Abstract:

This dataset provides an estimate of the relative abundance of active microbial eukaryotes within and across 3 treatments (oil, oil and Corexit, and a dilute oil and Corexit treatment) in the GOMCOAST mesocosm experiment based on 18S RNA sequences from RNAseq data. The data reflects the relative presence, activity and detectability of eukaryotic species captured on an 8 micron polycarbonate filter.

Suggested Citation:

Zoe V. Finkel, Andrew J. Irwin, Yue Liang, Chris M. Brown, Michaël Bradet-Legris, Laura Bretherton, Antonietta Quigg. 2018. Percent relative abundance of the eukaryotic microbial community in the GOMCOAST (Gulf of Mexico, Coastal Water) mesocosm from 8 micron polycarbonate filters. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/N7JD4V8Z

Publications:

Bretherton, L., Kamalanathan, M., Genzer, J., Hillhouse, J., Setta, S., Liang, Y., … Quigg, A. (2019). Response of natural phytoplankton communities exposed to crude oil and chemical dispersants during a mesocosm experiment. Aquatic Toxicology, 206, 43–53. doi:10.1016/j.aquatox.2018.11.004

Purpose:

To determine how oil and oil and dispersant mixtures influence eukaryotic microbial community composition.

Data Parameters and Units:

Raw data_GOMCOAST-taxonomic-composition-data-analysis.csv KPC: Kingdom, Phylum, or Class, derived from columns 2-4 and chosen to produce a useful decomposition of total diversity in a small number of categories Kingdom: The Kingdom of the taxon, from World Registry of Marine Species (WoRMS, marinespecies.org) Phylum: The phylum of the taxon from WoRMS Class: The class of the taxon from WoRMS Genus: The genus ot the taxon from WoRMS target_id: The ID from the silva SSU non-redundant database version 123 qval.WAF: The multiple-comparisons corrected p-value for the hypothesis that relative abundance is the same in control and WAF treatments b.WAF: The log2 fold change in relative abundance between WAF and control (> 0 means higher relative abundance in WAF) sig.WAF: A flag of 0 or 1 to indicate if there is a change due to the treatment (1= change); used to produce the summary table qval.dCEWAF: The multiple-comparisons corrected p-value for the hypothesis that relative abundance is the same in control and dilute CEWAF treatments b.dCEWAF: The log2 fold change in relative abundance between dilute CEWAF and control (> 0 means higher relative abundance in WAF) sig.dCEWAF: A flag of 0 or 1 to indicate if there is a change due to the treatment (1= change); used to produce the summary table qval.CEWAF: The multiple-comparisons corrected p-value for the hypothesis that relative abundance is the same in control and CEWAF treatments b.CEWAF: The log2 fold change in relative abundance between CEWAF and control (> 0 means higher relative abundance in WAF) sig.CEWAF: A flag of 0 or 1 to indicate if there is a change due to the treatment (1= change); used to produce the summary table control: The estimated relative abundance (as %) of this taxon from kallisto in the control WAF: The estimated relative abundance of this taxon from kallisto in the WAF treatment DCEWAF: The estimated relative abundance of this taxon from kallisto in the dilute CEWAF treatment CEWAF: The estimated relative abundance of this taxon from kallisto in the CEWAF treatment Summary_GOMCOAST-taxonomic-composition-data-analysis.csv Number with significant differential abundance relative to control: Category, n, WAF, dCEWAF, CEWAF Proportion of total abundance (%): Control, WAF, DCEWAF, CEWAF Kingdom-Phylum-Class: The taxonomic affiliation derived from column 1 of the raw data table Count of Genus: The number of taxa (rows) aggregated from the raw data table to produce this summary Sum of sig.WAF: The number of these taxa with significant differential abundance in WAF relative to control Sum of sig.dCEWAF: The number of these taxa with significant differential abundance in dilute CEWAF relative to control Sum of sig.CEWAF: The number of these taxa with significant differential abundance in CEWAF relative to control Sum of control: The proportion (as a %) of total abundance attributed to this taxon in the control Sum of WAF: The proportion (as a %) of total abundance attributed to this taxon in the WAF treatment Sum of DCEWAF: The proportion (as a %) of total abundance attributed to this taxon in the dilute CEWAF treatment Sum of CEWAF: The proportion (as a %) of total abundance attributed to this taxon in the CEWAF treatment

