PreMiEr Publications

Cummings CL, Landreville KD and Kuzma J (2025) Natural vs. genetically engineered microbiomes: understanding public attitudes for indoor applications and pathways for future engagement. Front. Genet. 16:1560601. doi: 10.3389/fgene.2025.1560601

This study examines public preferences for natural microbiomes and support for genetically engineered (GE) microbiomes in the built environment, focusing on the demographic, sociographic, and attitudinal factors that influence these preferences. Using data from a nationally representative survey of 1,000 U.S. adults, we employed hierarchical regression analyses to assess the relative contribution of these variables. While demographic and sociographic factors explained limited variance, topic-specific attitudes, including positive perceptions of microbiome engineering’s potential to improve quality of life, were the most significant predictors of support. Conversely, age, distrust in science, and perceived knowledge negatively influenced support for GE microbiomes, reflecting skepticism among some audiences. The findings highlight the potential of the Responsible Research and Innovation (RRI) framework to align the development of microbiome engineering with societal values and to address diverse public perspectives. This research provides actionable insights for policymakers, researchers, and communicators seeking to navigate the complexities of public engagement with emerging biotechnologies.

Kim Y, Ihrie V, Shi P, Ihrie MW, JT, Meares A, Granek J, Gunsch C, Ingram J. 2025. Glucagon-like peptide 1 receptor (Glp1r) deficiency does not appreciably alter airway inflammation or gut-lung microbiome axis in a mouse model of obese allergic airways disease and bariatric surgery. J Asthma Allergy 18:285-305. https://doi.org/10.2147/JAA.S478329

Purpose: High body mass index (≥ 30 kg/m²) is associated with asthma severity, and nearly 40% of asthma patients exhibit obesity. Furthermore, over 40% of patients with obesity and asthma that receive bariatric surgery no longer require asthma medication. Increased levels of glucagon-like peptide 1 (GLP-1) occur after bariatric surgery, and recent studies suggest that GLP-1 receptor (GLP-1R) signaling may regulate the gut microbiome and have anti-inflammatory properties in the lung. Thus, we hypothesized that increased GLP-1R signaling following metabolic surgery in obese and allergen-challenged mice leads to gut/lung microbiome alterations, which together contribute to improved features of allergic airways disease.
Methods: Male and female Glp1r-deficient (Glp1r^−/−) and replete (Glp1r^+/+) mice were administered high fat diet (HFD) to induce obesity with simultaneous intranasal challenge with house dust mite (HDM) allergen to model allergic airway disease with appropriate controls. Mice on HFD received either no surgery, sham surgery, or vertical sleeve gastrectomy (VSG) on week 10 and were sacrificed on week 13. Data were collected with regard to fecal and lung tissue microbiome, lung histology, metabolic markers, and respiratory inflammation.
Results: HFD led to metabolic imbalance characterized by lower GLP-1 and higher leptin levels, increased glucose intolerance, and alterations in gut microbiome composition. Prevalence of bacteria associated with short chain fatty acid (SCFA) production, namely Bifidobacterium, Lachnospiraceae UCG-001, and Parasutterella, was reduced in mice fed HFD and positively associated with serum GLP-1 levels. Intranasal HDM exposure induced airway inflammation. While Glp1r^−/− genotype affected fecal microbiome beta diversity metrics, its effect was limited.
Conclusion: Herein, GLP-1R deficiency had surprisingly little effect on host gut and lung microbiomes and health, despite recent studies suggesting that GLP-1 receptor agonists are protective against lung inflammation.

Plain Language Summary: Asthma is a chronic lung disease that affects over 20 million Americans. In addition, obesity is a major worldwide healthcare problem whose impact is increasing each year. Asthma patients with obesity have worse symptoms, use more medications and have reduced asthma control than lean patients. Also, both asthma and obesity are linked to changes in the normal bacteria that live in the gut and the airways. Weight loss, either through bariatric surgery or through diet and exercise, improves asthma symptoms, through unknown mechanisms. One way that bariatric surgery may impact asthma is through the increase in beneficial gut hormones (GLP-1) and restoration of altered bacteria in the gut and airways. Our study shows that overall, obese mice experience changes in hormones involved in weight gain and glucose metabolism as well as gut bacteria, particularly those that may produce important factors that could improve health. We did not find any links between changes in the bacteria in the gut or lung and features of airway disease or effects from bariatric surgery in the mice. Mice that lacked GLP-1 activity did not differ from normal mice for weight, glucose tolerance, or prevalence of bacteria in the lung. Future studies will explore the timeline of these bacterial changes and the role of the gut and lung microbiomes in human obesity-associated asthma.

Sun G and Zhou Y. 2024. Predicting microbiome growth dynamics under environmental perturbations. Appl. Microbiol. 4:948-958. https://doi.org/10.3390/applmicrobiol4020064

MicroGrowthPredictor is a model that leverages Long Short-Term Memory (LSTM) networks to predict dynamic changes in microbiome growth in response to varying environmental perturbations. In this article, we present the innovative capabilities of MicroGrowthPredictor, which include the integration of LSTM modeling with a novel confidence interval estimation technique. The LSTM network captures the complex temporal dynamics of microbiome systems, while the novel confidence intervals provide a robust measure of prediction uncertainty. We include two examples—one illustrating the human gut microbiota composition and diversity due to recurrent antibiotic treatment and the other demonstrating the application of MicroGrowthPredictor on an artificial gut dataset. The results demonstrate the enhanced accuracy and reliability of the LSTM-based predictions facilitated by MicroGrowthPredictor. The inclusion of specific metrics, such as the mean square error, validates the model’s predictive performance. Our model holds immense potential for applications in environmental sciences, healthcare, and biotechnology, fostering advancements in microbiome research and analysis. Moreover, it is noteworthy that MicroGrowthPredictor is applicable to real data with small sample sizes and temporal observations under environmental perturbations, thus ensuring its practical utility across various domains.

Background: Understanding the cause vs consequence relationship of gut inflammation and microbial dysbiosis in inflammatory bowel diseases (IBD) requires a reproducible mouse model of human-microbiota-driven experimental colitis.

Results: Our study demonstrated that human fecal microbiota transplant (FMT) transfer efficiency is an underappreciated source of experimental variability in human microbiota-associated (HMA) mice. Pooled human IBD patient fecal microbiota engrafted germ-free (GF) mice with low amplicon sequence variant (ASV)-level transfer efficiency, resulting in high recipient-to-recipient variation of microbiota composition and colitis severity in HMA Il-10^-/- mice. In contrast, mouse-to-mouse transfer of mouse-adapted human IBD patient microbiota transferred with high efficiency and low compositional variability resulting in highly consistent and reproducible colitis phenotypes in recipient Il-10^-/- mice. Engraftment of human-to-mouse FMT stochastically varied with individual transplantation events more than mouse-adapted FMT. Human-to-mouse FMT caused a population bottleneck with reassembly of microbiota composition that was host inflammatory environment specific. Mouse-adaptation in the inflamed Il-10^-/- host reassembled a more aggressive microbiota that induced more severe colitis in serial transplant to Il-10^-/- mice than the distinct microbiota reassembled in non-inflamed WT hosts.

Conclusions: Our findings support a model of IBD pathogenesis in which host inflammation promotes aggressive resident bacteria, which further drives a feed-forward process of dysbiosis exacerbated by gut inflammation. This model implies that effective management of IBD requires treating both the dysregulated host immune response and aggressive inflammation-driven microbiota. We propose that our mouse-adapted human microbiota model is an optimized, reproducible, and rigorous system to study human microbiome-driven disease phenotypes, which may be generalized to mouse models of other human microbiota-modulated diseases, including metabolic syndrome/obesity, diabetes, autoimmune diseases, and cancer.

Cummings C, Landreville KD and Kuzma J. 2024. Taking the temperature of the United States public regarding microbiome engineering. Front. Public Health 12:1477377. doi: 10.3389/fpubh.2024.1477377

This paper presents the first representative survey of U.S. adults’ opinions on microbiome engineering within the built environment, revealing public awareness, perceived benefits and risks, and attitudes toward genetically engineered microbiomes. Using data from a cross-sectional survey of 1,000 nationally representative U.S. residents over 18 years of age, we examined demographic and cultural factors influencing public sentiment. Results indicate that younger generations report higher knowledge levels, optimism, and perceived benefits of microbiome engineering, while older generations exhibit more caution and concern about risks. Political affiliation, education level, and trust in science also shape public attitudes, with Democrats, college-educated individuals, and those with higher trust in science more likely to view microbiome engineering positively. Notably, nearly half of respondents across demographic groups remain uncertain about the technology’s benefits and risks, and a majority of participants support government oversight to ensure ethical and responsible development. These insights provide a foundation for policymakers and researchers to foster informed public engagement and guide responsible innovation in microbiome engineering for built environments.

