Posters - Schedules

Posters Home

The development of low-cost sequencing technologies has generated a massive amount of microbiome datasets in public repositories during the last 20 years. However, their reuse raises many difficulties making their comparison and integration very limited, even for a given ecosystem.
In this study, we present an integrative bioinformatics approach focusing on public metagenetic 16S datasets targeting lacto-fermented vegetables. This ecosystem needs to be better characterized regarding how microbial communities interact and evolve dynamically.
We have developed a workflow to explore, compare, and integrate public 16S datasets to conduct meta-analyses in the microbiota field. The workflow includes searching and selecting public time-series datasets and constructing Amplicon Sequence Variants (ASV) association networks based on co-abundance metrics. Microbial communities detection is achieved by comparison and clustering of ASVs networks (Figure 1). We applied the workflow to ten public datasets and demonstrated its value in monitoring precisely the fermentation with the identification of the bacterial communities succession (Figure 2) and of putative core-consortia shared by different plant fermentation types (Figure 3).
Our integrative analysis demonstrates that the reuse and integration of microbiome datasets can provide new insights into a little-known biotope and add value to the independent analysis of individual studies.

Understanding interactions between microbial taxa is crucial for unraveling the complexity and dynamics of ecological communities. In this study, we present an innovative approach that combines social network analysis and machine learning techniques to predict the assignment of microbial taxa to their respective communities based on their functional profiles.

Using MultiDimensional Feature Selection (MDFS) we construct a dataset of taxon interactions, which is used to identify potential relationships that may predict the presence of one taxon based on the presence of others. Using social network analysis, specifically Louvain community detection, we partition the microbial interaction network into distinct communities to gain insights into the underlying structure of the complex microbial ecosystem.

Next, we employ machine learning algorithms to predict community assignments of taxa based on their functional profiles. By training the model on a set of known community assignments, we can evaluate its performance and generalize its predictions to previously unseen taxa.

Our research demonstrates the potential of combining social network analysis and machine learning techniques to better understand and predict microbial community dynamics. This interdisciplinary approach paves the way for future studies aiming to uncover the intricate relationships between microbial taxa and their functional roles within ecological communities

Genome mining has become a key technology for discovering novel natural products with therapeutic potential. Such analysis involves searching for biosynthetic gene clusters (BGCs) within the genome of an organism to identify genes responsible for the production of natural products. Bacteria and fungi are particularly attractive targets for genome mining due to their high genetic diversity and ability to produce a wide range of bioactive compounds. Microbial genomes are relatively small compared to higher eukaryotes, making them easier to sequence and analyze. In the pre-genomic era, a limited number of microbes were studied for drug discovery. The growing availability of genomic data has led to a significant increase in the number of novel BGCs identified through genome sequence mining. The limited computational methods for prioritization, as well as a lack of comparison across virulence risk classes of species (e.g. COGEM), have created challenges for further understanding and identification of species. In this work, we examine 81 species of bacteria and fungi from the COGEM list, to identify genomic traits which could prioritize species and clades drug discovery. Based on our analysis, we propose novel methods and suggestions for improving genome mining that can improve efficiency of discovery of new biomolecules.

The Sequence Read Archive is the central repository for genomics experiments and a treasure trove of over 70 petabases of sequence data. However, its massive size presents a significant challenge to traditional search methods. Bloom-filter and sketching-based methods have been proposed as scalable alternatives, but their sensitivity is limited.

We present Petasearch, a tool for quickly and accurately searching protein sequences within large databases. Petasearch's algorithm involves three stages: First, the sequences in the database are pre-processed, sorted, and stored in a compressed k-mer index. Then, similar query k-mers are extracted and matched with database k-mers, filtering out non-homologous sequences early. Finally, high-scoring k-mer matches are aligned with a SIMD-accelerated banded Smith-Waterman.
We optimize Petasearch using modern CPU caching and prefetching, advanced Linux IO techniques, and high read-bandwidth NVMe-SSDs. Across 21 NVMe-SSDs, Petasearch is 15 and 145 times faster than current search algorithms for a 450GB and 9.3TB dataset, respectively. Petasearch maintains comparable sensitivity to state-of-the-art algorithms, detecting sequence identities as low as 60%, and identifying homology using its profile-search down to 40% in a SCOP25 benchmark.

Petasearch is available at petasearch.mmseqs.com as free open-source software for analysis and comparison of protein sequences at scale.

Metagenomics is widely used and is especially useful when working with organisms, that are hard to isolate. Nanopore sequencing can be used to quickly assess the content of a metagenome and the real time data it offers can be used to monitor metagenomes by frequently taking samples. However, there exist a huge variety of bioinformatics tools and choosing the right tools for answering biological question can be hard for lab researchers without bioinformatics experience. We developed MMonitor, an open source software with a focus on tracking metagenomes. It combines existing bioinformatics software with a user-friendly GUI and offers interactive visualizations on a web server that can be viewed remotely while sequencing.

