DeepARG: a deep learning approach for predicting antibiotic resistance genes from metagenomic data

Role of artificial intelligence in early diagnosis and treatment of infectious diseases

Infectious diseases remain a global health challenge, necessitating innovative approaches for their early diagnosis and effective treatment. Artificial Intelligence (AI) has emerged as a transformative force in healthcare, offering promising solutions to address this challenge. This review article provides a comprehensive overview of the pivotal role AI can play in the early diagnosis and treatment of infectious diseases. It explores how AI-driven diagnostic tools, including machine learning algorithms, deep learning, and image recognition systems, enhance the accuracy and efficiency of disease detection and surveillance. Furthermore, it delves into the potential of AI to predict disease outbreaks, optimise treatment strategies, and personalise interventions based on individual patient data and how AI can be used to gear up the drug discovery and development (D3) process.The ethical considerations, challenges, and limitations associated with the integration of AI in infectious disease management are also examined. By harnessing the capabilities of AI, healthcare systems can significantly improve their preparedness, responsiveness, and outcomes in the battle against infectious diseases.

Distinctive gut antibiotic resistome, potential health risks and underlying pathways upon cerebral ischemia-reperfusion injury

Antibiotic resistance genes (ARGs) as emerging pollutants pose health risks to humans and the environment. Gut microbiota is an important reservoir for ARGs and hotspot for ARG acquisition and dissemination. Non-antibiotic factors (such as disease pathophysiology) affect ARG emergence and dissemination. Cerebral ischemia-reperfusion injury (I/R) commonly occurs in stroke patients. However, effects of I/R on ARG emergence and dissemination are unknown. Therefore, metagenomics was used to unveil selective collection of gut antibiotic resistome and its health risks, key ARG hosts and underlying pathways upon I/R. Changes in gut antibiotic resistome upon I/R were characterized by tetracycline ARG accumulation and decreases in aminoglycoside and glycopeptide ARGs. Besides, changes in gut antibiotic resistome were corrected with those in gut microbiota from phylum to species, serum lipid accumulation and glucose depletion upon I/R. Additionally, health risks of gut microbial multidrug ARGs (such as abem, adek and TolC), macA, aph(3')-I and carO, co-localized with mobile gene elements, were increased upon I/R. Moreover, phyla Firmicutes (especially order Eubacteriales, class Clostridia) and Bacteroidota were key ARG hosts in gut microbiota of I/R gerbils. Furthermore, suppression of vancomycin resistance, and lantibiotic biosynthesis and immunity, disturbances in peptidoglycan biosynthesis and hydrolysis, activation of antimicrobial peptide resistance, lipopolysaccharide biosynthesis, teichoic acid biosynthesis, arabinogalactan biosynthesis, aromatic compound degradation, oxidative phosphorylation, the tricarboxylic acid cycle and its anaplerotic pathways were observed in upon I/R. This study provides novel insights and intervention targets related to selective collection of gut antibiotic resistome and its potential health risks upon I/R.

Urbanization enhances consumer protist-driven ARGs dissemination in riverine ecosystems

Despite the emergence of antibiotic resistance genes (ARGs) and antibiotic resistant bacteria (ARBs), how biological inter-trophic interactions, modulated by watershed urbanization, shape the resistome remains unexplored. We collected water samples from the highly urbanized (western: 65 % built land, sewage-affected) and lesser-urbanized (northern: 25 % built land, drinking water source) downstream tributaries of the Jiulong River in southeast China over dry and wet seasons. We utilized metagenomic and amplicon (16S and 18S rDNA) sequencing to investigate the relationships among microeukaryotic algae, consumer protists, bacterial communities, and the resistome. Metagenomic results showed that ARG-MGE-carrying contigs (mobile ARGs), rather than ARG-carrying contigs (non-mobile ARGs), exhibited more pronounced discrepancies between tributaries. A higher total abundance of ARGs and a greater number of co-shared ARGs between pathogen and non-pathogen bacteria were observed in the more urbanized western tributary. Structural equation modeling revealed that consumer protist-bacteria and algae-bacteria cohesions predominantly influenced the resistome in the western and northern tributaries, respectively. Additionally, consumer protists had more significant associations (511 out of 634) with bacteria carrying mobile ARGs in western tributary, while algae had more significant associations (73 out of 105) in northern tributary. These results highlight the distinct inter-trophic driving factors of the resistome modulated by watershed urbanization.

Determinants of antimicrobial resistance in biosolids: A systematic review, database, and meta-analysis

Biosolids can provide a nutrient rich soil amendment, particularly for poor soils and semi-arid or drought-prone areas. However, there are concerns that sludge and biosolids could be a source of propagation and exposure to AMR determinants such as antibiotic resistant bacteria (ARB), and antibiotic resistance genes (ARGs). To inform risk assessment efforts, a systematic literature review was performed to build a comprehensive spreadsheet database of ARB and ARG concentrations in biosolids (and some sludges specified as intended for land application), along with 69 other quantitative and qualitative meta-data fields from 68 published studies describing sampling information and processing methods that can be used for modeling purposes. Mean ARG concentrations per gram in positive samples of biosolids ranged from -5.7 log10(gene copies [gc]/g) to 12.92 log10(gc/g) (with these range values reported per dry weight), and aqueous concentrations ranged from 0.9 log10(gc/L) to 14.6 log10(gc/L). Mean ARB concentrations per gram of biosolids ranged from 2.02 log10 (colony forming units [CFU]/g) to 9.00 log10 (CFU/g) (dry weight), and aqueous concentrations ranged from 3.23 log10 (CFU/L) to 12.0 log10 (CFU/L). ARG log removal values (LRVs) during sewage sludge stabilization were calculated from a meta-analysis of mean concentrations before and after stabilization from 31 studies, ranging from -2.05 to 5.52 logs. The classes of resistance most relevant for a risk assessment corresponded to sulfonamide (sul1 and sul2), tetracycline (tetZ, tetX, tetA and tetG), beta-lactam (blaTEM), macrolide (ermB and ermF), aminoglycoside (strA and aac(6')-Ib-cr), and integron-associated (intI1). The resistance classes most relevant for ARB risk assessment included sulfonamides (sulfamethoxazole and sulfamethazine), cephalosporin (cephalothin and cefoxitin), penicillin (ampicillin), and ciprofloxin (ciprofloxacin). Considerations for exposure assessment are discussed to highlight risk assessment needs relating to antimicrobial resistance (AMR) associated with biosolids application. This study aids in prioritization of resources for reducing the spread of AMR within a One Health framework.

Critical perspectives on advancing antibiotic resistant gene (ARG) detection technologies in aquatic ecosystems

The spread of antibiotic resistance genes (ARGs) in aquatic ecosystems poses a serious risk to environmental and public health, making advanced detection and monitoring methods essential. This review provides a fresh perspective and a critical evaluation of recent advances in detecting and monitoring ARGs in aquatic environments. It highlights the latest innovations in molecular, bioinformatic, and environmental techniques. While traditional methods like culture-based assays and polymerase chain reaction (PCR) remain important, they are increasingly being supplemented by high-throughput sequencing technologies applied to metagenomics. These technologies offer comprehensive insights into the diversity and distribution of ARGs in aquatic environments. The integration of bioinformatic tools and databases has improved the accuracy and efficiency of ARG detection, enabling the analysis of complex datasets and tracking the evolution of ARGs in aquatic settings. Additionally, new environmental monitoring methods, including novel biosensors, geographic information systems (GIS) applications, and remote sensing technologies, have emerged as powerful tools for real-time ARG surveillance in water systems. This review critically examines the challenges of standardizing these methodologies and emphasizes the need for interdisciplinary approaches to enhance ARG monitoring across different aquatic ecosystems. By assessing the strengths and limitations of various methods, this review aims to guide future research and the development of more effective strategies for managing antibiotic resistance in aquatic environments.

Spatiotemporal profiles and underlying mechanisms of the antibiotic resistome in two water-diversion lakes

Human-induced interventions have altered the local characteristics of the lake ecosystems through changes in hydraulic exchange, which in turn impacts the ecological processes of antibiotic resistance genes (ARGs) in the lakes. However, the current understanding of the spatiotemporal patterns and driving factors of ARGs in water-diversion lakes is still seriously insufficient. In the present study, we investigated antibiotic resistome in the main regulation and storage hubs, namely Nansi Lake and Dongping Lake, of the eastern part of the South-to-North Water Diversion project in Shandong Province (China) using a metagenomic-based approach. A total of 653 ARG subtypes belonging to 25 ARG types were detected with a total abundance of 0.125-0.390 copies/cell, with the dominance of bacitracin, multidrug, and macrolide-lincosamide streptogramin resistance genes. The ARG compositions were sensitive to seasonal variation and also interfered by artificial regulation structures along the way. Human pathogenic bacteria such as Acinetobacter calcoaceticus, Acinetobacter lwoffii, Klebsiella pneumoniae, along with the multidrug resistance genes they carried, were the focus of risk control in the two studied lakes, especially in summer. Plasmids were the key mobile genetic elements (MGEs) driving the horizontal gene transfer of ARGs, especially multidrug and sulfonamide resistance genes. The null model revealed that stochastic process was the main driver of ecological drift for ARGs in the lakes. The partial least squares structural equation model further determined that seasonal changes of pH and temperature drove a shift in the bacterial community, which in turn shaped the profile of ARGs by altering the composition of MGEs, antibacterial biocide- and metal-resistance genes (BMGs), and virulence factor genes (VFGs). Our results highlighted the importance of seasonal factors in determining the water transfer period. These findings can aid in a deeper understanding of the spatiotemporal variations of ARGs in lakes and their driving factors, offering a scientific basis for antibiotic resistance management.

Enriching the future of public health microbiology with hybridization bait capture

SUMMARYPublic health microbiology focuses on microorganisms and infectious agents that impact human health. For years, this field has relied on culture or molecular methods to investigate complex samples of public health importance. However, with the increase in accuracy and decrease in sequencing cost over the last decade, there has been a transition to the use of next-generation sequencing in public health microbiology. Nevertheless, many available sequencing methods (e.g., shotgun metagenomics and amplicon sequencing) do not work well in complex sample types, require deep sequencing, or have inherent biases associated with them. Hybridization bait capture, also known as target enrichment, brings in solutions for such limitations. It is an increasingly popular technique to simultaneously characterize many thousands of genetic elements while reducing the amount of sequencing needed (thereby reducing the sequencing costs). Here, we summarize the concept of hybridization bait capture for public health, reviewing a total of 35 bait sets designed in six key topic areas for public health microbiology [i.e., antimicrobial resistance (AMR), bacteria, fungi, parasites, vectors, and viruses], and compare hybridization bait capture to previously relied upon methods. Furthermore, we provide an in-depth comparison of the three most popular bait sets designed for AMR by evaluating each of them against three major AMR databases: Comprehensive Antibiotic Resistance Database, Microbial Ecology Group Antimicrobial Resistance Database, and Pathogenicity Island Database. Thus, this article provides a review of hybridization bait capture for public health microbiologists.

Research on antibiotic resistance genes in wild and artificially bred green turtles (Chelonia mydas)

Sea turtles, vital to marine ecosystems, face population decline. Artificial breeding is a recovery strategy, yet it risks introducing antibiotic resistance genes (ARGs) to wild populations and ecosystems. This study employed metagenomic techniques to compare the distribution characteristics of ARGs in the guts of wild and artificially bred green turtles (Chelonia mydas). The findings revealed that the total abundance of ARGs in C. mydas that have been artificially bred was significantly higher than that in wild individuals. Additionally, the abundance of mobile genetic elements (MGEs) co-occurring with ARGs in artificially bred C. mydas was significantly higher than in wild C. mydas. In the analysis of bacteria carrying ARGs, wild C. mydas exhibited greater bacterial diversity. Furthermore, in artificially bred C. mydas, we discovered 23 potential human pathogenic bacteria (HPB) that contain antibiotic resistance genes. In contrast, in wild C. mydas, only one type of HPB carrying an antibiotic resistance gene was found. The findings of this study not only enhance our understanding of the distribution and dissemination of ARGs within the gut microbial communities of C. mydas, but also provide vital information for assessing the potential impact of releasing artificially bred C. mydas on the spread of antibiotic resistance.

Predicting antimicrobial resistance in E. coli with discriminative position fused deep learning classifier

Escherichia coli (E. coli) has become a particular concern due to the increasing incidence of antimicrobial resistance (AMR) observed worldwide. Using machine learning (ML) to predict E. coli AMR is a more efficient method than traditional laboratory testing. However, further improvement in the predictive performance of existing models remains challenging. In this study, we collected 1937 high-quality whole genome sequencing (WGS) data from public databases with an antimicrobial resistance phenotype and modified the existing workflow by adding an attention mechanism to enable the modified workflow to focus more on core single nucleotide polymorphisms (SNPs) that may significantly lead to the development of AMR in E. coli. While comparing the model performance before and after adding the attention mechanism, we also performed a cross-comparison among the published models using random forest (RF), support vector machine (SVM), logistic regression (LR), and convolutional neural network (CNN). Our study demonstrates that the discriminative positional colors of Chaos Game Representation (CGR) images can selectively influence and highlight genome regions without prior knowledge, enhancing prediction accuracy. Furthermore, we developed an online tool (https://github.com/tjiaa/E.coli-ML/tree/main) for assisting clinicians in the rapid prediction of the AMR phenotype of E. coli and accelerating clinical decision-making.

MetaCompare 2.0: differential ranking of ecological and human health resistome risks

While numerous environmental factors contribute to the spread of antibiotic resistance genes (ARGs), quantifying their relative contributions remains a fundamental challenge. Similarly, it is important to differentiate acute human health risks from environmental exposure, versus broader ecological risk of ARG evolution and spread across microbial taxa. Recent studies have proposed various methods for achieving such aims. Here, we introduce MetaCompare 2.0, which improves upon original MetaCompare pipeline by differentiating indicators of human health resistome risk (potential for human pathogens of acute resistance concern to acquire ARGs) from ecological resistome risk (overall mobility of ARGs and potential for pathogen acquisition). The updated pipeline's sensitivity was demonstrated by analyzing diverse publicly-available metagenomes from wastewater, surface water, soil, sediment, human gut, and synthetic microbial communities. MetaCompare 2.0 provided distinct rankings of the metagenomes according to both human health resistome risk and ecological resistome risk, with both scores trending higher when influenced by anthropogenic impact or other stress. We evaluated the robustness of the pipeline to sequence assembly methods, sequencing depth, contig count, and metagenomic library coverage bias. The risk scores were remarkably consistent despite variations in these technological aspects. We packaged the improved pipeline into a publicly-available web service (http://metacompare.cs.vt.edu/) that provides an easy-to-use interface for computing resistome risk scores and visualizing results.

Exploring protein natural diversity in environmental microbiomes with DeepMetagenome

Protein natural diversity offers a vast sequence space for protein engineering, and deep learning enables its detection from metagenomes/proteomes without prior assumptions. DeepMetagenome, a Python-based method, explores protein diversity through modules for training and analyzing sequence datasets. The deep learning model includes Embedding, Conv1D, LSTM, and Dense layers, with sequence feature analysis for data cleaning. Applied to metallothioneins from a database of over 146 million coding features, DeepMetagenome identified over 500 high-confidence metallothionein sequences, outperforming DIAMOND and CNN-based models. It showed stable performance compared to a Transformer-based model over 25 epochs. Among 23 synthesized sequences, 20 exhibited metal resistance. The tool also successfully explored the diversity of three additional protein families and is freely available on GitHub with detailed instructions.

Occurrence of antibiotic-resistant bacteria in the sea surface microlayer of coastal waters in the southern North Sea

The emergence of antibiotic-resistant bacteria in coastal waters is a global health problem posing potential risks to the health of humans who depend on coastal resources. Monitoring and increased efforts are needed to maintain the health of marine ecosystems. The sea surface microlayer (SML) is poorly studied for antibiotic resistance of the inhabiting bacteria. Therefore, we examined the antibiotic resistance patterns of 41 bacterial strains isolated from the SML in a harbor in the southern North Sea. The strains are affiliated with 17 genera typically found in the marine environment. Using the disc diffusion assay, we found extensive resistance, particularly to gentamycin, kanamycin, nalidixic acid, penicillin, sulfadimidine, and streptomycin. A broth microdilution assay showed high minimum inhibitory concentrations (MICs) for most isolates for amikacin, aztreonam, ceftazidime, cefepime, minocycline, and tobramycin. Genome analysis of three strains affiliated with the genera Pseudoseohaeicola, Nereida, and Vibrio, all showing a highly resistant phenotype, revealed the presence of 57, 42, and 90 genes, respectively, associated with antibiotic resistance. Over 50 % of these genes are multidrug efflux pumps. Our study shows that the SML in anthropogenic-influenced coastal regions harbors a wide diversity of antibiotic-resistant bacteria equipped with a broad range of multidrug efflux pumps.

