Article Figures & Data
Figures
- FIG 1
Dendrogram of bacterial community dissimilarity across University of Wisconsin–Milwaukee RAS components. The dendrogram was created using average-linkage Bray-Curtis dissimilarity of bacterial communities between RAS components. Leaves are labeled by date and sample site within the RAS. Sample site references are as follows, in order of process flow: Tank, rearing tank; pH, pH buffering tank; Clar, clarifier sludge; ClarOut, clarifier effluent; Bfsand, biofilter sand substrate, BfOut, biofilter effluent; Degas, carbon dioxide degassing tower.
- FIG 2
Nonmetric multidimensional scaling (nMDS) plot of bacterial community composition. The plot indicates Bray-Curtis dissimilarity between samples (V4-V5 16S rRNA gene data) as ordinated by the metaMDS function with dimensions k = 5 and stress = 0.078. Data in ellipses are illustrated as follows: (A) all samples from a particular site (ADONIS R2 = 0.355, P ≤ 0.001) and (B) all samples from a particular water source (ADONIS R2 = 0.110, P ≤ 0.001). Ellipses were added with the ordiellipse function in vegan (70). The component origin of each sample is indicated via the shape of the sample point as indicated in the figure key. UWM, University of Wisconsin—Milwaukee.
- FIG 3
Heat map of the most abundant 16S rRNA gene amplicon sequence variants across component types. The plot indicates bacterial community composition relative to abundance data ordered by distance-based clustering (Euclidian) of samples containing only the illustrated amplicon sequence variants (ASVs). The top 10 most abundant ASVs (mean relative abundances) for each major component type, biofilter, solids, and rearing tank are plotted (heatmap2 from the ggplots R package). The ASVs are also ordered by distance-based clustering (Euclidian) of their relative abundance patterns. Gene relative abundances are indicated via color-coding as listed in the plot key. Each sample’s component type and sample site origin are indicated below the heat map. The sample name abbreviations are as follows: Bell, Bell Aquaculture; DW, Discovery World; PF, PortFish; UFP, UrbanFarmProject; MAR, Marinette. Family-level and genus-level taxonomic assignments are listed for each ASV. When these levels were not assigned, then the most refined taxonomic level is provided. Identical taxonomic assignments are given a number (indicated in square brackets) to facilitate references in the text. na, unassigned taxonomy at this level.
- FIG 4
Presence/absence heat map of nitrifying microorganism amplicon sequence variants (ASVs). Samples are included as rows, and sample facilities are indicated. Columns represent unique ASVs (>95% nucleotide sequence identity) ordered by taxonomic affiliation as indicated. The presence of an ASV in a sample is indicated by a colored square. ASV designations are indicated along the x axis. Sequence alignments and phylogenies for Betaproteobacteria amoA, Nitrospira nxrB, and Nitrospira amoA ASVs can be found at https://doi.org/10.6084/m9.figshare.7777232.v1, https://doi.org/10.6084/m9.figshare.7777241.v2, and https://doi.org/10.6084/m9.figshare.7777247.v1.
- FIG 5
Richness distribution for recovered nxrB genotypes between RAS and aquaponic samples. Pirate plots (“yarrr” package in R) are indicated for nxrB genotype richness for each sample class (aquaponic or RAS). A black line indicates the richness mean, a colored box indicates the Bayesian highest-density interval, and all points are listed with corresponding indication of data distribution/density via figure shape.
- FIG 6
Generalized component process diagram. Black arrows indicate the flow of the water through the University of Wisconsin—Milwaukee (UWM) system. The dashed green arrows represent the generalized flow of water in the aquaponic systems.
Tables
- TABLE 1
Alpha-diversity metrics across UWM RAS sites
Sitea H' ± SDb J ± SDc nd Sample
categoryeBiofilter effluent 4.22 ± 0.32 0.68 ± 0.05 4 Plankton Biofilter sand 4.66 ± 0.20 0.75 ± 0.04 6 Biofilm Biofilter water 4.26 ± 0.74 0.68 ± 0.12 4 Plankton Clarifier 4.27 ± 0.21 0.68 ± 0.03 6 Sludge Clarifier effluent 4.38 ± 0.35 0.70 ± 0.06 4 Plankton Degasser 4.20 ± 0.31 0.67 ± 0.05 5 Plankton pH buffer tank 3.54 ± 1.41 0.57 ± 0.22 3 Plankton Rearing tank 4.13 ± 0.13 0.66 ± 0.02 4 Plankton ↵a All alpha-diversity metrics were calculated across the University of Wisconsin—Milwaukee Recirculating Aquaculture System sites using the V6 16S rRNA gene sequence data set.
↵b Data represent Shannon-Weaver diversity index (H’) values. Standard deviation (SD) values are indicated after the mean values.
↵c Data represent Pielou’s evenness metric (J) values. Standard deviation (SD) values are indicated after the mean values.
↵d The number of samples at each site is indicated.
↵e Sample categories used in Kruskall-Wallis rank sum hypothesis testing are indicated.
- TABLE 2
PCR primers used in this study
Gene target Forward primer(s)
(5′–3′)Reverse primer(s)
(5′–3′)Component(s)
surveyedSample
site(s)Reference(s) Bacterial 16S rRNA gene
V6 regionCTAACCGANGAACCTYACC,
CNACGCGAAGAACCTTANC,
CAACGCGMARAACCTTACC,
ATACGCGARGAACCTTACCCGACRRCCATGCANCACCT All UWM RAS 72 Bacterial 16S rRNA gene
V4-V5 regionCCAGCAGCYGCGGTAAN CCGTCAATTCNTTTRAGT,
CCGTCAATTTCTTTGAGT,
CCGTCTATTCCTTTGANTAll All 73 Archaeal 16S rRNA gene
V4-V5 regionGCCTAAAGCATCCGTAGC,
GCCTAAARCGTYCGTAGC,
GTCTAAAGGGTCYGTAGC,
GCTTAAAGNGTYCGTAGC,
GTCTAAARCGYYCGTAGCCCGGCGTTGANTCCAATT Biofilters All biofilter samples 74 Betaproteobacterial amoA GGGGHTTYTACTGGTGGT CCCCTCKGSAAAGCCTTCTTC Biofilters All Biofilter Samples 75, 76 Comammox amoA GGAYTTYTGGNTNGATTGGA WRKTNNGACCACCASKACCA Biofilters All biofilter samples Modified from
reference 58Nitrospira nxrB TACATGTGGTGGAACA CGGTTCTGGTCRATCA Biofilters All biofilter samples 53
TABLE S1
Sample list and associated metadata for the V6 and V4-V5 16S rRNA gene sequence datasets. Download Table S1, XLSX file, 0.1 MB.
Copyright © 2019 Bartelme et al.
This content is distributed under the terms of the Creative Commons Attribution 4.0 International license.
