(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Symbiotic bacteria may support calcium carbonate precipitation in the Gulf toadfish [1]

['Anthony M. Bonacolta', 'Department Of Marine Biology', 'Ecology', 'Rosenstiel School Of Marine', 'Atmospheric', 'Earth Science', 'University Of Miami', 'Miami', 'Florida', 'United States Of America']

Date: 2026-05 

Abstract Marine fish play a significant yet understudied role in the oceanic carbon cycle through the production of magnesium-rich calcium carbonate (CaCO 3 ) precipitates known as ichthyocarbonates. These deposits form in the gut of marine teleost fish in response to salinity, serving as part of their osmoregulation strategy. Through this, marine fish may contribute as much as 9.04 Pg of CaCO 3 per year in global new carbonate production, being equivalent to or potentially higher than the production by coccolithophores and pelagic foraminifera. Despite their ecological relevance, the biological mechanisms driving ichthyocarbonate precipitation remain to be fully resolved. Intriguingly, bacteria are consistently found in intimate association with ichthyocarbonate precipitates. Given the widespread capacity of prokaryotes to mediate CaCO₃ precipitation, this association points to a previously unexplored microbial contribution to the process. To investigate the potential role of bacteria in ichthyocarbonate production, we subjected Gulf toadfish (Opsanus beta) to salinity treatments common to their native range and known to elicit changes in CaCO 3 precipitation. To assess the respective contributions of the host and its microbiota to ichthyocarbonate formation in the gut, we characterized the microbiome across the toadfish gut and performed meta-transcriptomic analysis. Across the toadfish gut, we identify a high abundance of vibrios associated with ichthyocarbonates with the metabolic potential for CaCO 3 precipitation. Specifically, we observe the expression of the transcriptional activator of urease (ureR) by Photobacterium damselae subsp. damselae, which can induce the precipitation of CaCO 3 via the production of bicarbonate. We demonstrate that CaCO₃ precipitation in marine fish may not solely be a host-driven process, but potentially the result of a functional symbiosis with gut-associated Vibrio bacteria. We hypothesize that just as photosymbionts enable corals to build reefs, fish hosts, along with their microbial partners, may synergistically contribute to oceanic carbonate production. This discovery, if confirmed, expands the role of symbiosis in marine biomineralization and underscores its broader influence on global biogeochemical cycles.

Citation: Bonacolta AM, Kravitz T, Mozo R, Baker LJ, Heuer RM, Grosell M, et al. (2026) Symbiotic bacteria may support calcium carbonate precipitation in the Gulf toadfish. PLoS Biol 24(5): e3003764. https://doi.org/10.1371/journal.pbio.3003764 Academic Editor: Holly M. Bik, University of Georgia, UNITED STATES OF AMERICA Received: October 20, 2025; Accepted: April 2, 2026; Published: May 18, 2026 Copyright: © 2026 Bonacolta et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The raw reads for the project have been deposited on NCBI SRA (BioProject: PRJNA1308130). Code used for transcriptomic analysis, single-gene trees, as well as ASV count tables, taxonomy tables, and nucleotide sequences can be found on both GitHub (https://github.com/delCampoLab/toadfish_microbiome) and Zenodo (https://doi.org/10.5281/zenodo.18867155) along with the underlying data for all Figures. Funding: M.G. is a Maytag Professor of Ichthyology. This work was supported by start-up funds from the University of Miami and was supported by Grant PID2023-152522NB-I00, funded by MICIU/AEI/10.13039/501100011033 and by ERDF, EU provided to J.d.C. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: AAALAC, Association for Assessment and Accreditation of Laboratory Animal Care; ANCOM-BC, Analysis of compositions of microbiomes with bias correction; ANOSIM, Analysis of similarities; ASVs, amplicon sequence variants; AWA, Animal Welfare Act; CA, carbonic anhydrase; CaCO 3 , calcium carbonate; CFTR, cystic fibrosis transmembrane regulator; CG-C, Guanylin cyclase-C; DIC, dissolved inorganic carbon; EPA, Evolutionary Placement Algorithm; etfB, electron transfer flavoprotein subunit beta; GN, Guanylin peptides; MRA, mean relative abundance; napAB, Periplasmic nitrate reductase; narGHI, Respiratory nitrate reductase; PHS, Public Health Service; PICRUSt2, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States; sAC, soluble adenyl cyclase; TAE, Tris-acetate-EDTA buffer; Ure, Urease

