Data from: DNA metabarcoding as a tool to study plankton responses to warming and salinity change in mesocosms
Data files
Sep 01, 2025 version files 1.35 MB
-
all_plakton_micro_wide.csv
24.82 KB
-
Aquacosm2022_DataTZS.csv
5.79 KB
-
long_table_with_taxonomy.csv
695.33 KB
-
mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv
289.57 KB
-
mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv
327.57 KB
-
new_metadata.tsv
957 B
-
README.md
8.76 KB
Abstract
Climate change is transforming marine ecosystems, with rising temperatures and changing salinity patterns expected to reshape plankton communities in the Baltic Sea. As key components of marine food webs and biogeochemical cycles, plankton are highly sensitive to environmental change. Here, we examined the effects of warming and salinity change on plankton communities using a mesocosm experiment at the Tvärminne Zoological Station, Finland. We employed both traditional microscopy-based identification and DNA metabarcoding (18S rRNA and COI markers) to assess shifts in phytoplankton, ciliates, and mesozooplankton. Our findings indicate that salinity primarily affected higher trophic levels, while warming influenced lower ones. Warmer conditions increased community evenness and favoured mixotrophic and heterotrophic taxa, whereas salinity effects were most pronounced in rotifers and copepods, reflecting species-specific tolerances. Interactive effects varied, with salinity sometimes buffering warming impacts and other times intensifying them, highlighting complex stressor interactions. Microscopy allowed for a more precise quantification of plankton abundance, whereas metabarcoding captured a broader taxonomic diversity. Our results suggest that freshening and warming in the Baltic Sea may lead to a shift towards smaller, mixotrophic and bloom-forming plankton species, with potential consequences for ecosystem functioning. This study highlights metabarcoding’s value in mesocosm research while emphasising the need to refine molecular techniques for ecological interpretations.
https://doi.org/10.5061/dryad.bvq83bkkq
Description of the data and file structure
This dataset contains species abundance data from a mesocosm experiment conducted at Tvärminne Zoological Station, Finland (59.84° N, 23.25° E), between September and October 2022. The experiment investigated the effects of warming and salinity changes on plankton communities using microscopy-based counts, DNA metabarcoding (18S rRNA and COI markers), and associated environmental metadata.
Files and variables
File: new_metadata.tsv
Description: This file contains metadata associated with the mesocosm experiment to use with metabarcoding tables, including sample identifiers, environmental conditions, and treatment groups.
Variables
- sample_metab: Unique identifier for each sample.
- Treatment: Identifier for the experimental mesocosm.
- Date: Timepoint of sample collection (Start, Mid, End).
- Temperature: Temperature treatment (Low, High).
- Salinity: Salinity treatment in PSU (Practical Salinity Units).
File: all_plakton_micro_wide.csv
Description: This file contains plankton abundance data obtained from microscopy, with individual species abundances recorded for each sample in wide format with species names.
Variables
- date: Timepoint of sample collection (Start, Mid, End).
- sample: Unique identifier for each sample.
- salinity: Salinity treatment in PSU (Practical Salinity Units).
- temperature: Temperature treatment (Low, High).
- mesocosm_id: Identifier for the experimental mesocosm.
- Other columns are individual species names
File: Aquacosm2022_DataTZS.csv
Description: This file contains extra environmental data from the mesocosm experiment, including temperature, salinity and chlorophyll measurements.
Variables
- Mesocosm: Identifier for the experimental mesocosm.
- Temperature_treatment: Temperature treatment (Low, High).
- Salinity_treatment: Salinity treatment in PSU (Practical Salinity Units).
- DOY: Day of the year.
- Date: Date of sample collection (YYYY-MM-DD).
- Temperature: Actual water temperature (°C).
- Salinity: Actual salinity (PSU).
- O2_mgL: Dissolved oxygen concentration (mg/L).
- Fluorescence: Fluorescence in RFU (Relative Fluorescence Units).
File: mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv
Description: This file contains DNA metabarcoding data from the 18S rRNA gene marker, including taxonomic assignments and sequence abundance across samples.
Variables
- amplicon: Unique identifier for each ASV (Amplicon Sequence Variant).
