An in-depth evaluation of metagenomic classifiers for soil microbiomes.

BACKGROUND: Recent endeavours in metagenomics, exemplified by projects such as the human microbiome project and TARA Oceans, have illuminated the complexities of microbial biomes. A robust bioinformatic pipeline and meticulous evaluation of their methodology have contributed to the success of these projects. The soil environment, however, with its unique challenges, requires a specialized methodological exploration to maximize microbial insights. A notable limitation in soil microbiome studies is the dearth of soil-specific reference databases available to classifiers that emulate the complexity of soil communities. There is also a lack of in-vitro mock communities derived from soil strains that can be assessed for taxonomic classification accuracy. RESULTS: In this study, we generated a custom in-silico mock community containing microbial genomes commonly observed in the soil microbiome. Using this mock community, we simulated shotgun sequencing data to evaluate the performance of three leading metagenomic classifiers: Kraken2 (supplemented with Bracken, using a custom database derived from GTDB-TK genomes along with its own default database), Kaiju, and MetaPhlAn, utilizing their respective default databases for a robust analysis. Our results highlight the importance of optimizing taxonomic classification parameters, database selection, as well as analysing trimmed reads and contigs. Our study showed that classifiers tailored to the specific taxa present in our samples led to fewer errors compared to broader databases including microbial eukaryotes, protozoa, or human genomes, highlighting the effectiveness of targeted taxonomic classification. Notably, an optimal classifier performance was achieved when applying a relative abundance threshold of 0.001% or 0.005%. The Kraken2 supplemented with bracken, with a custom database demonstrated superior precision, sensitivity, F1 score, and overall sequence classification. Using a custom database, this classifier classified 99% of in-silico reads and 58% of real-world soil shotgun reads, with the latter identifying previously overlooked phyla using a custom database. CONCLUSION: This study underscores the potential advantages of in-silico methodological optimization in metagenomic analyses, especially when deciphering the complexities of soil microbiomes. We demonstrate that the choice of classifier and database significantly impacts microbial taxonomic profiling. Our findings suggest that employing Kraken2 with Bracken, coupled with a custom database of GTDB-TK genomes and fungal genomes at a relative abundance threshold of 0.001% provides optimal accuracy in soil shotgun metagenome analysis.

Critical evaluation of the use of artificial data for machine learning based de novo peptide identification.

Proteins are essential components of all living cells and so the study of their in situ expression, proteomics, has wide reaching applications. Peptide identification in proteomics typically relies on matching high resolution tandem mass spectra to a protein database but can also be performed de novo. While artificial spectra have been successfully incorporated into database search pipelines to increase peptide identification rates, little work has been done to investigate the utility of artificial spectra in the context of de novo peptide identification. Here, we perform a critical analysis of the use of artificial data for the training and evaluation of de novo peptide identification algorithms. First, we classify the different fragment ion types present in real spectra and then estimate the number of spurious matches using random peptides. We then categorise the different types of noise present in real spectra. Finally, we transfer this knowledge to artificial data and test the performance of a state-of-the-art de novo peptide identification algorithm trained using artificial spectra with and without relevant noise addition. Noise supplementation increased artificial training data performance from 30% to 77% of real training data peptide recall. While real data performance was not fully replicated, this work provides the first steps towards an artificial spectrum framework for the training and evaluation of de novo peptide identification algorithms. Further enhanced artificial spectra may allow for more in depth analysis of de novo algorithms as well as alleviating the reliance on database searches for training data.

Impacts of metal stress on extracellular microbial products, and potential for selective metal recovery.

Harnessing microbial capabilities for metal recovery from secondary waste sources is an eco-friendly and sustainable approach for the management of metal-containing wastes. Soluble microbial products (SMP) and extracellular polymeric substances (EPS) are the two main groups of extracellular compounds produced by microorganisms in response to metal stress that are of great importance for remediation and recovery of metals. These include various high-, and low, molecular weight components, which serve various functional and structural roles. These compounds often contain functional groups with metal binding potential that can attenuate metal stress by sequestering metal ions, making them less bioavailable. Microorganisms can regulate the content and composition of EPS and SMP in response to metal stress in order to increase the compounds specificity and capacity for metal binding. Thus, EPS and SMP represent ideal candidates for developing technologies for selective metal recovery from complex wastes. To discover highly metal-sorptive compounds with specific metal binding affinity for metal recovery applications, it is necessary to investigate the metal binding affinity of these compounds, especially under metal stressed conditions. In this review we critically reviewed microbial EPS and SMP production as a response to metal stress with a particular emphasis on the metal binding properties of these compounds and their role in altering metal bioavailability. Furthermore, for the first time, we compiled the available data on potential application of these compounds for selective metal recovery from waste streams.