Introduction The marine environment is central to the global carbon cycle as the ocean is the largest reservoir of CO 2, which can be readily exchanged with the atmosphere, thus having a significant role in determining atmospheric CO 2 concentrations over decadal to millennial timescales [1]. Two mechanisms, the biological carbon pump and the carbonate pump, are integral to this process in the ocean. The biological carbon pump sequesters atmospheric CO 2 in the form of photosynthetically fixed carbon, whereas the carbonate pump transfers CO 2 back into the atmosphere via marine biogenic calcification. Marine fish play an important role not only in the biological carbon pump through prey consumption, fecal pellet production, and deadfall, but also in the carbonate pump through the production of calcium carbonate (CaCO 3 ) precipitates (ichthyocarbonates), which form in their intestines from metabolic, endogenous CO 2 derived from their diet [2,3]. Through this process, marine fish are one of the dominant carbonate producers in the ocean, providing an estimated 0.33–9.03 Pg CaCO 3 per year in global new carbonate production [3], relying mainly on photosynthetic fixed carbon. This makes their contribution equivalent to or higher than more well-known sources such as coccolithophores and pelagic foraminifera [3–5], which largely derive carbon from dissolved inorganic carbon (DIC). In contrast to those sources, marine fish carbonate excretion is anticipated to play an increasingly critical role in the inorganic carbon cycle under the developing conditions of global climate change as they increase ichthyocarbonate production in response to both rising temperatures and higher CO 2 levels [6–9]. Marine fish produce ichthyocarbonates as part of their osmoregulatory strategy. Briefly, to maintain osmotic homeostasis while in a concentrated external environment, marine teleost fish ingest seawater to replace osmotic and urinary fluid loss [10]. This seawater is first desalinated by the water-impermeable esophagus [11], followed by the anterior intestine, where NaCl uptake drives water absorption [12]. This NaCl is later excreted across the gill epithelium [10]. Bicarbonate (HCO 3 −) transported into the intestinal lumen facilitates water absorption [13]. In the intestinal fluids, CaCO 3 is precipitated, removing Ca2+, Mg2+, and HCO 3 − ions to lower the osmotic pressure. This precipitated CaCO 3 then gets excreted into the marine environment. Several fish enzymes and transporters have been identified that play a role in this calcification process, including, but not limited to, carbonic anhydrases, guanylin peptides, sodium-bicarbonate co-transporters, Na-H exchangers, V-type H-ATPases [14–16]. However, the role of microbes, specifically bacteria, in this calcification process has gone unexamined since the initial observation of bacteria being intimately associated with ichthyocarbonates three decades ago [17]. Relatedly, CaCO 3 precipitation has been extensively documented as a ubiquitous process among various bacterial groups in a diverse array of environments [18]. Symbiotic microbes can contribute to host physiology and its ability to adapt to certain environmental stressors [19], even co-evolving with their hosts over long periods [20]. While microbiome research has gained traction in the last two decades, only recently have fish microbiomes received the same amount of attention, despite representing the greatest species diversity among vertebrates [21]. Fish gut microbiomes are distinct from those of the water column and can vary significantly from fish to fish [22]. Proteobacteria and Firmicutes are the dominant taxa of most fish gut microbiota, but environmental factors such as salinity play a major role in the microbiota composition [23,24]. In marine fish, Proteobacteria are enriched compared to freshwater fish gut microbiomes. At the family level, Moraxellaceae, Vibrionaceae, Enterobacteriaceae, and Alcaligenaceae bacteria are significantly more common in marine fish compared to freshwater fish [23]. Vibrios (family: Vibrionaceae), specifically, have been shown to dominate the gut microbiomes of marine fish species and also increase in abundance across a salinity gradient [25,26], indicating a likely physiological link between saltwater fish guts and these bacteria. The Gulf toadfish, Opsanus beta, is an established marine teleost fish model for osmoregulatory and acid-base physiology, and the calcification process within their intestinal lumen has been extensively analyzed [13–15], making it the ideal subject for this investigation. We aim first to characterize the gut microbiome of the Gulf toadfish and then determine any possible role