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

Associations of the skin, oral and gut microbiome with aging, frailty and infection risk reservoirs in older adults

Older adults represent a vulnerable population with elevated risk for numerous morbidities. To explore the association of the microbiome with aging and age-related susceptibilities including frailty and infectious disease risk, we conducted a longitudinal study of the skin, oral, and gut microbiota in 47 community- or skilled nursing facility-dwelling older adults vs. younger adults. We found that microbiome changes were not associated with chronological age so much as frailty: we identified prominent changes in microbiome features associated with susceptibility to pathogen colonization and disease risk, including diversity, stability, heterogeneity, and biogeographic determinism, which were moreover associated with a loss of Cutibacterium (C.) acnes in the skin microbiome. Strikingly, the skin microbiota were also the primary reservoir for antimicrobial resistance, clinically important pathobionts, and nosocomial strains, suggesting a potential role particularly for the skin microbiome in disease risk and dissemination of multidrug resistant pathogens.

Trade-off for survival: Microbiome response to chemical exposure combines activation of intrinsic resistances and adapted metabolic activity

The environmental microbiota is increasingly exposed to chemical pollution. While the emergence of multi-resistant pathogens is recognized as a global challenge, our understanding of antimicrobial resistance (AMR) development from native microbiomes and the risks associated with chemical exposure is limited. By implementing a lichen asa bioindicatororganism and model for a native microbiome, we systematically examined responses towards antimicrobials (colistin, tetracycline, glyphosate, and alkylpyrazine). Despite an unexpectedly high resilience, we identified potential evolutionary consequences of chemical exposure in terms of composition and functioning of native bacterial communities. Major shifts in bacterial composition were observed due to replacement of naturally abundant taxa; e.g. Chthoniobacterales by Pseudomonadales. A general response, which comprised activation of intrinsic resistance and parallel reduction of metabolic activity at RNA and protein levels was deciphered by a multi-omics approach. Targeted analyses of key taxa based on metagenome-assembled genomes reflected these responses but also revealed diversified strategies of their players. Chemical-specific responses were also observed, e.g., glyphosate enriched bacterial r-strategists and activated distinct ARGs. Our work demonstrates that the high resilience of the native microbiota toward antimicrobial exposure is not only explained by the presence of antibiotic resistance genes but also adapted metabolic activity as a trade-off for survival. Moreover, our results highlight the importance of native microbiomes as important but so far neglected AMR reservoirs. We expect that this phenomenon is representative for a wide range of environmental microbiota exposed to chemicals that potentially contribute to the emergence of antibiotic-resistant bacteria from natural environments.

Microbial Dark Matter: from Discovery to Applications

With the rapid increase of the microbiome samples and sequencing data, more and more knowledge about microbial communities has been gained. However, there is still much more to learn about microbial communities, including billions of novel species and genes, as well as countless spatiotemporal dynamic patterns within the microbial communities, which together form the microbial dark matter. In this work, we summarized the dark matter in microbiome research and reviewed current data mining methods, especially artificial intelligence (AI) methods, for different types of knowledge discovery from microbial dark matter. We also provided case studies on using AI methods for microbiome data mining and knowledge discovery. In summary, we view microbial dark matter not as a problem to be solved but as an opportunity for AI methods to explore, with the goal of advancing our understanding of microbial communities, as well as developing better solutions to global concerns about human health and the environment.

Metagenomic insights into the antibiotic resistomes of typical Chinese dairy farm environments

Antibiotic resistance genes (ARGs) in the environment pose a threat to human and animal health. Dairy cows are important livestock in China; however, a comprehensive understanding of antibiotic resistance in their production environment has not been well clarified. In this study, we used metagenomic methods to analyze the resistomes, microbiomes, and potential ARG bacterial hosts in typical dairy farm environments (including feces, wastewater, and soil). The ARGs resistant to tetracyclines, MLS, β-lactams, aminoglycoside, and multidrug was dominant in the dairy farm ecosystem. The abundance and diversity of total ARGs in dairy feces and wastewater were significantly higher than in soil (P < 0.05). The same environmental samples from different dairy have similar resistomes and microbiomes. A high detection rate of tet(X) in wastewater and feces (100% and 71.4%, respectively), high abundance (range from 5.74 to 68.99 copies/Gb), and the finding of tet(X5) challenged the clinical application of the last antibiotics resort of tigecycline. Network analysis identified Bacteroides as the dominant genus in feces and wastewater, which harbored the greatest abundance of their respective total ARG coverage and shared ARGs. These results improved our understanding of ARG profiles and their bacterial hosts in dairy farm environments and provided a basis for further surveillance.

Exploration of machine learning algorithms for predicting the changes in abundance of antibiotic resistance genes in anaerobic digestion

The land application of digestate from anaerobic digestion (AD) is considered a significant route for transmitting antibiotic resistance genes (ARGs) and mobile genetic elements (MGEs) to ecosystems. To date, efforts towards understanding complex non-linear interactions between AD operating parameters with ARG/MGE abundances rely on experimental investigations due to a lack of mechanistic models. Herein, three different machine learning (ML) algorithms, Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN), were compared for their predictive capacities in simulating ARG/MGE abundance changes during AD. The models were trained and cross-validated using experimental data collected from 33 published literature. The comparison of model performance using coefficients of determination (R²) and root mean squared errors (RMSE) indicated that ANN was more reliable than RF and XGBoost. The mode of operation (batch/semi-continuous), co-digestion of food waste and sewage sludge, and residence time were identified as the three most critical features in predicting ARG/MGE abundance changes. Moreover, the trained ANN model could simulate non-linear interactions between operational parameters and ARG/MGE abundance changes that could be interpreted intuitively based on existing knowledge. Overall, this study demonstrates that machine learning can enable a reliable predictive model that can provide a holistic optimization tool for mitigating the ARG/MGE transmission potential of AD.

Deciphering environmental resistome and mobilome risks on the stone monument: A reservoir of antimicrobial resistance genes

Antimicrobial resistance (AMR) in the environment has attracted increasing attention as an emerging global threat to public health. Stone is an essential ecosystem in nature and also an important material for human society, having architectural and aesthetic values. However, little is known about the AMR in stone ecosystems, particularly in the stone monument, where antimicrobials are often applied against biodeterioration. Here, we provide the first detailed metagenomic study of AMR genes across different types of biodeteriorated stone monuments, which revealed abundant and diverse AMR genes conferring resistance to drugs (antibiotics), biocides, and metals. Totally, 132 AMR subtypes belonging to 27 AMR types were detected including copper-, rifampin-, and aminocoumarins-resistance genes, of which diversity was mainly explained by the spatial turnover (replacement of genes between samples) rather than nestedness (loss of nested genes between samples). Source track analysis confirms that stone resistomes are likely driven by anthropogenic activities across stone heritage areas. We also detected various mobile genetic elements (namely mobilome, e.g., prophages, plasmids, and insertion sequences) that could accelerate replication and horizontal transfer of AMR genes. Host-tracking analysis further identified multiple biodeterioration-related bacterial genera such as Pseudonocardia, Sphingmonas, and Streptomyces as the major hosts of resistome. Taken together, these findings highlight that stone microbiota is one of the natural reservoirs of antimicrobial-resistant hazards, and the diverse resistome and mobilome carried by active biodeteriogens may improve their adaptation on stone and even deactivate the antimicrobials applied against biodeterioration. This enhanced knowledge may also provide novel and specific avenues for environmental management and stone heritage protection.

Metagenome Analysis Reveals a Response of the Antibiotic Resistome to Mars-like Extraterrestrial Conditions

The spread of antibiotic resistance is becoming a serious global health concern. Numerous studies have been done to investigate the dynamics of antibiotic resistance genes (ARGs) in both indoor and outdoor environments. Nonetheless, few studies are available about the dynamics of the antibiotic resistome (total content of ARGs in the microbial cultures or communities) under stress in outer space environments. In this study, we aimed to experimentally investigate the dynamics of ARGs and metal resistance genes (MRGs) in Kombucha Mutualistic Community (KMC) samples exposed to Mars-like conditions simulated during the BIOMEX experiment outside the International Space Station with analysis of the metagenomics data previously produced. Thus, we compared them with those of the respective non-exposed KMC samples. The antibiotic resistome responded to the Mars-like conditions by enriching its diversity with ARGs after exposure, which were not found in non-exposed samples (i.e., tet and van genes against tetracycline and vancomycin, respectively). Furthermore, ARGs and MRGs were correlated; therefore, their co-selection could be assumed as a mechanism for maintaining antibiotic resistance in Mars-like environments. Overall, these results highlight the high plasticity of the antibiotic resistome in response to extraterrestrial conditions and in the absence of anthropogenic stresses.

