Biology subjects

Pruden, A.

Publications and source records attributed to Pruden, A..

18 recordsLinked to original sources

bioRxiv2026

Genomic diversification underlies the broad ecological range of Salmonella enterica serotype Typhimurium

Salmonella Typhimurium is a versatile foodborne pathogen with a broad ecological range, making it an ideal model to better understand pathogen adaptations that allow them to infect multiple hosts and persist across environments. Here, we analyzed 595 genomes of S. Typhimurium representing three food animal sources (bovine, swine, and poultry) and one non-food animal source (wild birds). We found that S. Typhimurium from food animal sources generally had a more open pangenome and harbored more antimicrobial resistance genes (ARGs). Notably, swine isolates exhibited the most open pangenome and the highest prevalence of ARGs, patterns that were associated with a greater presence of mobile genetic elements, particularly plasmids. Despite similar core genome sizes, S. Typhimurium from different sources displayed distinct patterns of positive selection in the core genome that varied in frequency and targeted functional categories. In contrast, although accessory genome sizes varied substantially across sources, the frequency of positive selection remained similar. Using machine learning, we identified source-predictive genetic variants, many of which are associated with stress-response functions. These findings suggest that gain and loss of accessory genes and positive selection acting on core genes support differential adaptation in S. Typhimurium, potentially contributing to its broad ecological range.

genomics↗

bioRxiv2025

Assessment of a Passive Sampler for Aquifer Microbial Community Profiling and Comparison of Porous Media Applied in a Managed Recharge Scenario

Microbial communities play an important role in aquifer systems, as both hazards (e.g., pathogens, carriers of antibiotic resistance genes) and contributors to contaminant remediation. However, monitoring native microbial communities in situ remains challenging. To address this challenge, we developed a passive sampler with that employs removable cartridges containing solid media designed to re-create the aquifer environment in a controlled fashion and support consistent, repeatable, time-series sampling. With a bench-scale, continuous-loop setup circulating advanced-treated wastewater, we compared microbial community dynamics across three candidate porous media: native aquifer sediment, zirconia beads, and laboratory-grade silica sand. 16S rRNA gene amplicon sequencing revealed that native aquifer sediment best reflected influent microbial composition and temporal shifts. Native aquifer sediment also exhibited more spatial consistency in microbial diversity along the sampler. These findings suggest that native aquifer sediment is an optimal porous medium for long-term, in situ monitoring of microbial communities for managed aquifer recharge and other applications in groundwater systems.

ecology↗

bioRxiv2025

ARGContextProfiler: Extracting and Scoring the Genomic Contexts of Antibiotic Resistance Genes using Assembly Graphs

Antibiotic resistance (AR) presents a global health challenge, necessitating an improved understanding of the ecology, evolution, and dissemination of antibiotic resistance genes (ARGs). Several tools, databases, and algorithms are now available to facilitate the identification of ARGs in metagenomic sequencing data; however, direct annotation of short-read data provides limited contextual information. Knowledge of whether an ARG is carried in the chromosome or on a specific mobile genetic element (MGE) is critical to understanding mobility, persistence, and potential for co-selection. Here we developed ARGContextProfiler, a pipeline designed to extract and visualize ARG genomic contexts. By leveraging the assembly graph for genomic neighborhood extraction and validating contexts through read mapping, ARGContextProfiler minimizes chimeric errors that are a common artifact of assembly outputs. Testing on real, synthetic, and semi-synthetic data, including long-read sequencing data from environmental samples, demonstrated that ARGContextProfiler offers superior accuracy, precision, and sensitivity compared to conventional assembly-based methods. ARGContextProfiler thus provides a powerful tool for uncovering the genomic context of ARGs in metagenomic sequencing data, which can be of value to both fundamental and applied research aimed at understanding and stemming the spread of AR. The source code of ARGContextProfiler is publicly available at GitHub.