for the gut microbiota in the precipitation of ichthyocarbonates in this species. Using 16S rRNA gene metabarcoding and laboratory aquaria, we profiled the microbiome of nine O. beta individuals across a gradient of salinity treatments (9 [brackish], 35 [marine], and 60 [hypersaline] parts per thousand [ppt]) designed to induce precipitate formation at the higher salinities in the lab, while also representing possible salinity levels this species may be exposed to in its natural habitat [27,28]. Given that different microenvironments of the intestine serve distinct roles in ichthyocarbonate production [16], we split our sampling across four specific portions of the toadfish gut: the anterior intestine, posterior intestine, intestinal fluid, and ichthyocarbonates (when present; Fig 1A). This was followed by transcriptomic sequencing targeting bacteria and host reads from select samples to understand the metabolic contributions of each to CaCO 3 precipitation. Phylogenetic analysis of genes involved in calcium carbonate precipitation confirmed the origin of essential genes detected in the transcriptomic data. PNG larger image

TIFF original image Download: Fig 1. Prokaryotic community of the gulf toadfish (O. beta). (A) Diagram of sampled regions of the O. beta intestinal tract, including the anterior and posterior intestines, intestinal fluid, and ichthyocarbonates, when observed. (B) Bubble plot of prokaryotic genera across sampled regions. Only genera with a mean relative abundance (MRA) above 0.01% are shown. Genera are colored by phylum, with upper bound and lower standard deviations indicated by bubble shading. IS – incertae sedis. (C) Upset plot depicting unique and shared ASVs across sampled regions of the Gulf toadfish gut. ASV intersection size represents ASVs common to each dot-connected grouping. Set size represents the total ASVs unique to each area. The data underlying this figure can be found on Zenodo (https://doi.org/10.5281/zenodo.18867155). https://doi.org/10.1371/journal.pbio.3003764.g001

Discussion This is the first study in 30 years to investigate the potential contribution of microbes to ichthyocarbonate production. We identified both the microbes and prokaryotic gene expression capable of supporting CaCO 3 precipitation in the gut of O. beta. The prokaryotic microbiome of O. beta was stable across sampled regions of the intestine and distinct from that of the tank water (Figs 1B, S2). Consistent with previous literature analyzing the gut composition of marine fish, we saw Proteobacteria (especially vibrios) and Firmicutes (Mycoplasma spp.) as significant members of the intestinal community [23,24] (Fig 1B). Mycoplasma spp. have previously been observed in clusters on the intestinal mucosa of fish and appear adapted to the nutrient-rich environment of the gut [43]. Brevinema spirochetes, which were most common in fluid and posterior intestine samples (Fig 1B), have also been noted as common inhabitants of fish guts before and are presumed to have a similar nutrient scavenging role [43,44]. Importantly, we note the high abundance of various Vibrionaceae within the gut of O. beta, including Vibrio spp., Photobacterium spp., and Thaumasiovibrio spp. (Figs 1B, 2A–2C). The surface of the ichthyocarbonates, which form in higher salinities as part of the toadfish’s osmoregulatory process, is dominated by vibrios, specifically ASVs related to V. ponticus, V. rhodolitus, P. damselae, and V. stylophorae (Fig 2A–2C). V. ponticus and V. rhodolitus are members of the Ponticus vibrio clade [45] and were shown here to be more abundant in ichthyocarbonates from the highest salinity treatment. Members of this clade have been isolated from diseased fish, crustaceans, and bivalves [46]. We did not note any signs of illness within our sampled fish, although the high abundance of these vibrios in the 60 ppt treated fish could be due to the increased stress on the fish holobiont caused by this extreme salinity treatment. P. damselae has also been reported as a pathogen of a variety of marine animals and even in humans [47]; although, it has also been collected from marine fish not showing any signs of disease [48], like the O. beta used in this study. V. stylophorae was first isolated from a reef-building coral in Taiwan [49]. The unique clustering of our fish-associated Vibrio ASVs as sister to V. stylophorae suggests a possible fish host-specific divergence in this lineage. Along with the high abundance of vibrios on ichthyocarbonates, we also identify vibrio-specific gene expression within these samples that would aid in the precipitation of calcium carbonate. Two of these vibrio-expressed genes, fadR and fadB, are synonymous to the ysiA and ysiB genes in Bacillus subtilis, which have been shown to induce calcium carbonate biomineralization [29]. As these genes are primarily involved in fatty acid metabolism [50], an intermediate in this process likely contributes to the biomineralization process [29]. Likewise, napA and napB, which encode periplasmic nitrate reductases, were also found to be expressed by vibrios within the gut of the gulf toadfish (Table 1). Furthermore, narG and narH, which encode respiratory nitrate reductases, were expressed by nonVibrio bacteria within the gut of O. beta as well (Table 1). As nitrate reduction coupled with an organic carbon source generates CO 2, which equilibrates with water to form bicarbonate ions, the expression of these genes by the intestinal microbiota of O. beta is capable of promoting CaCO 3 precipitation [32]. More directly, the genomic potential of urease accessory proteins combined with the observed expression of ureR (urease operon transcriptional activator) by vibrios (Fig 3), specifically P. damselae, which we also found as highly abundant on ichthyocarbonates, supports a microbial contribution to ichthyocarbonate production. Urease is the enzyme that catalyzes the hydrolysis of urea into ammonia and carbamic acid. Carbamic acid spontaneously decomposes into additional ammonia and carbonic acid [51]. Carbonic acid dissociates to form bicarbonate, which likely reacts with Ca2+ ions within the fish gut to precipitate CaCO 3 (Fig 3B). The ability to hydrolyze urea is a key characteristic of P. damselae subsp. damselae, which discriminates it from P. damselae subsp. piscicida [52], further discerning ASV 2 as most likely P. damselae subsp. damselae. This P. damselae pathway, combined with the expression of carbonic anhydrases (which catalyzes the hydration of carbon dioxide, increasing bicarbonate concentration [30]) by the toadfish host, further induces ichthyocarbonate production. Vibrios have been hypothesized to play an active role in CaCO 3 precipitation in their natural habitats [53], and here we show for the first time the potential for this within the fish gut microenvironment. Potential caveats of this work include the limited sample size for each salinity treatment. Furthermore, our study was limited to a single fish species; thus, the prevalence of this symbiosis across other marine teleost remains to be examined. The microbial community in vertebrate guts plays critical roles in immune system maturation, host metabolism, and even reproductive organ development [54–57]. The gut microbiota also provides vertebrate hosts with the flexibility to digest diverse substrates, allowing for rapid adaptation to shifts in diet [58]. In fish, the gut microbiota similarly contributes to intestinal development, stimulates fatty acid uptake, and influences nutrient absorption, thereby helping regulate energy homeostasis within the fish holobiont [54,59]. As for marine vibrios, the most well-studied bacteria-metazoan symbiosis in the ocean is arguably that between the bobtail squid (Euprymna scolopes) and its bioluminescent Vibrio fischeri [60]. In fish, vibrios encoding chitinases and alginate lyases are hypothesized to aid fish host digestion of crustaceans and algae [61,62], respectively. Our data support the role of intestinal microbiota, specifically vibrios, in the precipitation of CaCO 3 in the guts of marine fish. The abundance of P. damselae subsp. damselae on ichthyocarbonates, and their expression of ureR suggests a partnership with the fish host, whereby microbial production of bicarbonate works alongside host produced HCO 3 −, H+-elimination, and enzymatic activity to drive CaCO 3 precipitation. We also report several other vibrio-encoded genes involved in fatty acid metabolism (fadR and fadB) and nitrate reduction (napA and napB), which also contribute to biomineralization conditions in fish intestines. The interplay between host transporters and enzymes and gut-associated vibrios creates the necessary conditions for CaCO 3 precipitation, unveiling a novel symbiotic pathway in marine biomineralization. Based on these findings, we hypothesize that fish gut bacteria may function as significant microbial contributors to global carbonate cycling alongside coccolithophores, foraminifera, and coral symbionts. Direct measurements of the contribution of vibrio-derived bicarbonate to ichthyocarbonate composition, as well as ecological surveys to determine the prevalence and specificity of this partnership across fish species, would further validate the findings reported here and yield important insights into this novel symbiosis.

