Posters - Schedules

Posters Home

Microorganisms form a complex network of ecological interactions. These interactions can have a positive, negative, or no impact on the species involved. Mutualism is a form of symbiotic interaction in which partnership quality varies from high to low; some symbiont genotypes increase plant fitness, while others may hoard benefits. In this study, we leverage genetic variation from a model symbiosis in the wild and use an integrated approach that combines dual RNA-seq transcriptomics with quantitative symbiosis traits and long-read genome assemblies to predict the underlying molecular pathways/genes responsible for varying symbiotic partner quality. This study used 20 wild Sinorhizobium meliloti strains in single-strain inoculation experiments with the Medicago truncatula (DZA) host. We explored the transcriptomic data from both partners in planta (from symbiotic nodules) and measured plant traits. We found gene clusters on symbiosis plasmid (pSymA) that are exclusively absent in either high- or low-quality partners. We also explored the neighborhood of these clusters and identified known symbiosis-related genes, transporters and transposases suggesting co-regulation. WGCNA analysis of plants and bacteria genes revealed several modules significantly correlated with each other and host fitness.

Motivation: The microbiome has been shown to impact both host development, normal metabolic processes, as well as the pathogenesis of various diseases, and the engineering the gut microbiome for the treatment of such diseases is an exciting new direction in medical science. The understanding of how to modulate a patient’s microbiome requires accurate modeling of the dynamic nature of the microbiome community in conjunction with multiple host characteristics and environmental factors. Here we present DiRLaM, a deep-learning framework combining an autoencoder with a novel Beta-diversity regularization and a neural network for modeling microbiome dynamics.

Results: Using synthetic and real-world longitudinal datasets, we show that DiRLaM provides a more robust interpolation under increasing levels of noise compared to standard B-Spline interpolations. DiRLaM also outperforms the state-of-the-art dynamic Bayesian network model for predicting subsequent microbiome communities in longitudinal data. Additionally, we demonstrate DiRLaM’s ability to identify significant host characteristics and environmental factors contributing to the dynamics of the microbiome community.

Conclusion: DiRLaM utilizes an autoencoder with a novel Beta-diversity regularization and deep neural network to model microbiome dynamics. DiRLaM was more robust and accurate when modeling longitudinal microbiome data and was able to infer informative interactions in real-world longitudinal microbiome datasets.

Carbohydrate Active EnZymes (CAZymes) are important for microbial communities to thrive in carbohydrate-rich environments such as animal guts, agricultural soils, forest floors, and ocean sediments. Since 2017, microbiome sequencing and assembly have produced numerous metagenome-assembled genomes (MAGs). We have updated our dbCAN-seq database (https://bcb.unl.edu/dbCAN_seq) to include the following new data and features: (i) ∼498 000 CAZymes and ∼169 000 CAZyme gene clusters (CGCs) from 9421 MAGs of four ecological (human gut, human oral, cow rumen, and marine) environments; (ii) Glycan substrates for 41 447 (24.54%) CGCs inferred by two novel approaches (dbCAN-PUL homology search and eCAMI subfamily majority voting) (the two approaches agreed on 4183 CGCs for substrate assignments); (iii) A redesigned CGC page to include the graphical display of CGC gene compositions, the alignment of query CGC and subject PUL (polysaccharide utilization loci) of dbCAN-PUL, and the eCAMI subfamily table to support the predicted substrates; (iv) A statistics page to organize all the data for easy CGC access according to substrates and taxonomic phyla; and (v) A batch download page. In summary, this updated dbCAN-seq database highlights glycan substrates predicted for CGCs from microbiome. Future work will implement the substrate prediction function in our dbCAN2 web server.

Beta-diversity quantitatively measures the difference among microbial communities, thus enlightening the association between microbiome composition and environment properties or host phenotypes. The beta-diversity analysis mainly relies on distances among microbiomes that are calculated by all microbial features. However, in some cases, only a small fraction of members in a community plays crucial roles. Such tiny proportion is insufficient to alter the overall distance, which is always missed by end-to-end comparison. On the other hand, beta-diversity pattern can also be interfered due to the data sparsity when only focusing on non-abundant microbes. Here we develop Flex Meta-Storms (FMS) distance algorithm that implements the “local alignment” of microbiomes for the first time. Using a flexible extraction that considers the weighted phylogenetic and functional relations of microbes, FMS produces a normalized phylogenetic distance among members of interest for microbiome pairs. We demonstrated the advantage of FMS in detecting the subtle variations of microbiomes among different states using artificial and real datasets, which were neglected by regular distance metrics. Therefore, FMS elucidates beta-diversity with higher sensitivity and flexibility, thus contributing to in-depth comprehension of microbe-host interactions, as well as promoting the utilization of microbiome data such as disease screening and prediction.

Beta-diversity quantitatively measures the difference among microbial communities, thus enlightening the association between microbiome composition and environment properties or host phenotypes. The beta-diversity analysis mainly relies on distances among microbiomes that are calculated by all microbial features. However, in some cases, only a small fraction of members in a community plays crucial roles. Such tiny proportion is insufficient to alter the overall distance, which is always missed by end-to-end comparison. On the other hand, beta-diversity pattern can also be interfered due to the data sparsity when only focusing on non-abundant microbes. Here we develop Flex Meta-Storms (FMS) distance algorithm that implements the “local alignment” of microbiomes for the first time. Using a flexible extraction that considers the weighted phylogenetic and functional relations of microbes, FMS produces a normalized phylogenetic distance among members of interest for microbiome pairs. We demonstrated the advantage of FMS in detecting the subtle variations of microbiomes among different states using artificial and real datasets, which were neglected by regular distance metrics. Therefore, FMS elucidates beta-diversity with higher sensitivity and flexibility, thus contributing to in-depth comprehension of microbe-host interactions, as well as promoting the utilization of microbiome data such as disease screening and prediction.