Data-Driven Approaches in Antimicrobial Resistance: Machine Learning Solutions

Background/Objectives: The emergence of antimicrobial resistance (AMR) due to the misuse and overuse of antibiotics has become a critical threat to global public health. There is a dire need to forecast AMR to understand the underlying mechanisms of resistance for the development of effective interventions. This paper explores the capability of machine learning (ML) methods, particularly unsupervised learning methods, to enhance the understanding and prediction of AMR. It aims to determine the patterns from AMR gene data that are clinically relevant and, in public health, capable of informing strategies. Methods: We analyzed AMR gene data in the PanRes dataset by applying unsupervised learning techniques, namely K-means clustering and Principal Component Analysis (PCA). These techniques were applied to identify clusters based on gene length and distribution according to resistance class, offering insights into the resistance genes' structural and functional properties. Data preprocessing, such as filtering and normalization, was conducted prior to applying machine learning methods to ensure consistency and accuracy. Our methodology included the preprocessing of data and reduction of dimensionality to ensure that our models were both accurate and interpretable. Results: The unsupervised learning models highlighted distinct clusters of AMR genes, with significant patterns in gene length, including their associated resistance classes. Further dimensionality reduction by PCA allows for clearer visualizations of relationships among gene groupings. These patterns provide novel insights into the potential mechanisms of resistance, particularly the role of gene length in different resistance pathways. Conclusions: This study demonstrates the potential of ML, specifically unsupervised approaches, to enhance the understanding of AMR. The identified patterns in resistance genes could support clinical decision-making and inform public health interventions. However, challenges remain, particularly in integrating genomic data and ensuring model interpretability. Further research is needed to advance ML applications in AMR prediction and management.

Unnoticed antimicrobial resistance risk in Tibetan cities unveiled by sewage metagenomic surveillance: Compared to the eastern Chinese cities

Sewage surveillance is a cost-effective tool for assessing antimicrobial resistance (AMR) in urban populations. However, research on sewage AMR in remote areas is still limited. Here, we used shotgun metagenomic sequencing to profile antibiotic resistance genes (ARGs) and ARG-carrying pathogens (APs) across 15 cities in Tibetan Plateau (TP) and the major cities in eastern China. Notable regional disparities in sewage ARG composition were found, with a significantly higher ARG abundance in TP (2.97 copies/cell). A total of 542 and 545 APs were identified in sewage from TP and the East, respectively, while more than 40 % carried mobile genetic elements (MGEs). Moreover, 65 MGEs-carrying APs were identified as World Health Organization (WHO) priority-like bacterial and fungal pathogens. Notably, a fungal zoonotic pathogen, Enterocytozoon bieneusi, was found for the first time to carry a nitroimidazole resistance gene (nimJ). Although distinct in AP compositions, the relative abundances of APs were comparable in these two regions. Furthermore, sewage in TP was found to be comparable to the cities in eastern China in terms of ARG mobility and AMR risks. These findings provide insights into ARGs and APs distribution in Chinese sewage and stress the importance of AMR surveillance and management strategies in remote regions.

Multi-omics illuminates the functional significance of previously unknown species in a full-scale landfill leachate treatment plant

Landfill leachate treatment plants (LLTPs) harbor a vast reservoir of uncultured microbes, yet limited studies have systematically unraveled their functional potentials within LLTPs. Combining 36 metagenomic and 18 metatranscriptomic datasets from a full-scale LLTP, we unveiled a double-edged sword role of unknown species in leachate biotreatment and environmental implication. We identified 655 species-level genome bins (SGBs) spanning 47 bacterial and 3 archaeal phyla, with 75.9 % unassigned to any known species. Over 90 % of up-regulated functional genes in biotreatment units, compared to the leachate influent, were carried by unknown species and actively participated in carbon, nitrogen, and sulfur cycles. Approximately 79 % of the 37,366 carbohydrate active enzymes (CAZymes), with ∼90 % novelty and high expression, were encoded by unknown species, exhibiting great potential in biodegrading carbohydrate compounds linked to human meat-rich diets. Unknown species offered a valuable genetic resource of thousands of versatile, abundant, and actively expressed metabolic gene clusters (MGCs) and biosynthetic gene clusters (BGCs) for enhancing leachate treatment. However, unknown species may contribute to the emission of hazardous N2O/H2S and represented significant reservoirs for antibiotic-resistant pathogens that posed environmental safety risks. This study highlighted the significance of considering both positive and adverse effects of LLTP microbes to optimize LLTP performance.

AI in microbiome-related healthcare

Artificial intelligence (AI) has the potential to transform clinical practice and healthcare. Following impressive advancements in fields such as computer vision and medical imaging, AI is poised to drive changes in microbiome-based healthcare while facing challenges specific to the field. This review describes the state-of-the-art use of AI in microbiome-related healthcare. It points out limitations across topics such as data handling, AI modelling and safeguarding patient privacy. Furthermore, we indicate how these current shortcomings could be overcome in the future and discuss the influence and opportunities of increasingly complex data on microbiome-based healthcare.

Decreasing light exposure increases the abundance of antibiotic resistance genes in the cecum and feces of laying hens

The gut microbiome plays a crucial role in maintaining animal health and is influenced by various factors, including light exposure; however, the response in laying hens of the gut microbiome to intermittent light regimes and the related impact on antibiotic resistance genes (ARGs) remain poorly understood. In this study, we divided 20-week-old laying hens into two groups. These groups were exposed to either continuous normal light or intermittent light for 8 weeks. The feces and cecal contents of laying hens were collected for analysis. Metagenomic analysis of both feces and cecal content samples revealed significant shifts in the microbial composition and abundance of ARGs under intermittent light exposure compared to normal light exposure (P < 0.05). Furthermore, metabolomic analysis of the cecal contents revealed substantial alterations in the abundance and composition of ARGs and mobile genetic elements (MGEs) in response to intermittent light exposure (P < 0.05). Network analysis revealed intricate co-occurrence patterns among bacterial communities, metabolites, and ARGs, highlighting correlations between Bacteroidetes species, ARGs, and metabolites. Although certain bacterial species showed differential associations, the dominant bacteria carrying ARGs or MGEs had relatively low numbers, suggesting that other bacterial communities may have had a greater influence on ARG dissemination. Moreover, our observations highlight the crucial role of metabolites as mediators between bacterial communities and ARGs, providing novel insights into the dynamics of antibiotic resistance development. Our findings underscore the impact of intermittent light exposure on ARG proliferation in poultry farming and emphasize interconnections among ARGs, bacterial communities, and metabolic pathways. The results underscore the importance of considering both microbial communities and metabolic processes to understand antibiotic resistance in agricultural settings.

Artificial intelligence in predicting pathogenic microorganisms' antimicrobial resistance: challenges, progress, and prospects

The issue of antimicrobial resistance (AMR) in pathogenic microorganisms has emerged as a global public health crisis, posing a significant threat to the modern healthcare system. The advent of Artificial Intelligence (AI) and Machine Learning (ML) technologies has brought about revolutionary changes in this field. These advanced computational methods are capable of processing and analyzing large-scale biomedical data, thereby uncovering complex patterns and mechanisms behind the development of resistance. AI technologies are increasingly applied to predict the resistance of pathogens to various antibiotics based on gene content and genomic composition. This article reviews the latest advancements in AI and ML for predicting antimicrobial resistance in pathogenic microorganisms. We begin with an overview of the biological foundations of microbial resistance and its epidemiological research. Subsequently, we highlight the main AI and ML models used in resistance prediction, including but not limited to Support Vector Machines, Random Forests, and Deep Learning networks. Furthermore, we explore the major challenges in the field, such as data availability, model interpretability, and cross-species resistance prediction. Finally, we discuss new perspectives and solutions for research into microbial resistance through algorithm optimization, dataset expansion, and interdisciplinary collaboration. With the continuous advancement of AI technology, we will have the most powerful weapon in the fight against pathogenic microbial resistance in the future.