bioinformatics↗

bioRxiv2025

Influence of Copper Dose on Mycobacterium avium and Legionella pneumophila Growth in Premise Plumbing

Effects of copper at 0, 4, 30, 250, or 2000 {micro}g/L on microbial communities were examined in an 11-month study using triplicate 120-mL water heater microcosms with PEX-b pipes containing mature biofilms to simulate premise plumbing. Effluent total cell counts (TCCs) and Mycobacterium avium peaked at 250 {micro}g/L, reflecting the dual role of copper as a nutrient and antimicrobial. TCCs and M. avium were relatively consistent among replicate microcosms at each dose, but Legionella pneumophila (Lp) diverged among biological triplicates at 250 {micro}g/L, consistently producing high culturable Lp (average 2.5 log MPN/mL) in one microcosm and low/non-detectable levels in the other two. Repeated cross-inoculations and a reinoculation failed to normalize the microbial community composition across 250 {micro}g/L and other triplicate microcosms. 16S rRNA gene amplicon sequencing revealed that the 250 {micro}g/L replicate with high Lp was characterized by a distinct microbial community composition relative to the two replicates. At 2,000 {micro}g/L copper, microbial diversity and TCCs initially decreased, but then TCCs subsequently increased and ultimately were not statistically different from the 250 {micro}g/L microcosms. This study provides insight into mechanisms underlying non-linear effects of copper dosing when applied as a disinfectant to premise plumbing for opportunistic pathogen control. SYNOPSISNon-linear effects of copper on Legionella pneumophila and microbial communities are identified, providing deeper understanding to the efficacy of its use as a premise plumbing disinfectant.

microbiology↗

bioRxiv2025

ARGfore: A multivariate framework for forecasting antibiotic resistance gene abundances using time-series metagenomic datasets

BackgroundThe global spread of antibiotic resistance presents a significant threat to human, animal, and plant health. Metagenomic sequencing is increasingly being utilized to profile antibiotic resistance genes (ARGs) in various environments, but presently a mechanism for predicting future trends in ARG occurrence patterns is lacking. Capability of forecasting ARG abundance trends could be extremely valuable towards informing policy and practice aimed at mitigating the evolution and spread of ARGs. ResultsHere we propose ARGfore, a multivariate forecasting model for predicting ARG abundances from time-series metagenomic data. ARGfore extracts features that capture inherent relationships among ARGs and is trained to recognize patterns in ARG trends and seasonality. ConclusionARGfore outperformed standard time-series forecasting methods, such as N-HiTS, LSTM, and ARIMA, exhibiting the lowest mean absolute percentage error when applied to different wastewater datasets. Additionally, ARGfore demonstrated enhanced computational efficiency, making it a promising candidate for a variety of ARG surveillance applications. The rapid prediction of future trends can facilitate early detection and deployment of mitigation efforts if necessary. ARGfore is publicly available at https://github.com/joungmin-choi/ARGfore.

bioinformatics↗

bioRxiv2024

CIWARS: a web server for waterborne antibiotic resistance surveillance using longitudinal metagenomic data

The rise of antibiotic resistance (AR) is a major global health crisis, exacerbated by the overuse and misuse of antibiotics, leading to the rapid spread of antibiotic resistance genes (ARGs) in bacterial pathogens. This phenomenon poses significant threats to human and animal health, food security, and economic stability. Water bodies, particularly wastewater treatment plants (WWTPs), serve as critical reservoirs for ARGs, creating environments that favor the proliferation of resistant bacteria. Wastewater-based surveillance (WBS) has emerged as a cost-effective strategy for monitoring AR at the population level, providing real-time data to guide public health and policy decisions. Despite advancements in WBS, there are no comprehensive online analytical platforms for continuous environmental AR surveillance. This paper introduces CIWARS, a web server designed for AR analyses of longitudinal metagenomic data. CIWARS offers comprehensive ARG profiling, taxonomic annotation, and anomalous AR risk points detection. We demonstrate its capabilities through an interactive temporal data visualization, showcasing its potential for enhancing AR risk monitoring and guiding effective mitigation strategies. CIWARS is broadly applicable to longitudinal metagenomic data generated from any environment and aims to support global efforts in addressing the AR crisis by providing cyberinfrastructure for continuous AR surveillance. The web server is freely available at https://ciwars.cs.vt.edu/.

bioinformatics↗

bioRxiv2024

Evidence of horizontal gene transfer and environmental selection impacting antibiotic resistance evolution in soil-dwelling Listeria

Soil has been identified as an important reservoir of antibiotic resistance genes (ARGs) and there is a need to understand how corresponding environmental changes influence their emergence, evolution, and spread. As a soil-dwelling bacterial genus containing important pathogens, Listeria, including L. monocytogenes, the causative agent of listeriosis in humans, could serve as a key model for establishing this understanding. Notably, acquired antibiotic resistance among L. monocytogenes isolated from foods and the environment has been observed in some regions over the past decade. Here we characterized ARGs using 594 genomes representing 19 Listeria species that we previously isolated from soils across the United States. Among the five putatively functional ARGs identified, lin, which confers resistance to lincomycin, was the most prevalent, followed by mprF, sul, fosX, and norB. ARGs were found to be predominant in Listeria sensu stricto species and species more closely related to L. monocytogenes tended to harbor more ARGs. Notably, lin, fosX, and norB showed evidence of recent horizontal gene transfer (HGT) across species, likely through transformation as opposed to conjugation and transduction, while mprF and sul appear to have undergone positive selection. In addition, soil properties and surrounding land use were identified as the most important factors associated with ARG richness and genetic divergence, respectively. Using machine learning, we demonstrated that the presence of ARGs can be predicted from environmental variables with good accuracy (mean auROC of 0.76). Collectively, our data suggest that recent HGT and environmental selection played a vital role in the acquisition and diversification of ARGs in the soil environment.

microbiology↗

bioRxiv2024

Phage-plasmid hybrids are found throughout diverse environments and encode niche-specific functional traits

Phage-plasmids are unique mobile genetic elements that function as plasmids and temperate phages. While it has been observed that such elements often encode antibiotic resistance genes and defense system genes, little else is known about other functional traits they encode. Further, no study to date has documented their environmental distribution and prevalence. Here, we performed genome sequence mining of public databases of phages and plasmids utilizing a random forest classifier to identify phage-plasmids. We recovered 5,742 unique phage-plasmid genomes from a remarkable array of disparate environments, including human, animal, plant, fungi, soil, sediment, freshwater, wastewater, and saltwater environments. The resulting genomes were used in a comparative sequence analysis, revealing functional traits/accessory genes associated with specific environments. Host-associated elements contained the most defense systems (including CRISPR and anti-CRISPR systems) as well as antibiotic resistance genes, while other environments, such as freshwater and saltwater systems, tended to encode components of various biosynthetic pathways. Interestingly, we identified genes encoding for certain functional traits, including anti-CRISPR systems and specific antibiotic resistance genes, that were enriched in phage-plasmids relative to both plasmids and phages. Our results highlight that phage-plasmids are found across a wide-array of environments and likely play a role in shaping microbial ecology in a multitude of niches. IMPORTANCEPhage-plasmids are a novel, hybrid class of mobile genetic element which retain aspects of both phages and plasmids. However, whether phage-plasmids represent merely a rarity or are instead important players in horizontal gene transfer and other important ecological processes has remained a mystery. Here, we document that these hybrids are encountered across a broad range of distinct environments and encode niche-specific functional traits, including the carriage of antibiotic biosynthesis genes and both CRISPR and anti-CRISPR defense systems. These findings highlight phage-plasmids as an important class of mobile genetic element with diverse roles in multiple distinct ecological niches.

genomics↗

bioRxiv2024

ProtAlign-ARG: Antibiotic Resistance Gene Characterization Integrating Protein Language Models and Alignment-Based Scoring