Beta-diversity quantitatively measures the difference among microbial communities, thus enlightening the association between microbiome composition and environment properties or host phenotypes. The beta-diversity analysis mainly relies on distances among microbiomes that are calculated by all microbial features. However, in some cases, only a small fraction of members in a community plays crucial roles. Such tiny proportion is insufficient to alter the overall distance, which is always missed by end-to-end comparison. On the other hand, beta-diversity pattern can also be interfered due to the data sparsity when only focusing on non-abundant microbes. Here we develop Flex Meta-Storms (FMS) distance algorithm that implements the “local alignment” of microbiomes for the first time. Using a flexible extraction that considers the weighted phylogenetic and functional relations of microbes, FMS produces a normalized phylogenetic distance among members of interest for microbiome pairs. We demonstrated the advantage of FMS in detecting the subtle variations of microbiomes among different states using artificial and real datasets, which were neglected by regular distance metrics. Therefore, FMS elucidates beta-diversity with higher sensitivity and flexibility, thus contributing to in-depth comprehension of microbe-host interactions, as well as promoting the utilization of microbiome data such as disease screening and prediction.

Recently, interest in the black soldier fly larvae (BSFL) gut microbiome has received increased attention primarily due to their role in waste bioconversion. However, there is a lack of information on the activities of the gut microbiomes and enzymes (CAZyme families) acting on lignocellulose. In this study, BSFL were subjected to lignocellulose-rich diets: chicken feed (CF), chicken manure (CM), brewers’ spent grain (BSG), and water hyacinth (WH). mRNA libraries were prepared, and RNA-Sequencing was conducted using the PCR-cDNA approach through the MinION sequencing platform. Our results demonstrated that BSFL bred on BSG and WH had the highest abundance of Bacteroides and Dysgonomonas. The presence of GH51 and GH43_16 enzyme families containing both α-L-arabinofuranosidases and exo-alpha-L-arabinofuranosidase 2 activity were common in guts of BSFL reared on the highly lignocellulosic WH and BSG diets. Gene clusters that encode hemicellulolytic arabinofuranosidases in the CAZy family GH51 were also identified. These findings provide novel insight into the shift of gut microbiomes and the potential role of BSFL in the bioconversion of various highly lignocellulosic substrates to fermentable sugars for subsequent value-added products e.g. bioethanol. Further research on the role of these enzymes to improve existing technologies and their biotechnological applications is crucial.

Nitrous oxide (N₂O) emissions from soils, with agricultural land as a significant source, have become one of the primary causes of global warming. It is suggested that, in addition to physical and chemical processes, the nitrogen metabolism carried out by soil microorganisms plays a crucial role in this phenomenon. In order to investigate this relationship, we have launched a citizen science project named "Soil in a Bottle", and have been collecting the bacterial community and N2O emission data for diverse soil samples in Japan. To date, we have collected and analyzed more than 1,400 soil samples with the cooperation of over 650 citizen scientists, and are progressing in exploring soil microorganisms involved in N2O emissions using multivariate analysis and machine learning techniques. At this conference, we would also like to discuss the quality assessment, visualization, and research applications for the large-scale microbiome dataset constructed using a citizen science approach.

In Bangladesh, there has been a simultaneous increase in Dengue and COVID-19 cases over the past two years, presenting a significant public health challenge. Dengue cases surged in Bangladesh in June 2021 and 2022, coinciding with COVID-19 surges caused by the Delta and Omicron variants, respectively. Dengue and COVID-19 have almost identical symptoms, leading to misdiagnosis and mistreatment. To address this double threat, computational analysis, epidemiological studies, serotype surveillance, and structured community participation are critical. Our research focuses on the incremental cases and patterns of both Dengue and COVID-19 in rural and urban areas of Bangladesh. In 2022, Bangladesh experienced the second-highest number of hospitalized Dengue cases on record, with 62,382 reported cases (the highest number of 101,354 cases was in 2019), and also the highest number of Dengue-related deaths in its history, with 281 fatalities reported (the second-highest number of 164 was in 2019). We are developing a computational pipeline to analyze both diseases and separate their subtle patterns to aid management. Additionally, we aim to design a cost-effective dual-pathogen RT-PCR kit that can simultaneously detect SARS-CoV-2 and Dengue in a single reaction and an amplicon primer scheme for sequencing all Dengue virus serotypes.

Plants in natural ecosystems face diverse pathogens that can negatively impact their health and productivity. Predicting the disease and health status of plants using computational approaches that link microbiome composition to host phylotypes can help develop experimental strategies to control pathogens. In this study, we analyzed the microbiome of Arabidopsis thaliana samples from natural populations, infected and uninfected with the pathogen Albugo laibachii, using machine learning algorithms. We trained six supervised machine learning classifiers to differentiate infected and uninfected plants based on microbiome data. Classification achieved highest performance with 88% accuracy being the highest among the models. We identified key microbes that discriminate between infected and uninfected plants using recursive feature elimination approaches. We further examined pairwise interactions between microbes using microbial networks, totaling 2543 microbes. We evaluated experimentally the predictive model's results by selecting candidates from distinct groups of microbes to compete against the pathogen A. laibachii. This study highlights the potential of machine learning algorithms in accurately distinguishing microbiome patterns of plant infections and identifying key microbes that contribute to disease resistance. Our findings provide insights into the complex microbial mechanisms underlying plant-microbe interactions and may have implications for the development of microbiome-based strategies for disease control.