Microbial communities play essential roles in various biological processes, and manipulating their composition and structure can enhance the production of valuable products and improve human health. The coexistence theory provides insights into the mechanisms that allow multiple species to coexist in the same community. Understanding these mechanisms is essential for predicting how community may change in response to environmental disturbances and developing effective strategies for conserving and managing biodiversity. However, to apply this theory we need to identify the niche for each specie and model trade-offs. Recently we have proposed techniques FBA-PRCC to apply global sensitivity analysis to the whole-genome bacterial metabolic models. We modelled the sensitivity of bacterial growth to the presence of various external metabolites via FBA-PRCC and created the community sensitivity graph. In that graph, each bacterial species is connected to the sensitive metabolites. Based upon AGORA reconstruction we have prepared the sensitivity graph of the human gut microbiome. In addition, we have collected publicly available human gut metagenomics data from MG-RAST, in which species of the AGORA collection cover more than 50% of the DNA abundance. We are developing graph-based techniques to evaluate the stability of the community composition from its sensitivity graph.

The human gut microbiota is a complex community of micro-organisms that interact within each other but also with their host. These interactions occur essentially at the metabolic level, building a commensal relationship between the host and its microbiota through beneficial exchanges, metabolic niches and competition for nutrients. Disruption of the balance can occur following perturbations such as infections by pathogens. Salmonella enterica Typhimurium is an example of pathogen which benefits from the host immune response and its consequences in the luminal environment in order to build a metabolic niche favourable to its own growth.
In this work, we construct a tri-partite model of Salmonella's infection dynamics in the gut, by relying on genome-scale metabolic modelling. We model the cross-talk between the human epithelial cells, commensals represented by a butyrate-producing bacterium, and Salmonella, and reproduce the mechanisms of the infection by integrating constraint-based modelling predictions into a dynamic framework. Our results accurately predict the known metabolic interactions between the species and the important role of specific metabolites. Overall, our work paves the way to mechanistic and dynamic modelling of complex ecosystems that will help us better understanding the complex network of metabolic interactions between species.

Metagenomic analysis is crucial for understanding microbial communities, necessitating innovative approaches for accurate classification. In this study, we developed a deep learning model with transformer-based architecture, trained end-to-end across multiple microbial taxonomic levels. We focused on long reads, simulating various lengths and introducing errors to accommodate long-read sequencing technologies. Preprocessing included minimizers and tokenization vocabularies. Training and validation datasets were simulated using the NCBI Reference Sequence database.
We evaluated the model's generalization capabilities by validating on genomes not included in the training set and assessing performance on genomes with taxonomic IDs absent from the training sample. We experimented with different read lengths, emphasizing the importance of longer reads. The validation results demonstrate high accuracy in taxonomic classification at higher taxonomic levels. Additionally, we assessed the model's performance on real datasets by examining abundance estimations, finding that our model's estimates aligned well with alternative non-ML tools.
Our research highlights the potential of using transformer-based deep learning models for taxonomic classification in metagenomics. These models capture complex relationships and contextual information across taxonomic levels, providing a foundation for further development. As we continue optimizing these models, they hold promise for improved classification accuracy and a deeper understanding of microbial communities.

Intestinal dysbiosis, an imbalance in the gut microbiome, has been associated with several health conditions, including inflammatory bowel disease, obesity, and diabetes. Microbiome sequencing has emerged as a powerful tool to identify dysbiosis in clinical contexts. However, there is little agreement on which alpha diversity metrics are the most relevant, and few characterizations of a healthy microbiome exist.
We systematically compared the range of alpha metrics and its association with health and disease. We analyzed data from patients in well-curated repositories for different diseases and a healthy population from our Hospital stool bank. We also assessed whether differences in alpha metrics before and after a fecal microbiota transplantation showed predictable responses. Our goal was to provide a justified choice of metric or set of metrics that best represented one's current state of microbiome diversity.
Our results offer insights into the most informative metrics for characterizing microbiome diversity, which can aid in establishing diagnoses, finding suitable donors or recipients of stool transplants, monitoring disease progression, and evaluating the effects of therapy. This study contributes to the growing body of knowledge on dysbiosis and its link to various health conditions, ultimately paving the way for personalized microbiome-based therapies.

Summary: A major bottleneck in understanding newly discovered microbial genes is the lack of functional annotations. Genes participating in the same biological pathways are often found co-localized in the genome as conserved gene clusters. Conservation of gene neighborhood between two genes over a long evolutionary distance is a strong indicator of function association, in addition to sequence homology. Spacedust (Spatially conserved gene cluster search tool) is a fast and sensitive tool for systematic, de novo discovery of conserved gene clusters across multiple genomes, which does not require experimental or functional information. Given a set of genomes, it systematically finds all clusters of matched homologous genes showing significant neighborhood conservation between two genomes. Matches to homologous genes are found by sensitive structure similarity search with Foldseek. Additionally, we improved search speed against a large target database, making it fast enough to support all-vs-all searches of a large number of genomes. We anticipate that spacedust will provide insights on the function and evolutionary conservation for existing complete prokaryotic genomes and novel (meta)genomic contigs.
Availability and implementation: Spacedust is available as an open-source (GPLv3), user-friendly command-line software for Linux and macOS (https://github.com/soedinglab/spacedust).