Nature-based bioreactors: Tackling antibiotic resistance in urban wastewater treatment

The overuse and misuse of antibiotics have accelerated the selection of antibiotic-resistant bacteria, significantly impacting human, animal, and environmental health. As aquatic environments are vulnerable to antibiotic resistance, suitable management practices should be adopted to tackle this phenomenon. Here we show an effective, nature-based solution for reducing antibiotic resistance from actual wastewater. We utilize a bioreactor that relies on benthic (biofilms) and planktonic microbial communities to treat secondary effluent from a small urban wastewater treatment plant (<10,000 population equivalent). This treated effluent is eventually released into the local aquatic ecosystem. We observe high removal efficiency for genes that provide resistance to commonly used antibiotic families, as well as for mobile genetic elements that could potentially aid in their spread. Importantly, we notice a buildup of sulfonamide (sul1 and sul2) and tetracycline (tet(C), tet(G), and tetR) resistance genes specifically in biofilms. This advancement marks the initial step in considering this bioreactor as a nature-based, cost-effective tertiary treatment option for small UWWTPs facing antibiotic resistance challenges.

Traditionally produced tempeh harbors more diverse bacteria with more putative health-promoting properties than industrially produced tempeh

In recent years, there has been a significant shift towards industrialization in food production, resulting in the implementation of higher hygiene standards globally. Our study focused on examining the impact of hygiene standards on tempeh, a popular Rhizopus-based fermented soybean product native to Indonesia, and now famous around the world. We observed that tempeh produced with standardized hygiene measures exhibited a microbiome with comparable bacterial abundances but a markedly different community structure and function than traditionally produced tempeh. In detail, we found a decreased bacterial abundance of lactobacilli and enterobacteria, bacterial diversity, different indicator taxa, and significantly changed community structure in industrial tempeh. A similar picture was found for functional analysis: the quantity of bacterial genes was similar but qualitative changes were found for genes associated with human health. The resistome of tempeh varied based on its microbiome composition. The higher number of antimicrobial resistance genes in tempeh produced without standardized hygiene measures mainly belong to multidrug efflux pumps known to occur in plant-based food. Our findings were confirmed by functional insights into genomes and metagenome-assembled genomes from the dominant bacteria, e.g. Leuconostoc, Limosilactobacillus, Lactobacillus, Enterococcus, Paenibacillus, Azotobacter and Enterobacter. They harboured an impressive spectrum of genes important for human health, e.g. for production of vitamin B1, B7, B12, and K, iron and zinc transport systems and short chain fatty acid production. In conclusion, industrially produced tempeh harbours a less diverse microbiome than the traditional one. Although this ensures production at large scales as well as biosafety, in the long-term it can lead to potential effects for human gut health.

Overlooked dissemination risks of antimicrobial resistance through green tide proliferation

Green tides, particularly those induced by Enteromorpha, pose significant environmental challenges, exacerbated by climate change, coastal eutrophication, and other anthropogenic impacts. More concerningly, these blooms may influence the spread of antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs) within ecosystems. However, the manner in which Enteromorpha blooms affect the distribution and spread of antimicrobial resistance (AMR) remains uncertain. This study investigated ARG profiles, dynamic composition, and associated health risks within the Enteromorpha phycosphere and surrounding seawater in typical bays (Jiaozhou, Aoshan, and Lingshan) in the South Yellow Sea. The Enteromorpha phycosphere exhibited significantly higher ARG abundance (p < 0.05) but lower diversity compared to the surrounding seawater. Source-tracking and metagenomic analyses revealed that the phycosphere was the main contributor to the resistome of surrounding seawater. Moreover, resistant pathogens, especially ESKAPE pathogens, with horizontal gene transfer (HGT) potential, were more abundant in the phycosphere than in the surrounding seawater. The phycosphere released high-risk ARGs to the surrounding seawater during Enteromorpha blooms, posing serious health and ecological AMR risks in marine environments. This study highlights the significant role of Enteromorpha blooms in ARG spread and associated risks, urging a reassessment of AMR burden from a public health perspective.

Feedstock-dependent antibiotic resistance gene patterns and expression profiles in industrial scale biogas plants revealed by meta-omics technology

This study investigated antimicrobial resistance in the anaerobic digesters of two industrial-scale biogas plants processing agricultural biomass and municipal wastewater sludge. A combination of deep sequencing and genome-centric workflow was implemented for metagenomic and metatranscriptomics data analysis to comprehensively examine potential antimicrobial resistance in microbial communities. Anaerobic microbes were found to harbour numerous antibiotic resistance genes (ARGs), with 58.85% of the metagenome-assembled genomes (MAGs) harbouring antibiotic resistance. A moderately positive correlation was observed between the abundance and expression of ARGs. ARGs were located primarily on bacterial chromosomes. A higher expression of resistance genes was observed on plasmids than on chromosomes. Risk index assessment suggests that most ARGs identified posed a significant risk to human health. However, potentially pathogenic bacteria showed lower ARG expression than non-pathogenic ones, indicating that anaerobic treatment is effective against pathogenic microbes. Resistomes at the gene category level were associated with various antibiotic resistance categories, including multidrug resistance, beta-lactams, glycopeptides, peptides, and macrolide-lincosamide-streptogramin (MLS). Differential expression analysis revealed specific genes associated with potential pathogenicity, emphasizing the importance of active gene expression in assessing the risks associated with ARGs.

Distinct species turnover patterns shaped the richness of antibiotic resistance genes on eight different microplastic polymers

Elucidating the formation mechanism of plastisphere antibiotic resistance genes (ARGs) on different polymers is necessary to understand the ecological risks of plastisphere ARGs. Here, we explored the turnover and assembly mechanism of plastisphere ARGs on 8 different microplastic polymers (4 biodegradable (bMPs) and 4 non-biodegradable microplastics (nMPs)) by metagenomic sequencing. Our study revealed the presence of 479 ARGs with abundance ranging from 41.37 to 58.17 copies/16S rRNA gene in all plastispheres. These ARGs were predominantly multidrug resistance genes. The richness of plastisphere ARGs on different polymers had a significant correlation with the contribution of species turnover to plastisphere ARGs β diversity. Furthermore, polymer type was the most critical factor affecting the composition of plastisphere ARGs. More opportunistic pathogens carrying diverse ARGs on BMPs (PBAT, PBS, and PHA) with higher horizontal gene transfer potential may further magnify the ecological risks and human health threats. For example, the opportunistic pathogens Riemerella anatipestifer, Vibrio campbellii, and Vibrio cholerae are closely related to human production and life, which were the important potential hosts of many plastisphere ARGs and mobile genetic elements on BMPs. Thus, we emphasize the urgency of developing the formation mechanism of plastisphere ARGs and the necessity of controlling BMPs and ARG pollution, especially BMPs, with ever-increasing usage in daily life.

Prediction of antibiotic resistance mechanisms using a protein language model

MOTIVATION

Antibiotic resistance has emerged as a major global health threat, with an increasing number of bacterial infections becoming difficult to treat. Predicting the underlying resistance mechanisms of antibiotic resistance genes (ARGs) is crucial for understanding and combating this problem. However, existing methods struggle to accurately predict resistance mechanisms for ARGs with low similarity to known sequences and lack sufficient interpretability of the prediction models.

RESULTS

In this study, we present a novel approach for predicting ARG resistance mechanisms using ProteinBERT, a protein language model (pLM) based on deep learning. Our method outperforms state-of-the-art techniques on diverse ARG datasets, including those with low homology to the training data, highlighting its potential for predicting the resistance mechanisms of unknown ARGs. Attention analysis of the model reveals that it considers biologically relevant features, such as conserved amino acid residues and antibiotic target binding sites, when making predictions. These findings provide valuable insights into the molecular basis of antibiotic resistance and demonstrate the interpretability of pLMs, offering a new perspective on their application in bioinformatics.

AVAILABILITY AND IMPLEMENTATION

The source code is available for free at https://github.com/hmdlab/ARG-BERT. The output results of the model are published at https://waseda.box.com/v/ARG-BERT-suppl.

