Predicting the abundance of metal resistance genes in subtropical estuaries using amplicon sequencing and machine learning

Oxygen mediated mobilization and co-occurrence of antibiotic resistance in lab-scale bioreactor using metagenomic binning

Sub-lethal levels of antibiotic stimulate bacteria to generate reactive oxygen species (ROS) that promotes emergence and spread of antibiotic resistance mediated by mobile genetic elements (MGEs). Nevertheless, the influence of dissolved oxygen (DO) levels on mobility of antibiotic resistance genes (ARGs) in response to ROS-induced stress remains elusive. Thus, the study employs metagenomic assembly and binning approaches to decipher mobility potential and co-occurrence frequency of ARGs and MGEs under hyperoxic (5.5-7 mgL- 1), normoxic (2.5-4 mgL- 1), and hypoxic (0.5-1 mgL- 1) conditions in lab-scale bioreactor for 6 months. Among 163 high-quality metagenome-assembled genomes (MAGs) recovered from 13 metagenomes, 42 MAGs harboured multiple ARGs and were assigned to priority pathogen group. Total ARG count increased by 4.3 and 2.5% in hyperoxic and normoxic, but decreased by 0.53% in hypoxic conditions after 150 days. On contrary, MGE count increased by 7.3-1.3% in all the DO levels, with only two ARGs showed positive correlation with MGEs in hypoxic compared to 20 ARGs under hyperoxic conditions. Opportunistic pathogens (Escherichia, Klebsiella, Clostridium, and Proteus) were detected as potential hosts of ARGs wherein co-localisation of critical ARG gene cassette (sul1, dfr1,adeF, and qacC) were identified in class 1 integron/Tn1 family transposons. Thus, enhanced co-occurrence frequency of ARGs with MGEs in pathogens suggested promotion of ARGs mobility under oxidative stress. The study offers valuable insights into ARG dissemination and hosts dynamics that is essential for controlling oxygen-related stress for mitigating MGEs and ARGs in the environment.

Mapping the spread and mobility of antibiotic resistance in wastewater due to COVID-19 surge

Large amounts of antibiotics have been discharged into wastewater during the COVID-19 pandemic due to overuse and misuse of antibiotics to treat patients. Wastewater-based surveillance can be used as an early warning for antibiotic resistance (AR) emergence. The present study analyzed municipal wastewater corresponding to the major pandemic waves (WW1, WW2, and WW3) in India along with hospital wastewater (Ho) taken as a benchmark for AR. Commonly prescribed antibiotics during a pandemic, azithromycin and cefixime residues, were found in the range of 2.1-2.6 μg/L in Ho and WW2. Total residual antibiotic concentration was less in WW2; however, the total antibiotic resistance gene (ARG) count was 1065.6 ppm compared to 85.2 ppm in Ho. Metagenome and RT-qPCR analysis indicated a positive correlation between antibiotics and non-corresponding ARGs (blaOXA, aadA, cat, aph3, and ere), where 7.2-7.5% was carried by plasmid in the bacterial community of WW1 and WW2. Moreover, as the abundance of the dfrA and int1 genes varied most among municipal wastewater, they can be suggested as AR markers for the pandemic. The common pathogens Streptococcus, Escherichia, Shigella, and Aeromonas were putative ARG hosts in metagenome-assembled genomes. The ARG profile and antibiotic levels varied between municipal wastewaters but were fairly similar for WW2 and Ho, suggesting the impact of the pandemic in shaping the resistome pattern. The study provides insights into the resistome dynamic, AR markers, and host-ARG association in wastewater during the COVID-19 surge. Continued surveillance and identification of intervention points for AR beyond the pandemic are essential to curbing the environmental spread of ARGs in the near future.

Assembly processes of bacterial and fungal communities in metal(loid)s smelter soil

There are few studies on concurrent bacterial and fungal community assembly processes that govern the metal(loid)s biogeochemical cycles at smelters. Here, a systematic investigation combined geochemical characterization, co-occurrence patterns, and assembly mechanisms of bacterial and fungal communities inhabiting soils around an abandoned arsenic smelter. Acidobacteriota, Actinobacteriota, Chloroflexi, and Pseudomonadota were dominant in bacterial communities, whereas Ascomycota and Basidiomycota dominated fungal communities. The random forest model indicated the bioavailable fractions of Fe (9.58%) were the main positive factor driving the beta diversity of bacterial communities, and the total N (8.09%) was the main negative factor for fungal communities. Microbe-contaminant interactions demonstrate the positive impact of the bioavailable fractions of certain metal(loid)s on bacteria (Comamonadaceae and Rhodocyclaceae) and fungi (Meruliaceae and Pleosporaceae). The fungal co-occurrence networks exhibited more connectivity and complexity than the bacterial networks. The keystone taxa were identified in bacterial (including Diplorickettsiaceae, norank_o_Candidatus_Woesebacteria, norank_o_norank_c_AT-s3-28, norank_o_norank_c_bacteriap25, and Phycisphaeraceae) and fungal (including Biatriosporaceae, Ganodermataceae, Peniophoraceae, Phaeosphaeriaceae, Polyporaceae, Teichosporaceae, Trichomeriaceae, Wrightoporiaceae, and Xylariaceae) communities. Meanwhile, community assembly analysis revealed that deterministic processes dominated the microbial community assemblies, which were highly impacted by pH, total N, and total and bioavailable metal(loid) content. This study provides helpful information to develop bioremediation strategies for the mitigation of metal(loid)s-polluted soils.