Globally distributed mining-impacted environments are underexplored hotspots of multidrug resistance genes

Mining is among the human activities with widest environmental impacts, and mining-impacted environments are characterized by high levels of metals that can co-select for antibiotic resistance genes (ARGs) in microorganisms. However, ARGs in mining-impacted environments are still poorly understood. Here, we conducted a comprehensive study of ARGs in such environments worldwide, taking advantage of 272 metagenomes generated from a global-scale data collection and two national sampling efforts in China. The average total abundance of the ARGs in globally distributed studied mine sites was 1572 times per gigabase, being rivaling that of urban sewage but much higher than that of freshwater sediments. Multidrug resistance genes accounted for 40% of the total ARG abundance, tended to co-occur with multimetal resistance genes, and were highly mobile (e.g. on average 16% occurring on plasmids). Among the 1848 high-quality metagenome-assembled genomes (MAGs), 85% carried at least one multidrug resistance gene plus one multimetal resistance gene. These high-quality ARG-carrying MAGs considerably expanded the phylogenetic diversity of ARG hosts, providing the first representatives of ARG-carrying MAGs for the Archaea domain and three bacterial phyla. Moreover, 54 high-quality ARG-carrying MAGs were identified as potential pathogens. Our findings suggest that mining-impacted environments worldwide are underexplored hotspots of multidrug resistance genes.

Influence of timing of maternal antibiotic administration during caesarean section on infant microbial colonisation: a randomised controlled trial

OBJECTIVE

Revised guidelines for caesarean section (CS) advise maternal antibiotic administration prior to skin incision instead of after umbilical cord clamping, unintentionally exposing the infant to antibiotics antenatally. We aimed to investigate if timing of intrapartum antibiotics contributes to the impairment of microbiota colonisation in CS born infants.

DESIGN

In this randomised controlled trial, women delivering via CS received antibiotics prior to skin incision (n=20) or after umbilical cord clamping (n=20). A third control group of vaginally delivering women (n=23) was included. Faecal microbiota was determined from all infants at 1, 7 and 28 days after birth and at 3 years by 16S rRNA gene sequencing and whole-metagenome shotgun sequencing.

RESULTS

Compared with vaginally born infants, profound differences were found in microbial diversity and composition in both CS groups in the first month of life. A decreased abundance in species belonging to the genera Bacteroides and Bifidobacterium was found with a concurrent increase in members belonging to the phylum Proteobacteria. These differences could not be observed at 3 years of age. No statistically significant differences were observed in taxonomic and functional composition of the microbiome between both CS groups at any of the time points.

CONCLUSION

We confirmed that microbiome colonisation is strongly affected by CS delivery. Our findings suggest that maternal antibiotic administration prior to CS does not result in a second hit on the compromised microbiome. Future, larger studies should confirm that antenatal antibiotic exposure in CS born infants does not aggravate colonisation impairment and impact long-term health.

Metagenomic binning and assembled genome analysis revealed the distinct composition of resistome and mobilome in the Ili River

Rivers play an important role in receiving and transporting the resistome among different environmental compartments. However, the difference in resistome and mobilome between the water and sediment and their underlying mechanisms were still poorly understood. In this study, the Ili River, an important water source in the arid area of Central Asia, was selected as the studied target. The comprehensive profile of resistome and mobilome and their host in water and sediment were studied based on metagenomic binning and assembled genome (MAG) analysis. The relative abundance of resistome and mobilome in sediment were 28.0 - 67.8 × /Gb and 46.5 - 121.1 × /Gb, respectively, which were significantly higher than those in water (23.1 - 52.8 ×/Gb and 25.3 - 67.7 ×/Gb). Multidrug and macrolides-lincosamides-streptogramin (MLS) resistance genes were the main ARG types in both water and sediment from relative abundance. Transposases dominated the relative abundance of mobilome, followed by insert elements and integrases. Strong correlations were found between the relative abundance of resistome and mobilome (r > 0.6 and p < 0.01) in both water and sediment, indicating the mobilome played an important role in the propagation of resistome in the Ili River. The main hosts for multidrug resistance genes via MAG analysis differed in water (Alphaproteobacteria and Gammaproteobacteria) and sediment (Gammaproteobacteria). Distinct compositions of resistome and mobilome existed between water and sediment in the Ili River. Specificity-occupancy analysis of the differential resistome and mobilome showed that occurrence frequencies and habitat selections of the differential ARGs shaped the resistome of water and sediment. In contrast, habitat was the main driver that shaped the mobilome in the Ili River.

Guts of the Urban Ecosystem: Microbial Ecology of Sewer Infrastructure

Microbes have inhabited the oceans and soils for millions of years and are uniquely adapted to their habitat. In contrast, sewer infrastructure in modern cities dates back only ~150 years. Sewer pipes transport human waste and provide a view into public health, but the resident organisms that likely modulate these features are relatively unexplored. Here, we show that the bacterial assemblages sequenced from untreated wastewater in 71 U.S. cities were highly coherent at a fine sequence level, suggesting that urban infrastructure separated by great spatial distances can give rise to strikingly similar communities. Within the overall microbial community structure, temperature had a discernible impact on the distribution patterns of closely related amplicon sequence variants, resulting in warm and cold ecotypes. Two bacterial genera were dominant in most cities regardless of their size or geographic location; on average, Arcobacter accounted for 11% and Acinetobacter 10% of the entire community. Metagenomic analysis of six cities revealed these highly abundant resident organisms carry clinically important antibiotic resistant genes blaCTX-M, blaOXA, and blaTEM. In contrast, human fecal bacteria account for only ~13% of the community; therefore, antibiotic resistance gene inputs from human sources to the sewer system could be comparatively small, which will impact measurement capabilities when monitoring human populations using wastewater. With growing awareness of the metabolic potential of microbes within these vast networks of pipes and the ability to examine the health of human populations, it is timely to increase our understanding of the ecology of these systems. IMPORTANCE Sewer infrastructure is a relatively new habitat comprised of thousands of kilometers of pipes beneath cities. These wastewater conveyance systems contain large reservoirs of microbial biomass with a wide range of metabolic potential and are significant reservoirs of antibiotic resistant organisms; however, we lack an adequate understanding of the ecology or activity of these communities beyond wastewater treatment plants. The striking coherence of the sewer microbiome across the United States demonstrates that the sewer environment is highly selective for a particular microbial community composition. Therefore, results from more in-depth studies or proven engineering controls in one system could be extrapolated more broadly. Understanding the complex ecology of sewer infrastructure is critical for not only improving our ability to treat human waste and increasing the sustainability of our cities but also to create scalable and effective sewage microbial observatories, which are inevitable investments of the future to monitor health in human populations.

Capturing the antibiotic resistome of preterm infants reveals new benefits of probiotic supplementation

BACKGROUND

Probiotic use in preterm infants can mitigate the impact of antibiotic exposure and reduce rates of certain illnesses; however, the benefit on the gut resistome, the collection of antibiotic resistance genes, requires further investigation. We hypothesized that probiotic supplementation of early preterm infants (born < 32-week gestation) while in hospital reduces the prevalence of antibiotic resistance genes associated with pathogenic bacteria in the gut. We used a targeted capture approach to compare the resistome from stool samples collected at the term corrected age of 40 weeks for two groups of preterm infants (those that routinely received a multi-strain probiotic during hospitalization and those that did not) with samples from full-term infants at 10 days of age to identify if preterm birth or probiotic supplementation impacted the resistome. We also compared the two groups of preterm infants up to 5 months of age to identify persistent antibiotic resistance genes.

RESULTS

At the term corrected age, or 10 days of age for the full-term infants, we found over 80 antibiotic resistance genes in the preterm infants that did not receive probiotics that were not identified in either the full-term or probiotic-supplemented preterm infants. More genes associated with antibiotic inactivation mechanisms were identified in preterm infants unexposed to probiotics at this collection time-point compared to the other infants. We further linked these genes to mobile genetic elements and Enterobacteriaceae, which were also abundant in their gut microbiomes. Various genes associated with aminoglycoside and beta-lactam resistance, commonly found in pathogenic bacteria, were retained for up to 5 months in the preterm infants that did not receive probiotics.