The evolution and spread of antibiotic resistance pose a global health challenge. Whole genome and metagenomic sequencing pose a promising approach to monitoring the spread, but typical alignment-based approaches for antibiotic resistance gene (ARG) detection are inherently limited in the ability to detect new variants. Large protein language models could present a powerful alternative but are limited by databases available for training. Here we introduce ProtAlign-ARG, a novel hybrid model combining a pre-trained protein language model and an alignment scoring-based model to expand the capacity for ARG detection from DNA sequencing data. ProtAlign-ARG learns from vast unannotated protein sequences, utilizing raw protein language model embeddings to improve the accuracy of ARG classification. In instances where the model lacks confidence, ProtAlign-ARG employs an alignment-based scoring method, incorporating bit scores and e-values to classify ARGs according to their corresponding classes of antibiotics. ProtAlign-ARG demonstrated remarkable accuracy in identifying and classifying ARGs, particularly excelling in recall compared to existing ARG identification and classification tools. We also extended ProtAlign-ARG to predict the functionality and mobility of ARGs, highlighting the models robustness in various predictive tasks. A comprehensive comparison of ProtAlign-ARG with both the alignment-based scoring model and the pre-trained protein language model demonstrated the superior performance of ProtAlign-ARG.

bioinformatics↗

bioRxiv2024

Resistance genes are distinct in protein-protein interaction networks according to drug class and gene mobility

With growing calls for increased surveillance of antibiotic resistance as an escalating global health threat, improved bioinformatic tools are needed for tracking antibiotic resistance genes (ARGs) across One Health domains. Most studies to date profile ARGs using sequence homology, but such approaches provide limited information about the broader context or function of the ARG in bacterial genomes. Here we introduce a new pipeline for identifying ARGs in genomic data that employs machine learning analysis of Protein-Protein Interaction Networks (PPINs) as a means to improve predictions of ARGs while also providing vital information about the context, such as gene mobility. A random forest model was trained to effectively differentiate between ARGs and nonARGs and was validated using the PPINs of ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter cloacae), which represent urgent threats to human health because they tend to be multi-antibiotic resistant. The pipeline exhibited robustness in discriminating ARGs from nonARGs, achieving an average area under the precision-recall curve of 88%. We further identified that the neighbors of ARGs, i.e., genes connected to ARGs by only one edge, were disproportionately associated with mobile genetic elements, which is consistent with the understanding that ARGs tend to be mobile compared to randomly sampled genes in the PPINs. This pipeline showcases the utility of PPINs in discerning distinctive characteristics of ARGs within a broader genomic context and in differentiating ARGs from nonARGs through network-based attributes and interaction patterns. The code for running the pipeline is publicly available at https://github.com/NazifaMoumi/PPI-ARG-ESKAPE

bioinformatics↗

bioRxiv2024

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

BackgroundWhile there is increasing recognition of numerous environmental contributions to the spread of antibiotic resistance, quantifying the relative contributions of various sources remains a fundamental challenge. Similarly, there is a need to differentiate acute human health risks corresponding to exposure to a given environment, versus broader ecological risk of evolution and spread of antibiotic resistance genes (ARGs) across microbial taxa. Recent studies have proposed various methods of harnessing the rich information housed by metagenomic data for achieving such aims. Here, we introduce MetaCompare 2.0, which improves upon the original MetaCompare pipeline by differentiating indicators of human health resistome risk (i.e., potential for human pathogens to acquire ARGs) from ecological resistome risk (i.e., overall mobility of ARGs across a given microbiome). ResultsTo demonstrate the sensitivity of the MetaCompare 2.0 pipeline, we analyzed publicly available metagenomes representing a broad array of environments, including wastewater, surface water, soil, sediment, and human gut. We also assessed the effect of sequence assembly methods on the risk scores. We further evaluated the robustness of the pipeline to sequencing depth, contig count, and metagenomic library coverage bias through comparative analysis of a range of subsamples extracted from a set of deeply sequenced wastewater metagenomes. The analysis utilizing samples from different environments demonstrated that MetaCompare 2.0 consistently produces lower risk scores for environments with little human influence and higher risk scores for human contaminated environments affected by pollution or other stressors. We found that the ranks of risk scores were not measurably affected by different assemblers employed. The Meta-Compare 2.0 risk scores were remarkably consistent despite varying sequencing depth, contig count, and coverage. ConclusionMetaCompare 2.0 successfully ranked a wide array of environments according to both human health and ecological resistome risks, with both scores being strongly impacted by anthropogenic stress. We packaged the improved pipeline into a publicly-available web service that provides an easy-to-use interface for computing resistome risk scores and visualizing results. The web service is available at http://metacompare.cs.vt.edu/