Application of a Novel Hybrid CNN-GNN for Peptide Ion Encoding.

Almost all state-of-the-art de novo peptide sequencing algorithms now use machine learning models to encode fragment peaks and hence identify amino acids in mass spectrometry (MS) spectra. Previous work has highlighted how the inherent MS challenges of noise and missing peptide peaks detrimentally affect the performance of these models. In the present research we extracted and evaluated the encoding modules from 3 state-of-the-art de novo peptide sequencing algorithms. We also propose a convolutional neural network-graph neural network machine learning model for encoding peptide ions in tandem MS spectra. We compared the proposed encoding module to those used in the state-of-the-art de novo peptide sequencing algorithms by assessing their ability to identify b-ions and y-ions in MS spectra. This included a comprehensive evaluation in both real and artificial data across various levels of noise and missing peptide peaks. The proposed model performed best across all data sets using two different metrics (area under the receiver operating characteristic curve (AUC) and average precision). The work also highlighted the effect of including additional features such as intensity rank in these encoding modules as well as issues with using the AUC as a metric. This work is of significance to those designing future de novo peptide identification algorithms as it is the first step toward a new approach.

The impact of noise and missing fragmentation cleavages on de novo peptide identification algorithms.

Proteomics aims to characterise system-wide protein expression and typically relies on mass-spectrometry and peptide fragmentation, followed by a database search for protein identification. It has wide ranging applications from clinical to environmental settings and virtually impacts on every area of biology. In that context, de novo peptide sequencing is becoming increasingly popular. Historically its performance lagged behind database search methods but with the integration of machine learning, this field of research is gaining momentum. To enable de novo peptide sequencing to realise its full potential, it is critical to explore the mass spectrometry data underpinning peptide identification. In this research we investigate the characteristics of tandem mass spectra using 8 published datasets. We then evaluate two state of the art de novo peptide sequencing algorithms, Novor and DeepNovo, with a particular focus on their performance with regard to missing fragmentation cleavage sites and noise. DeepNovo was found to perform better than Novor overall. However, Novor recalled more correct amino acids when 6 or more cleavage sites were missing. Furthermore, less than 11% of each algorithms' correct peptide predictions emanate from data with more than one missing cleavage site, highlighting the issues missing cleavages pose. We further investigate how the algorithms manage to correctly identify peptides with many of these missing fragmentation cleavages. We show how noise negatively impacts the performance of both algorithms, when high intensity peaks are considered. Finally, we provide recommendations regarding further algorithms' improvements and offer potential avenues to overcome current inherent data limitations.

Evolutionary trade-offs between growth and survival: The delicate balance between reproductive success and longevity in bacteria.

All living cells strive to allocate cellular resources in a way that promotes maximal evolutionary fitness. While there are many competing demands for resources the main decision making process centres on whether to proceed with growth and reproduction or to "hunker down" and invest in protection and survival (or to strike an optimal balance between these two processes). The transcriptional programme active at any given time largely determines which of these competing processes is dominant. At the top of the regulatory hierarchy are the sigma factors that commandeer the transcriptional machinery and determine which set of promoters are active at any given time. The regulatory inputs controlling their activity are therefore often highly complex, with multiple layers of regulation, allowing relevant environmental information to produce the most beneficial response. The tension between growth and survival is also evident in the developmental programme necessary to promote biofilm formation, which is typically associated with low growth rates and enhanced long-term survival. Nucleotide second messengers and energy pools (ATP/ADP levels) play critical roles in determining the fate of individual cells. Regulatory small RNAs frequently play important roles in the decision making processes too. In this review we discuss the trade-off that exists between reproduction and persistence in bacteria and discuss some of the recent advances in this fascinating field.

