The Source and Evolutionary History of a Microbial Contaminant Identified Through Soil Metagenomic Analysis

Ecological genomics in the Northern krill uncovers loci for local adaptation across ocean basins

Krill are vital as food for many marine animals but also impacted by global warming. To learn how they and other zooplankton may adapt to a warmer world we studied local adaptation in the widespread Northern krill (Meganyctiphanes norvegica). We assemble and characterize its large genome and compare genome-scale variation among 74 specimens from the colder Atlantic Ocean and warmer Mediterranean Sea. The 19 Gb genome likely evolved through proliferation of retrotransposons, now targeted for inactivation by extensive DNA methylation, and contains many duplicated genes associated with molting and vision. Analysis of 760 million SNPs indicates extensive homogenizing gene-flow among populations. Nevertheless, we detect signatures of adaptive divergence across hundreds of genes, implicated in photoreception, circadian regulation, reproduction and thermal tolerance, indicating polygenic adaptation to light and temperature. The top gene candidate for ecological adaptation was nrf-6, a lipid transporter with a Mediterranean variant that may contribute to early spring reproduction. Such variation could become increasingly important for fitness in Atlantic stocks. Our study underscores the widespread but uneven distribution of adaptive variation, necessitating characterization of genetic variation among natural zooplankton populations to understand their adaptive potential, predict risks and support ocean conservation in the face of climate change.

Two-tiered mutualism improves survival and competitiveness of cross-feeding soil bacteria

Metabolic cross-feeding is a pervasive microbial interaction type that affects community stability and functioning and directs carbon and energy flows. The mechanisms that underlie these interactions and their association with metal/metalloid biogeochemistry, however, remain poorly understood. Here, we identified two soil bacteria, Bacillus sp. BP-3 and Delftia sp. DT-2, that engage in a two-tiered mutualism. Strain BP-3 has low utilization ability of pyruvic acid while strain DT-2 lacks hexokinase, lacks a phosphotransferase system, and is defective in glucose utilization. When strain BP-3 is grown in isolation with glucose, it releases pyruvic acid to the environment resulting in acidification and eventual self-killing. However, when strain BP-3 is grown together with strain DT-2, strain DT-2 utilizes the released pyruvic acid to meet its energy requirements, consequently rescuing strain BP-3 from pyruvic acid-induced growth inhibition. The two bacteria further enhance their collective competitiveness against other microbes by using arsenic as a weapon. Strain DT-2 reduces relatively non-toxic methylarsenate [MAs(V)] to highly toxic methylarsenite [MAs(III)], which kills or suppresses competitors, while strain BP-3 detoxifies MAs(III) by methylation to non-toxic dimethylarsenate [DMAs(V)]. These two arsenic transformations are enhanced when strains DT-2 and BP-3 are grown together. The two strains, along with their close relatives, widely co-occur in soils and their abundances increase with the soil arsenic concentration. Our results reveal that these bacterial types employ a two-tiered mutualism to ensure their collective metabolic activity and maintain their ecological competitive against other soil microbes. These findings shed light on the intricateness of bacterial interactions and their roles in ecosystem functioning.

SemiBin2: self-supervised contrastive learning leads to better MAGs for short- and long-read sequencing

MOTIVATION

Metagenomic binning methods to reconstruct metagenome-assembled genomes (MAGs) from environmental samples have been widely used in large-scale metagenomic studies. The recently proposed semi-supervised binning method, SemiBin, achieved state-of-the-art binning results in several environments. However, this required annotating contigs, a computationally costly and potentially biased process.

RESULTS

We propose SemiBin2, which uses self-supervised learning to learn feature embeddings from the contigs. In simulated and real datasets, we show that self-supervised learning achieves better results than the semi-supervised learning used in SemiBin1 and that SemiBin2 outperforms other state-of-the-art binners. Compared to SemiBin1, SemiBin2 can reconstruct 8.3-21.5% more high-quality bins and requires only 25% of the running time and 11% of peak memory usage in real short-read sequencing samples. To extend SemiBin2 to long-read data, we also propose ensemble-based DBSCAN clustering algorithm, resulting in 13.1-26.3% more high-quality genomes than the second best binner for long-read data.

AVAILABILITY AND IMPLEMENTATION

SemiBin2 is available as open source software at https://github.com/BigDataBiology/SemiBin/ and the analysis scripts used in the study can be found at https://github.com/BigDataBiology/SemiBin2_benchmark.