Accurate prediction of huanglongbing occurrence in citrus plants by machine learning-based analysis of symbiotic bacteria

Huanglongbing (HLB), the most prevalent citrus disease worldwide, is responsible for substantial yield and economic losses. Phytobiomes, which have critical effects on plant health, are associated with HLB outcomes. The development of a refined model for predicting HLB outbreaks based on phytobiome markers may facilitate early disease detection, thus enabling growers to minimize damages. Although some investigations have focused on differences in the phytobiomes of HLB-infected citrus plants and healthy ones, individual studies are inappropriate for generating common biomarkers useful for detecting HLB on a global scale. In this study, we therefore obtained bacterial information from several independent datasets representing hundreds of citrus samples from six continents and used these data to construct HLB prediction models based on 10 machine learning algorithms. We detected clear differences in the phyllosphere and rhizosphere microbiomes of HLB-infected and healthy citrus samples. Moreover, phytobiome alpha diversity indices were consistently higher for healthy samples. Furthermore, the contribution of stochastic processes to citrus rhizosphere and phyllosphere microbiome assemblies decreased in response to HLB. Comparison of all constructed models indicated that a random forest model based on 28 bacterial genera in the rhizosphere and a bagging model based on 17 bacterial species in the phyllosphere predicted the health status of citrus plants with almost 100% accuracy. Our results thus demonstrate that machine learning models and phytobiome biomarkers may be applied to evaluate the health status of citrus plants.

Using machine learning models to predict the effects of seasonal fluxes on Plesiomonas shigelloides population density

Seasonal variations (SVs) affect the population density (PD), fate, and fitness of pathogens in environmental water resources and the public health impacts. Therefore, this study is aimed at applying machine learning intelligence (MLI) to predict the impacts of SVs on P. shigelloides population density (PDP) in the aquatic milieu. Physicochemical events (PEs) and PDP from three rivers acquired via standard microbiological and instrumental techniques across seasons were fitted to MLI algorithms (linear regression (LR), multiple linear regression (MR), random forest (RF), gradient boosted machine (GBM), neural network (NN), K-nearest neighbour (KNN), boosted regression tree (BRT), extreme gradient boosting (XGB) regression, support vector regression (SVR), decision tree regression (DTR), M5 pruned regression (M5P), artificial neural network (ANN) regression (with one 10-node hidden layer (ANN10), two 6- and 4-node hidden layers (ANN64), and two 5- and 5-node hidden layers (ANN55)), and elastic net regression (ENR)) to assess the implications of the SVs of PEs on aquatic PDP. The results showed that SVs significantly influenced PDP and PEs in the water (p < 0.0001), exhibiting a site-specific pattern. While MLI algorithms predicted PDP with differing absolute flux magnitudes for the contributing variables, DTR predicted the highest PDP value of 1.707 log unit, followed by XGB (1.637 log unit), but XGB (mean-squared-error (MSE) = 0.0025; root-mean-squared-error (RMSE) = 0.0501; R² =0.998; medium absolute deviation (MAD) = 0.0275) outperformed other models in terms of regression metrics. Temperature and total suspended solids (TSS) ranked first and second as significant factors in predicting PDP in 53.3% (8/15) and 40% (6/15), respectively, of the models, based on the RMSE loss after permutations. Additionally, season ranked third among the 7 models, and turbidity (TBS) ranked fourth at 26.7% (4/15), as the primary significant factor for predicting PDP in the aquatic milieu. The results of this investigation demonstrated that MLI predictive modelling techniques can promisingly be exploited to complement the repetitive laboratory-based monitoring of PDP and other pathogens, especially in low-resource settings, in response to seasonal fluxes and can provide insights into the potential public health risks of emerging pathogens and TSS pollution (e.g., nanoparticles and micro- and nanoplastics) in the aquatic milieu. The model outputs provide low-cost and effective early warning information to assist watershed managers and fish farmers in making appropriate decisions about water resource protection, aquaculture management, and sustainable public health protection.

Arsenic shapes the microbial community structures in tungsten mine waste rocks

Tungsten (W) is a critical material that is widely used in military applications, electronics, lighting technology, power engineering and the automotive and aerospace industries. In recent decades, overexploitation of W has generated large amounts of mine waste rocks, which generate elevated content of toxic elements and cause serious adverse effects on ecosystems and public health. Microorganisms are considered important players in toxic element migrations from waste rocks. However, the understanding of how the microbial community structure varies in W mine waste rocks and its key driving factors is still unknown. In this study, high-throughput sequencing methods were used to determine the microbial community profiles along a W content gradient in W mine waste rocks. We found that the microbial community structures showed clear differences across the different W levels in waste rocks. Notably, arsenic (As), instead of W and nutrients, was identified as the most important predictor influencing microbial diversity. Furthermore, our results also showed that As is the most important environmental factor that regulates the distribution patterns of ecological clusters and keystone ASVs. Importantly, we found that the dominant genera have been regulated by As and were widely involved in As biogeochemical cycling in waste rocks. Taken together, our results have provided useful information about the response of microbial communities to W mine waste rocks.