bioinformatics↗

bioRxiv2023

Kairos infers in situ horizontal gene transfer in longitudinally sampled microbiomes through microdiversity-aware sequence analysis

Horizontal gene transfer (HGT) occurring within microbiomes is linked to complex environmental and ecological dynamics that are challenging to replicate in controlled settings. Consequently, most extant studies of microbiome HGT are either simplistic experimental settings with tenuous relevance to real microbiomes or correlative studies that assume that HGT potential is a function of the relative abundance of mobile genetic elements (MGEs), the vehicles of HGT. Here we introduce Kairos as a bioinformatic tool deployed in nextflow for detecting HGT events "in situ," i.e., within a microbiome, through analysis of time-series metagenomic sequencing data. The in-situ framework proposed here leverages available metagenomic data from a longitudinally sampled microbiome to assess whether the chronological occurrence of potential donors, recipients, and putatively transferred regions could plausibly have arisen due to HGT over a range of defined time periods. The centerpiece of the Kairos workflow is a novel competitive read alignment method that enables discernment of even very similar genomic sequences, such as those produced by MGE-associated recombination. A key advantage of Kairos is its reliance on assemblies rather than metagenome assembled genomes (MAGs), which avoids systematic exclusion of accessory genes associated with the binning process. In an example test-case of real world data, use of assemblies directly produced a 264-fold increase in the number of antibiotic resistance genes included in the analysis of HGT compared to analysis of MAGs with MetaCHIP. Further, in silico evaluation of contig taxonomy was performed to assess the accuracy of classification for both chromosomally- and MGE-derived sequences, indicating a high degree of accuracy even for conjugative plasmids up to the level of class or order. Thus, Kairos enables the analysis of very recent HGT events, making it suitable for studying rapid prokaryotic adaptation in environmental systems without disturbing the ornate ecological dynamics associated with microbiomes. Current versions of the Kairos workflow are available here: https://github.com/clb21565/kairos.

bioinformatics↗

bioRxiv2021

mobileOG-db: a manually curated database of protein families mediating the life cycle of bacterial mobile genetic elements

Currently available databases of bacterial mobile genetic elements (MGEs) contain both "core" and accessory MGE functional modules, the latter of which are often only transiently associated with the element. The presence of these accessory genes, which are often close homologs to primarily immobile genes, limits the usability of these databases for MGE annotation. To overcome this limitation, we analysed 10,776,212 protein sequences derived from seven MGE databases to compile a comprehensive database of 6,140 manually curated protein families that are linked to the "life cycle" (integration, excision, replication/recombination/repair, transfer, and stability/defense) of all major classes of bacterial MGEs. We overlay experimental information where available to create a tiered annotation scheme of high-quality annotations and annotations inferred exclusively through bioinformatic evidence. We additionally provide an MGE-class label for each entry (e.g., plasmid, integrative element) derived from the source database, and assign a list of keywords to each entry to delineate different MGE functional modules and to facilitate annotation. The resulting database, mobileOG-db (for mobile orthologous groups), provides a simple and readily interpretable foundation for an array of MGE-centred analyses. mobileOG-db can be accessed at mobileogdb.flsi.cloud.vt.edu/, where users can browse and design, refine, and analyse custom subsets of the dynamic mobilome.

genomics↗

bioRxiv2019

Whole genome sequence analysis reveals broad distribution of the RtxA type 1 secretion system and four novel type 1 secretion systems throughout the Legionella genus

Type 1 secretion systems (T1SSs) are broadly distributed among bacteria and translocate effectors with diverse function across the bacterial cell membrane. Legionella pneumophila, the species most commonly associated with Legionellosis, encodes a T1SS at the lssXYZABD locus which is responsible for the secretion of the virulence factor RtxA. Many investigations have failed to detect lssD, the gene encoding the membrane fusion protein of the RtxA T1SS, in non-pneumophila Legionella, suggesting that this system is a conserved virulence factor in L. pneumophila. Here we discovered RtxA and its associated T1SS in a novel Legionella taurinensis strain, leading us to question whether this system may be more widespread than previously thought. Through a bioinformatic analysis of publicly available data, we classified and determined the distribution of four T1SSs including the RtxA T1SS and four novel T1SSs among diverse Legionella spp. The ABC transporter of the novel Legionella T1SS Legonella repeat protein secretion system (LRPSS) shares structural similarity to those of diverse T1SS families, including the alkaline protease T1SS in Pseudomonas aeruginosa. The Legionella bacteriocin (1-3) secretion systems (LB1SS-LB3SS) T1SSs are novel putative bacteriocin transporting T1SSs as their ABC transporters include C-39 peptidase domains in their N-terminal regions, with LB2SS and LB3SS likely constituting a nitrile hydratase leader peptide transport T1SSs. The LB1SS is more closely related to the colicin V T1SS in Escherichia coli. Of 45 Legionella spp. whole genomes examined, 19 (42%) were determined to possess lssB and lssD homologs. Of these 19, only 7 (37%) are known pathogens. There was no difference in the proportions of disease associated and non-disease associated species that possessed the RtxA T1SS (p = 0.4), contrary to the current consensus regarding the RtxA T1SS. These results draw into question the nature of RtxA and its T1SS as a genetic virulence determinant.

microbiology↗

bioRxiv2019

MetaMLP: A fast word embedding based classifier to profile target gene databases in metagenomic samples

The functional profile of metagenomic samples allows the understanding of the role of the microbes in their environment. Such analysis consists of assigning short sequencing reads to a particular functional category. Normally, manually curated databases are used for functional assignment where genes are arranged into different classes. Sequence alignment has been widely used to profile metagenomic samples against curated databases. However, this method is time consuming and requires high computing resources. Although several alignment free methods based on k-mer composition have been developed in the recent years, they still require a large amount of memory. In this paper, MetaMLP (Metagenomics Machine Learning Profiler) a machine learning method that represents sequences into numerical vectors (embeddings) and uses a simple one hidden layer neural network to profile functional categories is proposed. Unlike other methods, MetaMLP enables partial matching by using a reduced alphabet to build sequence embeddings from full and partial kmers. MetaMLP is able to identify a slightly larger number of reads compared to Diamond (one of the fastest sequence alignment method) as well as to perform accurate predictions with 0.99 precision and 0.99 recall. MetaMLP can process 100M reads in around 10 minutes in a laptop computer which is 50x faster than Diamond. MetaMLP is free for use, available at https://bitbucket.org/gaarangoa/metamlp/src/master/.

bioinformatics↗

bioRxiv2018

NanoARG: A web service for identification of antimicrobial resistance elements from nanopore-derived environmental metagenomes