Survival of Escherichia coli and Listeria innocua on Lettuce after Irrigation with Contaminated Water in a Temperate Climate.

Microbial disease outbreaks related to fresh produce consumption, including leafy green vegetables, have increased in recent years. Where contamination occurs, pathogen persistence may represent a risk for consumers' health. This study analysed the survival of E. coli and L. innocua on lettuce plants watered with contaminated irrigation water via a single irrigation event and within stored irrigation water. Separate lettuce plants (Lactuca sativa var. capitata) were irrigated with water spiked with Log10 7 cfu/mL of each of the two strains and survival assessed via direct enumeration, enrichment and qPCR. In parallel, individual 20 L water microcosms were spiked with Log10 7 cfu/mL of the individual strains and sampled at similar time points. Both strains were observed to survive on lettuce plants up to 28 days after inoculation. Direct quantification by culture methods showed a Log10 4 decrease in the concentration of E. coli 14 days after inoculation, and a Log10 3 decrease in the concentration of L. innocua 10 days after inoculation. E. coli was detected in water samples up to 7 days after inoculation and L. innocua was detected up to 28 days by direct enumeration. Both strains were recovered from enriched samples up to 28 days after inoculation. These results demonstrate that E. coli and L. innocua strains are able to persist on lettuce after a single contamination event up until the plants reach a harvestable state. Furthermore, the persistence of E. coli and L. innocua in water for up to 28 days after inoculation illustrates the potential for multiple plant contamination events from stored irrigation water, emphasising the importance of ensuring that irrigation water is of a high quality.

Combined Stochastic and Deterministic Processes Drive Community Assembly of Anaerobic Microbiomes During Granule Flotation.

Advances in null-model approaches have resulted in a deeper understanding of community assembly mechanisms for a variety of complex microbiomes. One under-explored application is assembly of communities from the built-environment, especially during process disturbances. Anaerobic digestion for biological wastewater treatment is often underpinned by retaining millions of active granular biofilm aggregates. Flotation of granules is a major problem, resulting in process failure. Anaerobic aggregates were sampled from three identical bioreactors treating dairy wastewater. Microbiome structure was analysed using qPCR and 16S rRNA gene amplicon sequencing from DNA and cDNA. A comprehensive null-model approach quantified assembly mechanisms of floating and settled communities. Significant differences in diversity were observed between floating and settled granules, in particular, we highlight the changing abundances of Methanosaeta and Lactococcus. Both stochastic and deterministic processes were important for community assembly. Homogeneous selection was the primary mechanism for all categories, but dispersal processes also contributed. The lottery model was used to identify clade-level competition driving community assembly. Lottery "winners" were identified with different winners between floating and settled groups. Some groups changed their winner status when flotation occurred. Spirochaetaceae, for example, was only a winner in settled biomass (cDNA-level) and lost its winner status during flotation. Alternatively, Arcobacter butzerli gained winner status during flotation. This analysis provides a deeper understanding of changes that occur during process instabilities and identified groups which may be washed out-an important consideration for process control.

Application of plasma activated water for decontamination of alfalfa and mung bean seeds.

Microbial contamination of fresh produce is a major public health concern, with the number of associated disease outbreaks increasing in recent years. The consumption of sprouted beans and seeds is of particular concern, as these foodstuffs are generally consumed raw, and are produced in conditions favourable for the growth of zoonotic pathogens, if present in seeds prior to sprouting or in irrigation water. This work aimed to evaluate the activity of plasma activated water (PAW) as a disinfecting agent for alfalfa (Medicago sativa) and mung bean (Vigna radiata) seeds, during seed soaking. Each seed type was inoculated with Escherichia coli O157, E. coli O104, Listeria monocytogenes or Salmonella Montevideo, and treated with PAW for different times. A combination of PAW and ultrasound treatment was also evaluated. The germination and growth rate of both seeds were assessed after PAW treatments. PAW was demonstrated to have disinfecting ability on sprouted seeds, with reductions of up to Log10 1.67 cfu/g in alfalfa seeds inoculated with E. coli O104, and a reduction of Log10 1.76 cfu/g for mung bean seeds inoculated with E. coli O157 observed. The germination and growth rate of alfalfa and mung bean sprouts were not affected by the PAW treatments. The combination of a PAW treatment and ultrasound resulted in increased antimicrobial activity, with a reduction of Log10 3.48 cfu/g of S. Montevideo in mung bean seeds observed. These results demonstrate the potential for PAW to be used for the inactivation of pathogenic microorganisms which may be present on sprouted seeds and beans, thereby providing greater assurance of produce safety.