Using strain-resolved analysis to identify contamination in metagenomics data

BACKGROUND

Metagenomics analyses can be negatively impacted by DNA contamination. While external sources of contamination such as DNA extraction kits have been widely reported and investigated, contamination originating within the study itself remains underreported.

RESULTS

Here, we applied high-resolution strain-resolved analyses to identify contamination in two large-scale clinical metagenomics datasets. By mapping strain sharing to DNA extraction plates, we identified well-to-well contamination in both negative controls and biological samples in one dataset. Such contamination is more likely to occur among samples that are on the same or adjacent columns or rows of the extraction plate than samples that are far apart. Our strain-resolved workflow also reveals the presence of externally derived contamination, primarily in the other dataset. Overall, in both datasets, contamination is more significant in samples with lower biomass.

CONCLUSION

Our work demonstrates that genome-resolved strain tracking, with its essentially genome-wide nucleotide-level resolution, can be used to detect contamination in sequencing-based microbiome studies. Our results underscore the value of strain-specific methods to detect contamination and the critical importance of looking for contamination beyond negative and positive controls. Video Abstract.

PhyloPlus: a Universal Tool for Phylogenetic Interrogation of Metagenomic Communities

Phylogeny is a powerful tool that can be incorporated into quantitative descriptions of community diversity, yet its use has been limited largely due to the difficulty in constructing phylogenies which incorporate the wide genomic diversity of microbial communities. Here, we describe the development of a web portal, PhyloPlus, which enables users to generate customized phylogenies that may be applied to any bacterial or archaeal communities. We demonstrate the power of phylogeny by comparing metrics that employ phylogeny with those that do not when applied to data sets from two metagenomic studies (fermented food, n = 58; human microbiome, n = 60). This example shows how inclusion of all bacterial species identified by taxonomic classifiers (Kraken2 and Kaiju) made the phylogeny perfectly congruent to the corresponding classification outputs. Our phylogeny-based approach also enabled the construction of more constrained null models which (i) shed light into community structure and (ii) minimize potential inflation of type I errors. Construction of such null models allowed for the observation of under-dispersion in 44 (75.86%) food samples, with the metacommunity defined as bacteria that were found in different food matrices. We also observed that closely related species with high abundance and uneven distribution across different sites could potentially exaggerate the dissimilarity between phylogenetically similar communities if they were measured using traditional species-based metrics (P_adj. = 0.003), whereas this effect was mitigated by incorporating phylogeny (P_adj. = 1). In summary, our tool can provide additional insights into microbial communities of interest and facilitate the use of phylogeny-based approaches in metagenomic analyses. IMPORTANCE There has been an explosion of interest in how microbial diversity affects human health, food safety, and environmental functions among many other processes. Accurately measuring the diversity and structure of those communities is central to understanding their effects. Here, we describe the development of a freely available online tool, PhyloPlus, which allows users to generate custom phylogenies that may be applied to any data set, thereby removing a major obstacle to the application of phylogeny to metagenomic data analysis. We demonstrate that the genetic relatedness of the organisms within those communities is a critical feature of their overall diversity, and that using a phylogeny which captures and quantifies this diversity allows for much more accurate descriptions while preventing misleading conclusions based on estimates that ignore evolutionary relationships.

A deep siamese neural network improves metagenome-assembled genomes in microbiome datasets across different environments

Metagenomic binning is the step in building metagenome-assembled genomes (MAGs) when sequences predicted to originate from the same genome are automatically grouped together. The most widely-used methods for binning are reference-independent, operating de novo and enable the recovery of genomes from previously unsampled clades. However, they do not leverage the knowledge in existing databases. Here, we introduce SemiBin, an open source tool that uses deep siamese neural networks to implement a semi-supervised approach, i.e. SemiBin exploits the information in reference genomes, while retaining the capability of reconstructing high-quality bins that are outside the reference dataset. Using simulated and real microbiome datasets from several different habitats from GMGCv1 (Global Microbial Gene Catalog), including the human gut, non-human guts, and environmental habitats (ocean and soil), we show that SemiBin outperforms existing state-of-the-art binning methods. In particular, compared to other methods, SemiBin returns more high-quality bins with larger taxonomic diversity, including more distinct genera and species.

A walk on the dirt: soil microbial forensics from ecological theory to the crime lab

Forensics aims at using physical evidence to solve investigations with science-based principles, thus operating within a theoretical framework. This however is often rather weak, the exception being DNA-based human forensics that is well anchored in theory. Soil is a most commonly encountered, easily and unknowingly transferred evidence but it is seldom employed as soil analyses require extensive expertise. In contrast, comparative analyses of soil bacterial communities using nucleic acid technologies can efficiently and precisely locate the origin of forensic soil traces. However, this application is still in its infancy, and is very rarely used. We posit that understanding the theoretical bases and limitations of their uses is essential for soil microbial forensics to be judiciously implemented. Accordingly, we review the ecological theory and experimental evidence explaining differences between soil microbial communities, i.e. the generation of beta diversity, and propose to integrate a bottom-up approach of interactions at the microscale, reflecting historical contingencies with top-down mechanisms driven by the geographic template, providing a potential explanation as to why bacterial communities map according to soil types. Finally, we delimit the use of soil microbial forensics based on the present technologies and ecological knowledge, and propose possible venues to remove existing bottlenecks.

Adhesion of Pseudomonas aeruginosa, Achromobacter xylosoxidans, Delftia acidovorans, Stenotrophomonas maltophilia to contact lenses under the influence of an artificial tear solution

Corneal infection is a devastating sight-threatening complication that is associated with contact lens (CL) wear, commonly caused by Pseudomonas aeruginosa. Lately, Achromobacter xylosoxidans, Delftia acidovorans, and Stenotrophomonas maltophilia have been associated with corneal infection. This study investigated the adhesion of these emerging pathogens to CLs, under the influence of an artificial tear solution (ATS) containing a variety of components commonly found in human tears. Two different CL materials, etafilcon A and senofilcon A, either soaked in an ATS or phosphate buffered saline, were exposed to the bacteria. Bacterial adhesion was investigated using a radio-labeling technique (total counts) and plate count method (viable counts). The findings from this study revealed that in addition to P. aeruginosa, among the emerging pathogens evaluated, A. xylosoxidans showed an increased propensity for adherence to both CL materials and S. maltophilia showed lower viability. ATS influenced the viable counts more than the total counts on CLs.

Issues and current standards of controls in microbiome research

Good scientific practice is important in all areas of science. In recent years this has gained more and more attention, especially considering the 'scientific reproducibility crisis'. While most researchers are aware of the issues with good scientific practice, not all of these issues are necessarily clear, and the details can be very complicated. For many years it has been accepted to perform and publish sequencing based microbiome studies without including proper controls. Although in recent years more scientists realize the necessity of implementing controls, this poses a problem due to the complexity of the field. Another concern is the inability to properly interpret the information gained from controls in microbiome studies. Here, we will discuss these issues and provide a comprehensive overview of problematic points regarding controls in microbiome research, and of the current standards in this area.

Recentrifuge: Robust comparative analysis and contamination removal for metagenomics

Metagenomic sequencing is becoming widespread in biomedical and environmental research, and the pace is increasing even more thanks to nanopore sequencing. With a rising number of samples and data per sample, the challenge of efficiently comparing results within a specimen and between specimens arises. Reagents, laboratory, and host related contaminants complicate such analysis. Contamination is particularly critical in low microbial biomass body sites and environments, where it can comprise most of a sample if not all. Recentrifuge implements a robust method for the removal of negative-control and crossover taxa from the rest of samples. With Recentrifuge, researchers can analyze results from taxonomic classifiers using interactive charts with emphasis on the confidence level of the classifications. In addition to contamination-subtracted samples, Recentrifuge provides shared and exclusive taxa per sample, thus enabling robust contamination removal and comparative analysis in environmental and clinical metagenomics. Regarding the first area, Recentrifuge's novel approach has already demonstrated its benefits showing that microbiomes of Arctic and Antarctic solar panels display similar taxonomic profiles. In the clinical field, to confirm Recentrifuge's ability to analyze complex metagenomes, we challenged it with data coming from a metagenomic investigation of RNA in plasma that suffered from critical contamination to the point of preventing any positive conclusion. Recentrifuge provided results that yielded new biological insight into the problem, supporting the growing evidence of a blood microbiota even in healthy individuals, mostly translocated from the gut, the oral cavity, and the genitourinary tract. We also developed a synthetic dataset carefully designed to rate the robust contamination removal algorithm, which demonstrated a significant improvement in specificity while retaining a high sensitivity even in the presence of cross-contaminants. Recentrifuge's official website is www.recentrifuge.org. The data and source code are anonymously and freely available on GitHub and PyPI. The computing code is licensed under the AGPLv3. The Recentrifuge Wiki is the most extensive and continually-updated source of documentation for Recentrifuge, covering installation, use cases, testing, and other useful topics.

Genome-resolved metagenomics of eukaryotic populations during early colonization of premature infants and in hospital rooms

BACKGROUND

Fungal infections are a significant cause of mortality and morbidity in hospitalized preterm infants, yet little is known about eukaryotic colonization of infants and of the neonatal intensive care unit as a possible source of colonizing strains. This is partly because microbiome studies often utilize bacterial 16S rRNA marker gene sequencing, a technique that is blind to eukaryotic organisms. Knowledge gaps exist regarding the phylogeny and microdiversity of eukaryotes that colonize hospitalized infants, as well as potential reservoirs of eukaryotes in the hospital room built environment.

RESULTS

Genome-resolved analysis of 1174 time-series fecal metagenomes from 161 premature infants revealed fungal colonization of 10 infants. Relative abundance levels reached as high as 97% and were significantly higher in the first weeks of life (p = 0.004). When fungal colonization occurred, multiple species were present more often than expected by random chance (p = 0.008). Twenty-four metagenomic samples were analyzed from hospital rooms of six different infants. Compared to floor and surface samples, hospital sinks hosted diverse and highly variable communities containing genomically novel species, including from Diptera (fly) and Rhabditida (worm) for which genomes were assembled. With the exception of Diptera and two other organisms, zygosity of the newly assembled diploid eukaryote genomes was low. Interestingly, Malassezia and Candida species were present in both room and infant gut samples.

CONCLUSIONS

Increased levels of fungal co-colonization may reflect synergistic interactions or differences in infant susceptibility to fungal colonization. Discovery of eukaryotic organisms that have not been sequenced previously highlights the benefit of genome-resolved analyses, and low zygosity of assembled genomes could reflect inbreeding or strong selection imposed by room conditions.

Identification and Removal of Contaminant Sequences From Ribosomal Gene Databases: Lessons From the Census of Deep Life

Earth’s subsurface environment is one of the largest, yet least studied, biomes on Earth, and many questions remain regarding what microorganisms are indigenous to the subsurface. Through the activity of the Census of Deep Life (CoDL) and the Deep Carbon Observatory, an open access 16S ribosomal RNA gene sequence database from diverse subsurface environments has been compiled. However, due to low quantities of biomass in the deep subsurface, the potential for incorporation of contaminants from reagents used during sample collection, processing, and/or sequencing is high. Thus, to understand the ecology of subsurface microorganisms (i.e., the distribution, richness, or survival), it is necessary to minimize, identify, and remove contaminant sequences that will skew the relative abundances of all taxa in the sample. In this meta-analysis, we identify putative contaminants associated with the CoDL dataset, recommend best practices for removing contaminants from samples, and propose a series of best practices for subsurface microbiology sampling. The most abundant putative contaminant genera observed, independent of evenness across samples, were Propionibacterium, Aquabacterium, Ralstonia, and Acinetobacter. While the top five most frequently observed genera were Pseudomonas, Propionibacterium, Acinetobacter, Ralstonia, and Sphingomonas. The majority of the most frequently observed genera (high evenness) were associated with reagent or potential human contamination. Additionally, in DNA extraction blanks, we observed potential archaeal contaminants, including methanogens, which have not been discussed in previous contamination studies. Such contaminants would directly affect the interpretation of subsurface molecular studies, as methanogenesis is an important subsurface biogeochemical process. Utilizing previously identified contaminant genera, we found that ∼27% of the total dataset were identified as contaminant sequences that likely originate from DNA extraction and DNA cleanup methods. Thus, controls must be taken at every step of the collection and processing procedure when working with low biomass environments such as, but not limited to, portions of Earth’s deep subsurface. Taken together, we stress that the CoDL dataset is an incredible resource for the broader research community interested in subsurface life, and steps to remove contamination derived sequences must be taken prior to using this dataset.