Materials and methods Sample collection Gulf toadfish were obtained as bycatch from local shrimp fishermen in Biscayne Bay, FL, and transported to laboratory holding tanks following brief (5 min) freshwater baths and malachite green treatment to treat for any ectoparasites [36]. Holding tanks (62 L) were continuously provided with flow-through, sand-filtered seawater from Biscayne Bay (32–35 ppt salinity, 25–32 °C) and constant aeration. Gulf toadfish were fed to satiation twice weekly with meals consisting of chopped squid and shrimp. After approximately one month in holding tanks, Gulf toadfish were transferred to experimental tanks (62 L) containing either sand-filtered Biscayne Bay seawater, 60 ppt or 9 ppt water, and were allowed to acclimate for 1 week. Three fish per salinity treatment were held together in one tank. Sixty ppt water was produced by adding Instant Ocean sea salt to seawater, while 9 ppt water was produced by diluting seawater with deionized water. Common for all three salinity treatments was renewal of the water every second day and constant aeration. Samples were obtained from the intestinal lumen and from epithelial scrapings of individual toadfish exposed to different salinity treatments (n = 3 fish per treatment, mass = 95.5 ± 3.9 g). Fish were kept at three different salinity treatments because salinity has been observed to impact carbonate precipitation [13]. At 9 ppt, no ichthyocarbonates were found to have formed. Whereas the fish kept at 35 ppt and 60 ppt salinity, corresponding to normal seawater and hyper-salinity, had ample ichthyocarbonate production. O. beta sampling occurred at four locations: the anterior intestine, posterior intestine, intestinal fluid, and ichthyocarbonates (Fig 1A). Water samples were also collected from each respective salinity treatment. Samples were flash frozen and stored at −80 °C until nucleic acid extraction. All samples were extracted using the Qiagen All Prep DNA/RNA Minikit (Qiagen, Hilden, Germany) within 6 months of sampling. 16S rRNA gene metabarcoding The V4 region of the 16S rRNA gene was amplified according to the Earth Microbiome protocol using 515F (5′ – GTGYCAGCMGCCGCGGTAA – 3′) and 806R (5′ – GGACTACNVGGGTWTCTAAT – 3′) primers from Apprill and Parada [63,64]. PCR mixture for one sample contained 10.0 μl PCR master mix (2x), 0.5 μl each of the forward and reverse primers (diluted at 1:10), 1.0 μl template DNA, and reaction volumes made up to 25.0 μl with PCR-grade water (13.0 μl). The PCR program of the thermal cycler was set up as follows: an initial denaturation step at 94 ºC for 3 min, subsequent 35 cycles of denaturation at 94 ºC for 45 s, 35 cycles of annealing at 50 ºC for 60 s, and 35 cycles of elongation at 72 ºC for 90 s, all of them followed by a final elongation at 72 ºC for 10 min. Amplification was checked by gel electrophoresis. Electrophoresis gel was made with 1 g of ultrapure agarose dissolved in 100 ml of 1X Tris-acetate-EDTA buffer (TAE) and stained with 5 μl of SYBR safe DNA gel stain. The gel was run for about 20 min at 100 V at room temperature, then visualized under a blue light transilluminator. Amplified PCR products were cleaned using AMPure magnetic beads (Thermofisher). DNA concentrations from purified PCR samples were checked using a Qubit fluorometer (Thermofisher) before being sent to the Integrated Microbiome Resource facility at the Centre for Comparative Genomics and Evolutionary Bioinformatics at Dalhousie University for sequencing on an Illumina MiSeq using 300 + 300 bp paired end V3 chemistry. Samples that failed the first round of sequencing were further purified using AMPure magnetic beads and sent to Novogene for another round of sequencing on an Illumina MiSeq following the same library prep instructions. Microbiome analysis Primers and low-quality bases were removed from reads using Cutadapt v3.1 [65]. The trimmed reads were then processed in R using DADA2 [66]. Chimeras were removed using the ‘removeBimeraDenovo’ command with the “consensus” option. For the 16S rRNA gene amplicons, forward reads were truncated at 250 bp and reverse reads at 210 bp, corresponding to a general drop-off in read quality past these points of the sequences. Truncated reads were then denoised and merged into ASVs before assigning taxonomy using SILVA v138 in DADA2 [67]. Sequence tables, taxonomy tables, and metadata for the 16S rRNA gene amplicon datasets were uploaded