Data and code from: Assessment of preservation and extraction methods for environmental DNA in sediments
Data files
Apr 30, 2026 version files 7.14 GB
-
AOYG-95_Euka03_corrected.zip
5.58 KB
-
AOYG-96_Euka03_corrected.zip
5.74 KB
-
AOYG-97_Bact01_corrected.zip
5.58 KB
-
AOYG-98_Bact01_corrected.zip
5.73 KB
-
bact_river_cleaned_TSE.RData.xz
554.52 KB
-
bact_sea_cleaned_TSE.RData.xz
773.80 KB
-
euka_river_cleaned_TSE.RData.xz
322.50 KB
-
euka_sea_cleaned_TSE.RData.xz
457.78 KB
-
Figures_papier_TSE_Jan2025.zip
5.59 KB
-
OBIToolsScript_Bact01.sh
2.77 KB
-
OBIToolsScript_Euka03.sh
2.77 KB
-
R_script_fit_models_TSE_2025.zip
4.38 KB
-
raw_data.tar.xz
7.14 GB
-
README.md
3.96 KB
-
TSE_Metadata.xlsx
34.38 KB
Abstract
Sediments host diverse benthic organisms, but natural and anthropogenic disturbances threaten their biodiversity and the ecosystem services they provide. Accurate sediment biodiversity assessments are pivotal for biomonitoring and conservation. Environmental DNA (eDNA) allows studies of multiple taxonomic groups to be performed at broad spatial scales. However, standardized operational protocols are needed from DNA collection to data analysis to ensure data reliability and replicability. Here, we assessed the effects of eDNA preservation methods on the ecological results obtained for bacterial, fungal, microeukaryotic and metazoan communities in marine and freshwater ecosystems. We used preservation methods such as freezing, room temperature (RT), silica gel, and Longmire buffer for one and three weeks, and compared eDNA metabarcoding results to those obtained when eDNA was extracted on the day of sampling (control). For each strategy, we extracted DNA using two protocols targeting extracellular (Phosphate) or both extracellular and intracellular DNA (PowerMax). Both protocols led to similar ecological conclusions, with controls showing shared eukaryote taxa across extraction methods, while Phosphate extracted most bacterial taxa. Preservation methods slightly affected community composition, with patterns differing across groups and habitats. Community composition was shaped by preservation strategies, with heterogeneous responses across bacterial, fungal, and eukaryotic assemblages. Preservation methods significantly influenced alpha diversity, with effects varying across taxonomic groups, habitats, and extraction protocols. These results highlight the importance of choosing appropriate preservation methods for eDNA in sediments, depending on the study goal, and offer insights into cost-effective available extraction techniques for sediment biomonitoring.
https://doi.org/10.5061/dryad.9zw3r22nf
This dataset contains raw sequence data, filtered data, metadata and scripts to reproduce the bioinformatic and statistical analyses of the study.
Description of the data and file structure
For both Bact01 and Euka03 primer pairs, two libraries were sequenced with a paired-end approach (2x150 bp) and they are provided in a .fastq format. Each library is associated to a .ngs file containing information about sample identity. Raw sequence reads were analyzed using the OBITools software suite (Boyer et al., 2016)(see Code to reproduce bioinformatic analyses).
raw_data.tar.xz
|- Bact01/
| |- 211224_SN1126_A_L001_AOYG-97_R1.fastq/
| |- 211224_SN1126_A_L001_AOYG-97_R2.fastq/
| |- 211224_SN1126_A_L001_AOYG-98_R1.fastq/
| |- 211224_SN1126_A_L001_AOYG-98_R2.fastq/
| |- AOYG-97_Bact01.ngs/
| |- AOYG-98_Bact01.ngs/
|- Euka03/
| |- 211220_SN7280_A_L001_AOYG-95_R1.fastq/
| |- 211220_SN7280_A_L001_AOYG-95_R2.fastq/
| |- 211220_SN7280_A_L001_AOYG-96_R1.fastq/
| |- 211220_SN7280_A_L001_AOYG-96_R2.fastq/
| |- AOYG-95_Euka03.ngs/
| |- AOYG-96_Euka03.ngs/
After the bioinformatic step, data curation was conducted in R (R Core Team, 2022) with the metabaR package (Zinger et al., 2021) to remove sequences that may bias ecological conclusions. MOTUs with the following characteristics were excluded from the datasets:
(1) MOTUs with a best identity with the reference database < 80%;
(2) MOTUs that were more numerous in the PCR negative controls than in true PCR replicates (contaslayer function with the "max" method);
(3) MOTUs with a relative frequency <3% within a PCR replicate (tagjumpslayer function ran with default setting).
PCR replicates with < 1000 reads were also eliminated before aggregating the replicates belonging to the same sample with the aggregate_pcrs function. At this stage, sequences observed <10 times per sample were set to zero, and samples with < 1000 reads overall were discarded.
Filtered datasets are provided for both Bact01 and Euka03 as R Workspace (.RData):
|-bact_river_cleaned_TSE.RData.xz
|-bact_sea_cleaned_TSE.RData.xz
|-euka_river_cleaned_TSE.RData.xz
|-euka_sea_cleaned_TSE.RData.xz
Metadata are provided in the Excel (.xlsx) format:
|-TSE_Metadata.xlsx
Description of metadata variables
Each row corresponds to a single sample.
- ID_Argaly: Unique identifier assigned to each sample within the Argaly project database.
- ID_client: Identifier associated with the client.
- Ecosystem: Type of ecosystem from which the sample was collected.
- Site: Sampling site code.
- Extraction: DNA extraction method used for the sample.
- Preservation_expected: Expected preservation method applied to the sample according to the experimental design.
- Preservation_observed: Preservation method actually applied during sample processing.
- Time: Duration of preservation condition (T0: immediately extracted; T1: one week of preservation; T2: three weeks of preservation)
- Replicate: Biological sampling replicate at each sampling Site
- Method: General methodological category combining preservation and duration.
All ecological analyses were performed using the R statistical software (version 4.3.0; R Core Team 2022)(see Core to reproduce statistical analyses).
Code/Software
All commands used to perform bioinfomatic analyses are provided here :
|-OBIToolsScript_Bact01.sh
|-OBIToolsScript_Euka03.sh
Information about sample identity are provided here:
|-AOYG-95_Euka03_corrected.zip
|-AOYG-96_Euka03_corrected.zip
|-AOYG-97_Bact01_corrected.zip
|-AOYG-98_Bact01_corrected.zip
All commands and packages used to perform statistical analyses are provided here :
|-R_script_fit_models_TSE_2025.zip
|-Figures_papier_TSE_Jan2025.zip
Two libraries per marker were prepared by Fasteris (Geneva, Switzerland; https://www.fasteris.com/en-us/NGS) following the MetaFast protocol (Taberlet et al., 2018) and sequenced with the Illumina MiSeq platform using a paired-end approach (2x150 bp for Euka03 and 2x250 bp for Bact01). Raw sequence reads were analyzed using the OBITools software suite (Boyer et al., 2016) specially dedicated to the handling of metabarcoding data. All ecological analyses were performed using the R statistical software (version 4.3.0; R Core Team 2022).