CONCLUSIONS

This pilot survey of preterm infants shows that probiotics administered after preterm birth during hospitalization reduced the diversity and prevented persistence of antibiotic resistance genes in the gut microbiome. The benefits of probiotic use on the microbiome and the resistome should be further explored in larger groups of infants. Due to its high sensitivity and lower sequencing cost, our targeted capture approach can facilitate these surveys to further address the implications of resistance genes persisting into infancy without the need for large-scale metagenomic sequencing. Video Abstract.

Machine learning algorithm to characterize antimicrobial resistance associated with the International Space Station surface microbiome

BACKGROUND

Antimicrobial resistance (AMR) has a detrimental impact on human health on Earth and it is equally concerning in other environments such as space habitat due to microgravity, radiation and confinement, especially for long-distance space travel. The International Space Station (ISS) is ideal for investigating microbial diversity and virulence associated with spaceflight. The shotgun metagenomics data of the ISS generated during the Microbial Tracking-1 (MT-1) project and resulting metagenome-assembled genomes (MAGs) across three flights in eight different locations during 12 months were used in this study. The objective of this study was to identify the AMR genes associated with whole genomes of 226 cultivable strains, 21 shotgun metagenome sequences, and 24 MAGs retrieved from the ISS environmental samples that were treated with propidium monoazide (PMA; viable microbes).

RESULTS

We have analyzed the data using a deep learning model, allowing us to go beyond traditional cut-offs based only on high DNA sequence similarity and extending the catalog of AMR genes. Our results in PMA treated samples revealed AMR dominance in the last flight for Kalamiella piersonii, a bacteria related to urinary tract infection in humans. The analysis of 226 pure strains isolated from the MT-1 project revealed hundreds of antibiotic resistance genes from many isolates, including two top-ranking species that corresponded to strains of Enterobacter bugandensis and Bacillus cereus. Computational predictions were experimentally validated by antibiotic resistance profiles in these two species, showing a high degree of concordance. Specifically, disc assay data confirmed the high resistance of these two pathogens to various beta-lactam antibiotics.

CONCLUSION

Overall, our computational predictions and validation analyses demonstrate the advantages of machine learning to uncover concealed AMR determinants in metagenomics datasets, expanding the understanding of the ISS environmental microbiomes and their pathogenic potential in humans. Video Abstract.

Using Culture-Enriched Phenotypic Metagenomics for Targeted High-Throughput Monitoring of the Clinically Important Fraction of the β-Lactam Resistome

High bacterial community diversity and complexity greatly challenge the cost-efficient monitoring of clinically prevalent antibiotic-resistant bacteria, which are usually present as rare and important populations involved in the environmental dissemination of clinical resistance. Here, we introduce culture-enriched phenotypic metagenomics that integrates culture enrichment, phenotypic screening, and metagenomic analyses as an emerging standardized methodology for targeted resistome monitoring and apply it to decipher the extended-spectrum β-lactam resistome in a municipal wastewater treatment plant (WWTP) and its receiving river. The results showed that clinically prevalent carbapenemase genes (e.g., the NDM and KPC families) and extended-spectrum β-lactamase genes (e.g., the CTX-M, TEM, and OXA families) were prevalent in the WWTP and showed prominent potential in horizontal dissemination. Strikingly, carbapenem and polymyxin resistance genes co-occurred in the highly virulent nosocomial pathogens Enterobacter kobei and Citrobacter freundii. Overall, this study exemplifies phenotypic metagenomics for high-throughput surveillance of a targeted clinically important fraction of antibiotic resistomes and substantially expands current knowledge on extended-spectrum β-lactam resistance in WWTPs.

Deciphering Resistome in Patients With Chronic Obstructive Pulmonary Diseases and Clostridioides difficile Infections

Antibiotics alter the gut microbiome and cause dysbiosis leading to antibiotic-resistant organisms. Different patterns of antibiotic administration cause a difference in bacterial composition and resistome in the human gut. We comprehensively investigated the association between the distribution of antibiotic resistance genes (ARGs), bacterial composition, and antibiotic treatments in patients with chronic obstructive pulmonary diseases (COPD) and Clostridioides difficile infections (CDI) who had chronic or acute intermittent use of antibiotics and compared them with healthy individuals. We analyzed the gut microbiomes of 61 healthy individuals, 16 patients with COPD, and 26 patients with CDI. The COPD patients were antibiotic-free before stool collection for a median of 40 days (Q1: 9.5; Q3: 60 days), while the CDI patients were antibiotic-free for 0 days (Q1: 0; Q3: 0.3). The intra-group beta diversity measured by the median Bray-Curtis index was the lowest for the healthy individuals (0.55), followed by the COPD (0.69) and CDI groups (0.72). The inter-group beta diversity was the highest among the healthy and CDI groups (median index = 0.89). The abundance of ARGs measured by the number of reads per kilobase per million reads (RPKM) was 684.2; 1,215.2; and 2,025.1 for the healthy, COPD, and CDI groups. It was negatively correlated with the alpha diversity of bacterial composition. For the prevalent ARG classes, healthy individuals had the lowest diversity and abundance of aminoglycoside, β-lactam, and macrolide-lincosamide-streptogramin (MLS) resistance genes, followed by the COPD and CDI groups. The abundances of Enterococcus and Escherichia species were positively correlated with ARG abundance and the days of antibiotic treatment, while Bifidobacterium and Ruminococcus showed negative correlations for the same. In addition, we analyzed the mobilome patterns of aminoglycoside and β-lactam resistance gene carriers using metagenomic sequencing data. In conclusion, the ARGs were significantly enhanced in the CDI and COPD groups than in healthy individuals. In particular, aminoglycoside and β-lactam resistance genes were more abundant in the CDI and COPD groups, but the dominant mobile genetic elements that enable the transfer of such genes showed similar prevalence patterns among the groups.

Metagenomic tracking of antibiotic resistance genes through a pre-harvest vegetable production system: an integrated lab-, microcosm- and greenhouse-scale analysis

Prior research demonstrated the potential for agricultural production systems to contribute to the environmental spread of antibiotic resistance genes (ARGs). However, there is a need for integrated assessment of critical management points for minimizing this potential. Shotgun metagenomic sequencing data were analysed to comprehensively compare total ARG profiles characteristic of amendments (manure or compost) derived from either beef or dairy cattle (with and without dosing antibiotics according to conventional practice), soil (loamy sand or silty clay loam) and vegetable (lettuce or radish) samples collected across studies carried out at laboratory-, microcosm- and greenhouse-scale. Vegetables carried the greatest diversity of ARGs (n = 838) as well as the most ARG-mobile genetic element co-occurrences (n = 945). Radishes grown in manure- or compost-amended soils harboured a higher relative abundance of total (0.91 and 0.91 ARGs/16S rRNA gene) and clinically relevant ARGs than vegetables from other experimental conditions (average: 0.36 ARGs/16S rRNA gene). Lettuce carried the highest relative abundance of pathogen gene markers among the metagenomes examined. Total ARG relative abundances were highest on vegetables grown in loamy sand receiving antibiotic-treated beef amendments. The findings emphasize that additional barriers, such as post-harvest processes, merit further study to minimize potential exposure to consumers.

Antibiotic resistance genes are differentially mobilized according to resistance mechanism

BACKGROUND

Screening for antibiotic resistance genes (ARGs) in especially environmental samples with (meta)genomic sequencing is associated with false-positive predictions of phenotypic resistance. This stems from the fact that most acquired ARGs require being overexpressed before conferring resistance, which is often caused by decontextualization of putative ARGs by mobile genetic elements (MGEs). Consequent overexpression of ARGs can be caused by strong promoters often present in insertion sequence (IS) elements and integrons and the copy number effect of plasmids, which may contribute to high expression of accessory genes.

RESULTS

Here, we screen all complete bacterial RefSeq genomes for ARGs. The genetic contexts of detected ARGs are investigated for IS elements, integrons, plasmids, and phylogenetic dispersion. The ARG-MOB scale is proposed, which indicates how mobilized detected ARGs are in bacterial genomes. It is concluded that antibiotic efflux genes are rarely mobilized and even 80% of β-lactamases have never, or very rarely, been mobilized in the 15,790 studied genomes. However, some ARGs are indeed mobilized and co-occur with IS elements, plasmids, and integrons.