Preference and regulation mechanism mediated via mobile genetic elements for antibiotic and metal resistomes during composting amended with nano ZVI loaded on biochar

This study assessed the effectiveness of nano zero-valent iron loaded on biochar (BC-nZVI) during swine manure composting. BC-nZVI significantly reduced the abundance of antibiotic resistance genes (ARGs), metal resistance genes (MRGs), and mobile genetic elements (MGEs). BC-nZVI modified the preference of MGEs to carry ARGs and MRGs, and the corrosion products of BC-nZVI could destroy cell structure, hinder electron transfer between cells, and weaken the association between ARGs, MRGs, and host bacteria. Functional genes analysis revealed that BC-nZVI down-regulated the abundance of genes affecting the transmission and metabolism of ARGs and MRGs, including type IV secretion systems, transporter systems, two-component systems, and multidrug efflux pumps. Furthermore, the BC-nZVI decreased genes related to flagella and pili production and cell membrane permeability, thereby hindering the transfer of ARGs, MRGs, and MGEs in the environment. Redundancy analysis demonstrated that changes in the microbial community induced by BC-nZVI were pivotal factors impacting the abundance of ARGs, MRGs, and MGEs. Overall, this study confirmed the efficacy of BC-nZVI in reducing resistance genes during swine manure composting, offering a promising environmental strategy to mitigate the dissemination of these contaminants.

Healthcare as a driver, reservoir and amplifier of antimicrobial resistance: opportunities for interventions

Antimicrobial resistance (AMR) is a global health challenge that threatens humans, animals and the environment. Evidence is emerging for a role of healthcare infrastructure, environments and patient pathways in promoting and maintaining AMR via direct and indirect mechanisms. Advances in vaccination and monoclonal antibody therapies together with integrated surveillance, rapid diagnostics, targeted antimicrobial therapy and infection control measures offer opportunities to address healthcare-associated AMR risks more effectively. Additionally, innovations in artificial intelligence, data linkage and intelligent systems can be used to better predict and reduce AMR and improve healthcare resilience. In this Review, we examine the mechanisms by which healthcare functions as a driver, reservoir and amplifier of AMR, contextualized within a One Health framework. We also explore the opportunities and innovative solutions that can be used to combat AMR throughout the patient journey. We provide a perspective on the current evidence for the effectiveness of interventions designed to mitigate healthcare-associated AMR and promote healthcare resilience within high-income and resource-limited settings, as well as the challenges associated with their implementation.

Exploring microbial diversity and biosynthetic potential in zoo and wildlife animal microbiomes

Understanding human, animal, and environmental microbiota is essential for advancing global health and combating antimicrobial resistance (AMR). We investigate the oral and gut microbiota of 48 animal species in captivity, comparing them to those of wildlife animals. Specifically, we characterize the microbiota composition, metabolic pathways, AMR genes, and biosynthetic gene clusters (BGCs) encoding the production of specialized metabolites. Our results reveal a high diversity of microbiota, with 585 novel species-level genome bins (SGBs) and 484 complete BGCs identified. Functional gene analysis of microbiomes shows diet-dependent variations. Furthermore, by comparing our findings to wildlife-derived microbiomes, we observe the impact of captivity on the animal microbiome, including examples of converging microbiome compositions. Importantly, our study identifies AMR genes against commonly used veterinary antibiotics, as well as resistance to vancomycin, a critical antibiotic in human medicine. These findings underscore the importance of the 'One Health' approach and the potential for zoonotic transmission of pathogenic bacteria and AMR. Overall, our study contributes to a better understanding of the complexity of the animal microbiome and highlights its BGC diversity relevant to the discovery of novel antimicrobial compounds.

Prediction of antimicrobial resistance of Klebsiella pneumoniae from genomic data through machine learning

Antimicrobials, such as antibiotics or antivirals are medications employed to prevent and treat infectious diseases in humans, animals, and plants. Antimicrobial Resistance occurs when bacteria, viruses, and parasites no longer respond to these medicines. This resistance renders antibiotics and other antimicrobial drugs ineffective, making infections challenging or impossible to treat. This escalation in drug resistance heightens the risk of disease spread, severe illness, disability, and mortality. With datasets now containing hundreds or even thousands of pathogen genomes, machine learning techniques are on the rise for predicting antibiotic resistance in pathogens, prediction based on gene content and genome composition. Aim of this work is to combine and incorporate machine learning methods on bacterial genomic data to predict antimicrobial resistance, we will focus on the case of Klebsiella pneumoniae in order to support clinicians in selecting appropriate therapy.

Unde venis? Bacterial resistance from environmental reservoirs to lettuce: tracking microbiome and resistome over a growth period

Fresh produce is suggested to contribute highly to shaping the gut resistome. We investigated the impact of pig manure and irrigation water quality on microbiome and resistome of field-grown lettuce over an entire growth period. Lettuce was grown under four regimes, combining soil amendment with manure (with/without) with sprinkler irrigation using river water with an upstream wastewater input, disinfected by UV (with/without). Lettuce leaves, soil, and water samples were collected weekly and analysed by bacterial cultivation, 16S rRNA gene amplicon sequencing, and shotgun metagenomics from total community DNA. Cultivation yielded only few clinically relevant antibiotic-resistant bacteria (ARB), but numbers of ARB on lettuce increased over time, while no treatment-dependent changes were observed. Microbiome analysis confirmed a temporal trend. Antibiotic resistance genes (ARGs) unique to lettuce and water included multidrug and β-lactam ARGs, whereas lettuce and soil uniquely shared mainly glycopeptide and tetracycline ARGs. Surface water carried clinically relevant ARB (e.g. ESBL-producing Escherichia coli or Serratia fonticola) without affecting the overall lettuce resistome significantly. Resistance markers including biocide and metal resistance were increased in lettuce grown with manure, especially young lettuce (increased soil contact). Overall, while all investigated environments had their share as sources of the lettuce resistome, manure was the main source especially on young plants. We therefore suggest minimizing soil-vegetable contact to minimize resistance markers on fresh produce.

Sewage sludge fertilization affects microbial community structure and its resistome in agricultural soils

Global sewage sludge production is rapidly increasing, and its safe disposal is becoming an increasingly serious issue. One of the main methods of municipal sewage sludge management is based on its agricultural use. The wastewater and sewage sludge contain numerous antibiotic resistance genes (ARGs), and its microbiome differs significantly from the soil microbial community. The aim of the study was to assess the changes occurring in the soil microbial community and resistome after the addition of sewage sludge from municipal wastewater treatment plant (WWTP) in central Poland, from which the sludge is used for fertilizing agricultural soils on a regular basis. This study used a high-throughput shotgun metagenomics approach to compare the microbial communities and ARGs present in two soils fertilized with sewage sludge. The two soils represented different land uses and different physicochemical and granulometric properties. Both soils were characterized by a similar taxonomic composition of the bacterial community, despite dissimilarities between soils properties. Five phyla predominated, viz. Planctomycetes, Actinobacteria, Proteobacteria, Chloroflexi and Firmicutes, and they were present in comparable proportions in both soils. Network analysis revealed that the application of sewage sludge resulted in substantial qualitative and quantitative changes in bacterial taxonomic profile, with most abundant phyla being considerably depleted and replaced by Proteobacteria and Spirochaetes. In addition, the ratio of oligotrophic to copiotrophic bacteria substantially decreased in both amended soils. Furthermore, fertilized soils demonstrated greater diversity and richness of ARGs compared to control soils. The increased abundance concerned mainly genes of resistance to antibiotics most commonly used in human and animal medicine. The level of heavy metals in sewage sludge was low and did not exceed the standards permitted in Poland for sludge used in agriculture, and their level in fertilized soils was still inconsiderable.

Genome-centric metagenomes unveiling the hidden resistome in an anchialine cave

BACKGROUND

Antibiotic resistance is a critical global concern, posing significant challenges to human health and medical treatments. Studying antibiotic resistance genes (ARGs) is essential not only in clinical settings but also in diverse environmental contexts. However, ARGs in unique environments such as anchialine caves, which connect both fresh and marine water, remain largely unexplored despite their intriguing ecological characteristics.

RESULTS

We present the first study that comprehensively explores the occurrence and distribution of ARGs and mobile genetic elements (MGEs) within an anchialine cave. Utilizing metagenomic sequencing we uncovered a wide array of ARGs with the bacitracin resistance gene, bacA and multidrug resistance genes, being the most dominant. The cave's microbial community and associated resistome were significantly influenced by the salinity gradient. The discovery of novel β-lactamase variants revealed the cave's potential as a reservoir for previously undetected resistance genes. ARGs in the cave demonstrated horizontal transfer potential via plasmids, unveiling ecological implications.