Long-read assemblers struggle to distinguish between closely related strains and therefore tend to collapse them into a single sequence. This hinders metagenome analysis, as closely related strains present in a sample may have important functional differences. To solve this problem, we present a new pipeline, called HairSplitter, that phases a (partially or totally) collapsed assembly, thereby improving the reconstruction of the genomes of the different strains in a metagenome. The originality of the method lies in a custom variant-calling step that allows HairSplitter to filter out most sequencing errors from a long-read alignment. On simulated datasets comprising up to 10 strains of the same species and on a real dataset containing 5 strains of E. coli, HairSplitter improved on metaFlye, ouperforming Strainberry and stRainy in terms of k-mer completeness, both for low-quality Nanopore and high-quality Pacbio HiFi sequences.

Microbiome next-generation sequencing data are distorted by multiple laboratory and bioinformatic biases. Extraction bias, sequence errors and contamination are major factors blurring true biological signals, and could potentially be corrected by jointly optimizing experimental and computational workflows.
We compared dilution series (10^8-10^4 bacteria) of 3 mock communities with an even or staggered composition. DNA was extracted with 8 different extraction protocols (2 buffers, 2 extraction kits, 2 lysis conditions). Extracted DNA was sequenced (V1-V3 16S) together with corresponding DNA mocks. Sequences were denoised using DADA2, and annotated by exact matching against reference genomes.
Independent of the extraction protocol, contamination increased with less input cells, but interestingly, chimera formation increased with higher input cells. Microbiome composition was significantly different between extraction kits and lysis conditions, but not between buffers. Bias in microbiome composition compared to corresponding DNA mocks revealed that extraction protocols favored specific groups of bacteria. Strikingly, this extraction bias per groups of species was predictable by bacterial cell morphology.
We provide novel explanations that higher DNA density increases chimera formation during PCR amplification, and present a robust link between cell morphology and extraction bias. These findings pave the road for bioinformatic correction of biases in microbiome data.

Microbiome research is currently facing a reproducibility challenge. Numerous laboratory protocols exist for generating microbiome sequencing data, and each laboratory method creates protocol-specific biases in microbiome results, limiting their comparability. Even more concerningly, poor reporting of methods hampers the reproducibility of findings.
To evaluate the FAIR status, and to quantify the impact of specific laboratory choices on protocol bias, we performed a meta-analysis of bacterial mock communities from 52 published microbiome studies. We extracted 67 protocol variables from each study’s methods section, and jointly pre-processed corresponding raw mock sequencing data of 171 samples.
Regarding FAIR principles, we found no increase in raw data depositing over time. Key factors for protocol reproducibility, e.g. PCR details, were frequently not reported. The substantial protocol-specific bias on mock microbiome composition was mainly driven by choice of extraction kit, 16S region, primers, and PCR conditions, jointly explaining up to 96% of variation of bias per bacterial genus.
This unique meta-analysis approach using standardized mock controls revealed striking protocol-specific variation in microbiome data. We provide novel insights into the incomplete implementation of FAIR principles and reporting guidelines, highlighting the need for open and reproducible science for overcoming protocol biases in microbiome research.

Metagenome binning groups contigs assembled from metagenomic samples by their genome of origin, resulting in 'metagenome-assembled genomes' (MAGs). High-quality MAGs are essential for most downstream analyses of microbiomes. We present ’McDevol’ a metagenomic binner which applies a novel Bayesian distance measure using Poisson statistics, agglomerative clustering, and refinement through density-based clustering. Results on a CAMI II dataset demonstrate that McDevol performs better on almost all AMBER quality scores and provides more high-quality MAGs than the widely used binners MetaBAT2, MaxBin2, and CONCOCT. McDevol is fast and memory-efficient, making it a suitable tool for binning contigs of large metagenomic datasets.

Metagenome binning groups contigs assembled from metagenomic samples by their genome of origin, resulting in 'metagenome-assembled genomes' (MAGs). High-quality MAGs are essential for most downstream analyses of microbiomes. We present ’McDevol’ a metagenomic binner which applies a novel Bayesian distance measure using Poisson statistics, agglomerative clustering, and refinement through density-based clustering. Results on a CAMI II dataset demonstrate that McDevol performs better on almost all AMBER quality scores and provides more high-quality MAGs than the widely used binners MetaBAT2, MaxBin2, and CONCOCT. McDevol is fast and memory-efficient, making it a suitable tool for binning contigs of large metagenomic datasets.