Methods:

Twelve 100L mesocosm tanks were filled with Gulf of Mexico seawater collected from the Texas coastline, near TABS buoy R (29° 38.1000'N, 93° 38.5020'W) which is located ~100 miles away from Galveston (TX). Four treatments were prepared in triplicate. Control tanks were filled with seawater. Water accommodated fraction (WAF) of oil was prepared by mixing 25 mL (5 ml ~ every 30 min for 2.5 hrs) of Macondo surrogate oil into 130 L of seawater. Mixing ended 24 hrs. after the initial oil addition (Knap et al. 1986; Wade et al. 2017in preparation). The WAF was then introduced into the WAF mesocosm tanks and filled to 87 L and mixed. From these WAF tanks 6 L was removed for other experiments and analyses (2 L dark/light, 4 L hydrocarbon analyses). In order to make chemically enhanced water accommodated fraction (CEWAF), Corexit was mixed with oil in a ratio of 1:20 and 25 mL of this mixture (5 ml every 30 min for 2.5 hrs) of surrogate oil plus Corexit was added to 130 L of seawater. Mixing ended 24 hrs after the initial oil addition. The CEWAF was then introduced into the CEWAF mesocosm tanks and filled to 96 L and mixed. From these CEWAF tanks 13 L was removed for other experiments and analyses (7 L for the DCEWAF mesocosms, 2 L dark/light, 4 L hydrocarbon analyses). Diluted CEWAF (DCEWAF) was prepared by mixing 9 L of CEWAF with 78 L of the original seawater for a total volume of 87 L. From these DCEWAF tanks 6 L was removed for other experiments and analyses (2 L dark/light, 4 L hydrocarbon analyses). To the water in the 12 mesocosms, nutrients were added (final concentration f/20) and the tanks stirred. Banks of lights were placed behind each of the glass mesocosm tanks and a 12:12 light/dark cycle employed. Sampling commenced and defined as time zero. The estimated oil equivalents (EOE) were determined using Macondo surrogate oil as the calibration standard (Wade et al. 2011) for the fluorescence analyses (Horiba Scientific Aqualog Fluorometer). The EOE mean concentration of the three mesocosms for the control, WAF, DCEWAF and CEWAF at the start of the experiments were 0 mg/L, 0.26 mg/L, 2.74 mg/L and 41.5 mg/L, respectively. The EOE mean concentration of the three mesocosms for the in the control, WAF, DCEWAF and CEWAF after 72 hours were 0 mg/L , 0.06 mg/L, 1.03. and 17.3 mg/L, respectively. The estimate of the relative abundance of active microbial eukaryotes within and across treatments is based on 18S RNA sequences from RNAseq data and reflects the relative presence and activity and detectability of each species. RNA was harvested from each of the 12 mesocosm tanks (3 replicate tanks for each treatment: control, WAF, CEWAF, DECWAF) 72 hours after the initiation of the experiment. Several hundred mL (250 to 4000 mL) were rapidly and gently filtered onto two 47mm, 8 m polycarbonate filters. We limited the total filtration time to 20 minutes. It took more time to filter water from the CEWAF tanks, and therefore less volume was filtered from this treatment. The filters and denaturation solution (Ambion Simply RNA) was added to Y-matrix bead beater tubes (MoBio). The samples were lysed using a SuperFastprep2 bead beater (30 seconds at the maximum setting) and immediately stored in a -80°C freezer. RNA was extracted by exposing samples immediately after thawing to 2 additional 30 second rounds in the SuperFastprep2 bead beater. The Ambion Total RNA kit (ThermoFisher AM1910) was used to extract RNA followed by DNA removal with the Ambion Turbo DNAfree kit (ThermoFisher AM1907) as per manufacturer instructions. RNA was sequenced as 125 base pair paired-end reads using Illumina HiSeq 250 RNAseq by Genome Quebec. PolyA selection was used to remove the majority of the rRNA using the NEBNext Poly(A) mRNA magnetic isolation module kit from New England Biolabs. Approximately 2-10% of the original rRNA sequences pass through this step are sequenced (Abernathy and Overturf 2016). Trimmomatic was used to remove Illumina adapters and low quality bases were identified using Phred scores (Bolger, Lohse et al. 2014). Kraken and the Silva SSU rRNA Ref NR 99 database (release 123) were used to filter reads matching 18S sequences (Quast, Pruesse et al. 2013, Wood and Salzberg 2014). Kallisto and Sleuth were used to match, count and perform differential expression and relative abundance analysis for reads against the Silva 18S database (Bray, Pimentel et al. 2016, Pimentel, Bray et al. 2017). Three replicate mesocosms were sequenced for the control and WAF treatment, but following quality control only two replicates were available for the DCEWAF treatment and one replicate for the CEWAF treatment. Each 18S sequence in the Silva database is identified with an accession code and a hierarchical taxonomic identification. All non-Eukaryota, Metazoa and Embryophyta sequences were removed from the analyses. The taxonomic identification in Silva is hierarchical but the identifications do not all have the same number of levels in the hierarchy so the phylum, order, or class, for example, are not readily extracted. Where possible we use the taxonomic hierarchy used by WoRMS in 2016/17. The genus and species name for each taxon in the Silva database (the last entry in the hierarchy) was used to search the World Registry of Marine Species (WoRMS, marinespecies.org). A variety of secondary sources were used to identify the taxonomic hierarchy for taxa without a match in WoRMS including the Global Names Index (GNI), the Pan-European Species Infrastructure (PESI), the Paleobiology database (Paleo), and the taxonomic hierarchy used by Wikipedia/Wikispecies. An automatic search for all taxa was performed using lifewatch.be on 2016 December 31 using WoRMS, GNI, PESI, and Paleo.