CONCLUSIONS

In this study, ARGs in all complete bacterial genomes are classified by their association with MGEs, using the proposed ARG-MOB scale. These results have consequences for the design and interpretation of studies screening for resistance determinants, as mobilized ARGs pose a more concrete risk to human health. An interactive table of all results is provided for future studies targeting highly mobilized ARGs.

Supercarriers of antibiotic resistome in a world's large river

BACKGROUND

Antibiotic resistome has been found to strongly interact with the core microbiota in the human gut, yet little is known about how antibiotic resistance genes (ARGs) correlate with certain microbes in large rivers that are regarded as "terrestrial gut."

RESULTS

By creating the integral pattern for ARGs and antibiotic-resistant microbes in water and sediment along a 4300-km continuum of the Yangtze River, we found that human pathogen bacteria (HPB) share 13.4% and 5.9% of the ARG hosts in water and sediment but contribute 64% and 46% to the total number of planktonic and sedimentary ARGs, respectively. Moreover, the planktonic HPB harbored 79 ARG combinations that are dominated by "natural" supercarriers (e.g., Rheinheimera texasensis and Noviherbaspirillum sp. Root189) in river basins.

CONCLUSIONS

We confirmed that terrestrial HPB are the major ARG hosts in the river, rather than conventional supercarriers (e.g., Enterococcus spp. and other fecal indicator bacteria) that prevail in the human gut. The discovery of HPB as natural supercarriers in a world's large river not only interprets the inconsistency between the spatial dissimilarities in ARGs and their hosts, but also highlights the top priority of controlling terrestrial HPB in the future ARG-related risk management of riverine ecosystems globally. Video Abstract.

The Application of Artificial Intelligence in the Diagnosis and Drug Resistance Prediction of Pulmonary Tuberculosis

With the increasing incidence and mortality of pulmonary tuberculosis, in addition to tough and controversial disease management, time-wasting and resource-limited conventional approaches to the diagnosis and differential diagnosis of tuberculosis are still awkward issues, especially in countries with high tuberculosis burden and backwardness. In the meantime, the climbing proportion of drug-resistant tuberculosis poses a significant hazard to public health. Thus, auxiliary diagnostic tools with higher efficiency and accuracy are urgently required. Artificial intelligence (AI), which is not new but has recently grown in popularity, provides researchers with opportunities and technical underpinnings to develop novel, precise, rapid, and automated implements for pulmonary tuberculosis care, including but not limited to tuberculosis detection. In this review, we aimed to introduce representative AI methods, focusing on deep learning and radiomics, followed by definite descriptions of the state-of-the-art AI models developed using medical images and genetic data to detect pulmonary tuberculosis, distinguish the infection from other pulmonary diseases, and identify drug resistance of tuberculosis, with the purpose of assisting physicians in deciding the appropriate therapeutic schedule in the early stage of the disease. We also enumerated the challenges in maximizing the impact of AI in this field such as generalization and clinical utility of the deep learning models.

Metagenomic Characterization of Resistance Genes in Deception Island and Their Association with Mobile Genetic Elements

Antibiotic resistance genes (ARGs) are undergoing a remarkably rapid geographic expansion in various ecosystems, including pristine environments such as Antarctica. The study of ARGs and environmental resistance genes (ERGs) mechanisms could provide a better understanding of their origin, evolution, and dissemination in these pristine environments. Here, we describe the diversity of ARGs and ERGs and the importance of mobile genetic elements as a possible mechanism for the dissemination of resistance genes in Antarctica. We analyzed five soil metagenomes from Deception Island in Antarctica. Results showed that detected ARGs are associated with mechanisms such as antibiotic efflux, antibiotic inactivation, and target alteration. On the other hand, resistance to metals, surfactants, and aromatic hydrocarbons were the dominant ERGs. The taxonomy of ARGs showed that Pseudomonas, Psychrobacter, and Staphylococcus could be key taxa for studying antibiotic resistance and environmental resistance to stress in Deception Island. In addition, results showed that ARGs are mainly associated with phage-type mobile elements suggesting a potential role in their dissemination and prevalence. Finally, these results provide valuable information regarding the ARGs and ERGs in Deception Island including the potential contribution of mobile genetic elements to the spread of ARGs and ERGs in one of the least studied Antarctic ecosystems to date.

Antimicrobial Resistance Monitoring of Water Environments: A Framework for Standardized Methods and Quality Control

Antimicrobial resistance (AMR) is a grand societal challenge with important dimensions in the water environment that contribute to its evolution and spread. Environmental monitoring could provide vital information for mitigating the spread of AMR; this includes assessing antibiotic resistance genes (ARGs) circulating among human populations, identifying key hotspots for evolution and dissemination of resistance, informing epidemiological and human health risk assessment models, and quantifying removal efficiencies by domestic wastewater infrastructure. However, standardized methods for monitoring AMR in the water environment will be vital to producing the comparable data sets needed to address such questions. Here we sought to establish scientific consensus on a framework for such standardization, evaluating the state of the science and practice of AMR monitoring of wastewater, recycled water, and surface water, through a literature review, survey, and workshop leveraging the expertise of academic, governmental, consulting, and water utility professionals.

Antimicrobial Resistance Monitoring of Water Environments: A Framework for Standardized Methods and Quality Control

Antimicrobial resistance (AMR) is a grand societal challenge with important dimensions in the water environment that contribute to its evolution and spread. Environmental monitoring could provide vital information for mitigating the spread of AMR; this includes assessing antibiotic resistance genes (ARGs) circulating among human populations, identifying key hotspots for evolution and dissemination of resistance, informing epidemiological and human health risk assessment models, and quantifying removal efficiencies by domestic wastewater infrastructure. However, standardized methods for monitoring AMR in the water environment will be vital to producing the comparable data sets needed to address such questions. Here we sought to establish scientific consensus on a framework for such standardization, evaluating the state of the science and practice of AMR monitoring of wastewater, recycled water, and surface water, through a literature review, survey, and workshop leveraging the expertise of academic, governmental, consulting, and water utility professionals.

PCycDB: a comprehensive and accurate database for fast analysis of phosphorus cycling genes

BACKGROUND

Phosphorus (P) is one of the most essential macronutrients on the planet, and microorganisms (including bacteria and archaea) play a key role in P cycling in all living things and ecosystems. However, our comprehensive understanding of key P cycling genes (PCGs) and microorganisms (PCMs) as well as their ecological functions remains elusive even with the rapid advancement of metagenome sequencing technologies. One of major challenges is a lack of a comprehensive and accurately annotated P cycling functional gene database.

RESULTS

In this study, we constructed a well-curated P cycling database (PCycDB) covering 139 gene families and 10 P metabolic processes, including several previously ignored PCGs such as pafA encoding phosphate-insensitive phosphatase, ptxABCD (phosphite-related genes), and novel aepXVWPS genes for 2-aminoethylphosphonate transporters. We achieved an annotation accuracy, positive predictive value (PPV), sensitivity, specificity, and negative predictive value (NPV) of 99.8%, 96.1%, 99.9%, 99.8%, and 99.9%, respectively, for simulated gene datasets. Compared to other orthology databases, PCycDB is more accurate, more comprehensive, and faster to profile the PCGs. We used PCycDB to analyze P cycling microbial communities from representative natural and engineered environments and showed that PCycDB could apply to different environments.

CONCLUSIONS

We demonstrate that PCycDB is a powerful tool for advancing our understanding of microbially driven P cycling in the environment with high coverage, high accuracy, and rapid analysis of metagenome sequencing data. The PCycDB is available at https://github.com/ZengJiaxiong/Phosphorus-cycling-database . Video Abstract.

Emergence of nutriments as a nascent complementary therapy against antimicrobial resistance

With these growing and evolving years, antimicrobial resistance has become a great subject of interest. The idea of using natural productive ways can be an effective measure against antimicrobial resistance. The growing prevalence of antimicrobial resistance indicates that advanced natural approaches are a topic of concern for fighting the resistance. Many natural products including essential oils, flavonoids, alkaloids and botanicals have been demonstrated as effective bactericidal agents. In this review, we will discuss in detail about the relevance of such natural products to tackle the problem of antimicrobial resistance, antibiotic adjuvants that aim towards non-essential bacterial targets to reduce the prevalence of resistant bacterial infections, latest bioinformatics approach towards antibacterial drug discovery along with an understanding of biogenic nanoparticles in antimicrobial activity.