Metagenome binning groups contigs assembled from metagenomic samples by their genome of origin, resulting in 'metagenome-assembled genomes' (MAGs). High-quality MAGs are essential for most downstream analyses of microbiomes. We present ’McDevol’ a metagenomic binner which applies a novel Bayesian distance measure using Poisson statistics, agglomerative clustering, and refinement through density-based clustering. Results on a CAMI II dataset demonstrate that McDevol performs better on almost all AMBER quality scores and provides more high-quality MAGs than the widely used binners MetaBAT2, MaxBin2, and CONCOCT. McDevol is fast and memory-efficient, making it a suitable tool for binning contigs of large metagenomic datasets.

In 2019, we developed Autometa, an automated binning pipeline that is able to effectively recover metagenome-assembled genomes from complex environmental and non-model host-associated microbial communities. Autometa has gained widespread use in a variety of environments and has been applied in multiple research projects. However, this genome-binning workflow was at times overly complex and computationally demanding. As a consequence of its diverse application, non-technical and technical researchers alike have noted its burdensome installation and inefficient as well as error prone processes. Moreover its taxon-binning and genome-binning behaviors have remained obscure. For these reasons we set out to improve its accessibility, efficiency and efficacy to further enable the research community during their exploration of Earth’s environments. Here we present the highly augmented Autometa 2 release, available at https://github.com/KwanLab/Autometa. We vastly simplified installation, created a graphical user interface and refactored the workflow for transparency and reproducibility. Furthermore, we conducted a parameter sweep on standardized community datasets to show that it is possible for Autometa to achieve performance better than any other binning pipeline, as judged by Adjusted Rand Index. These improvements enhance the accessibility for non-bioinformatic oriented researchers, scalability for large-scale and highly-complex samples and interpretation of recovered microbial communities.

The gut microbiome influences the host's aging, with significant implications for overall health and disease progression. To unravel these complex interactions, our lab adopts a comprehensive, multi-layered approach, utilizing publicly available data from diverse host tissues to explore associations with age-related diseases and elucidate the impact of leaky gut on aging. We employ causal inference methods to decipher links between the gut microbiome and age-dependent changes in brain structure and function, integrating metabolome and transcriptome data for mechanistic insights into neurodegenerative disorders. Furthermore, we develop individual-specific ecological and network models for longitudinal microbiome profiles, identifying microbial taxa and community properties related to lifespan regulation in killifish, mice, and wild baboons, providing open-source software packages for future research. This poster offers an overview of our rigorous, multi-layered approach, aiming to unravel the intricate relationships between the microbiome and aging.

Secondary metabolites are biomolecules that are not essential for the growth or survival of organisms but offer ecological and physiological advantages to them. They have various applications in medicine, biotechnology, and agriculture. These metabolites are often produced by biosynthetic gene clusters (BGCs), groups of genes located close to each other in a microbial genome.

The advent and widespread availability of metagenomics have enabled the study of BGCs and their associated secondary metabolites directly from environmental samples, eliminating the need to culture individual microorganisms.

Here, we present the BGC Atlas, a web resource that facilitates exploring and analysing the diversity of biosynthetic gene clusters found in various environments through metagenomic sequencing. The BGC Atlas identifies and clusters BGCs from publicly available metagenomic datasets and provides a centralized database for exploring BGCs, their gene cluster families (GCFs), their associations with metadata, and the ability to search for similarities to the identified BGCs.

A web resource enabling researchers to easily explore and analyze the diversity of biosynthetic gene clusters in environmental samples can significantly enhance our understanding of secondary metabolites produced by microorganisms. Additionally, it can promote the identification of ecological and evolutionary factors that influence the biosynthetic potential of microbial communities.

Microbial abundance co-occurrence networks have been used in microbial ecology to infer interactions between taxa. Networks can reflect similar microbial responses to environmental variables, so we hypothesized that urban environmental stressors would decrease connectivity between microbes, and microbes in urban networks would co-occur due to shared tolerance to these stressors. We created microbial networks from urban and rural forest soils and found urban networks were more densely connected than rural networks (F(1,8) = 62.04, p = 4.88e-5). We then used enzyme genome abundances to identify genes enriched in associations. We categorized enriched genes into four categories based on their presence in associating microbes and the type of association between the microbes: shared environmental tolerances (genes present in both microbes in a positively associating pair), functional complementarity (genes present in one microbe in a positively associating pair), combat (genes present in one microbe in a negatively associating pair), and competitive exclusion (genes present in both microbes in a negatively associating pair). We found that most positive associations were likely explained by functional complementarity. Negative associations were likely explained by both combat and competitive exclusion. Future work will calculate the fraction of associations explained by biological vs. environmental mechanisms.

QIIME-2 has become a cornerstone of modern reproducible microbiome data analysis. QIIME-2’s modular Python-based framework includes a variety of analysis plug-ins, ranging from amplicon denoising to statistical analysis. In this contribution, we introduce two new QIIME-2 plugins, q2-classo and q2-gglasso, that enable sparse log-contrast regression, classification, and sparse microbial network estimation from compositional microbiome data. Both plugins connect to underlying stand-alone Python libraries, c-lasso and GGLasso, that provide efficient optimization routines for solving the underlying constrained lasso and graphical lasso problems, respectively. The plugins also contain multiple data transformation methods and interactive visualizations to aid in the analysis of microbial associations and latent variables.
To illustrate the usability and efficacy of our framework, we present several analysis workflows using publicly available soil and gut microbiome datasets and provide corresponding documentation and tutorials to ensure that our methods are easily accessible, highly reproducible, and user-friendly.
Using the described workflows, we also show preliminary results on an ongoing large-scale study, connecting changes in the human gut microbiome to common food allergies.