- upper_similarity_vsearch: Upper similarity threshold from VSEARCH taxonomic assignment.
- lower_similarity_vsearch: Lower similarity threshold from VSEARCH taxonomic assignment.
- references_vsearch: Accession numbers of reference sequences used for taxonomic assignment in VSEARCH.
- taxonomy_vsearch: Taxonomic classification based on VSEARCH.
- taxonomy_idtaxa: Taxonomic classification based on IDTAXA.
- confidence_idtaxa: Confidence score for IDTAXA classification.
- sequence: DNA sequence of the ASV.
- total: Total sequence reads per ASV.
- spread: Distribution of sequence reads across samples.
- Blank-22-9, Blank-6-10, Blank-8-9, etc: Control sample identifiers.
- m1-22-9, m2-6-10, etc: Mesocosm sample identifiers.
- pcr-0ctrl1, pcr-0ctrl3: PCR negative control sample identifiers.
File: mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv
Description: This file contains DNA metabarcoding data from the COI gene marker, including taxonomic assignments and sequence abundance across samples.
Variables
- amplicon: Unique identifier for each ASV (Amplicon Sequence Variant).
- upper_similarity_vsearch: Upper similarity threshold from VSEARCH taxonomic assignment.
- lower_similarity_vsearch: Lower similarity threshold from VSEARCH taxonomic assignment.
- references_vsearch: Accession numbers of reference sequences used for taxonomic assignment in VSEARCH.
- taxonomy_vsearch: Taxonomic classification based on VSEARCH.
- taxonomy_idtaxa: Taxonomic classification based on IDTAXA.
- confidence_idtaxa: Confidence score for IDTAXA classification.
- sequence: DNA sequence of the ASV.
- total: Total sequence reads per ASV.
- spread: Distribution of sequence reads across samples.
- Blank-22-9, Blank-6-10, Blank-8-9, etc: Control sample identifiers.
- m1-22-9, m2-6-10, etc: Mesocosm sample identifiers.
- pcr-0ctrl1, pcr-0ctrl3: PCR negative control sample identifiers.
File: long_table_with_taxonomy.csv
Description: This file contains long-format species abundance data with taxonomic classifications.
Variables
- date: Timepoint of sample collection (Start, Mid, End).
- sample: Unique identifier for each sample.
- salinity: Salinity treatment in PSU (Practical Salinity Units).
- temperature: Temperature treatment (Low, High).
- mesocosm_id: Identifier for the experimental mesocosm.
- species: Species/taxa names.
- units.per.l: Individuals/cells per litre.
- taxonomy: Full taxonomic classification of the species.
NA: In this dataset, NA represents values that were not available (e.g., unassigned taxonomy or unmeasured variables).
Code/software
Details about metabarcoding workflows for generating ASV and OTU tables are available at https://gitlab.com/tvarminne-metabarcoding/mesocosm-18s-bioinfo and https://gitlab.com/tvarminne-metabarcoding/mesocosm-coi-bioinfo. Zip files of these is also available here under 'mesocosm-18s-bioinfo-main-2.zip' and 'mesocosm-coi-bioinfo-main.zip'.
Code for analysis:
This dataset includes R scripts used for analysing mesocosm experiment data, including statistical modelling, visualisation and diversity analyses. The analyses were performed using R (version 4.4.2).
R packages required:
- tidyverse: Data manipulation
- vegan: NMDS, PERMANOVA and diversity analysis
- ggplot2: Visualisation of bar plots and graphs
- phyloseq: Analysis of metabarcoding data
- nlme: Linear models and mixed-effects models
- dplyr: Data manipulation
- ggpubr: Multiple plotting
- patchwork: Combines multiple ggplot figures into a single layout
- gridExtra: Arranges multiple grid-based plots
- cowplot: Multi-panel figures
- MicrobiomeStat: Statistical analysis for microbiome and community data (LinDA analysis)
- purrr: Handling lists and iteration
- RColorBrewer: Predefined color palettes for data visualisation
- data.table: Data manipulation and aggregation
- grid: Arranging plots
Description of R scripts:
-
Diversity_models_plots_micro_metaB.R: Runs diversity models and creates plots comparing microscopy and metabarcoding diversity data. Uses data: 'long_table_with_taxonomy.csv', 'mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv', 'mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv', and 'new_metadata.tsv'.