Microbial Tracking-2, a metagenomics analysis of bacteria and fungi onboard the International Space Station

BACKGROUND

The International Space Station (ISS) is a unique and complex built environment with the ISS surface microbiome originating from crew and cargo or from life support recirculation in an almost entirely closed system. The Microbial Tracking 1 (MT-1) project was the first ISS environmental surface study to report on the metagenome profiles without using whole-genome amplification. The study surveyed the microbial communities from eight surfaces over a 14-month period. The Microbial Tracking 2 (MT-2) project aimed to continue the work of MT-1, sampling an additional four flights from the same locations, over another 14 months.

METHODS

Eight surfaces across the ISS were sampled with sterile wipes and processed upon return to Earth. DNA extracted from the processed samples (and controls) were treated with propidium monoazide (PMA) to detect intact/viable cells or left untreated and to detect the total DNA population (free DNA/compromised cells/intact cells/viable cells). DNA extracted from PMA-treated and untreated samples were analyzed using shotgun metagenomics. Samples were cultured for bacteria and fungi to supplement the above results.

RESULTS

Staphylococcus sp. and Malassezia sp. were the most represented bacterial and fungal species, respectively, on the ISS. Overall, the ISS surface microbiome was dominated by organisms associated with the human skin. Multi-dimensional scaling and differential abundance analysis showed significant temporal changes in the microbial population but no spatial differences. The ISS antimicrobial resistance gene profiles were however more stable over time, with no differences over the 5-year span of the MT-1 and MT-2 studies. Twenty-nine antimicrobial resistance genes were detected across all samples, with macrolide/lincosamide/streptogramin resistance being the most widespread. Metagenomic assembled genomes were reconstructed from the dataset, resulting in 82 MAGs. Functional assessment of the collective MAGs showed a propensity for amino acid utilization over carbohydrate metabolism. Co-occurrence analyses showed strong associations between bacterial and fungal genera. Culture analysis showed the microbial load to be on average 3.0 × 10⁵ cfu/m² CONCLUSIONS: Utilizing various metagenomics analyses and culture methods, we provided a comprehensive analysis of the ISS surface microbiome, showing microbial burden, bacterial and fungal species prevalence, changes in the microbiome, and resistome over time and space, as well as the functional capabilities and microbial interactions of this unique built microbiome. Data from this study may help to inform policies for future space missions to ensure an ISS surface microbiome that promotes astronaut health and spacecraft integrity. Video Abstract.

Plastisphere community assemblage of aquatic environment: plastic-microbe interaction, role in degradation and characterization technologies

It is undeniable that plastics are ubiquitous and a threat to global ecosystems. Plastic waste is transformed into microplastics (MPs) through physical and chemical disruption processes within the aquatic environment. MPs are detected in almost every environment due to their worldwide transportability through ocean currents or wind, which allows them to reach even the most remote regions of our planet. MPs colonized by biofilm-forming microbial communities are known as the ''plastisphere". The revelation that this unique substrate can aid microbial dispersal has piqued interest in the ground of microbial ecology. MPs have synergetic effects on the development, transportation, persistence, and ecology of microorganisms. This review summarizes the studies of plastisphere in recent years and the microbial community assemblage (viz. autotrophs, heterotrophs, predators, and pathogens). We also discussed plastic-microbe interactions and the potential sources of plastic degrading microorganisms. Finally, it also focuses on current technologies used to characterize those microbial inhabitants and recommendations for further research.

SeqScreen: accurate and sensitive functional screening of pathogenic sequences via ensemble learning

The COVID-19 pandemic has emphasized the importance of accurate detection of known and emerging pathogens. However, robust characterization of pathogenic sequences remains an open challenge. To address this need we developed SeqScreen, which accurately characterizes short nucleotide sequences using taxonomic and functional labels and a customized set of curated Functions of Sequences of Concern (FunSoCs) specific to microbial pathogenesis. We show our ensemble machine learning model can label protein-coding sequences with FunSoCs with high recall and precision. SeqScreen is a step towards a novel paradigm of functionally informed synthetic DNA screening and pathogen characterization, available for download at www.gitlab.com/treangenlab/seqscreen .

DeephageTP: a convolutional neural network framework for identifying phage-specific proteins from metagenomic sequencing data

Bacteriophages (phages) are the most abundant and diverse biological entity on Earth. Due to the lack of universal gene markers and database representatives, there about 50-90% of genes of phages are unable to assign functions. This makes it a challenge to identify phage genomes and annotate functions of phage genes efficiently by homology search on a large scale, especially for newly phages. Portal (portal protein), TerL (large terminase subunit protein), and TerS (small terminase subunit protein) are three specific proteins of Caudovirales phage. Here, we developed a CNN (convolutional neural network)-based framework, DeephageTP, to identify the three specific proteins from metagenomic data. The framework takes one-hot encoding data of original protein sequences as the input and automatically extracts predictive features in the process of modeling. To overcome the false positive problem, a cutoff-loss-value strategy is introduced based on the distributions of the loss values of protein sequences within the same category. The proposed model with a set of cutoff-loss-values demonstrates high performance in terms of Precision in identifying TerL and Portal sequences (94% and 90%, respectively) from the mimic metagenomic dataset. Finally, we tested the efficacy of the framework using three real metagenomic datasets, and the results shown that compared to the conventional alignment-based methods, our proposed framework had a particular advantage in identifying the novel phage-specific protein sequences of portal and TerL with remote homology to their counterparts in the training datasets. In summary, our study for the first time develops a CNN-based framework for identifying the phage-specific protein sequences with high complexity and low conservation, and this framework will help us find novel phages in metagenomic sequencing data. The DeephageTP is available at https://github.com/chuym726/DeephageTP.

Application of Artificial Intelligence in Combating High Antimicrobial Resistance Rates

Artificial intelligence (AI) is a branch of science and engineering that focuses on the computational understanding of intelligent behavior. Many human professions, including clinical diagnosis and prognosis, are greatly useful from AI. Antimicrobial resistance (AMR) is among the most critical challenges facing Pakistan and the rest of the world. The rising incidence of AMR has become a significant issue, and authorities must take measures to combat the overuse and incorrect use of antibiotics in order to combat rising resistance rates. The widespread use of antibiotics in clinical practice has not only resulted in drug resistance but has also increased the threat of super-resistant bacteria emergence. As AMR rises, clinicians find it more difficult to treat many bacterial infections in a timely manner, and therapy becomes prohibitively costly for patients. To combat the rise in AMR rates, it is critical to implement an institutional antibiotic stewardship program that monitors correct antibiotic use, controls antibiotics, and generates antibiograms. Furthermore, these types of tools may aid in the treatment of patients in the event of a medical emergency in which a physician is unable to wait for bacterial culture results. AI’s applications in healthcare might be unlimited, reducing the time it takes to discover new antimicrobial drugs, improving diagnostic and treatment accuracy, and lowering expenses at the same time. The majority of suggested AI solutions for AMR are meant to supplement rather than replace a doctor’s prescription or opinion, but rather to serve as a valuable tool for making their work easier. When it comes to infectious diseases, AI has the potential to be a game-changer in the battle against antibiotic resistance. Finally, when selecting antibiotic therapy for infections, data from local antibiotic stewardship programs are critical to ensuring that these bacteria are treated quickly and effectively. Furthermore, organizations such as the World Health Organization (WHO) have underlined the necessity of selecting the appropriate antibiotic and treating for the shortest time feasible to minimize the spread of resistant and invasive resistant bacterial strains.

Inhalable Antibiotic Resistome from Wastewater Treatment Plants to Urban Areas: Bacterial Hosts, Dissemination Risks, and Source Contributions

Antibiotic resistance genes (ARGs) are commonly detected in the atmosphere, but questions remain regarding their sources and relative contributions, bacterial hosts, and corresponding human health risks. Here, we conducted a qPCR- and metagenomics-based investigation of inhalable fine particulate matter (PM_2.5) at a large wastewater treatment plant (WWTP) and in the ambient air of Hong Kong, together with an in-depth analysis of published data of other potential sources in the area. PM_2.5 was observed with increasing enrichment of total ARGs along the coastal-urban-WWTP gradient and clinically relevant ARGs commonly identified in urban and WWTP sites, illustrating anthropogenic impacts on the atmospheric accumulation of ARGs. With certain kinds of putative antibiotic-resistant pathogens detected in urban and WWTP PM_2.5, a comparable proportion of ARGs that co-occurred with MGEs was found between the atmosphere and WWTP matrices. Despite similar emission rates of bacteria and ARGs within each WWTP matrix, about 11-13% of the bacteria and >57% of the relevant ARGs in urban and WWTP PM_2.5 were attributable to WWTPs. Our study highlights the importance of WWTPs in disseminating bacteria and ARGs to the ambient air from a quantitative perspective and, thus, the need to control potential sources of inhalation exposure to protect the health of urban populations.