Metagenomic assembly results in fragmented contigs, which can be further grouped by their genome origins via binning. Many unsupervised binning methods use contig features like oligonucleotide composition and coverage patterns as the basis for sequence clustering. Additional information, such as assembly graphs, have not been intensively used to infer contig relationships for binning. Here, we introduce Mapbin, a novel binning refinement algorithm that enhances the binning results by building a multilayer network that combines the initial binning, assembly graph, and read-pairing information from paired-end sequencing data. Mapbin partitions the network using Infomap, a community-detection algorithm for sequence clustering. Mapbin has been tested on multiple simulated and empirical datasets. The results indicate a significant improvement in bin completeness without compromising purity.
Using Mapbin with the output from multiple binners, we obtained a substantial increase in the number of high-quality bins (>90% completeness, <5% contamination), as well as an overall improvement in the common binning quality metrics. We also demonstrate that Mapbin scales efficiently with growing input data volume, capable of processing a 10-sample assembly with hundreds of thousands of contigs in under 15 minutes. Mapbin is an open-source tool available at https://github.com/u-xixi/mapbin.

Coupling the drop in sequencing costs and the rise of initiatives such as the Human Microbiome Project has resulted in an increasing amount of multi-omics data currently available in public repositories. As microbiome analysis methodologies lack standardization, it is only through the integration of multiple datasets that we can capture the complexity of the microbiome by eliminating the biases associated with individual studies. For this reason, data and related metadata accessibility and reusability are the first pivotal points for current microbiome research.

Here, we present MADAME (MetADAta MicrobiomE), a bioinformatic open-source and easy-to-use tool to facilitate and automate the process of metadata and data retrieval and download from the European Nucleotide Archive (ENA). Moreover, MADAME provides users to search for related publications and visualize the downloaded information by generating a report with graphs and statistics. We applied MADAME to the specific case study of the skin microbiome, exploring the available metadata for a total of 33 projects and 7162 samples.

In conclusion, MADAME can easily provide data and all the related information for downstream analysis, with the ultimate aim of contributing to microbiome research advances by facilitating multiple datasets integration.

Next-generation sequencing technologies continue to deliver vast numbers of novel uncultivated viral sequences and (partial) genomes across all habitats of life, ranging from ocean environments to the human gut. Many of these viral sequences cannot be faithfully attributed to a specific viral taxonomy unit due to the absence of universal marker genes.

Here, we introduce a data-driven two-stage method, virNest, for estimating hierarchical relationships between viruses from viral metagenomic data. In the first step, we exploited the gene-sharing network approach by constructing a network of viral contigs with gene-sharing scores as edges. In the second step, we learn hierarchical partitions of the resulting gene-sharing network using the nested stochastic block model (nSBM), a generative model that infers hierarchical organizations of networks across multiple scales. The hierarchical partitioning of viral sequences not only enabled a data-driven grouping of viruses but also multi-scale downstream statistical analysis.

We demonstrated the validity and competitive performance of virNest on several mock datasets with known viral taxonomy. Furthermore, using an IBD gut microbiome dataset where viral sequences are associated with patients' health metadata, we showed that pairing virNest with statistical tree-aggregation methods can uncover robust associations between viral abundance patterns and patients' health status.

Background
In recent years, the emerging omics technologies have provided unprecedented opportunity to better understand the structural and functional properties of microbial communities. Consequently, there is a growing need for bioinformatic workflows that integrate multi-omics data and allow for comprehensive characterization of microbiomes. Previously, we introduced gNOMO, a bioinformatic pipeline specifically designed to process and analyze multi-omics data in an integrative manner.

Objectives
Here, we present gNOMO2 pipeline to facilitate reproducible and modular analysis of up to four omics levels -16s rRNA gene amplicon sequencing, metagenomics, metatranscriptomics and metaproteomics- of microbiome data in an integrative manner.

Methods:
The gNOMO2 has been developed using the workflow management framework Snakemake in order to obtain an automated and reproducible pipeline. New analysis modules have been developed and integrated to the existing gNOMO pipeline.

Results:
gNOMO2 pipeline includes three new modules that allow analysis of 16S rRNA sequencing data, generation of a custom protein database for metaproteomic analyses and integrated visualization of omics analysis results. Thus, gNOMO2 provides a modular and reproducible tool for extensive taxonomic and functional analyses of microbial communities in both model and non-model organisms and paves the way for new insights in microbiome investigations.