Direct selection pressures imposed by antibiotics, indirect pressures by co-selective agents, and horizontal gene transfer are fundamental drivers of the evolution and spread of antibiotic resistance. Therefore, effective environmental monitoring tools should ideally capture not only antibiotic resistance genes (ARGs), but also mobile genetic elements (MGEs) and indicators of co-selective forces, such as metal resistance genes (MRGs). Further, a major challenge towards characterizing potential human risk is the ability to identify bacterial host organisms, especially human pathogens. Historically, short reads yielded by next-generation sequencing technology has hampered confidence in assemblies for achieving these purposes. Here we introduce NanoARG, an online computational resource that takes advantage of long reads produced by MinION nanopore sequencing. Specifically, long nanopore reads enable identification of ARGs in the context of relevant neighboring genes, providing relevant insight into mobility, co-selection, and pathogenicity. NanoARG allows users to upload sequence data online and provides various means to analyze and visualize the data, including quantitative and simultaneous profiling of ARG, MRG, MGE, and pathogens. NanoARG is publicly available and freely accessible at http://bench.cs.vt.edu/nanoARG.

bioinformatics↗

bioRxiv2018

ARG-miner: A web platform for crowdsourcing-based curation of antibiotic resistance genes

Curation of antibiotic resistance gene (ARG) databases is a labor-intensive process that requires expert knowledge to manually collect, correct, and/or annotate individual genes. Correspondingly, updates to existing databases tend to be infrequent, commonly requiring years for completion and often containing inconsistences. Further, because of limitations of manual curation, most existing ARG databases contain only a small proportion of known ARGs (~5k genes). A new approach is needed to achieve a truly comprehensive ARG database, while also maintaining a high level of accuracy. Here we propose a new web-based curation system, ARG-miner, which supports annotation of ARGs at multiple levels, including: gene name, antibiotic category, resistance mechanism, and evidence for mobility and occurrence in clinically-important bacterial strains. To overcome limitations of manual curation, we employ crowdsourcing as a novel strategy for expanding curation capacity towards achieving a truly comprehensive, up-to-date database. We develop and validate the approach by comparing performance of multiple cohorts of curators with varying levels of expertise, demonstrating that ARG-miner is more cost effective and less time-consuming relative to traditional expert curation. We further demonstrate the reliability of a trust validation filter for rejecting confounding input generated by spammers. Crowdsourcing was found to be as accurate as expert annotation, with an accuracy >90% for the annotation of a diverse test set of ARGs. ARG-miner provides a public API and database available at http://bench.cs.vt.edu/argminer.

bioinformatics↗

bioRxiv2017

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

Growing concerns regarding increasing rates of antibiotic resistance call for global monitoring efforts. Monitoring of environmental media (e.g., wastewater, agricultural waste, food, and water) is of particular interest as these media can serve as sources of potential novel antibiotic resistance genes (ARGs), as hot spots for ARG exchange, and as pathways for the spread of ARGs and human exposure. Next-generation sequence-based monitoring has recently enabled direct access and profiling of the total metagenomic DNA pool, where ARGs are identified or predicted based on the \"best hits\" of homology searches against existing databases. Unfortunately, this approach tends to produce high rates of false negatives. To address such limitations, we propose here a deep leaning approach, taking into account a dissimilarity matrix created using all known categories of ARGs. Two models, deepARG-SS and deepARG-LS, were constructed for short read sequences and full gene length sequences, respectively. Performance evaluation of the deep learning models over 30 classes of antibiotics demonstrates that the deepARG models can predict ARGs with both high precision (>0.97) and recall (>0.90) for most of the antibiotic resistance categories. The models show advantage over the traditional best hit approach by having consistently much lower false negative rates and thus higher overall recall (>0.9). As more data become available for under-represented antibiotic resistance categories, the deepARG models performance can be expected to be further enhanced due to the nature of the underlying neural networks. The deepARG models are available both in command line version and via a Web server at http://bench.cs.vt.edu/deeparg. Our newly developed ARG database, deepARG-DB, containing predicted ARGs with high confidence and high degree of manual curation, greatly expands the current ARG repository. DeepARG-DB can be downloaded freely to benefit community research and future development of antibiotic resistance-related resources.\n\nAbbreviations

bioinformatics↗