A Resistome Roadmap: From the Human Body to Pristine Environments

A comprehensive characterization of the human body resistome [sets of antibiotic resistance genes (ARGs)] is yet to be done and paramount for addressing the antibiotic microbial resistance threat. Here, we study the resistome of 771 samples from five major body parts (skin, nares, vagina, gut, and oral cavity) of healthy subjects from the Human Microbiome Project (HMP) and addressed the potential dispersion of ARGs in pristine environments. A total of 28,714 ARGs belonging to 235 different ARG types were found in the HMP proteome dataset (n = 9.1 × 10⁷ proteins analyzed). Our study reveals a distinct resistome profile (ARG type and abundance) between body sites and high interindividual variability. Nares had the highest ARG load (≈5.4 genes/genome) followed by the oral cavity, whereas the gut showed one of the highest ARG richness (shared with nares) but the lowest abundance (≈1.3 genes/genome). The fluroquinolone resistance genes were the most abundant in the human body, followed by macrolide–lincosamide–streptogramin (MLS) or tetracycline. Most ARGs belonged to common bacterial commensals and multidrug resistance trait were predominant in the nares and vagina. Many ARGs detected here were considered as low risk for human health, whereas only a few of them, such as BlaZ, dfrA14, dfrA17, or tetM, were classified as high-risk ARG. Our data also provide hope, since the spread of common ARG from the human body to pristine environments (n = 271 samples; 77 Gb of sequencing data and 2.1 × 10⁸ proteins analyzed) thus far remains very unlikely (only one case found in an autochthonous bacterium from a pristine environment). These findings broaden our understanding of ARG in the context of the human microbiome and the One-Health Initiative of WHO uniting human host–microbes and environments as a whole.

A neural network-based framework to understand the type 2 diabetes-related alteration of the human gut microbiome

The identification of microbial markers adequate to delineate the disease-related microbiome alterations from the complex human gut microbiota is of great interest. Here, we develop a framework combining neural network (NN) and random forest, resulting in 40 marker species and 90 marker genes identified from the metagenomic data set (185 healthy and 183 type 2 diabetes [T2D] samples), respectively. In terms of these markers, the NN model obtained higher accuracy in classifying the T2D-related samples than other methods; the interaction network analyses identified the key species and functional modules; the regression analysis determined that fasting blood glucose is the most significant factor (p < 0.05) in the T2D-related alteration of the human gut microbiome. We also observed that those marker species varied little across the case and control samples greatly shift in the different stages of the T2D development, suggestive of their important roles in the T2D-related microbiome alteration. Our study provides a new way of identifying the disease-related biomarkers and analyzing the role they may play in the development of the disease.

ROCker Models for Reliable Detection and Typing of Short-Read Sequences Carrying β-Lactamase Genes

Identification of genes encoding β-lactamases (BLs) from short-read sequences remains challenging due to the high frequency of shared amino acid functional domains and motifs in proteins encoded by BL genes and related non-BL gene sequences. Divergent BL homologs can be frequently missed during similarity searches, which has important practical consequences for monitoring antibiotic resistance. To address this limitation, we built ROCker models that targeted broad classes (e.g., class A, B, C, and D) and individual families (e.g., TEM) of BLs and challenged them with mock 150-bp- and 250-bp-read data sets of known composition. ROCker identifies most-discriminant bit score thresholds in sliding windows along the sequence of the target protein sequence and hence can account for nondiscriminative domains shared by unrelated proteins. BL ROCker models showed a 0% false-positive rate (FPR), a 0% to 4% false-negative rate (FNR), and an up-to-50-fold-higher F1 score [2 × precision × recall/(precision + recall)] compared to alternative methods, such as similarity searches using BLASTx with various e-value thresholds and BL hidden Markov models, or tools like DeepARG, ShortBRED, and AMRFinder. The ROCker models and the underlying protein sequence reference data sets and phylogenetic trees for read placement are freely available through http://enve-omics.ce.gatech.edu/data/rocker-bla. Application of these BL ROCker models to metagenomics, metatranscriptomics, and high-throughput PCR gene amplicon data should facilitate the reliable detection and quantification of BL variants encoded by environmental or clinical isolates and microbiomes and more accurate assessment of the associated public health risk, compared to the current practice.

IMPORTANCE Resistance genes encoding β-lactamases (BLs) confer resistance to the widely prescribed antibiotic class β-lactams. Therefore, it is important to assess the prevalence of BL genes in clinical or environmental samples for monitoring the spreading of these genes into pathogens and estimating public health risk. However, detecting BLs in short-read sequence data is technically challenging. Our ROCker model-based bioinformatics approach showcases the reliable detection and typing of BLs in complex data sets and thus contributes toward solving an important problem in antibiotic resistance surveillance. The ROCker models developed substantially expand the toolbox for monitoring antibiotic resistance in clinical or environmental settings.

Sustainability of the rice-crayfish co-culture aquaculture model: microbiome profiles based on multi-kingdom analyses

While the rice-crayfish culture (RCFP) model, an important aquaculture model in Asia, is generally considered a sustainable model, its sustainability in terms of microbial community profiles has not been evaluated. In this study, multi-kingdom analyses of microbiome profiles (i.e., bacteria, archaea, viruses, and eukaryotes) were performed using environmental (i.e., water and sediment) and animal gut (i.e., crayfish and crab gut) microbial samples from the RCFP and other aquaculture models, including the crab-crayfish co-culture, crayfish culture, and crab culture models, to evaluate the sustainability of the RCFP systematically. Results showed that RCFP samples are enriched with a distinct set of microbes, including Shewanella, Ferroplasma, Leishmania, and Siphoviridae, when compared with other aquaculture models. Additionally, most microbes in the RCFP samples, especially microbes from different kingdoms, were densely and positively connected, which indicates their robustness against environmental stress. Whereas microbes in different aquaculture models demonstrated moderate levels of horizontal gene transfer (HGT) across kingdoms, the RCFP showed relatively lower frequencies of HGT events, especially those involving antibiotic resistance genes. Finally, environmental factors, including pH, oxidation-reduction potential, temperature, and total nitrogen, contributed profoundly to shaping the microbial communities in these aquaculture models. Interestingly, compared with other models, the microbial communities of the RCFP model were less influenced by these environmental factors, which suggests that microbes in the latter have stronger ability to resist environmental stress. The findings collectively reflect the unique multi-kingdom microbial patterns of the RCFP model and suggest that this model is a sustainable model from the perspective of microbiome profiles.

AMR-meta: a k-mer and metafeature approach to classify antimicrobial resistance from high-throughput short-read metagenomics data

BACKGROUND

Antimicrobial resistance (AMR) is a global health concern. High-throughput metagenomic sequencing of microbial samples enables profiling of AMR genes through comparison with curated AMR databases. However, the performance of current methods is often hampered by database incompleteness and the presence of homology/homoplasy with other non-AMR genes in sequenced samples.

RESULTS

We present AMR-meta, a database-free and alignment-free approach, based on k-mers, which combines algebraic matrix factorization into metafeatures with regularized regression. Metafeatures capture multi-level gene diversity across the main antibiotic classes. AMR-meta takes in reads from metagenomic shotgun sequencing and outputs predictions about whether those reads contribute to resistance against specific classes of antibiotics. In addition, AMR-meta uses an augmented training strategy that joins an AMR gene database with non-AMR genes (used as negative examples). We compare AMR-meta with AMRPlusPlus, DeepARG, and Meta-MARC, further testing their ensemble via a voting system. In cross-validation, AMR-meta has a median f-score of 0.7 (interquartile range, 0.2-0.9). On semi-synthetic metagenomic data-external test-on average AMR-meta yields a 1.3-fold hit rate increase over existing methods. In terms of run-time, AMR-meta is 3 times faster than DeepARG, 30 times faster than Meta-MARC, and as fast as AMRPlusPlus. Finally, we note that differences in AMR ontologies and observed variance of all tools in classification outputs call for further development on standardization of benchmarking data and protocols.