Microorganisms communities studies have been revolutionized by metagenomics and metabarcoding methods. These methods involve high-throughput sequencing of environmental DNA, which gives an in-depth view of the composition of the studied environment with a large number of sequences belonging to different microorganisms. In such studies, researchers very often aim to identify a stable and permanent microbiota related to the studied environment or host, known as the core microbiota. However, current approaches to define core microbiota use arbitrary cut-offs of local and regional abundances, which makes them less statistically robust and subject to various biases. To address this issue, an R package called corebiota is introduced. It is based on the species abundance distribution modeling and partition between satellite and core microbiome. The package provides a function to identify the core microbiome using an index of dispersion and to model whether lineages follow a Poisson distribution. A convenient function is also provided to help make core-satellite plots for visualizing the species abundance distribution in the studied microbiome. The tool was tested on real data and performed well on both time and correctness. The package provides reliable tools for advancing microbiome research and is available at https://github.com/Ebedthan/corebiota.git.

In the past decade, there has been an increase in both private and government space exploration. Recent studies have revealed the existence of microbial species surviving harsh conditions beyond Earth, including the International Space Station (ISS). Microorganisms (including extremophiles), evolve quickly across short generations and can survive in habitats considered extreme, based on temperature, salinity, pressure, pollution, and/or radiation. Many studies have focused on microbial communities for model organisms for simulating life in the extreme. In this work, we explore biosynthetic gene clusters (BGCs) of species from the extreme environment as a new model for understanding the evolution and survival of extremophiles. BGCs have historically been associated with the production of Secondary metabolites (SM) for species survival. We further explored the BGCs from two extreme locations in this study: Lake Hillier in Australia and Gowanus Canal in New York City (NYC) and found that more than 98% of the identified BGCs in these locations are novel and have not been previously characterized, underscoring the capacity of these environments to create novel adaptations to selection pressures.

Metagenome interpretation depends on the recovery of high-quality genomes from short-read data through the application of performant assemblers followed by binners. The Initiative for the Critical Assessment of Metagenome Interpretation (CAMI) analyzed the results of a large number of methods thoroughly on benchmark datasets. Several binners in the CAMI II challenges performed well on gold standards representing perfect assemblies. However, performance was impacted when binners were applied on assemblies of MEGAHIT, the only assembler analyzed in combination with binners. To investigate whether certain methods are particularly capable of recovering high-quality genomes, here we benchmark various combinations of assemblers and binners, as well as present metrics and a framework for this purpose.

Research has shown that skin properties, such as dryness, are related to the composition of the skin microbiome. Therefore, the regular use of mild cleansers, which plays an important role in maintaining the chemical and physical integrity of the stratum corneum, is expected to have a measurable effect on the microbiome. Indeed, in-vitro experiments have shown an association of skin product properties such as pH or presence of fatty acids with the composition of the microbiome. However, the relation between product properties and microbiome composition in in-vivo data can be confounded by the high variability common to such data.
Here we present a novel analysis approach based on a combination of categorical embedding and a feedforward artificial neural network. We apply the method to data from two cohort studies, where different treatments were applied to skin. Microbiome samples were collected and processed for metataxonomic assessment. A numerical analysis of the embedding vector statistics results in a ranking of the treatments which is consistent with clinical measures of product mildness.

Species interactions are essential determinants for microbial ecosystem structure and functioning. To understand how these systems assemble and shape their environment, knowledge of the underlying ecological relationships is key.
Despite their popularity, statistical network prediction methods often overlook the dynamic nature of microbial interactions: species cooperating in one community may be fierce competitors in another, depending on environmental conditions or the presence of other “modulating” species. Such higher-order interactions (HOIs), though potentially common, have not yet been extensively quantified in natural communities.
Here, we introduce a computational framework that identifies HOIs in such complex, natural microbial communities. It uses statistical tests that identify context-dependent co-occurrences to predict potential modulating species and environmental factors, while also minimizing confounding effects like shared-habitat biases.
Upon screening the Human- and Earth Microbiome Project datasets, we found widespread signatures of HOIs across numerous habitats. These HOIs were often centered around uncharacterized modulating hub species, which could potentially have significant, yet unrecognized, impact on their communities. We furthermore observed striking environment-specific signatures, including a pronounced mitigation of predicted positive interactions within the human gut, indicative of frequent swaps of cross-feeding partners. Our method thus provides deeper insights on the dynamic nature of complex, natural microbial networks.

Short (~250-400 nt) amplicons provide low taxonomic resolution for the majority of sequences in microbiome studies, limiting insights into community diversity. Moreover, PCR reactions needed to amplify the target region distort original abundances. To tackle these challenges, we developed an experimental and computational approach using long amplicons sequenced with Oxford Nanopore Technologies (ONT), and unique molecular identifiers (UMIs). UMIs serve two purposes: to trace back unique biological fragments and improve microbial abundance estimates, and second, reducing sequence errors by allowing the generation of consensus sequences from sequence reads with the same UMI (PCR duplicates). In this way, we double the number of sequences annotated at species level.
With this approach, we analyzed high resolution temporal profiles of Wadden Sea protist (micro-eukaryotic) communities in surface waters. Marine protists produce and recycle most of the ocean’s primary organic matter, thus heavily impacting global biogeochemical cycles and marine food-webs. An intricate interplay of microbial interactions is needed for these tasks, but these interactions are poorly understood. We performed causal network modelling to retrieve candidate protist interactions. These predictions help disentangling metabolic dependencies between different protists and dependencies on environmental conditions, potentially even allowing predictions of how protist communities respond to climate change.