-
Fluorescence_graph.R: Generates fluorescence graph visualisations. Uses data: 'Aquacosm2022_DataTZS.csv'
-
LinDA script final.R: Performs differential abundance analysis (LinDA) on ASV/OTU data. Uses data: 'mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv', 'mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv,' and 'new_metadata.tsv'.
-
linear_models_micro.R: Runs statistical models on microscopy-based abundance data. Uses data: 'long_table_with_taxonomy.csv'.
-
NMDS PERMANOVA.R: Conducts Non-Metric Multidimensional Scaling (NMDS) and PERMANOVA analyses. Uses data: 'all_plakton_micro_wide.csv', 'mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv', 'mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv', and 'new_metadata.tsv'.
-
zooplankton_ciliate_protist relative abundance barplot ASV and micro.R: Generates ASV/OTU and species relative abundance bar plots for zooplankton, ciliates, and protists. Uses data: 'long_table_with_taxonomy.csv', 'mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv', 'mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv,' and 'new_metadata.tsv'.
-
zooplankton_ciliate_protist richness barplot ASV and micro.R: Generates ASV/OTU and unique species richness bar plots for zooplankton, ciliates, and protists. Uses data: 'long_table_with_taxonomy.csv', 'mesocosm-coi_dada2_v1.0.filtered.table.d4.lulu.with.taxo.tsv', 'mesocosm-18s_dada2_v1.0.filtered.table.lulu.with.taxo.tsv,' and 'new_metadata.tsv'.
Samples were collected from mesocosms at a depth of 1 m. Temperature, salinity, fluorescence, and dissolved oxygen were measured every two days using a calibrated digital water meter (MU 6100 H, VWR). Water samples (4 L) were taken every two days, stored in a 10 L plastic container, and transported to the laboratory for further analysis.
Microscopy-based plankton identification: Phytoplankton, ciliates, and zooplankton samples were preserved with Lugol’s iodine (1%) and counted using Utermöhl chambers under an inverted microscope (Olympus CKx41). Phytoplankton and ciliates were identified to genus level, while zooplankton were identified to genus or species where possible. Enumeration methods followed standard protocols, with abundance reported as cells or individuals per liter.
Metabarcoding: DNA samples for metabarcoding were collected every two weeks following modified protocols from Minamoto et al. (2019). A 1000 mL water sample was filtered onto 47 mm diameter, 0.7µm pore size GF/F filters (Whatman, USA) using a vacuum pump. Filters were stored at -80°C and transported on dry ice to the Molecular Ecology and Systematics Laboratory, Finland, for sequencing. DNA was extracted using the DNeasy® Blood and Tissue kit (Qiagen) with modifications, including Buffer ATL and a 6-hour incubation. Extracted DNA was quantified via a NanoDrop 1000 spectrophotometer, and integrity was assessed by electrophoresis. The cytochrome c oxidase subunit I (COI) gene was amplified with primers mlCOIintF and HCO2198, while the 18S rRNA V9 region was amplified with primers 1391F and EukBr. PCR products were dual-indexed, pooled, and sequenced on an Illumina MiSeq platform (MiSeq V3 600-cycle flow cell) with paired-end reads (326 bp Read 1 and 278 bp Read 2). Raw sequencing reads were processed using Cutadapt v3.1 for primer removal and quality-controlled with DADA2 in R. Forward and reverse reads were merged, and chimeric sequences were removed. COI ASVs were clustered into OTUs using Swarm (local similarity threshold d = 4) and refined with LULU. Taxonomic classification used PR2 5.0.0 with IDTAXA (40% confidence threshold) for 18S ASVs and MIDORI2 (GB254) with VSEARCH (≥80% similarity) for COI OTUs. Raw sequencing data is publicly available in the European Nucleotide Archive (ENA: PRJEB79753). Details about workflows are available at https://gitlab.com/tvarminne-metabarcoding/mesocosm-18s-bioinfo and https://gitlab.com/tvarminne-metabarcoding/mesocosm-coi-bioinfo.