Simultaneous adsorption and biodegradation of trichloroethylene occurs in a biochar packed column treating contaminated landfill leachate.

Trichloroethylene (TCE) is a human carcinogen that is commonly found in landfill leachate. Contaminated leachate plumes may be intercepted prior to reaching groundwater and treated in situ using permeable reactive barriers (PRB). This study used a packed column system containing herbal pomace and spruce biochar, previously shown to have TCE adsorptive capabilities. Influent containing raw or autoclaved landfill leachate was used to investigate the potential for environmental micro-organisms to establish a TCE-dechlorinating biofilm on the biochar, in order to prolong the operational life span of the system. TCE removal ≥ 99.7 % was observed by both biochars. No dichloroethylene (DCE) isomers were present in the column effluents, but cis-1,2 DCE was adsorbed to the biochar treating raw landfill leachate, indicating that dechlorination was occurring biologically in these columns. Known microbial species that are individually capable of complete dechlorination of TCE to ethene were not detected by 16S rRNA gene sequencing, but several species capable of partial TCE dechlorination (Desulfitobacterium spp., Sulfurospirillium spp. and Desulfuromonas spp) were present in the biofilms of the columns treating raw landfill leachate. These data demonstrate that biochar from waste material may be capable of supporting a dechlorinating biofilm to promote bioremediation of TCE.

Easy phylotyping of Escherichia coli via the EzClermont web app and command-line tool.

The Clermont PCR method for phylotyping Escherichia coli remains a useful classification scheme even though genome sequencing is now routine, and higher-resolution sequence typing schemes are now available. Relating present-day whole-genome E. coli classifications to legacy phylotyping is essential for harmonizing the historical literature and understanding of this important organism. Therefore, we present EzClermont - a novel in silico Clermont PCR phylotyping tool to enable ready application of this phylotyping scheme to whole-genome assemblies. We evaluate this tool against phylogenomic classifications, and an alternative software implementation of Clermont typing. EzClermont is available as a web app at www.ezclermont.org, and as a command-line tool at https://nickp60.github.io/EzClermont/.

A Small Study of Bacterial Contamination of Anaerobic Digestion Materials and Survival in Different Feed Stocks.

If pathogens are present in feedstock materials and survive in anaerobic digestion (AD) formulations at 37 °C, they may also survive the AD process to be disseminated in digestate spread on farmland as a fertilizer. The aim of this study was to investigate the prevalence of Salmonella spp., Escherichia coli O157, Listeria monocytogenes, Enterococcus faecalis and Clostridium spp. in AD feed and output materials and survival/growth in four formulations based on food waste, bovine slurry and/or grease-trap waste using International Organization for Standardization (ISO) or equivalent methods. The latter was undertaken in 100 mL Ramboldi tubes, incubated at 37 °C for 10 d with surviving cells enumerated periodically and the T90 values (time to achieve a 1 log reduction) calculated. The prevalence rates for Salmonella spp., Escherichia coli O157, Listeria monocytogenes, Enterococcus faecalis and Clostridium spp. were 3, 0, 5, 11 and 10/13 in food waste, 0, 0, 2, 3 and 2/3 in bovine slurry, 1, 0, 8, 7 and 8/8 in the mixing tank, 5, 1, 17, 18 and 17 /19 in raw digestate and 0, 0, 0, 2 and 2/2 in dried digestate, respectively. Depending on the formulation, T90 values ranged from 1.5 to 2.8 d, 1.6 to 2.8 d, 3.1 to 23.5 d, 2.2 to 6.6 d and 2.4 to 9.1 d for Salmonella Newport, Escherichia coli O157, Listeria monocytogenes, Enterococcus faecalis and Clostridium sporogenes, respectively. It was concluded that AD feed materials may be contaminated with a range of bacterial pathogens and L. monocytogenes may survive for extended periods in the test formulations incubated at 37 °C.

Reactor configuration influences microbial community structure during high-rate, low-temperature anaerobic treatment of dairy wastewater.

Low temperature anaerobic digestion remains in its infancy, despite increasing interest for the treatment of complex wastewaters. In this study, the feasibility of low-temperature anaerobic treatment of dairy wastewater was assessed during a 443-day laboratory-scale bioreactor trial. The bioreactors were operated in triplicate at organic loading rates of 7.5-9 kgCODm-3d-1 throughout five operational phases. The structure of the microbial community was analysed using quantitative real-time PCR and amplicon sequencing of 16S rRNA genes from DNA and rRNA. The results indicated that low-temperature treatment of dairy wastewater is feasible at 15 °C, but that reactor configuration remains extremely important. The upflow anaerobic sludge bed (UASB) configuration out-performed the expanded granular sludge bed (EGSB)-based configurations. Decreased temperatures resulted in significant reductions in microbiome diversity. Methanosaeta was identified as a dominant genus throughout the trial, while Lactococcus was identified as an important bacterial genus at low-temperatures. However, the relative abundance of Lactococcus was significantly influenced by reactor configuration.

Chordomics: a visualization tool for linking function to phylogeny in microbiomes.

SUMMARY: The overarching aim of microbiome analysis is to uncover the links between microbial phylogeny and function in order to access ecosystem functioning. This can be done using several experimental strategies targeting different biomolecules, including DNA (metagenomics), RNA (metatranscriptomics) and proteins (metaproteomics). Despite the importance of linking microbial function to phylogeny there are currently no visualization tools that effectively integrate this information. Chordomics is a Shiny-based application for linked -omics data analysis, allowing users to visualize microbial function and phylogeny on a single plot and compare datasets across time and environments. AVAILABILITY AND IMPLEMENTATION: Chordomics is available on GitHub: https://github.com/kevinmcdonnell6/chordomics; software is coded in R and JavaScript and a demonstration version is available at https://kmcd.shinyapps.io/ChordomicsDemo/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Impact of beef extract used for sample concentration on the detection of Escherichia coli DNA in water samples via qPCR.

There is increasing interest in methodologies for the simultaneous concentration and detection of multiple targets in individual samples. The aim of this study was to investigate the potential presence of E. coli DNA in beef extract powder used as part of a procedure to concentrate water samples for the simultaneous detection of bacteria, viruses and protozoa. DNA from E. coli was detected in five out of six beef extract lots tested, demonstrating the limitations of its inclusion when being used in assays that will be used for the detection of E. coli in water samples. Further work is required to clarify if this phenomenon also occurs for other microorganisms of interest in water.

Pyrolysed waste materials show potential for remediation of trichloroethylene-contaminated water.

Trichloroethylene (TCE) is an Environmental Protection Agency priority pollutant associated with cancer in humans. With numerous industrial applications and regular landfill disposal, TCE is a common landfill leachate pollutant. In situ treatment barriers use costly fill materials such as granular activated carbon (GAC). Here, we show that while a range of untreated waste materials had little ability to adsorb TCE, waste-derived biochar showed excellent capacity for TCE adsorption. TCE removal efficiencies by spruce and oak-derived biochars were > 99.5 %, outperforming GAC (95 %) and herbal pomace biochar (93 %). A contact time of at least 32 h was required to reach equilibrium for all of these media. Assessment of pollution swapping potential revealed release of phosphate by all biochars. Analysis of media surface characteristics by Fourier Transform Infrared Spectroscopy (FTIR) predicted that GAC should have the highest ability to adsorb TCE, followed by Oak Biochar, Herbal Pomace Biochar 1, and Spruce Biochar 2, which was not in agreement with the experimental adsorption data. These data demonstrate the potential for pyrolysed waste material to be used as an alternative fill material for in situ remediation applications, thereby also addressing the European Circular Economy Strategy.