A wide range of differential abundance analysis tools have been proposed addressing one of the key questions in microbiome research, namely the identification of microbial taxa that change between various conditions. However, the increasing number of available tools also poses a challenge for researchers in selecting the most appropriate one for their analysis. To address this issue, we are conducting a comprehensive benchmark study that aims at evaluating the performance of differential abundance tools for microbiome data in a most unbiased fashion. To do so we adhere to established principles and best-practices of study design and analysis that have been proven in the context of clinical trials. We incorporate a diverse collection of real microbiome datasets as well as generated synthetic datasets based on known ground truth compositions, allowing for a systematic evaluation. Additionally we characterize the datasets using a carefully selected set of properties enabling an in depth characterization of the performances of the studied differential abundance tools.

Motivation: Amoebae are identified as potential carriers and hosts of pathogenic bacteria within dental unit water sources. Understanding the relationship between amoebae and the dental unit microbiome is crucial for predicting the risk of amoebal infection and developing effective infection control strategies.
Methods: This study utilized ridge regression to predict the presence of amoebal infection in dental unit water sources based on the unit’s microbiome composition. Water samples from dental units were collected and analyzed for their microbial communities. High-throughput sequencing and q-PCR were performed to identify bacterial species and the presence of amoebae, respectively. Then the probability of amoebae presence was predicted based on the microbiome using ridge regression.
Results: The analysis revealed a significant association between amoebal infection and specific microbial patterns in the dental unit water microbiome. The developed machine learning models exhibited high accuracy (AUC= 0.867) in predicting amoebal presence, demonstrating the predictive potential of the microbial composition. Key microbial taxa including Legionella and Pseudomonas contributing to the predictive models were identified, highlighting their potential as biomarkers for amoebal presence. By identifying specific microbial signatures associated with amoebal presence, this study could contribute to enhancing preventive measures and improving patient safety in dental practice settings.

Marine microorganisms have evolved strategies for acclimating to dynamic environmental conditions. Some of these acclimation strategies can be detected in the differential expression of genes, measured in situ by sequencing the RNA transcripts of microbial communities. To date, the analysis of these environmental metatranscriptomes consists primarily of pairwise comparisons of single genes, complicated by challenges of sparse, noisy data, variable organism abundances, and limited prior knowledge about gene functions. Here we introduce a novel multi-way gene clustering method for unsupervised exploration of gene expression patterns in environmental metatranscriptomes. Built on sparse tensor decomposition, the method faithfully recovers multiple overlapping clusters from noisy simulated datasets. We applied the method to cyanobacterial metatranscriptomes and recovered dozens of robust clusters representing correlated gene expression patterns exhibited by Prochlorococcus and Synechococcus communities found in the North Pacific. These clusters are enriched for ecologically relevant processes such as nitrogen assimilation and acclimation to iron scarcity, and they exhibit distinctive spatial and temporal patterns. In addition to presenting a proof-of-concept applying tensor decomposition to environmental gene expression data, these results enhance our understanding of the molecular processes that govern acclimation to variable marine environments, and offer testable hypotheses about the functions of previously uncharacterized genes.

The gut bacterial population (microbiome) is known to play a significant role in the host health and homeostasis. Disruption of the gut microbiome has been reported in the neurodegenerative disorder Huntington’s disease (HD) however, the fungal community (gut mycobiome) has largely been overlooked despite its role in host homeostasis.

Here we report the first evidence of gut mycobiome disruption in a mouse model of Huntington’s disease. Briefly, shotgun metagenomics sequencing was performed on mice faecal samples were collected from 4 to 12 weeks of age which coincided with young and early disease stage respectively. Overall, the gut mycobiome composition was significantly different in the HD mice only at the early disease stage before overt motor symptoms appear but not in any early life stages. Key discriminatory fungal species at early disease stage were identified using Sparse Partial Least Square Discriminant Analysis (sPLS-DA). Integration of bacterial and fungal metagenomics dataset using DIABLO revealed perturbations in associations between Bacteroides spp. and the signature of discriminatory fungal species. Modelling of temporal trajectory of each gut fungal species using Linear Mixed Modelling Splines and sparse-PCA identified fungal clusters with similar temporal pattern of abundance that were different between groups.

