based on the replacement of the dysbiotic ecosystem by means of fecal microbiota transplantation from healthy donors or with the introduction of synthetic communities specifically designed to achieve the desired therapeutic outcome. Finally, we provide a perspective for future translational investigations through the implementation of bioinformatics approaches, including machine and deep learning, to predict health risks and therapeutic outcomes.

Application of Machine Learning Techniques to an Agent-Based Model of Pantoea

Agent-based modeling (ABM) is a powerful simulation technique which describes a complex dynamic system based on its interacting constituent entities. While the flexibility of ABM enables broad application, the complexity of real-world models demands intensive computing resources and computational time; however, a metamodel may be constructed to gain insight at less computational expense. Here, we developed a model in NetLogo to describe the growth of a microbial population consisting of Pantoea. We applied 13 parameters that defined the model and actively changed seven of the parameters to modulate the evolution of the population curve in response to these changes. We efficiently performed more than 3,000 simulations using a Python wrapper, NL4Py. Upon evaluation of the correlation between the active parameters and outputs by random forest regression, we found that the parameters which define the depth of medium and glucose concentration affect the population curves significantly. Subsequently, we constructed a metamodel, a dense neural network, to predict the simulation outputs from the active parameters and found that it achieves high prediction accuracy, reaching an R² coefficient of determination value up to 0.92. Our approach of using a combination of ABM with random forest regression and neural network reduces the number of required ABM simulations. The simplified and refined metamodels may provide insights into the complex dynamic system before their transition to more sophisticated models that run on high-performance computing systems. The ultimate goal is to build a bridge between simulation and experiment, allowing model validation by comparing the simulated data to experimental data in microbiology.

Bacteria-driven phthalic acid ester biodegradation: Current status and emerging opportunities

The extensive use of phthalic acid esters (PAEs) has led to their widespread distribution across various environments. As PAEs pose significant threats to human health, it is urgent to develop efficient strategies to eliminate them from environments. Bacteria-driven PAE biodegradation has been considered as an inexpensive yet effective strategy to restore the contaminated environments. Despite great advances in bacterial culturing and sequencing, the inherent complexity of indigenous microbial community hinders us to mechanistically understand in situ PAE biodegradation and efficiently harness the degrading power of bacteria. The synthetic microbial ecology provides us a simple and controllable model system to address this problem. In this review, we focus on the current progress of PAE biodegradation mediated by bacterial isolates and indigenous bacterial communities, and discuss the prospective of synthetic PAE-degrading bacterial communities in PAE biodegradation research. It is anticipated that the theories and approaches of synthetic microbial ecology will revolutionize the study of bacteria-driven PAE biodegradation and provide novel insights for developing effective bioremediation solutions.

Micro on a macroscale: relating microbial-scale soil processes to global ecosystem function

Soil microorganisms play a key role in driving major biogeochemical cycles and in global responses to climate change. However, understanding and predicting the behavior and function of these microorganisms remains a grand challenge for soil ecology due in part to the microscale complexity of soils. It is becoming increasingly clear that understanding the microbial perspective is vital to accurately predicting global processes. Here, we discuss the microbial perspective including the microbial habitat as it relates to measurement and modeling of ecosystem processes. We argue that clearly defining and quantifying the size, distribution and sphere of influence of microhabitats is crucial to managing microbial activity at the ecosystem scale. This can be achieved using controlled and hierarchical sampling designs. Model microbial systems can provide key data needed to integrate microhabitats into ecosystem models, while adapting soil sampling schemes and statistical methods can allow us to collect microbially-focused data. Quantifying soil processes, like biogeochemical cycles, from a microbial perspective will allow us to more accurately predict soil functions and address long-standing unknowns in soil ecology.

Microscopy deep learning predicts virus infections and reveals mechanics of lytic-infected cells

Imaging across scales reveals disease mechanisms in organisms, tissues, and cells. Yet, particular infection phenotypes, such as virus-induced cell lysis, have remained difficult to study. Here, we developed imaging modalities and deep learning procedures to identify herpesvirus and adenovirus (AdV) infected cells without virus-specific stainings. Fluorescence microscopy of vital DNA-dyes and live-cell imaging revealed learnable virus-specific nuclear patterns transferable to related viruses of the same family. Deep learning predicted two major AdV infection outcomes, non-lytic (nonspreading) and lytic (spreading) infections, up to about 20 hr prior to cell lysis. Using these predictive algorithms, lytic and non-lytic nuclei had the same levels of green fluorescent protein (GFP)-tagged virion proteins but lytic nuclei enriched the virion proteins faster, and collapsed more extensively upon laser-rupture than non-lytic nuclei, revealing impaired mechanical properties of lytic nuclei. Our algorithms may be used to infer infection phenotypes of emerging viruses, enhance single cell biology, and facilitate differential diagnosis of non-lytic and lytic infections.

•

Artificial intelligence identifies HSV- and AdV-infected cells without specific probes.

•

Imaging lytic-infected cells reveals nuclear envelope rupture and AdV dissemination.

•

Live cell imaging and neural networks presciently pinpoint lytic-infected cells.

•

Lytic-infected cell nuclei have mechanical properties distinct from non-lytic nuclei.

Fermented food products in the era of globalization: tradition meets biotechnology innovations

Omics tools offer the opportunity to characterize and trace traditional and industrial fermented foods. Bioinformatics, through machine learning, and other advanced statistical approaches, are able to disentangle fermentation processes and to predict the evolution and metabolic outcomes of a food microbial ecosystem. By assembling microbial artificial consortia, the biotechnological advances will also be able to enhance the nutritional value and organoleptics characteristics of fermented food, preserving, at the same time, the potential of autochthonous microbial consortia and metabolic pathways, which are difficult to reproduce. Preserving the traditional methods contributes to protecting the hidden value of local biodiversity, and exploits its potential in industrial processes with the final aim of guaranteeing food security and safety, even in developing countries.