CONCLUSIONS

AMR-meta is a fast, accurate classifier that exploits non-AMR negative sets to improve sensitivity and specificity. The differences in AMR ontologies and the high variance of all tools in classification outputs call for the deployment of standard benchmarking data and protocols, to fairly compare AMR prediction tools.

Linkage and driving mechanisms of antibiotic resistome in surface and ground water: Their responses to land use and seasonal variation

Antibiotic resistance in the environment, mostly mediated by antibiotic resistance genes (ARGs), has posed a threat to ecological and human health. Contamination of surface water and groundwater with ARGs has become a serious environmental concern. However, the distinctions and similarities across ARG profiles, the various ecological processes associated with ARGs, the driving mechanisms for ARG profiles in surface water and groundwater, and how they respond to land use and seasonal variation remain unknown. To tackle these issues, the contamination of ARGs in surface water and groundwater in central China was investigated using metagenomic technology. The results indicated that seasonal changes in ARG abundance and diversity were inconsistent across surface water and groundwater, and that the relationship between ARGs in surface water and groundwater was stronger during the rainy season. Land use had a greater effect on ARGs in surface water than in groundwater and was stronger during the dry season than during the rainy season. More interestingly, the ideal buffer zones with the greatest impact of land use on the ARGs of surface water and groundwater had distinct radii: 1500 m for both dry and rainy seasons in surface water, and 1000 m for dry season and 500 m for rainy season in groundwater. Furthermore, stochastic mechanisms mediated by mobile gene elements (MGEs) contribute significantly more to ARG assemblages than deterministic processes, particularly in groundwater. Furthermore, our results also showed that ARG enrichment in microbial communities was host- dependent, and the risk of ARGs in groundwater was greater both during the rainy season and dry season. In conclusion, the findings have improved our understanding of the relationship and driving mechanisms of ARGs in surface and ground water, as well as their responses to land use and seasonal variation, which may be beneficial for limiting ARG pollution in a watershed with high levels of anthropogenic activity.

Metagenomic insights into surface water microbial communities of a South Asian mangrove ecosystem

Estuaries are one of the most productive ecosystems and their productivity is maintained by resident microbial communities. Recent alterations driven by climate change have further escalated these stressors leading to the propagation of traits such as antibiotic resistance and heavy metal resistance in microbial communities. Surface water samples from eleven stations along the Thakuran and Matla estuaries of the Sundarbans Biosphere Reserve (SBR) of Sundarbans mangrove located in South Asia were sampled in monsoon (June) 2019 to elucidate resident microbial communities based on Nanopore sequencing. Metagenomic analyses revealed the widespread dominance of Proteobacteria across all the stations along with a high abundance of Firmicutes. Other phyla, including Euryarchaeota, Thaumarchaeota, Actinobacteria, Bacteroidetes and Cyanobacteria showed site-specific trends in abundance. Further taxonomic affiliations showed Gammaproteobacteria and Alphaproteobacteria to be dominant classes with high abundances of Bacilli in SBR_Stn58 and SBR_Stn113. Among the eukaryotic communities, the most abundant classes included Prasinophyceae, Saccharyomycetes and Sardariomycetes. Functional annotation showed metabolic activities such as carbohydrate, amino acid, nitrogen and phosphorus metabolisms to be uniformly distributed across all the studied stations. Pathways such as stress response, sulphur metabolism and motility-associated genes appeared in low abundances in SBR. Functional traits such as antibiotic resistance showed overwhelming dominance of genes involved in multidrug resistance along with widespread resistance towards commonly used antibiotics including Tetracycline, glycopeptide and aminoglycoside. Metal resistance genes including arsenic, nickel and copper were found in comparable abundances across the studied stations. The prevalence of ARG and MRG might indicate presence of pollutants and hint toward deteriorating ecosystem health status of Sundarbans mangrove.

Evaluation of Experimental Protocols for Shotgun Whole-Genome Metagenomic Discovery of Antibiotic Resistance Genes

Shotgun metagenomics has enabled the discovery of antibiotic resistance genes (ARGs). Although there have been numerous studies benchmarking the bioinformatics methods for shotgun metagenomic data analysis, there has not yet been a study that systematically evaluates the performance of different experimental protocols on metagenomic species profiling and ARG detection. In this study, we generated 35 whole genome shotgun metagenomic sequencing data sets for five samples (three human stool and two microbial standard) using seven experimental protocols (KAPA or Flex kits at 50ng, 10ng, or 5ng input amounts; XT kit at 1ng input amount). Using this comprehensive resource, we evaluated the seven protocols in terms of robust detection of ARGs and microbial abundance estimation at various sequencing depths. We found that the data generated by the seven protocols are largely similar. The inter-protocol variability is significantly smaller than the variability between samples or sequencing depths. We found that a sequencing depth of more than 30M is suitable for human stool samples. A higher input amount (50ng) is generally favorable for the KAPA and Flex kits. This systematic benchmarking study sheds light on the impact of sequencing depth, experimental protocol, and DNA input amount on ARG detection in human stool samples.

Knowledge integration and decision support for accelerated discovery of antibiotic resistance genes

We present a machine learning framework to automate knowledge discovery through knowledge graph construction, inconsistency resolution, and iterative link prediction. By incorporating knowledge from 10 publicly available sources, we construct an Escherichia coli antibiotic resistance knowledge graph with 651,758 triples from 23 triple types after resolving 236 sets of inconsistencies. Iteratively applying link prediction to this graph and wet-lab validation of the generated hypotheses reveal 15 antibiotic resistant E. coli genes, with 6 of them never associated with antibiotic resistance for any microbe. Iterative link prediction leads to a performance improvement and more findings. The probability of positive findings highly correlates with experimentally validated findings (R2 = 0.94). We also identify 5 homologs in Salmonella enterica that are all validated to confer resistance to antibiotics. This work demonstrates how evidence-driven decisions are a step toward automating knowledge discovery with high confidence and accelerated pace, thereby substituting traditional time-consuming and expensive methods.

Unveiling the occurrence, hosts and mobility potential of antibiotic resistance genes in the deep ocean

As emerging microbial contaminants, antibiotic resistance genes (ARGs) are widely reported in the neritic zone. However, the profiles of ARGs in the deep ocean have not yet been fully resolved. In this study, the distribution, hosts, and mobility potential of ARGs at different water depths in the Western Pacific (WP) were investigated and compared to those in Bohai Sea (BH) waters using environmental parameter measurements, amplicon sequencing, metagenomic assembly and binning approaches. Our results showed that the top eight most abundant known ARG types in WP and BH waters were multidrug (39.85%), peptide (14.98%), aminoglycoside (11.33%), macrolide-lincosamide-streptogramin (MLS, 4.06%), tetracycline (3.74%), beta-lactam (3.12%), fluoroquinolone (1.79%) and rifamycin (1.24%). The ARGs observed in mesopelagic and bathypelagic waters were abundant and diverse as those observed in neritic waters, indicating that deep-sea water could be another environmental reservoir for ARGs. For deep-sea ARGs, members from classes Gammaproteobacteria (70%) and Alphaproteobacteria (21.1%) were the most important potential hosts. In addition, mobile genetic element analysis suggested that the ARG migration potential in dee sea water (> 1000 m) was relatively high. Overall, our findings expanded the understanding of ARGs in deep seawater and provided guidance for ARG pollution control and risk prediction.

Metagenomics-enabled microbial surveillance

Lessons learnt from the COVID-19 pandemic include increased awareness of the potential for zoonoses and emerging infectious diseases that can adversely affect human health. Although emergent viruses are currently in the spotlight, we must not forget the ongoing toll of morbidity and mortality owing to antimicrobial resistance in bacterial pathogens and to vector-borne, foodborne and waterborne diseases. Population growth, planetary change, international travel and medical tourism all contribute to the increasing frequency of infectious disease outbreaks. Surveillance is therefore of crucial importance, but the diversity of microbial pathogens, coupled with resource-intensive methods, compromises our ability to scale-up such efforts. Innovative technologies that are both easy to use and able to simultaneously identify diverse microorganisms (viral, bacterial or fungal) with precision are necessary to enable informed public health decisions. Metagenomics-enabled surveillance methods offer the opportunity to improve detection of both known and yet-to-emerge pathogens.