CONCLUSIONS

These findings highlight the need for further exploration of the resistome in unique environments like anchialine caves. The interconnected dynamics of ARGs and MGEs within anchialine caves offer valuable insights into potential reservoirs and mechanisms of antibiotic resistance in natural ecosystems. This study not only advances our fundamental understanding but also highlights the need for a comprehensive approach to address antibiotic resistance in diverse ecological settings.

Disentangling abiotic and biotic effects of treated wastewater on stream biofilm resistomes enables the discovery of a new planctomycete beta-lactamase

BACKGROUND

Environmental reservoirs of antibiotic resistance pose a threat to human and animal health. Aquatic biofilms impacted by wastewater effluent (WW) are known environmental reservoirs for antibiotic resistance; however, the relative importance of biotic factors and abiotic factors from WW on the abundance of antibiotic resistance genes (ARGs) within aquatic biofilms remains unclear. Additionally, experimental evidence is limited within complex aquatic microbial communities as to whether genes bearing low sequence similarity to validated reference ARGs are functional as ARGs.

RESULTS

To disentangle the effects of abiotic and biotic factors on ARG abundances, natural biofilms were previously grown in flume systems with different proportions of stream water and either ultrafiltered or non-ultrafiltered WW. In this study, we conducted deep shotgun metagenomic sequencing of 75 biofilm, stream, and WW samples from these flume systems and compared the taxonomic and functional microbiome and resistome composition. Statistical analysis revealed an alignment of the resistome and microbiome composition and a significant association with experimental treatment. Several ARG classes exhibited an increase in normalized metagenomic abundances in biofilms grown with increasing percentages of non-ultrafiltered WW. In contrast, sulfonamide and extended-spectrum beta-lactamase ARGs showed greater abundances in biofilms grown in ultrafiltered WW compared to non-ultrafiltered WW. Overall, our results pointed toward the dominance of biotic factors over abiotic factors in determining ARG abundances in WW-impacted stream biofilms and suggested gene family-specific mechanisms for ARGs that exhibited divergent abundance patterns. To investigate one of these specific ARG families experimentally, we biochemically characterized a new beta-lactamase from the Planctomycetota (Phycisphaeraceae). This beta-lactamase displayed activity in the cleavage of cephalosporin analog despite sharing a low sequence identity with known ARGs.

CONCLUSIONS

This discovery of a functional planctomycete beta-lactamase ARG is noteworthy, not only because it was the first beta-lactamase to be biochemically characterized from this phylum, but also because it was not detected by standard homology-based ARG tools. In summary, this study conducted a metagenomic analysis of the relative importance of biotic and abiotic factors in the context of WW discharge and their impact on both known and new ARGs in aquatic biofilms. Video Abstract.

Antimicrobial resistance and its risks evaluation in wetlands on the Qinghai-Tibetan Plateau

Amidst the global antimicrobial resistance (AMR) crisis, antibiotic resistance has permeated even the most remote environments. To understand the dissemination and evolution of AMR in minimally impacted ecosystems, the resistome and mobilome of wetlands across the Qinghai-Tibetan Plateau and its marginal regions were scrutinized using metagenomic sequencing techniques. The composition of wetland microbiomes exhibits significant variability, with dominant phyla including Proteobacteria, Actinobacteria, Bacteroidetes, and Verrucomicrobia. Notably, a substantial abundance of Antibiotic Resistance Genes (ARGs) and Mobile Genetic Elements (MGEs) was detected, encompassing 17 ARG types, 132 ARG subtypes, and 5 types of MGEs (Insertion Sequences, Insertions Sequences, Genomic Islands, Transposons, and Integrative Conjugative Elements). No significant variance was observed in the prevalence of resistome and mobilome across different wetland types (i.e., the Yellow River, other rivers, lakes, and marshes) (R=-0.5882, P=0.607). The co-occurrence of 74 ARG subtypes and 22 MGEs was identified, underscoring the pivotal role of MGEs in shaping ARG pools within the Qinghai-Tibetan Plateau wetlands. Metagenomic binning and analysis of assembled genomes (MAGs) revealed that 93 out of 206 MAGs harbored ARGs (45.15 %). Predominantly, Burkholderiales, Pseudomonadales, and Enterobacterales were identified as the primary hosts of these ARGs, many of which represent novel species. Notably, a substantial proportion of ARG-carrying MAGs also contained MGEs, reaffirming the significance of MGEs in AMR dissemination. Furthermore, utilizing the arg_ranker framework for risk assessment unveiled severe contamination of high-risk ARGs across most plateau wetlands. Moreover, some prevalent human pathogens were identified as potential hosts for these high-risk ARGs, posing substantial transmission risks. This study aims to investigate the prevalence of resistome and mobilome in wetlands, along with evaluating the risk posed by high-risk ARGs. Such insights are crucial for informing environmental protection strategies and facilitating the management of water resources on the Qinghai-Tibetan Plateau.

Giant viruses as reservoirs of antibiotic resistance genes

Nucleocytoplasmic large DNA viruses (NCLDVs; also called giant viruses), constituting the phylum Nucleocytoviricota, can infect a wide range of eukaryotes and exchange genetic material with not only their hosts but also prokaryotes and phages. A few NCLDVs were reported to encode genes conferring resistance to beta‑lactam, trimethoprim, or pyrimethamine, suggesting that they are potential vehicles for the transmission of antibiotic resistance genes (ARGs) in the biome. However, the incidence of ARGs across the phylum Nucleocytoviricota, their evolutionary characteristics, their dissemination potential, and their association with virulence factors remain unexplored. Here, we systematically investigated ARGs of 1416 NCLDV genomes including those of almost all currently available cultured isolates and high-quality metagenome-assembled genomes from diverse habitats across the globe. We reveal that 39.5% of them carry ARGs, which is approximately 37 times higher than that for phage genomes. A total of 12 ARG types are encoded by NCLDVs. Phylogenies of the three most abundant NCLDV-encoded ARGs hint that NCLDVs acquire ARGs from not only eukaryotes but also prokaryotes and phages. Two NCLDV-encoded trimethoprim resistance genes are demonstrated to confer trimethoprim resistance in Escherichia coli. The presence of ARGs in NCLDV genomes is significantly correlated with mobile genetic elements and virulence factors.

Spatial analysis of murine microbiota and bile acid metabolism during amoxicillin treatment

Antibiotics cause collateral damage to resident microbes that is associated with various health risks. To date, studies have largely focused on the impacts of antibiotics on large intestinal and fecal microbiota. Here, we employ a gastrointestinal (GI) tract-wide integrated multiomic approach to show that amoxicillin (AMX) treatment reduces bacterial abundance, bile salt hydrolase activity, and unconjugated bile acids in the small intestine (SI). Losses of fatty acids (FAs) and increases in acylcarnitines in the large intestine (LI) correspond with spatially distinct expansions of Proteobacteria. Parasutterella excrementihominis engage in FA biosynthesis in the SI, while multiple Klebsiella species employ FA oxidation during expansion in the LI. We subsequently demonstrate that restoration of unconjugated bile acids can mitigate losses of commensals in the LI while also inhibiting the expansion of Proteobacteria during AMX treatment. These results suggest that the depletion of bile acids and lipids may contribute to AMX-induced dysbiosis in the lower GI tract.

A trait-based ecological perspective on the soil microbial antibiotic-related genetic machinery

Antibiotic resistance crisis dictates the need for resistance monitoring and the search for new antibiotics. The development of monitoring protocols is hindered by the great diversity of resistance factors, while the "streetlight effect" denies the possibility of discovering novel drugs based on existing databases. In this study, we address these challenges using high-throughput environmental screening viewed from a trait-based ecological perspective. Through an in-depth analysis of the metagenomes of 658 topsoil samples spanning Europe, we explored the distribution of 241 prokaryotic and fungal genes responsible for producing metabolites with antibiotic properties and 485 antibiotic resistance genes. We analyzed the diversity of these gene collections at different levels and modeled the distribution of each gene across environmental gradients. Our analyses revealed several nonparallel distribution patterns of the genes encoding sequential steps of enzymatic pathways synthesizing large antibiotic groups, pointing to gaps in existing databases and suggesting potential for discovering new analogues of known antibiotics. We show that agricultural activity caused a continental-scale homogenization of microbial antibiotic-related machinery, emphasizing the importance of maintaining indigenous ecosystems within the landscape mosaic. Based on the relationships between the proportion of the genes in the metagenomes with the main predictors (soil pH, land cover type, climate temperature and humidity), we illustrate how the properties of chemical structures dictate the distribution of the genes responsible for their synthesis across environments. With this understanding, we propose general principles to facilitate the discovery of antibiotics, including principally new ones, establish abundance baselines for antibiotic resistance genes, and predict their dissemination.