PreMiEr-Associated Project

Moon JF, Kunkleman S, Taylor W, Harris A, Gibas CJ, Schlueter JA. 2024. A gold standard dataset and evaluation of methods for lineage abundance estimation from wastewater. Science of The Total Environment:174515. doi.org/10.1016/j.scitotenv.2024.174515

During the SARS-CoV-2 pandemic, genome-based wastewater surveillance sequencing has been a powerful tool for public health to monitor circulating and emerging viral variants. As a medium, wastewater is very complex because of its mixed matrix nature, which makes the deconvolution of wastewater samples more difficult. Here we introduce a gold standard dataset constructed from synthetic viral control mixtures of known composition, spiked into a wastewater RNA matrix and sequenced on the Oxford Nanopore Technologies platform. We compare the performance of eight of the most commonly used deconvolution tools in identifying SARS-CoV-2 variants present in these mixtures. The software evaluated was primarily chosen for its relevance to the CDC wastewater surveillance reporting protocol, which until recently employed a pipeline that incorporates results from four deconvolution methods: Freyja, kallisto, Kraken 2/Bracken, and LCS. We also tested Lollipop, a deconvolution method used by the Swiss SARS-CoV-2 Sequencing Consortium, and three additional methods not used in the C-WAP pipeline: lineagespot, Alcov, and VaQuERo. We found that the commonly used software Freyja outperformed the other CDC pipeline tools in correct identification of lineages present in the control mixtures, and that the VaQuERo method was similarly accurate, with minor differences in the ability of the two methods to avoid false negatives and suppress false positives. Our results also provide insight into the effect of the tiling primer scheme and wastewater RNA extract matrix on viral sequencing and data deconvolution outcomes.

McCumber AW, Kim YJ, Granek J, Tighe RM, Gunsch CK. 2024. Soil exposure modulates the immune response to an influenza challenge in a mouse model. Science of The Total Environment 922:170865. DOI: 10.1016/j.scitotenv.2024.170865

There is increasing evidence that early life microbial exposure aids in immune system maturation, more recently known as the “old friends” hypothesis. To test this hypothesis, 4-week-old mice were exposed to soils of increasing microbial diversity for four weeks followed by an intranasal challenge with either live or heat inactivated influenza A virus and monitored for 7 additional days. Perturbations of the gut and lung microbiomes were explored through 16S rRNA amplicon sequencing. RNA-sequencing was used to examine the host response in the lung tissue through differential gene expression. We determined that compared to the gut microbiome, the lung microbiome is more susceptible to changes in beta diversity following soil exposure with Lachnospiraceae ASVs accounting for most of the differences between groups. While several immune system genes were found to be significantly differentially expressed in lung tissue due to soil exposures, there were no differences in viral load or weight loss. This study shows that exposure to diverse microbial communities through soil exposure alters the gut and lung microbiomes resulting in differential expression of specific immune system related genes within the lung following an influenza challenge.

Hardwick, A., Cummings, C., Graves, J. et al. Can societal and ethical implications of precision microbiome engineering be applied to the built environment? A systematic review of the literature. Environ Syst Decis (2024). https://doi.org/10.1007/s10669-024-09965-y

The goal of engineering the microbiome of the built environment is to create places and spaces that are better for human health. Like other emerging technologies, engineering the microbiome of the built environment may bring considerable benefits but there has been a lack of exploration on its societal implication and how to engineer in an ethical way. To date, this topic area has also not been pulled together into a singular study for any systematic review or analysis. This study fills this gap by providing the first a systematic review of societal and ethical implications of engineering microbiomes and the application of this knowledge to engineering the microbiome of the built environment. To organize and guide our analysis, we invoked four major ethical principles (individual good/non-maleficence, collective good/beneficence, autonomy, and justice) as a framework for characterizing and categorizing 15 distinct themes that emerged from the literature. We argue that these different themes can be used to explain and predict the social and ethical implications of engineering the microbiome of the built environment that if addressed adequately can help to improve public health as this field further develops at global scales.

