Machine learning approach identifies water sample source based on microbial abundance

Systematic tracking of nitrogen sources in complex river catchments: Machine learning approach based on microbial metagenomics

Tracking nitrogen pollution sources is crucial for the effective management of water quality; however, it is a challenging task due to the complex contaminative scenarios in the freshwater systems. The contaminative pattern variations can induce quick responses of aquatic microorganisms, making them sensitive indicators of pollution origins. In this study, the soil and water assessment tool, accompanied by a detailed pollution source database, was used to detect the main nitrogen pollution sources in each sub-basin of the Liuyang River watershed. Thus, each sub-basin was assigned to a known class according to SWAT outputs, including point source pollution-dominated area, crop cultivation pollution-dominated area, and the septic tank pollution-dominated area. Based on these outputs, the random forest (RF) model was developed to predict the main pollution sources from different river ecosystems using a series of input variable groups (e.g., natural macroscopic characteristics, river physicochemical properties, 16S rRNA microbial taxonomic composition, microbial metagenomic data containing taxonomic and functional information, and their combination). The accuracy and the Kappa coefficient were used as the performance metrics for the RF model. Compared with the prediction performance among all the input variable groups, the prediction performance of the RF model was significantly improved using metagenomic indices as inputs. Among the metagenomic data-based models, the combination of the taxonomic information with functional information of all the species achieved the highest accuracy (0.84) and increased median Kappa coefficient (0.70). Feature importance analysis was used to identify key features that could serve as indicators for sudden pollution accidents and contribute to the overall function of the river system. The bacteria Rhabdochromatium marinum, Frankia, Actinomycetia, and Competibacteraceae were the most important species, whose mean decrease Gini indices were 0.0023, 0.0021, 0.0019, and 0.0018, respectively, although their relative abundances ranged only from 0.0004 to 0.1 %. Among the top 30 important variables, functional variables constituted more than half, demonstrating the remarkable variation in the microbial functions among sites with distinct pollution sources and the key role of functionality in predicting pollution sources. Many functional indicators related to the metabolism of Mycobacterium tuberculosis, such as K24693, K25621, K16048, and K14952, emerged as significant important factors in distinguishing nitrogen pollution origins. With the shortage of pollution source data in developing regions, this suggested approach offers an economical, quick, and accurate solution to locate the origins of water nitrogen pollution using the metagenomic data of microbial communities.

Digitalizing river aquatic ecosystems

Traditional river health assessment relies on limited water quality indices and representative organism activity, but does not comprehensively obtain biotic and abiotic information of the ecosystem. Here, we propose a new approach to evaluate the ecological and health risks of river aquatic ecosystems. First, detailed physicochemical and biological characterization of a river ecosystem can be obtained through pollutant determination (especially emerging pollutants) and DNA/RNA sequencing. Second, supervised machine learning can be applied to perform classification analysis of characterization data and ascertain river ecosystem ecology and health. Our proposed methodology transforms river ecosystem health assessment and can be applied in river management.

Spatial and temporal distribution of the microbial community structure in the receiving rivers of the middle and lower reaches of the Yangtze River under the influence of different wastewater types

The gradual intensification of human activity has caused severe negative impacts on the ecosystems of the Yangtze River Basin. Treated effluents still affect the environment and health of receiving rivers, particularly in terms of microbial community structure. However, relatively few studies have been conducted on the differences in the effects of wastewater types on microbial community structure. Three sampling campaigns (237 samples) were conducted in the Nanjing and Wuhan sections of the Yangtze River Basin. Our results showed that the microbial community structure differed significantly among the water periods and could recover to its original state at > 500 m downstream of the outfall. The diversity of the receiving rivers under the influence of industrial wastewater was higher than that of the other wastewater types, although the number of taxa was lower than that of other wastewater types. Cyanobium_PCC-6307 and Rhodoferax were screened for biomarkers in samples affected by domestic and industrial wastewater, respectively. Although different kinds of wastewater influenced the microbial community structure, environmental factors, and geographical distance were still the main drivers. This study suggests that treated wastewater still poses a risk to ecosystems and highlights the importance of effective management strategies for assessing ecosystem health.