The harsh conditions of outer space create unique selective pressures on microorganisms. This study focuses on the functional annotation of seven Gram-positive bacterial isolates derived from the International Space Station (ISS) and Jet Propulsion Laboratory-Spacecraft Assembly Facilities (JPL-SAF) during the Mars 2020 mission, including a representative of a new genus. Using genome assembly and the machine learning-based functional annotation tool Deep Functional Residue Information (DeepFRI), as well as sequence-based homology and orthology analyses, we compared the predicted functional characteristics of these microorganisms with their closest Earth-bound relatives, including genes associated with radiation resistance, microgravity adaptation, stress response, and metabolic rearrangements. By analyzing the genomes and possible protein-coding sequences of these isolates, we have identified common features associated with adaptations to space conditions. These adaptations include the use of mechanosensitive channel proteins to mitigate microgravity-related hypoosmotic stress, DNA repair systems and mobile genetic elements to combat increased radiation exposure. Our study demonstrates the superior coverage of DeepFRI’s functional annotations compared to homology-based tools and highlights the potential of knowledge-based genome mining to enhance our understanding of previously-uncharacterized microbial adaptation to extreme conditions Our findings also provide a set of biomarkers that could aid in astrobiological studies targeting life on other planets.

Integration of multi-omics data is a challenging but necessary step to advance our understanding of the biology underlying human health and disease processes. To date, investigations seeking to integrate multi-omics (e.g., microbiome and metabolome) employ simple correlation-based network analyses; however, these methods are not always well-suited for microbiome analyses because they do not accommodate the excess zeros typically present in these data. In this paper, we introduce a bivariate zero-inflated negative binomial (BZINB) model-based network and module analysis method that addresses this limitation and improves microbiome-metabolome correlation-based model fitting by accommodating excess zeros. We use real and simulated data based on a multi-omics study of childhood oral health (ZOE 2.0; investigating early childhood dental caries, ECC) and find that the accuracy of the BZINB model-based correlation method is superior compared to Spearman's rank and Pearson correlations in terms of approximating the underlying relationships between microbial taxa and metabolites. The new method, BZINB-iMMPath, facilitates the construction of metabolite-species and species-species correlation networks using BZINB and identifies modules of (i.e., correlated) species by combining BZINB and similarity-based clustering.
This method was applied in the ZOE 2.0 study for oral microbiome-metabolome data analysis. This project has been supported by R03DE028983, UL1TR001111, and AAOF.

Bacteria-infecting viruses (phages) have been associated with gut microbiome homeostasis and diseases such as colorectal cancer (CRC). Previous work has shown qualitative differences in prophage composition. However, the activity of inducible prophages remains poorly described, primarily due to their genetic diversity and limited representation of their genomes in reference databases.
To address this issue, we collected and assembled publicly available metagenomic CRC datasets (n=1,378). We identified circularised sequences using a tool (mVIRs). Next, we compared results with other computational prophage prediction tools (VirSorter2, VIBRANT, phigaro). After assigning prophages to their metagenome-assembled genome of origin, we calculated phage-to-host (PtoH) ratios and determined differential phage induction levels between asymptomatic individuals and CRC patients.
Overall, a minority of predicted phages were present in circularised form across the samples. However, differential activity profiles could be identified for a number of prophages in asymptomatic individuals compared to CRC patients.
The signatures can be used to study the mechanisms behind these associations further. We aim to use functional prediction to characterize the potential role of prophages in disease development and identify possible interactions with bacterial and human host proteomes.
This study is, to our knowledge, the first one to quantify PtoH ratios in CRC.

The exploration of a large pig gut shotgun metagenomic dataset allowed us to characterize the microbiome composition of post-weaning piglets from a taxonomic, as well as from a functional perspective. The power of longitudinal sampling allowed for the reconstruction of over 51,000 MAGs, 12,400 of which were nearly complete. This allowed us to observe a strong phylogenetic succession within the piglet gut microbiome over the course of 5 weeks after weaning. Furthermore, MAGs and contigs were annotated at a functional level, with focus on the carbohydrate metabolism (doi: 10.1099/mgen.0.000501), as well as on other microbial metabolic processes, allowing the observation of functional succession and to a species to function mapping (new findings).

Gingivitis is a prevalent inflammatory disease resulting from oral bacterial accumulation, often leading to periodontitis and tooth loss. Chios Mastic, a natural resin from mastic trees, possesses potent antimicrobial properties and has been shown to reduce specific oral bacterial populations when chewed as gum. This study investigates the impact of Chios Mastic on the oral microbiome and explores the molecular mechanisms of the host immune response. Seventeen gingivitis and early-stage periodontitis patients were divided into three groups: a control group, a group chewing gum with 25% Chios Mastic, and a group using 100% Chios Mastic oil. Clinical evaluations and molecular analyses were performed using samples collected at various time points. Shotgun metagenomics and metatranscriptomics are employed to detect alterations in oral microbiome composition and gene expression of bacteria, while RNA and microRNA sequencing are used to examine host gene expression in gingival tissues and gingival crevicular fluid. Additionally, a meta-analysis of subgingival metatranscriptomics data from previous studies was conducted, comparing findings with our research. In conclusion, this comprehensive study combining clinical assessments, molecular analyses, and a meta-analysis contributes to our understanding of the therapeutic potential of Chios Mastic in modulating the oral microbiome.