The potential for polyphosphate metabolism in Archaea and anaerobic polyphosphate formation in Methanosarcina mazei.

Inorganic polyphosphate (polyP) is ubiquitous across all forms of life, but the study of its metabolism has been mainly confined to bacteria and yeasts. Few reports detail the presence and accumulation of polyP in Archaea, and little information is available on its functions and regulation. Here, we report that homologs of bacterial polyP metabolism proteins are present across the major taxa in the Archaea, suggesting that archaeal populations may have a greater contribution to global phosphorus cycling than has previously been recognised. We also demonstrate that polyP accumulation can be induced under strictly anaerobic conditions, in response to changes in phosphate (Pi) availability, i.e. Pi starvation, followed by incubation in Pi replete media (overplus), in cells of the methanogenic archaeon Methanosarcina mazei. Pi-starved M. mazei cells increased transcript abundance of the alkaline phosphatase (phoA) gene and of the high-affinity phosphate transport (pstSCAB-phoU) operon: no increase in polyphosphate kinase 1 (ppk1) transcript abundance was observed. Subsequent incubation of Pi-starved M. mazei cells under Pi replete conditions, led to a 237% increase in intracellular polyphosphate content and a > 5.7-fold increase in ppk1 gene transcripts. Ppk1 expression in M. mazei thus appears not to be under classical phosphate starvation control.

A robust, cost-effective method for DNA, RNA and protein co-extraction from soil, other complex microbiomes and pure cultures.

The soil microbiome is inherently complex with high biological diversity, and spatial heterogeneity typically occurring on the submillimetre scale. To study the microbial ecology of soils, and other microbiomes, biomolecules, that is, nucleic acids and proteins, must be efficiently and reliably co-recovered from the same biological samples. Commercial kits are currently available for the co-extraction of DNA, RNA and proteins but none has been developed for soil samples. We present a new protocol drawing on existing phenol-chloroform-based methods for nucleic acids co-extraction but incorporating targeted precipitation of proteins from the phenol phase. The protocol is cost-effective and robust, and easily implemented using reagents commonly available in laboratories. The method is estimated to be eight times cheaper than using disparate commercial kits for the isolation of DNA and/or RNA, and proteins, from soil. The method is effective, providing good quality biomolecules from a diverse range of soil types, with clay contents varying from 9.5% to 35.1%, which we successfully used for downstream, high-throughput gene sequencing and metaproteomics. Additionally, we demonstrate that the protocol can also be easily implemented for biomolecule co-extraction from other complex microbiome samples, including cattle slurry and microbial communities recovered from anaerobic bioreactors, as well as from Gram-positive and Gram-negative pure cultures.

Microbial Contamination of Fresh Produce: What, Where, and How?

Promotion of healthier lifestyles has led to an increase in consumption of fresh produce. Such foodstuffs may expose consumers to increased risk of foodborne disease, as often they are not subjected to processing steps to ensure effective removal or inactivation of pathogenic microorganisms before consumption. Consequently, reports of ready-to-eat fruit and vegetable related disease outbreak occurrences have increased substantially in recent years, and information regarding these events is often not readily available. Identifying the nature and source of microbial contamination of these foodstuffs is critical for developing appropriate mitigation measures to be implemented by food producers. This review aimed to identify the foodstuffs most susceptible to microbial contamination and the microorganisms responsible for disease outbreaks from information available in peer-reviewed scientific publications. A total of 571 outbreaks were identified from 1980 to 2016, accounting for 72,855 infections and 173 deaths. Contaminated leafy green vegetables were responsible for 51.7% of reported outbreaks. Contaminated soft fruits caused 27.8% of infections. Pathogenic strains of Escherichia coli and Salmonella, norovirus, and hepatitis A accounted for the majority of cases. Large outbreaks resulted in particular biases such as the observation that contaminated sprouted plants caused 31.8% of deaths. Where known, contamination mainly occurred via contaminated seeds, water, and contaminated food handlers. There is a critical need for standardized datasets regarding all aspects of disease outbreaks, including how foodstuffs are contaminated with pathogenic microorganisms. Providing food business operators with this knowledge will allow them to implement better strategies to improve safety and quality of fresh produce.