Identifying the drivers of chlorophyll-a dynamics in a landscape lake recharged by reclaimed water using interpretable machine learning

The water quality of lakes recharged by reclaimed water is affected by both the fluctuation of reclaimed water quality and the biochemical processes in the lakes, and therefore the main controlling factors of algal blooms are difficult to identify. Taking a typical landscape lake recharged by reclaimed water as an example and using the spatiotemporal distribution characteristics and correlation analysis of water quality indexes, we propose an interpretable machine learning framework based on random forest to predict chlorophyll-a (Chl-a). The model considered nutrient difference indexes between reclaimed water and lake water, and further used feature importance ranking and partial dependence plot to identify nutrient drivers. Results show that the NO3--N input from reclaimed water is the dominant nutrient driver for algal bloom especially at high temperatures, and the negative correlation between NO3--N and Chl-a in the lake water is the consequence of algal bloom rather than the cause. Our study provides new insights into the identification of eutrophication factors for lakes recharged by reclaimed water.

Developing a hybrid model for predicting the reaction kinetics between chlorine and micropollutants in water

Understanding the reactivities of chlorine towards micropollutants is crucial for assessing the fate of micropollutants in water chlorination. In this study, we integrated machine learning with kinetic modeling to predict the reaction kinetics between micropollutants and chlorine in deionized water and real surface water. We first established a framework to predict the apparent second-order rate constants for micropollutants with chlorine by combining Morgan molecular fingerprints with machine learning algorithms. The framework was tuned using Bayesian optimization and showed high prediction accuracy. It was validated through experiments and used to predict the unreported apparent second-order rate constants for 103 emerging micropollutants with chlorine. The framework also improved the understanding of the structure-dependence of micropollutants' reactivity with chlorine. We incorporated the predicted apparent second-order rate constants into the Kintecus software to establish a hybrid model to profile the time-dependent changes of micropollutant concentrations by chlorination. The hybrid model was validated by experiments conducted in real surface water in the presence of natural organic matter. The hybrid model could predict how much micropollutants were degraded by chlorination with varied chlorine contact times and/or initial chlorine dosages. This study advances fundamental understanding of the reaction kinetics between chlorine and emerging micropollutants, and also offers a valuable tool to assess the fate of micropollutants during chlorination of drinking water.

Climate Change Causes Salinity To Become Determinant in Shaping the Microeukaryotic Spatial Distribution among the Lakes of the Inner Mongolia-Xinjiang Plateau

Climate change greatly affects lake microorganisms in arid and semiarid zones, which alters ecosystem functions and the ecological security of lakes. However, the responses of lake microorganisms, especially microeukaryotes, to climate change are poorly understood. Here, using 18S ribosomal RNA (rRNA) high-throughput sequencing, we investigated the distribution patterns of microeukaryotic communities and whether and how climate change directly or indirectly affected the microeukaryotic communities on the Inner Mongolia-Xinjiang Plateau. Our results showed that climate change, as the main driving force of lake change, drives salinity to become a determinant of the microeukaryotic community among the lakes of the Inner Mongolia-Xinjiang Plateau. Salinity shapes the diversity and trophic level of the microeukaryotic community and further affects lake carbon cycling. Co-occurrence network analysis further revealed that increasing salinity reduced the complexity but improved the stability of microeukaryotic communities and changed ecological relationships. Meanwhile, increasing salinity enhanced the importance of deterministic processes in microeukaryotic community assembly, and the dominance of stochastic processes in freshwater lakes transformed into deterministic processes in salt lakes. Furthermore, we established lake biomonitoring and climate sentinel models by integrating microeukaryotic information, which would provide substantial improvements to our predictive ability of lake responses to climate change. IMPORTANCE Our findings have important implications for understanding the distribution patterns and the driving mechanisms of microeukaryotic communities among the lakes of the Inner Mongolia-Xinjiang Plateau and whether and how climate change directly or indirectly affects microeukaryotic communities. Our study also establishes the groundwork to use the lake microbiome for the assessment of aquatic ecological health and climate change, which is critical for ecosystem management and for projecting the ecological consequences of future climate warming.