PreMiEr-Associated Project

Prince J, Jones AD, 3rd. 2023. Heterogenous biofilm mass-transport model replicates periphery sequestration of antibiotics in Pseudomonas aeruginosa PAO1 microcolonies. Proc Natl Acad Sci U S A 120:e2312995120.

https://doi.org/10.1073/pnas.2312995120

A model for antibiotic accumulation in bacterial biofilm microcolonies utilizing heterogenous porosity and attachment site profiles replicated the periphery sequestration reported in prior experimental studies on Pseudomonas aeruginosa PAO1 biofilm cell clusters. These P. aeruginosa cell clusters are in vitro models of the chronic P. aeruginosa infections in cystic fibrosis patients which display recalcitrance to antibiotic treatments, leading to exacerbated morbidity and mortality. This resistance has been partially attributed to periphery sequestration, where antibiotics fail to penetrate biofilm cell clusters. The physical phenomena driving this periphery sequestration have not been definitively established. This paper introduces mathematical models to account for two proposed physical phenomena driving periphery sequestration: biofilm matrix attachment and volume-exclusion due to variable biofilm porosity. An antibiotic accumulation model which incorporated these phenomena better fit observed periphery sequestration data compared to previous models.

PreMiEr-Associated Project

Childs SK, Jones AAD. 2023. A microtiter peg lid with ziggurat geometry for medium-throughput antibiotic testing and in situ imaging of biofilms. Biofilm 6:100167.

doi.org/10.1016/j.bioflm.2023.100167

Bacteria biofilm responses to disinfectants and antibiotics are quantified and observed using multiple methods, though microscopy, particularly confocal laser scanning microscopy (CLSM) is preferred due to speed, a reduction in user error, and in situ analysis. CLSM can resolve biological and spatial heterogeneity of biofilms in 3D with limited throughput. The microplate peg-lid-based assay, described in ASTM E2799-22, is a medium-throughput method for testing biofilms but does not permit in situ imaging. Breaking off the peg, as recommended by the manufacturer, risks sample damage, and is limited to easily accessible pegs. Here we report modifications to the peg optimized for in situ visualization and visualization of all pegs. We report similar antibiotic challenge recovery via colony formation following the ASTM E2799-22 protocol and in situ imaging. We report novel quantifiable effects of antibiotics on biofilm morphologies, specifically biofilm streamers. The new design bridges the MBEC® assays design that selects for biofilm phenotypes with in situ imaging needs.

LaMontagne CD, Christenson EC, Rogers AT, Jacob ME, Stewart JR. 2023. Relating antimicrobial resistance and virulence in surface-water E. coli. Microorganisms 11:2647.

https://www.mdpi.com/2076-2607/11/11/2647

The role of the environment in the emergence and spread of antimicrobial resistance (AMR) is being increasingly recognized, raising questions about the public health risks associated with environmental AMR. Yet, little is known about pathogenicity among resistant bacteria in environmental systems. Existing studies on the association between AMR and virulence are contradictory, as fitness costs and genetic co-occurrence can be opposing influences. Using Escherichia coli isolated from surface waters in eastern North Carolina, we compared virulence gene prevalence between isolates resistant and susceptible to antibiotics. We also compared the prevalence of isolates from sub-watersheds with or without commercial hog operations (CHOs). Isolates that had previously been evaluated for phenotypic AMR were paired by matching isolates resistant to any tested antibiotic with fully susceptible isolates from the same sample date and site, forming 87 pairs. These 174 isolates were evaluated by conventional PCR for seven virulence genes (bfp, fimH, cnf-1, STa (estA), EAST-1 (astA), eae, and hlyA). One gene, fimH, was found in 93.1% of isolates. Excluding fimH, at least one virulence gene was detected in 24.7% of isolates. Significant negative associations were found between resistance to at least one antibiotic and presence of at least one virulence gene, tetracycline resistance and presence of a virulence gene, resistance and STa presence, and tetracycline resistance and STa presence. No significant associations were found between CHO presence and virulence, though some sub-significant associations merit further study. This work builds our understanding of factors controlling AMR dissemination through the environment and potential health risks.