Error Analysis:

There is no error quantified for the individual point estimates of relative abundance across treatments. The average relative abundance was estimated from replicate samples when available as described in the methods.

Provenance and Historical References:

The original raw sequence data from the Illumina sequencer have been deposited at NCBI’s Short read archive (SRA) under BioProject PRJNA338185. The data are archived as BioSample SAMN08118040 and are accessible at http://www.ncbi.nlm.nih.gov/biosample/8118040 Jason Abernathy & Ken Overturf (2016) Comparison of Ribosomal RNA Removal Methods for Transcriptome Sequencing Workflows in Teleost Fish, Animal Biotechnology, 27:1, 60-65, https://doi.org/10.1080/10495398.2015.1086365 Anthony M. Bolger, Marc Lohse, Bjoern Usadel; Trimmomatic: a flexible trimmer for Illumina sequence data, Bioinformatics, Volume 30, Issue 15, 1 August 2014, Pages 2114–2120, https://doi.org/10.1093/bioinformatics/btu170 Christian Quast, Elmar Pruesse, Pelin Yilmaz, Jan Gerken, Timmy Schweer, Pablo Yarza, Jörg Peplies, Frank Oliver Glöckner; The SILVA ribosomal RNA gene database project: improved data processing and web-based tools, Nucleic Acids Research, Volume 41, Issue D1, 1 January 2013, Pages D590–D596, https://doi.org/10.1093/nar/gks1219 Derek Wood and Steven Salzberg (2014) Kraken: ultrafast metagenomic sequence classification using exact alignments, Genome Biology, 15:R46. https://doi.org/10.1186/gb-2014-15-3-r46 Nicolas Bray, Harold Pimentel, Pall Melsted & Lior Pachter. Near-optimal probabilistic RNA-seq quantification (2016). Nature Biotechnology 34(5): 525-527. https://doi.org/10.1038/nbt.3519. Harold Pimentel, Nicolas Bray, Suzette Puente, Pall Melsted & Lior Pachter. Differential analysis of RNA-seq incorporating quantification uncertainty (2017). Nature Methods 14(7):687-690. https://doi.org/10.1038/nmeth

Individual Files: