Hakim JA , Schram JB , Galloway AWE , Morrow CD , Crowley MR , Watts SA , Bej AK .

P30AR050948 Comprehensive Cancer Center, University of Alabama at Birmingham, UL1TR000165 Center for Clinical and Translational Science, University of Alabama at Birmingham, P30 DK056336 NIDDK NIH HHS

	Figure 1. Sample collection site of S. purpuratus (purple sea urchins) from their natural rocky tide pool habitat along the coast of Oregon (43°18′14.3″N 124°24′05.1″W). (A) Satellite image of the collection site (red marker) provided through Google Earth Pro (v.7.3.2.5491) (Data SIO, NOAA, US Navy NGA, GEBCO, Image Landsat/Copernicus; US Dept. of State Geographer; image date: December 2015). (B) Overview of the tide pool collection site (labeled as tide pool 1) showing naturally occurring sea urchins. (C) Sea urchin congregates with the algal food source in view. Photographs by J.B Schram.
	Figure 3. Comparison of the observed taxa between the gut tissue (n = 3) and gut digesta (n = 3) using the rarefied OTU table data. Taxa observed at <1% were filtered from the graph. Standard deviation and relative abundances were determined through STAMP (v2.1.3), and the graph was generated through Microsoft Excel Software (Seattle, WA, USA).
	Figure 4. Per-group alpha diversity measurements calculated across all samples in the study. (A) Shannon and (B) Simpson alpha diversity histograms were smoothed by kernel density estimation. The Kruskal–Wallis H-tests were performed for the five groups and showed a significance value of p = 0.017 for the Shannon and p = 0.027 for Simpson diversity measurements, indicating significant differences between each group’s alpha diversity. The X-axis shows the diversity value of Shannon (values much greater than 0 are more diverse) and Simpson (values closer to 1 are more diverse). The histogram values of each sample were smoothed through kernel estimation to show the range of sample data points within each group. The Y-axis depicts the density function, which denotes the distribution of data points falling within this range (higher peak represents more clustered data points). Relevant p-values are listed in each graph. Plots were generated using the “diversity.py” command through PhyloToAST (v1.4.0).
	Figure 5. Beta diversity analysis of microbial communities observed across all samples in the study using Bray–Curtis similarity metrics determined for the rarefied OTU table. (A) A 2D multidimensional scale (MDS) plot analysis was performed to show sample cluster patterns based on observed OTUs, with a 40% and 60% Bray-Curtis similarity overlay, and the stress value (2D Stress = 0.04) was indicated. (B) Dendrogram analysis was also performed and each sample’s cluster patterns were based on group average. The OTU table was pretreated via standardization by the total and log transformation prior to Bray-Curtis analysis. Figure legends are shown in the 2D MDS plot. Data was generated and plotted through PRIMER-6 software (Primer-E Ltd, Plymouth Marine Laboratory, Plymouth UK, v6.1.2).
	Figure 6. Heatmap of the top 53 taxa at the highest resolution, determined using the rarefied OTU table and generated using R (v3.3.2). The heatmap.2 function from the gplots (v3.0.1) package (www.rdocumentation.org/packages/gplots) was used. Sample dendrogram was generated using Vegan (v2.4.3), employing the Bray-Curtis metric of the grouped biological replicate count data. Color palette selected using the RColorBrewer package and set from “sky blue” for less abundant, to “blue” for more abundant (shown in color key). Relative abundance values of each taxon are also indicated through a trace line (black). The associated table includes the most resolvable taxonomic assignment according to the GreenGenes (v13.8) database, which is color-coded to the phylum level assignments (class for Proteobacteria) as indicated in the key and corresponding to the relative abundances in the Figure 2A relative abundance graphs. The figure has been generated using scalable graphics and, therefore, regions of interest can be viewed at a higher resolution digitally by increasing the magnification.
	Figure 7. Linear discriminant analysis (LDA) effect size (LEfSe) performed on the microbial community relative abundance data at the of the gut tissue (n = 3) and gut digesta (n = 3). Grouped data were first analyzed using the Kruskal–Wallis test with a significance set to 0.05 to determine if the data was differentially distributed between groups, and those taxa that were differentially distributed were used for LDA model analysis to rank the relative abundance difference between groups. The LDA for significance was set to ±3, and the log(10) transformed score is shown to demonstrate the effect size. Data were analyzed and prepared through Hutlab Galaxy provided through the Huttenhower lab. The gut tissue group is shown as green, and the gut digesta group as red.
	Figure 8. Scatter plot analysis of the predicted KEGG Orthology (KO) metabolic functions determined through Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt v1.1.2) performed on the gut tissue (n = 3) and gut digesta (n = 3). Biological replicates were grouped, and analysis was performed for the (A) KEGG-Level-2 and (B) KEGG-Level-3 hierarchical functional categories. The linear regression value calculated for the two groups is shown for each scatter plot graph. Preferentially enriched categories for the gut tissue are shown as red, and for the gut digesta as brown. Those categories with clearly preferentially abundant categories have been labeled. Data were analyzed and visualized using STAMP (v2.1.3) analytical software. The node labels have been generated using scalable graphics, and therefore regions of interest can be viewed at a higher resolution digitally by increasing the magnification.
	Figure 9. Linear discriminant analysis (LDA) effect size (LEfSe) performed on the KEGG Orthology (KO) metabolic functions determined through PICRUSt (v1.1.2) for the gut tissue (n = 3) and gut digesta (n = 3). The KO Ids were determined through the “predict_metagenomes.py” command in PICRUSt (v1.1.2). Grouped data were analyzed using the Kruskal–Wallis with a significance set to 0.05, and the significantly differentially distributed KO Ids were used for LDA model analysis ranking the relative abundance significance, at an LDA threshold showing entries ranking at ± 2.4. The log(10) transformed score is shown as the effect size. Data were analyzed and prepared through Hutlab Galaxy provided through the Huttenhower lab. The gut tissue group is shown as green, and the gut digesta group as red. The KO Id labels have been generated using scalable graphics, and therefore regions of interest can be viewed at a higher resolution digitally by increasing the magnification.
	Figure 10. Co-occurrence patterns between taxonomic entries of the gut tissue and gut digesta, determined through Co-occurrence Network inference (CoNet v1.1.1), and analyzed through Cytoscape (v3.6.0). Taxonomic entries with a cumulative row sum of 200 or above with 2/3 of samples showing non-zero value entries were used through an ensemble approach that incorporated the Pearson, Spearman, Bray–Curtis, Kullback–Leibler, and mutual information metrics. The top and bottom 200 edges were selected and merged by the union method. (A) The network analysis shows the edges represented by the q-value (merged with the Brown method at p < 0.05 for each metric) and are shown as green (co-presence) and red (co-exclusion), with the nodes representing taxa were scaled according to relative abundance and colored according to the phyla (class for Proteobacteria) assignments. The final network was arranged using the yFiles (v1.0) Cytoscape (v3.6.0) add-on radial layout, and taxonomic entries shown at the highest resolution are denoted with the sample type (circle for gut tissue, “-gut”; diamond for gut digesta; “-dig”). The network has been generated using scalable graphics, and therefore nodes of interest can be viewed at a higher resolution digitally by increasing the magnification. (B) Scatter plot analysis was performed using topological metrics determined by NetworkAnalyzer (v2.7), to demonstrate patterns of key (keystone) species between taxonomic entries of the gut tissue and gut digesta based on closeness and betweenness centrality, as well as the degree (number of co-presence and co-exclusion edges). Linear regression between closeness and betweenness centrality was shown as logarithmic (R2 value = 0.7145), and the top 10 entries ranked by closeness centrality are depicted. Note, the taxa Rhodophyta and Rhodobacteraceae had the same closeness and betweenness centrality measurements, and their corresponding plot is indistinguishable. Linear regression determined through Microsoft Excel Software (Seattle, WA).

Achá, Importance of sulfate reducing bacteria in mercury methylation and demethylation in periphyton from Bolivian Amazon region. 2011, Pubmed

Achá, Importance of sulfate reducing bacteria in mercury methylation and demethylation in periphyton from Bolivian Amazon region. 2011, Pubmed
Altschul, Basic local alignment search tool. 1990, Pubmed
Assenov, Computing topological parameters of biological networks. 2008, Pubmed
Banerjee, Keystone taxa as drivers of microbiome structure and functioning. 2018, Pubmed
Beinart, Evidence for the role of endosymbionts in regional-scale habitat partitioning by hydrothermal vent symbioses. 2012, Pubmed
Beinart, The uptake and excretion of partially oxidized sulfur expands the repertoire of energy resources metabolized by hydrothermal vent symbioses. 2015, Pubmed
Berry, Deciphering microbial interactions and detecting keystone species with co-occurrence networks. 2014, Pubmed
Blaxter, Defining operational taxonomic units using DNA barcode data. 2005, Pubmed
Bokulich, Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. 2013, Pubmed
Bolyen, Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. 2019, Pubmed
Breezee, Subfreezing growth of the sea ice bacterium "Psychromonas ingrahamii". 2004, Pubmed
Brothers, Ocean warming alters predicted microbiome functionality in a common sea urchin. 2018, Pubmed , Echinobase
Callahan, DADA2: High-resolution sample inference from Illumina amplicon data. 2016, Pubmed
Callahan, Exact sequence variants should replace operational taxonomic units in marker-gene data analysis. 2017, Pubmed
Campbell, The versatile epsilon-proteobacteria: key players in sulphidic habitats. 2006, Pubmed
Caporaso, Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. 2012, Pubmed
Caporaso, Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. 2011, Pubmed
Caporaso, QIIME allows analysis of high-throughput community sequencing data. 2010, Pubmed
Cárcer, Evaluation of subsampling-based normalization strategies for tagged high-throughput sequencing data sets from gut microbiomes. 2011, Pubmed
Cardona, Bacterial community characterization of water and intestine of the shrimp Litopenaeus stylirostris in a biofloc system. 2016, Pubmed
Choma, Low abundance of Archaeorhizomycetes among fungi in soil metatranscriptomes. 2016, Pubmed
Cock, The Sanger FASTQ file format for sequences with quality scores, and the Solexa/Illumina FASTQ variants. 2010, Pubmed
Cole, The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. 2009, Pubmed
Dabdoub, PhyloToAST: Bioinformatics tools for species-level analysis and visualization of complex microbial datasets. 2016, Pubmed
DeSantis, Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. 2006, Pubmed
de Souza, Unlocking the bacterial and fungal communities assemblages of sugarcane microbiome. 2016, Pubmed
Durand, Microbial diversity associated with the hydrothermal shrimp Rimicaris exoculata gut and occurrence of a resident microbial community. 2010, Pubmed
Edgar, Search and clustering orders of magnitude faster than BLAST. 2010, Pubmed
Faust, Microbial co-occurrence relationships in the human microbiome. 2012, Pubmed
Faust, CoNet app: inference of biological association networks using Cytoscape. 2016, Pubmed
Faust, Microbial interactions: from networks to models. 2012, Pubmed
Frøslev, Algorithm for post-clustering curation of DNA amplicon data yields reliable biodiversity estimates. 2017, Pubmed
Groudieva, Psychromonas arctica sp. nov., a novel psychrotolerant, biofilm-forming bacterium isolated from Spitzbergen. 2003, Pubmed
Guerinot, N2-fixing vibrios isolated from the gastrointestinal tract of sea urchins. 1981, Pubmed , Echinobase
Hakim, An abundance of Epsilonproteobacteria revealed in the gut microbiome of the laboratory cultured sea urchin, Lytechinus variegatus. 2015, Pubmed , Echinobase
Hakim, The gut microbiome of the sea urchin, Lytechinus variegatus, from its natural habitat demonstrates selective attributes of microbial taxa and predictive metabolic profiles. 2016, Pubmed , Echinobase
Hill, Using ecological diversity measures with bacterial communities. 2003, Pubmed
Holland, A comparative study of gut mucous cells in thirty-seven species of the class Echinoidea (Echinodermata). 1970, Pubmed , Echinobase
Hosoya, Psychromonas agarivorans sp. nov., a novel agarolytic bacterium. 2009, Pubmed
Inagaki, Sulfurimonas autotrophica gen. nov., sp. nov., a novel sulfur-oxidizing epsilon-proteobacterium isolated from hydrothermal sediments in the Mid-Okinawa Trough. 2003, Pubmed
Kawasaki, Psychromonas marina sp. nov., a novel halophilic, facultatively psychrophilic bacterium isolated from the coast of the Okhotsk Sea. 2002, Pubmed
Konstantinidis, Genomic insights that advance the species definition for prokaryotes. 2005, Pubmed
Kozich, Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. 2013, Pubmed
Kumar, Getting started with microbiome analysis: sample acquisition to bioinformatics. 2014, Pubmed
Langille, Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. 2013, Pubmed
Ley, Microbial ecology: human gut microbes associated with obesity. 2006, Pubmed
Lozupone, UniFrac: an effective distance metric for microbial community comparison. 2011, Pubmed
Ma, Geographic patterns of co-occurrence network topological features for soil microbiota at continental scale in eastern China. 2016, Pubmed
Marcon, Generalization of the partitioning of shannon diversity. 2014, Pubmed
McDonald, An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. 2012, Pubmed
Nearing, Denoising the Denoisers: an independent evaluation of microbiome sequence error-correction approaches. 2018, Pubmed
Nichols, Prokaryotes and the input of polyunsaturated fatty acids to the marine food web. 2003, Pubmed
Parks, STAMP: statistical analysis of taxonomic and functional profiles. 2014, Pubmed
Pechal, Frozen: Thawing and Its Effect on the Postmortem Microbiome in Two Pediatric Cases. 2017, Pubmed
Pruesse, SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. 2007, Pubmed
Pruesse, SINA: accurate high-throughput multiple sequence alignment of ribosomal RNA genes. 2012, Pubmed
Rabus, The genome of Desulfotalea psychrophila, a sulfate-reducing bacterium from permanently cold Arctic sediments. 2004, Pubmed
Roalkvam, Physiological and genomic characterization of Arcobacter anaerophilus IR-1 reveals new metabolic features in Epsilonproteobacteria. 2015, Pubmed
Romero, Bacterial 16S rRNA gene analysis revealed that bacteria related to Arcobacter spp. constitute an abundant and common component of the oyster microbiota (Tiostrea chilensis). 2002, Pubmed
Sawabe, Alginate degradation by bacteria isolated from the gut of sea urchins and abalones. 1995, Pubmed , Echinobase
Segata, Metagenomic biomarker discovery and explanation. 2011, Pubmed
Shannon, Cytoscape: a software environment for integrated models of biomolecular interaction networks. 2003, Pubmed
Smith, Regulation of life span by the gut microbiota in the short-lived African turquoise killifish. 2017, Pubmed
Sodergren, The genome of the sea urchin Strongylocentrotus purpuratus. 2006, Pubmed , Echinobase
Song, Preservation Methods Differ in Fecal Microbiome Stability, Affecting Suitability for Field Studies. 2016, Pubmed
Spearman, The proof and measurement of association between two things. By C. Spearman, 1904. 1987, Pubmed
Tamar, The role of motility and chemotaxis in the bacterial colonization of protected surfaces. 2016, Pubmed
Thomas, Characterization of the first alginolytic operons in a marine bacterium: from their emergence in marine Flavobacteriia to their independent transfers to marine Proteobacteria and human gut Bacteroides. 2012, Pubmed
Troussellier, Sustaining Rare Marine Microorganisms: Macroorganisms As Repositories and Dispersal Agents of Microbial Diversity. 2017, Pubmed
Turnbaugh, An obesity-associated gut microbiome with increased capacity for energy harvest. 2006, Pubmed
Urakawa, Hydrothermal vent gastropods from the same family (Provannidae) harbour epsilon- and gamma-proteobacterial endosymbionts. 2005, Pubmed
Wang, Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. 2007, Pubmed
Wang, Lung microbiome dynamics in COPD exacerbations. 2016, Pubmed
Wang, Comparison of the levels of bacterial diversity in freshwater, intertidal wetland, and marine sediments by using millions of illumina tags. 2012, Pubmed
Wirsen, Characterization of an autotrophic sulfide-oxidizing marine Arcobacter sp. that produces filamentous sulfur. 2002, Pubmed
Ziegler, Origin and evolutionary plasticity of the gastric caecum in sea urchins (Echinodermata: Echinoidea). 2010, Pubmed , Echinobase