Coastal Water Quality Modelling Using E. coli, Meteorological Parameters and Machine Learning Algorithms

In this study, machine learning models were implemented to predict the classification of coastal waters in the region of Eastern Macedonia and Thrace (EMT) concerning Escherichia coli (E. coli) concentration and weather variables in the framework of the Directive 2006/7/EC. Six sampling stations of EMT, located on beaches of the regional units of Kavala, Xanthi, Rhodopi, Evros, Thasos and Samothraki, were selected. All 1039 samples were collected from May to September within a 14-year follow-up period (2009-2021). The weather parameters were acquired from nearby meteorological stations. The samples were analysed according to the ISO 9308-1 for the detection and the enumeration of E. coli. The vast majority of the samples fall into category 1 (Excellent), which is a mark of the high quality of the coastal waters of EMT. The experimental results disclose, additionally, that two-class classifiers, namely Decision Forest, Decision Jungle and Boosted Decision Tree, achieved high Accuracy scores over 99%. In addition, comparing our performance metrics with those of other researchers, diversity is observed in using algorithms for water quality prediction, with algorithms such as Decision Tree, Artificial Neural Networks and Bayesian Belief Networks demonstrating satisfactory results. Machine learning approaches can provide critical information about the dynamic of E. coli contamination and, concurrently, consider the meteorological parameters for coastal waters classification.

Dynamics of microbial community composition during degradation of silks in burial environment

The silk residues in the soil formed the unique niche, termed "silksphere." Here, we proposed a hypothesis that silksphere microbiota have great potential as a biomarker for unraveling the degradation of the ancient silk textiles with great archaeological and conservation values. To test our hypothesis, in this study, we monitored the dynamics of microbial community composition during silk degradation via both indoor soil microcosmos model and outdoor environment with amplicon sequencing against 16S and ITS gene. Microbial community divergence was evaluated with Welch two sample t-test, PCoA, negative binomial generalized log-linear model and clustering, etc. Community assembly mechanisms differences between silksphere and bulk soil microbiota were compared with dissimilarity-overlap curve (DOC) model, Neutral model and Null model. A well-established machine learning algorithm, random forest, was also applied to the screening of potential biomarkers of silk degradation. The results illustrated the ecological and microbial variability during the microbial degradation of silk. Vast majority of microbes populating the silksphere microbiota strongly diverged from those in bulk soil. Certain microbial flora can serve as an indicator of silk degradation, which would lead to a novel perspective to perform identification of archaeological silk residues in the field. To sum up, this study provides a new perspective to perform the identification of archaeological silk residue through the dynamics of microbial communities.

Effects of environmental and anthropogenic factors on the distribution and abundance of microplastics in freshwater ecosystems

Although microplastics are emerging marine pollutants that have recently attracted increasing attention, it is still difficult to identify their sources. This study reviewed 6487 articles to determine current research trends and found 237 effective concentration points after sorting, which were distributed in four regions and related to freshwater ecosystems. Results found that 15 environmental variables represented natural and anthropogenic environmental characteristics, of which seven environmental variables were selected for experimental modelling. Random forest models fitted sample data, thus facilitating the identification of regional microplastics distribution. The global random forest model had random forest importance scores (RFISs) for gross domestic product, population, and the proportion of agricultural land use were 15.76 %, 15.64 %, and 14.74 %, respectively; these indicate that human activities significantly affected the global distribution of microplastics. In Asia, agriculture and urban activities are the main sources of microplastics, with an RFIS of 11.58 % and 12.24 % for the proportion of agricultural and urban land use, respectively. Activities in urban areas were determined to be the main influencing factors in North America, with an RFIS of 13.92 % for the proportion of urban land use. Agricultural activities were the main influencing factors in Europe, with RFISs for the proportion of agricultural land use of 16.90 %. Our results indicate that region-specific policies are required to control microplastics in different regions, with soil composition being a latency factor that affects microplastics' distribution.