Mullowney, M.W., Duncan, K.R., Elsayed, S.S. et al. (…Reker D.). Artificial intelligence for natural product drug discovery. Nat Rev Drug Discov 22, 895–916 (2023).

https://doi.org/10.103/s41573-023-00774-7

Developments in computational omics technologies have provided new means to access the hidden diversity of natural products, unearthing new potential for drug discovery. In parallel, artificial intelligence approaches such as machine learning have led to exciting developments in the computational drug design field, facilitating biological activity prediction and de novo drug design for molecular targets of interest. Here, we describe current and future synergies between these developments to effectively identify drug candidates from the plethora of molecules produced by nature. We also discuss how to address key challenges in realizing the potential of these synergies, such as the need for high-quality datasets to train deep learning algorithms and appropriate strategies for algorithm validation.

Wiesner-Friedman C, Beattie RE, Stewart JR, Hristova KR, Serre ML. 2023. Identifying sources of antibiotic resistance genes in the environment using the microbial Find, Inform, and Test framework. Frontiers in Microbiology 14.

https://www.frontiersin.org/articles/10.3389/fmicb.2023.1223876/full

Introduction: Antimicrobial resistance (AMR) is an increasing public health concern for humans, animals, and the environment. However, the contributions of spatially distributed sources of AMR in the environment are not well defined.

Methods: To identify the sources of environmental AMR, the novel microbial Find, Inform, and Test (FIT) model was applied to a panel of five antibiotic resistance-associated genes (ARGs), namely, erm(B), tet(W), qnrA, sul1, and intI1, quantified from riverbed sediment and surface water from a mixed-use region.

Results: A one standard deviation increase in the modeled contributions of elevated AMR from bovine sources or land-applied waste sources [land application of biosolids, sludge, and industrial wastewater (i.e., food processing) and domestic (i.e., municipal and septage)] was associated with 34–80% and 33–77% increases in the relative abundances of the ARGs in riverbed sediment and surface water, respectively. Sources influenced environmental AMR at overland distances of up to 13 km.

Discussion: Our study corroborates previous evidence of offsite migration of microbial pollution from bovine sources and newly suggests offsite migration from land-applied waste. With FIT, we estimated the distance-based influence range overland and downstream around sources to model the impact these sources may have on AMR at unsampled sites. This modeling supports targeted monitoring of AMR from sources for future exposure and risk mitigation efforts.

Kelly, G.T., Granek, J., Gunsch C.K., Graves, J.L., Singleton, D. Advances in graduate training in Integrative Bioinformatics for Investigating and Engineering Microbiomes (IBIEM). Presented at 2023 ASEE Annual Conference and Exposition, Baltimore, Maryland. (2023).

https://peer.asee.org/42588

Innovations by engineers and physical scientists working at the frontiers of microbiome engineering and discovery requires in-depth understanding of microbiome systems with parallel skills to apply bioinformatics and biostatistics. Despite the importance of integrating bioinformatics and biology into graduate student training in fields outside traditional biological sciences, academic institutions remain challenged with including these disciplines across departmental boundaries. Furthermore, it is critical for students in engineering, bioinformatics, and biostatistics to understand fundamentals behind the biological systems they model, and for biology students to gain competencies to apply bioinformatics and biostatistics in quantitative biology arenas. To address these needs, the Integrative Bioinformatics for Investigating and Engineering Microbiomes (IBIEM) graduate training partnership between Duke University and North Carolina Agricultural and Technical State University was developed and funded by the National Science Foundation Research Traineeship (NRT) program. IBIEM’s goals include training interdisciplinary groups of students to: (a) transform conceptualization and develop skills for application of quantitative biology in microbiome areas; b) perform cutting edge research requiring interdisciplinary team skills; and to (b) communicate their research across disciplinary barriers and to diverse audiences. The pedagogical framework adapted to foster trainee engagement is learner-centered teaching which emphasizes the importance of self-directed learning with parallel ongoing assessment to optimize student outcomes. Since IBIEM trainee goals as well as entry-level knowledge and skills across disciplines varied greatly, program implementation was found to be challenging and required rigorous evaluation and refinements for effective training across disciplines and skill levels. A comprehensive program evaluation over five years found that the strongest learning and skills outcomes were linked to several “best practices”. Early provision of depth in fundamentals in R studio and Git Hub was found to be critical to “jump start” students without coding backgrounds. Addition of an overview of microbiome experimental design and analysis added important context as to how and where in the research process informatics fits into design progression and was highly motivating to students. Course modality was found to impact trainee outcomes with in-person classes that included hands-on practice and feedback showing greater improvements in training outcomes over hybrid, flipped and virtual course modalities. Furthermore, introduction of low, medium, and high level “challenges” along with in-person tutoring was found to be impactful in building a common foundation to span expertise levels and for engaging students across entry and advanced levels. Training impacts peaked during year four with cumulative implementation of revised strategies. Innovative training revisions and inclusion of critical elements was strongly linked to program satisfaction and ratings of advances in technical, professional and career skills as well as post-training carry over into trainees’ own research and leadership in their labs and careers. Furthermore, this training collaboration and partnership provided the foundation and training model for a newly funded NSF Engineering Research Center for Precision Microbiome Engineering (PreMiEr) for work in the critical area of engineering the microbiome in built environments.