Filamentous bacteria-induced sludge bulking can alter antibiotic resistance gene profiles and increase potential risks in wastewater treatment systems

Sludge bulking caused by filamentous bacteria is a prevalent issue in wastewater treatment systems. While previous studies have primarily concentrated on controlling sludge bulking, the biological risks associated with it have been overlooked. This study demonstrates that excessive growth of filamentous bacteria during sludge bulking can significantly increase the abundance of antibiotic resistance genes (ARGs) in activated sludge. Through metagenomic analysis, we identified specific ARGs carried by filamentous bacteria, such as Sphaerotilus and Thiothrix, which are responsible for bulking. Additionally, by examining over 1,000 filamentous bacterial genomes, we discovered a diverse array of ARGs across different filamentous bacteria derived from wastewater treatment systems. Our findings indicate that 74.84% of the filamentous bacteria harbor at least one ARG, with the occurrence frequency of ARGs in these bacteria being approximately 1.5 times higher than that in the overall bacterial population in activated sludge. Furthermore, genomic and metagenomic analyses have shown that the ARGs in filamentous bacteria are closely linked to mobile genetic elements and are frequently found in potentially pathogenic bacteria, highlighting potential risks posed by these filamentous bacteria. These insights enhance our understanding of ARGs in activated sludge and underscore the importance of risk management in wastewater treatment systems.

Hierarchical multi-task deep learning-assisted construction of human gut microbiota reactive oxygen species-scavenging enzymes database

In the process of oxygen reduction, reactive oxygen species (ROS) are generated as intermediates, including superoxide anion (O2-), hydrogen peroxide (H2O2), and hydroxyl radicals (OH-). ROS can be destructive, and an imbalance between oxidants and antioxidants in the body can lead to pathological inflammation. Inappropriate ROS production can cause oxidative damage, disrupting the balance in the body and potentially leading to DNA damage in intestinal epithelial cells and beneficial bacteria. Microorganisms have evolved various enzymes to mitigate the harmful effects of ROS. Accurately predicting the types of ROS-scavenging enzymes (ROSes) is crucial for understanding the oxidative stress mechanisms and formulating strategies to combat diseases related to the "gut-organ axis." Currently, there are no available ROSes databases (DBs). In this study, we propose a systematic workflow comprising three modules and employ a hierarchical multi-task deep learning approach to collect, expand, and explore ROSes-related entries. Based on this, we have developed the human gut microbiota ROSes DB (http://39.101.72.186/), which includes 7,689 entries. This DB provides user-friendly browsing and search features to support various applications. With the assistance of ROSes DB, various communication-based microbial interactions can be explored, further enabling the construction and analysis of the evolutionary and complex networks of ROSes DB in human gut microbiota species.IMPORTANCEReactive oxygen species (ROS) is generated during the process of oxygen reduction, including superoxide anion, hydrogen peroxide, and hydroxyl radicals. ROS can potentially cause damage to cells and DNA, leading to pathological inflammation within the body. Microorganisms have evolved various enzymes to mitigate the harmful effects of ROS, thereby maintaining a balance of microorganisms within the host. The study highlights the current absence of a ROSes DB, emphasizing the crucial importance of accurately predicting the types of ROSes for understanding oxidative stress mechanisms and developing strategies for diseases related to the "gut-organ axis." This research proposes a systematic workflow and employs a multi-task deep learning approach to establish the human gut microbiota ROSes DB. This DB comprises 7,689 entries and serves as a valuable tool for researchers to delve into the role of ROSes in the human gut microbiota.

Artificial intelligence tools for the identification of antibiotic resistance genes

The fight against bacterial antibiotic resistance must be given critical attention to avert the current and emerging crisis of treating bacterial infections due to the inefficacy of clinically relevant antibiotics. Intrinsic genetic mutations and transferrable antibiotic resistance genes (ARGs) are at the core of the development of antibiotic resistance. However, traditional alignment methods for detecting ARGs have limitations. Artificial intelligence (AI) methods and approaches can potentially augment the detection of ARGs and identify antibiotic targets and antagonistic bactericidal and bacteriostatic molecules that are or can be developed as antibiotics. This review delves into the literature regarding the various AI methods and approaches for identifying and annotating ARGs, highlighting their potential and limitations. Specifically, we discuss methods for (1) direct identification and classification of ARGs from genome DNA sequences, (2) direct identification and classification from plasmid sequences, and (3) identification of putative ARGs from feature selection.

Genome-resolved metagenomics: a game changer for microbiome medicine

Recent substantial evidence implicating commensal bacteria in human diseases has given rise to a new domain in biomedical research: microbiome medicine. This emerging field aims to understand and leverage the human microbiota and derivative molecules for disease prevention and treatment. Despite the complex and hierarchical organization of this ecosystem, most research over the years has relied on 16S amplicon sequencing, a legacy of bacterial phylogeny and taxonomy. Although advanced sequencing technologies have enabled cost-effective analysis of entire microbiota, translating the relatively short nucleotide information into the functional and taxonomic organization of the microbiome has posed challenges until recently. In the last decade, genome-resolved metagenomics, which aims to reconstruct microbial genomes directly from whole-metagenome sequencing data, has made significant strides and continues to unveil the mysteries of various human-associated microbial communities. There has been a rapid increase in the volume of whole metagenome sequencing data and in the compilation of novel metagenome-assembled genomes and protein sequences in public depositories. This review provides an overview of the capabilities and methods of genome-resolved metagenomics for studying the human microbiome, with a focus on investigating the prokaryotic microbiota of the human gut. Just as decoding the human genome and its variations marked the beginning of the genomic medicine era, unraveling the genomes of commensal microbes and their sequence variations is ushering us into the era of microbiome medicine. Genome-resolved metagenomics stands as a pivotal tool in this transition and can accelerate our journey toward achieving these scientific and medical milestones.

Subtask-Aware Representation Learning for Predicting Antibiotic Resistance Gene Properties via Gating-Controlled Mechanism

The crisis of antibiotic resistance has become a significant global threat to human health. Understanding properties of antibiotic resistance genes (ARGs) is the first step to mitigate this issue. Although many methods have been proposed for predicting properties of ARGs, most of these methods focus only on predicting antibiotic classes, while ignoring other properties of ARGs, such as resistance mechanisms and transferability. However, acquiring all of these properties of ARGs can help researchers gain a more comprehensive understanding of the essence of antibiotic resistance, which will facilitate the development of antibiotics. In this paper, the task of predicting properties of ARGs is modeled as a multi-task learning problem, and an effective subtask-aware representation learning-based framework is proposed accordingly. More specifically, property-specific expert networks and shared expert networks are utilized respectively to learn subtask-specific features for each subtask and shared features among different subtasks. In addition, a gating-controlled mechanism is employed to dynamically allocate weights to subtask-specific semantics and shared semantics obtained respectively from property-specific expert networks and shared expert networks, thus adjusting distinctive contributions of subtask-specific features and shared features to achieve optimal performance for each subtask simultaneously. Extensive experiments are conducted on publicly available data, and experimental results demonstrate the effectiveness of the proposed framework on the task of ARGs properties prediction.

TGC-ARG: Anticipating Antibiotic Resistance via Transformer-Based Modeling and Contrastive Learning

In various domains, including everyday activities, agricultural practices, and medical treatments, the escalating challenge of antibiotic resistance poses a significant concern. Traditional approaches to studying antibiotic resistance genes (ARGs) often require substantial time and effort and are limited in accuracy. Moreover, the decentralized nature of existing data repositories complicates comprehensive analysis of antibiotic resistance gene sequences. In this study, we introduce a novel computational framework named TGC-ARG designed to predict potential ARGs. This framework takes protein sequences as input, utilizes SCRATCH-1D for protein secondary structure prediction, and employs feature extraction techniques to derive distinctive features from both sequence and structural data. Subsequently, a Siamese network is employed to foster a contrastive learning environment, enhancing the model's ability to effectively represent the data. Finally, a multi-layer perceptron (MLP) integrates and processes sequence embeddings alongside predicted secondary structure embeddings to forecast ARG presence. To evaluate our approach, we curated a pioneering open dataset termed ARSS (Antibiotic Resistance Sequence Statistics). Comprehensive comparative experiments demonstrate that our method surpasses current state-of-the-art methodologies. Additionally, through detailed case studies, we illustrate the efficacy of our approach in predicting potential ARGs.