into a Phyloseq object in R for further filtering and analysis [68]. For the 16S rRNA gene amplicons, ASVs corresponding to Chloroplast and Mitochondria were removed. Furthermore, only ASVs present in the top 99% of at least one sample and with total counts above 150 reads were kept for further analysis. Bubble plots, alpha-diversity, and beta-diversity figures were constructed using ggplot2 and tidyverse packages [69]. Statistical analyses were conducted using vegan [70]. Analysis of similarities (ANOSIM) within vegan was used to test for differences in beta-diversity between groups using 999 permutations. Ampvis2 aided in distinguishing the most abundant and significant ASVs [71]. Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC) was used to find significantly enriched and differentially abundant taxa between treatment groups while adjusting for a false discovery rate [72]. PICRUSt2 was run on the filtered ASV dataset to assess the metagenomic potential of the prokaryotic community [73]. Vibrio phylogenetics and evolutionary placement of ASVs Phylogenetic analyses of the family Vibrionaceae were conducted to assess the evolutionary relationships between abundant ASVs from the ichthyocarbonates and closely related species. Due to the short length of the ASV sequences, these were placed onto a previously generated 16S rRNA maximum likelihood phylogenetic tree using the EPA. To construct the 16S phylogenetic tree, we compiled a dataset comprising a total of 87 Vibrionaceae sequences and 3 Shewanellaceae sequences as an outgroup. These context sequences were downloaded from the NCBI database. The 90 sequences were aligned using MAFFT v7.505 [74], and the alignment was inspected in AliView v1.28 [75]. For clean-up of the alignment, we employed trimAl v1.4.rev15 [76], which performed an automated alignment trimming. A maximum likelihood phylogenetic analysis was then performed using RAxML v8.2.12 [77] to construct the final Vibrionaceae phylogenetic tree. The most abundant Vibrio ASVs from the ichthyocarbonates were subsequently placed into this backbone reference tree using RAxML’s EPA. For this, the short ASV sequences were first aligned to the existing Vibrionaceae 16S rRNA gene alignment using MAFFT v7.505. The EPA analysis was then run to position the ASVs within the phylogeny. Finally, the resultant EPA tree was visualized and edited using the Interactive Tree of Life (iTOL) online platform. Host and microbial transcriptomic analysis Select samples were chosen for either host or prokaryotic community RNA sequencing through Novogene, depending on sample type and RNA quality (S1 Table). This resulted in 5 anterior gut samples for host mRNA sequencing via poly-A selection and 2 ichthyocabonate samples for prokaryotic mRNA sequencing via ribosomal depletion. Sequencing was conducted through Novogene on an Illumina NovaSeq targeting 6 GB of 150 + 150 PE reads per sample. Trim Galore! [78] was used to trim adaptors and poor-quality reads. Ribosomal RNAs were removed with sortmeRNA [79]. Cleaned reads were then aligned to a O. beta genomic reference using STAR to generate gene counts [80,81]. For the prokaryotic RNA samples, the unmapped reads were assembled using rnaSPAdes [82]. ORFs were identified, and then protein prediction was performed using TransDecoder with BlastP and hmmscan with the uniport and pfam databases [83–85]. Proteins were annotated using e-mapper [86]. Detected urease proteins were confirmed through SGTs. Briefly, SGTs were generated by first generating a protein alignment using linsi [74]. Alignments were trimmed with trimal and phylogenetic trees constructed using FastTree [76,87]. Hmmsearch was also used to detect urease genes within the predicted proteomes. Ethics statement This research was conducted under University of Miami Animal Care and Use Committee protocol number 22–150 adheres with all applicable rules, regulations, policies, standards, and guidelines that govern the humane, responsible, and judicious use of vertebrate animals under the Animal Welfare Act (AWA), the Public Health Service (PHS) Policy, and the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC).

Acknowledgments The authors acknowledge Cesar Cabot for help illustrating Fig 1A and Caleigh Weis for assistance with DNA clean-up for some microbiome samples.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3003764

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/