Short-term temporal variation of coastal marine eDNA
Jensen, Mads Reinholdt et al. (2022), Short-term temporal variation of coastal marine eDNA, Dryad, Dataset, https://doi.org/10.5061/dryad.q2bvq83kx
Temporal variation in eDNA signals is increasingly explored for understanding community ecology in aquatic habitats. Seasonal changes have been addressed using eDNA sampling, but very little is known regarding short-term temporal variation that spans hours to days. To address this, we filtered marine water samples from a single coastal site in Denmark every hour for 32 h. We used metabarcoding to target both fish and broader eukaryote diversity and evaluated temporal changes in this marine community. Results revealed variation in fish species richness (15–27) and eukaryote class richness (35–64) across the 32 h of sampling, and we further evaluated sampling efforts needed to reach different levels of diversity saturation. Relative read frequency data for both fish and eukaryotes indicated a clear diel change in community composition, with different communities detected during daylight versus dark hours. The abundance signals in our data reflected biological variation rather than stochastic variation, since replicates taken at the same hour were more similar to each other than those taken at different hours. Our compositional results indicated a dynamic community, rather than a static pool of eDNA—even across a few hours. The fish data showed a daily pattern of relative species abundances, and the uncoupling of fish and broader eukaryote data suggest that variation in eDNA profiles across a single day can provide valuable information reflecting diel changes, at least for highly mobile organism groups. However, our results also point to several pitfalls in current eDNA experimental design, in which samples are taken over large areas without relative time-consistency or short-term replication. Our findings shed new light on short-term variation in coastal eDNA and have wide implications for experimental study design and for incorporating temporality into project conceptualization for future aquatic biodiversity monitoring.
This dataset represents environmental DNA sequencing data from Skovshoved Harbour (Denmark) collected on 11th-12th of September, 2017. Samples were collected every hour for 32 hours straight, starting from 10 AM on the 11th and until 5 PM on the 12th (see connected publication for additional details).
DNA has been amplified with two different primer sets, namely the MiFish (for 12S fish data) and the 18S_allshorts (for 18S eukaryote data) primers. MiFish primers consists of the forward primer MiFish-U-F (5′-GTCGGTAAAACTCGTGCCAGC-3′) and the reverse primer MiFish-U-R (3′-GTTTGACCCTAATCTATGGGGTGATAC-5′), targeting a 163-185 bp fragment of 12S. Eukaryote amplification was done with forward primer 18S_allshorts forward (5’-TTTGTCTGSTTAATTSCG-3’) and reverse primer 18S_allshorts reverse (5’-CACAGACCTGTTATTGC-3’), targeting ca. 110 bp of 18S. The libraries have been sequenced using paired end NovaSeq 6000 sequencing (150 BP PE).
Libraries are named M1-M4 for the fish data and M1_KBJ-M4_KBJ for the eukaryote data.
We suggest you put the files in 8 separate folders in order to demultiplex. The files should be placed accordingly:
Folder 1: MiFish PCR replicate one
M1_MD5.txt (to check sums)
Folder 2: MiFish PCR replicate two
M2_MD5.txt (to check sums)
Folder 3: MiFish PCR replicate three
M3_MD5.txt (to check sums)
Folder 4: MiFish PCR replicate four
M4_MD5.txt (to check sums)
Folder 5: 18S_allshorts (eukaryote) PCR replicate one
M1_KBJ_MD5.txt (to check sums)
Folder 6: 18S_allshorts (eukaryote) PCR replicate two
M2_KBJ_MD5.txt (to check sums)
Folder 7: 18S_allshorts (eukaryote) PCR replicate three
M3_KBJ_MD5.txt (to check sums)
Folder 8: 18S_allshorts (eukaryote) PCR replicate four
M4_KBJ_MD5.txt (to check sums)
Following this folder structure, each folder will now contain two sequence data files (paired end sequencing), a barcode/tag file and an MD5 file for checking sums.
If you would like to demultiplex this data, all tag information needed is available in the ”list” files. Each tag file contains 52 samples, which are explained below:
10P-2.17P: Pooled samples (see associated article), with numbers denoting time of day the samples were taken. The prefix ”2.” denotes day two of sampling.
CON1 and CON2: Field blanks taken at 11.00h and 17.00h on day one, respectively.
12.1, 12.2, 12.3, 20.1, 20.2, 20.3, 04.1, 04.2 and 04.3: Individual samples (see associated article) sequenced to inspect what happens during pooling and to inspect biological variation vs. stochastic variation.
CNE_P5, CNE_P6, CNE_P7, CNE_P8, CNE_P9, CNE_P10, CNE_P11, and CNE_P12: Extraction blanks (one was included for each round of extraction).
NTC_X4: PCR blanks. Four PCR blanks were run in each PCR setup using the same tag, so these blanks will all appear in the same sample.
The ”list” files include the sample name followed by the PCR replicate number. The two following columns represent the tags used for each sample (both forward and reverse primer were tagged). Tags are consistent across PCR replicates.
After demultiplexing, you should be able to do as you please with the data.
If you want to follow the exact filtering and data analysis done in our study, we refer to the manuscript for further details after the demultiplex step. If you have any questions, feel free to send an email to Mads Reinholdt Jensen with any questions you may have.