PreMiEr-Associated Project

Jeje O, Ewunkem AJ, Jeffers-Francis LK, Graves JL. 2023. Serving two masters: Effect of Escherichia coli dual resistance on antibiotic susceptibility. Antibiotics 12:603.

https://www.mdpi.com/2079-6382/12/3/603

The prevalence of multidrug-resistant bacteria and their increased pathogenicity has led to a growing interest in metallic antimicrobial materials and bacteriophages as potential alternatives to conventional antibiotics. This study examines how resistance to excess iron (III) influences the evolution of bacteriophage resistance in the bacterium Escherichia coli. We utilized experimental evolution in E. coli to test the effect of the evolution of phage T7 resistance on populations resistant to excess iron (III) and populations without excess iron resistance. Phage resistance evolved rapidly in both groups. Dual-resistant (iron (III)/phage) populations were compared to their controls (excess iron (III)-resistant, phage-resistant, no resistance to either) for their performance against each stressor, excess iron (III) and phage; and correlated resistances to excess iron (II), gallium (III), silver (I) and conventional antibiotics. Excess iron (III)/phage-resistant populations demonstrated superior 24 h growth compared to all other populations when exposed to increasing concentrations of iron (II, III), gallium (III), ampicillin, and tetracycline. No differences in 24 h growth were shown between excess iron (III)/phage-resistant and excess iron (III)-resistant populations in chloramphenicol, sulfonamide, and silver (I). The genomic analysis identified selective sweeps in the iron (III) resistant (rpoB, rpoC, yegB, yeaG), phage-resistant (clpX →/→ lon, uvaB, yeaG, fliR, gatT, ypjF, waaC, rpoC, pgi, and yjbH) and iron (III)/phage resistant populations (rcsA, hldE, rpoB, and waaC). E. coli selected for resistance to both excess iron (III) and T7 phage showed some evidence of a synergistic effect on various components of fitness. Dual selection resulted in correlated resistances to ionic metals {iron (II), gallium (III), and silver (I)} and several conventional antibiotics. There is a likelihood that this sort of combination antimicrobial treatment may result in bacterial variants with multiple resistances.

Ji, Z., Ma, L. Controlling taxa abundance improves metatranscriptomics differential analysis. BMC Microbiol 23, 60 (2023).

https://doi.org/10.1186/s12866-023-02799-9

Song, K.; Zhou, Y.-H. Leveraging Scheme for Cross-Study Microbiome Machine Learning Prediction and Feature Evaluations. Bioengineering 2023, 10, 231.

https://doi.org/10.3390/bioengineering10020231

The microbiota has proved to be one of the critical factors for many diseases, and researchers have been using microbiome data for disease prediction. However, models trained on one independent microbiome study may not be easily applicable to other independent studies due to the high level of variability in microbiome data. In this study, we developed a method for improving the generalizability and interpretability of machine learning models for predicting three different diseases (colorectal cancer, Crohn’s disease, and immunotherapy response) using nine independent microbiome datasets. Our method involves combining a smaller dataset with a larger dataset, and we found that using at least 25% of the target samples in the source data resulted in improved model performance. We determined random forest as our top model and employed feature selection to identify common and important taxa for disease prediction across the different studies. Our results suggest that this leveraging scheme is a promising approach for improving the accuracy and interpretability of machine learning models for predicting diseases based on microbiome data.

Song, K., Zhou, YH. C3NA: correlation and consensus-based cross-taxonomy network analysis for compositional microbial data. BMC Bioinformatics 23, 468 (2022).

https://doi.org/10.1186/s12859-022-05027-9