Machine Learning and Deep Learning Applications in Metagenomic Taxonomy and Functional Annotation

Shotgun sequencing of environmental DNA (i.e., metagenomics) has revolutionized the field of environmental microbiology, allowing the characterization of all microorganisms in a sequencing experiment. To identify the microbes in terms of taxonomy and biological activity, the sequenced reads must necessarily be aligned on known microbial genomes/genes. However, current alignment methods are limited in terms of speed and can produce a significant number of false positives when detecting bacterial species or false negatives in specific cases (virus, plasmids, and gene detection). Moreover, recent advances in metagenomics have enabled the reconstruction of new genomes using de novo binning strategies, but these genomes, not yet fully characterized, are not used in classic approaches, whereas machine and deep learning methods can use them as models. In this article, we attempted to review the different methods and their efficiency to improve the annotation of metagenomic sequences. Deep learning models have reached the performance of the widely used k-mer alignment-based tools, with better accuracy in certain cases; however, they still must demonstrate their robustness across the variety of environmental samples and across the rapid expansion of accessible genomes in databases.

Metagenomic Analysis Reveals A Possible Association Between Respiratory Infection and Periodontitis

Periodontitis is an inflammatory disease that is characterized by progressive destruction of the periodontium and causes tooth loss in adults. Periodontitis is known to be associated with dysbiosis of the oral microflora, which is often linked to various diseases. However, the complexity of plaque microbial communities of periodontitis, antibiotic resistance, and enhanced virulence make this disease difficult to treat. In this study, using metagenomic shotgun sequencing, we investigated the etiology, antibiotic resistance genes (ARGs), and virulence genes (VirGs) of periodontitis. We revealed a significant shift in the composition of oral microbiota as well as several functional pathways that were represented significantly more abundantly in periodontitis patients than in controls. In addition, we observed several positively selected ARGs and VirGs with the Ka/Ks ratio > 1 by analyzing our data and a previous periodontitis dataset, indicating that ARGs and VirGs in oral microbiota may be subjected to positive selection. Moreover, 5 of 12 positively selected ARGs and VirGs in periodontitis patients were found in the genomes of respiratory tract pathogens. Of note, 91.8% of the background VirGs with at least one non-synonymous single-nucleotide polymorphism for natural selection were also from respiratory tract pathogens. These observations suggest a potential association between periodontitis and respiratory infection at the gene level. Our study enriches the knowledge of pathogens and functional pathways as well as the positive selection of antibiotic resistance and pathogen virulence in periodontitis patients, and provides evidence at the gene level for an association between periodontitis and respiratory infection.

Current progress and open challenges for applying deep learning across the biosciences

Deep Learning (DL) has recently enabled unprecedented advances in one of the grand challenges in computational biology: the half-century-old problem of protein structure prediction. In this paper we discuss recent advances, limitations, and future perspectives of DL on five broad areas: protein structure prediction, protein function prediction, genome engineering, systems biology and data integration, and phylogenetic inference. We discuss each application area and cover the main bottlenecks of DL approaches, such as training data, problem scope, and the ability to leverage existing DL architectures in new contexts. To conclude, we provide a summary of the subject-specific and general challenges for DL across the biosciences.

Metagenomics indicate that public health risk may be higher from flooding following dry versus rainy periods

Urban floodwater could lead to significant risk for public and environmental health from mobilization of microbial pathogens and overflow of wastewater treatment systems. Here, we attempted to assess this risk by obtaining metagenomic profiles of antibiotic resistance genes (ARGs), virulence factors (VFs) and pathogens present in floodwater samples collected in urban Atlanta, GA that were categorized in two distinct groups: floods that occurred after periods of drought and those after regular (seasonal) rain events. Even though no major (known) pathogens were present at the limit of detection of our sequencing effort (~3 Gbp/sample), we observed that floodwaters after drought showed a 2.5-fold higher abundance of both ARGs and VFs compared to floodwater after rainy days. These differences were mainly derived by several novel species of the Pseudomonas genus, which were more dominant in the former versus the latter samples and carried several genes to cope with osmotic stress in addition to ARGs and VFs. These results revealed that there are previously undescribed species that become mobilized after flooding events in the Southeast US urban settings and could represent an increased public health risk, especially after periods of drought, which warrants further attention.

Bacterial diversity and the antimicrobial resistome in the southwestern highlands of Saudi Arabia

Soil is a reservoir of microbial diversity and the most supportive habitat for acquiring and transmitting antimicrobial resistance. Resistance transfer usually occurs from animal to soil and vice versa, and it may ultimately appear in clinical pathogens. In this study, the southwestern highlands of Saudi Arabia were studied to assess the bacterial diversity and antimicrobial resistance that could be affected by the continuous development of tourism in the region. Such effects could have a long-lasting impact on the local environment and community. Culture-dependent, quantitative polymerase chain reaction (qPCR), and shotgun sequencing-based metagenomic approaches were used to evaluate the diversity, functional capabilities, and antimicrobial resistance of bacteria isolated from collected soil samples. Bacterial communities in the southwestern highlands were mainly composed of Proteobacteria, Bacteroidetes, and Actinobacteria. A total of 102 antimicrobial resistance genes (ARGs) and variants were identified in the soil microbiota and were mainly associated with multidrug resistance, followed by macrolide, tetracycline, glycopeptide, bacitracin, and beta-lactam antibiotic resistance. The mechanisms of resistance included efflux, antibiotic target alteration, and antibiotic inactivation. qPCR confirmed the detection of 18 clinically important ARGs. In addition, half of the 49 identified isolates were phenotypically resistant to at least one of the 15 antibiotics tested. Overall, ARGs and indicator genes of anthropogenic activities (human-mitochondrial [hmt] gene and integron-integrase [int1]) were found in relatively lower abundance. Along with a high diversity of bacterial communities, variation was observed in the relative abundance of bacterial taxa among sampling sites in the southwestern highlands of Saudi Arabia.

A multi-label learning framework for predicting antibiotic resistance genes via dual-view modeling

The increasing prevalence of antibiotic resistance has become a global health crisis. For the purpose of safety regulation, it is of high importance to identify antibiotic resistance genes (ARGs) in bacteria. Although culture-based methods can identify ARGs relatively more accurately, the identifying process is time-consuming and specialized knowledge is required. With the rapid development of whole genome sequencing technology, researchers attempt to identify ARGs by computing sequence similarity from public databases. However, these computational methods might fail to detect ARGs due to the low sequence identity to known ARGs. Moreover, existing methods cannot effectively address the issue of multidrug resistance prediction for ARGs, which is a great challenge to clinical treatments. To address the challenges, we propose an end-to-end multi-label learning framework for predicting ARGs. More specifically, the task of ARGs prediction is modeled as a problem of multi-label learning, and a deep neural network-based end-to-end framework is proposed, in which a specific loss function is introduced to employ the advantage of multi-label learning for ARGs prediction. In addition, a dual-view modeling mechanism is employed to make full use of the semantic associations among two views of ARGs, i.e. sequence-based information and structure-based information. Extensive experiments are conducted on publicly available data, and experimental results demonstrate the effectiveness of the proposed framework on the task of ARGs prediction.

Towards routine employment of computational tools for antimicrobial resistance determination via high-throughput sequencing

Antimicrobial resistance (AMR) is a growing threat to public health and farming at large. In clinical and veterinary practice, timely characterization of the antibiotic susceptibility profile of bacterial infections is a crucial step in optimizing treatment. High-throughput sequencing is a promising option for clinical point-of-care and ecological surveillance, opening the opportunity to develop genotyping-based AMR determination as a possibly faster alternative to phenotypic testing. In the present work, we compare the performance of state-of-the-art methods for detection of AMR using high-throughput sequencing data from clinical settings. We consider five computational approaches based on alignment (AMRPlusPlus), deep learning (DeepARG), k-mer genomic signatures (KARGA, ResFinder) or hidden Markov models (Meta-MARC). We use an extensive collection of 585 isolates with available AMR resistance profiles determined by phenotypic tests across nine antibiotic classes. We show how the prediction landscape of AMR classifiers is highly heterogeneous, with balanced accuracy varying from 0.40 to 0.92. Although some algorithms-ResFinder, KARGA and AMRPlusPlus-exhibit overall better balanced accuracy than others, the high per-AMR-class variance and related findings suggest that: (1) all algorithms might be subject to sampling bias both in data repositories used for training and experimental/clinical settings; and (2) a portion of clinical samples might contain uncharacterized AMR genes that the algorithms-mostly trained on known AMR genes-fail to generalize upon. These results lead us to formulate practical advice for software configuration and application, and give suggestions for future study designs to further develop AMR prediction tools from proof-of-concept to bedside.