Large language model for horizontal transfer of resistance gene: From resistance gene prevalence detection to plasmid conjugation rate evaluation

The burgeoning issue of plasmid-mediated resistance genes (ARGs) dissemination poses a significant threat to environmental integrity. However, the prediction of ARGs prevalence is overlooked, especially for emerging ARGs that are potentially evolving gene exchange hotspot. Here, we explored to classify plasmid or chromosome sequences and detect resistance gene prevalence by using DNABERT. Initially, the DNABERT fine-tuned in plasmid and chromosome sequences followed by multilayer perceptron (MLP) classifier could achieve 0.764 AUC (Area under curve) on external datasets across 23 genera, outperforming 0.02 AUC than traditional statistic-based model. Furthermore, Escherichia, Pseudomonas single genera based model were also be trained to explore its predict performance to ARGs prevalence detection. By integrating K-mer frequency attributes, our model could boost the performance to predict the prevalence of ARGs in an external dataset in Escherichia with 0.0281-0.0615 AUC and Pseudomonas with 0.0196-0.0928 AUC. Finally, we established a random forest model aimed at forecasting the relative conjugation transfer rate of plasmids with 0.7956 AUC, drawing on data from existing literature. It identifies the plasmid's repression status, cellular density, and temperature as the most important factors influencing transfer frequency. With these two models combined, they provide useful reference for quick and low-cost integrated evaluation of resistance gene transfer, accelerating the process of computer-assisted quantitative risk assessment of ARGs transfer in environmental field.

Deep learning for predicting 16S rRNA gene copy number

Culture-independent 16S rRNA gene metabarcoding is a commonly used method for microbiome profiling. To achieve more quantitative cell fraction estimates, it is important to account for the 16S rRNA gene copy number (hereafter 16S GCN) of different community members. Currently, there are several bioinformatic tools available to estimate the 16S GCN values, either based on taxonomy assignment or phylogeny. Here we present a novel approach ANNA16, Artificial Neural Network Approximator for 16S rRNA gene copy number, a deep learning-based method that estimates the 16S GCN values directly from the 16S gene sequence strings. Based on 27,579 16S rRNA gene sequences and gene copy number data from the rrnDB database, we show that ANNA16 outperforms the commonly used 16S GCN prediction algorithms. Interestingly, Shapley Additive exPlanations (SHAP) shows that ANNA16 can identify unexpected informative positions in 16S rRNA gene sequences without any prior phylogenetic knowledge, which suggests potential applications beyond 16S GCN prediction.

A systematic review of antibiotics and antibiotic resistance genes (ARGs) in mariculture wastewater: Antibiotics removal by microalgal-bacterial symbiotic system (MBSS), ARGs characterization on the metagenomic

Antibiotic residues in mariculture wastewater seriously affect the aquatic environment. Antibiotic Resistance Genes (ARGs) produced under antibiotic stress flow through the environment and eventually enter the human body, seriously affecting human health. Microalgal-bacterial symbiotic system (MBSS) can remove antibiotics from mariculture and reduce the flow of ARGs into the environment. This review encapsulates the present scenario of mariculture wastewater, the removal mechanism of MBSS for antibiotics, and the biomolecular information under metagenomic assay. When confronted with antibiotics, there was a notable augmentation in the extracellular polymeric substances (EPS) content within MBSS, along with a concurrent elevation in the proportion of protein (PN) constituents within the EPS, which limits the entry of antibiotics into the cellular interior. Quorum sensing stimulates the microorganisms to produce biological responses (DNA synthesis - for adhesion) through signaling. Oxidative stress promotes gene expression (coupling, conjugation) to enhance horizontal gene transfer (HGT) in MBSS. The microbial community under metagenomic detection is dominated by aerobic bacteria in the bacterial-microalgal system. Compared to aerobic bacteria, anaerobic bacteria had the significant advantage of decreasing the distribution of ARGs. Overall, MBSS exhibits remarkable efficacy in mitigating the challenges posed by antibiotics and resistant genes from mariculture wastewater.

Assessing intracellular and extracellular distribution of antibiotic resistance genes in the commercial organic fertilizers

Compost-based organic fertilizers often contain high levels of antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs). Previous studies focused on quantification of total ARGs and MGEs. For a more accurate risk assessment of the dissemination risk of antibiotic resistance, it is necessary to quantify the intracellular and extracellular distribution of ARGs and MGEs. In the present study, extracellular ARGs and MGEs (eARGs and eMGEs) and intracellular ARGs and MGEs (iARGs and iMGEs) were separately analyzed in 51 commercial composts derived from different raw materials by quantitative polymerase chain reaction (qPCR) and metagenomic sequencing. Results showed that eARGs and eMGEs accounted for 11-56% and 4-45% of the total absolute abundance of ARGs and MGEs, respectively. Comparable diversity, host composition and association with MGEs were observed between eARGs and iARGs. Contents of high-risk ARGs were similar between eARGs and iARGs, with high-risk ARGs in the two forms accounting for 6.7% and 8.2% of the total abundances, respectively. Twenty-four percent of the overall ARGs were present in plasmids, while 56.7% of potentially mobile ARGs were found to be associated with plasmids. Variation partitioning analysis, null model and neutral community model indicated that the compositions of both eARGs and iARGs were largely driven by deterministic mechanisms. These results provide important insights into the cellular distribution of ARGs in manure composts that should be paid with specific attention in risk assessment and management.

Application of machine learning for antibiotic resistance in water and wastewater: A systematic review

Antibiotic resistance (AR) is considered one of the greatest global threats in the current century, which can only be overcome if all interconnected areas of humans, animals and the environment are taken into account as part of the One Health concept proposed by the World Health Organization (WHO). Water and wastewater are among the most important environmental media of AR sources, where the phenomena are generally non-linear. Therefore, the aim of this study was to investigate the application of machine learning-based methods (MLMs) to solve AR-induced problems in water and wastewater. For this purpose, most relevant databases were searched in the period between 1987 and 2023 to systematically analyze and categorize the applications. Accordingly, the results showed that out of 12 applications, 11 (91.6%) were for shallow learning and 1 (8.3%) for deep learning. In shallow learning category, n = 6, 50% of the applications were regression and n = 4, 33.3% were classification, mainly using artificial neural networks, decision trees and Bayesian methods for the following objectives: Predicting the survival of antibiotic-resistant bacteria (ARB), determining the order of influencing parameters on AR-based scores, and identifying the major sources of antibiotic resistance genes (ARGs). In addition, only one study (8.3%) was found for clustering and no study for association. Surprisingly, deep learning had been used in only one study (8.3%) to predict ARGs sequences. Therefore, working on the knowledge gaps of AR, especially using clustering, association and deep learning methods, would be a promising option to analyze more aspects of the related problems. However, there is still a long way to go to consider and apply MLMs as unique approaches to study different aspects of AR in water and wastewater.

Parity in bacterial communities and resistomes: Microplastic and natural organic particles in the Tyrrhenian Sea

Petroleum-based microplastic particles (MPs) are carriers of antimicrobial resistance genes (ARGs) in aquatic environments, influencing the selection and spread of antimicrobial resistance. This research characterized MP and natural organic particle (NOP) bacterial communities and resistomes in the Tyrrhenian Sea, a region impacted by plastic pollution and climate change. MP and NOP bacterial communities were similar but different from the free-living planktonic communities. Likewise, MP and NOP ARG abundances were similar but different (higher) from the planktonic communities. MP and NOP metagenome-assembled genomes contained ARGs associated with mobile genetic elements and exhibited co-occurrence with metal resistance genes. Overall, these findings show that MPs and NOPs harbor potential pathogenic and antimicrobial resistant bacteria, which can aid in the spread of antimicrobial resistance. Further, petroleum-based MPs do not represent novel ecological niches for allochthonous bacteria; rather, they synergize with NOPs, collectively facilitating the spread of antimicrobial resistance in marine ecosystems.