The forecasting power of the microbiome

Microorganisms are informative biological integrators of past and present environmental abiotic and biotic conditions. At the same time, they are directly involved in ecosystem processes. Unfortunately, the complexity of microbial communities has so far resulted in most studies being descriptive. Here, we suggest that signals in the microbiome data can be used to forecast future ecosystem processes. The combination of omics with various statistical learning approaches, selected based on accuracy-interpretability and bias-variance trade-offs, will be key to attain this goal, as exemplified by recent studies. The time is ripe for microbial ecologists to fully exploit the forecasting power of microbiomes.

Improving kitchen waste composting maturity by optimizing the processing parameters based on machine learning model

As a novel analytical method based on big data, machine learning model can explore the relationship between different parameters and draw universal conclusions, which was used to predict composting maturity and identify key parameters in this study. The results showed that the Stacking model exhibited excellent prediction accuracy. SHapley Additive exPlanations (SHAP) and Partial Dependence Analysis (PDA) were performed to evaluate the importance of different parameters as well as their optimal interval. Optimal starting conditions should be maintained in the mesophilic state (temperature: 30-45℃, moisture content: 55-65%, pH: 6.3-8.0), and nutrients (total nitrogen > 2.3%, total organic carbon > 35%) should be adjusted in the thermophilic state. Experiments revealed that model-based optimization strategies could improve composting maturity because they could optimize compost microbial flora and perform complex carbon cycle functions. In conclusion, this study provides new insights into the enhancement of the composting process.

Advancing prediction of emerging contaminants in a tropical reservoir with general water quality indicators based on a hybrid process and data-driven approach

Monitoring and predicting the occurrence and dynamic distributions of emerging contaminants (ECs) in the aquatic environment has always been a great challenge. This study aims to explore the potential of fully utilizing the advantages of combining traditional process-based models (PBMs) and data-driven models (DDMs) with general water quality indicators in terms of improving the accuracy and efficiency of predicting ECs in aquatic ecosystems. Two representative ECs, namely Bisphenol A (BPA) and N, N-diethyltoluamide (DEET), in a tropical reservoir were chosen for this study. A total of 36 DDMs based on different input datasets using Artificial Neural Networks (ANN) and Random Forests (RF) were examined in three case studies. The models were applied in prognosis validation based on easily accessible data on water quality indicators. Our results revealed that all the models yielded good fits when compared to the observed data. These new insights into the advantages using the combination of traditional PBMs and DDMs with general water quality datasets help to overcome the constraints in terms of model accuracy and efficiency as well as technical and budget limitations due to monitoring surveys and laboratory experiments in the study of fate and transport of ECs in aquatic environments.

Interfacing Machine Learning and Microbial Omics: A Promising Means to Address Environmental Challenges

Microbial communities are ubiquitous and carry an exceptionally broad metabolic capability. Upon environmental perturbation, microbes are also amongst the first natural responsive elements with perturbation-specific cues and markers. These communities are thereby uniquely positioned to inform on the status of environmental conditions. The advent of microbial omics has led to an unprecedented volume of complex microbiological data sets. Importantly, these data sets are rich in biological information with potential for predictive environmental classification and forecasting. However, the patterns in this information are often hidden amongst the inherent complexity of the data. There has been a continued rise in the development and adoption of machine learning (ML) and deep learning architectures for solving research challenges of this sort. Indeed, the interface between molecular microbial ecology and artificial intelligence (AI) appears to show considerable potential for significantly advancing environmental monitoring and management practices through their application. Here, we provide a primer for ML, highlight the notion of retaining biological sample information for supervised ML, discuss workflow considerations, and review the state of the art of the exciting, yet nascent, interdisciplinary field of ML-driven microbial ecology. Current limitations in this sphere of research are also addressed to frame a forward-looking perspective toward the realization of what we anticipate will become a pivotal toolkit for addressing environmental monitoring and management challenges in the years ahead.