Data from: The quantity of deposited environmental DNA in plant-insect interactions depends on taxon, abundance, and interaction time

De Vivo, Mattia 1 ; Krehenwinkel, Henrik1

Published Oct 28, 2024 on Dryad. https://doi.org/10.5061/dryad.4mw6m90jh

Abstract

eDNA metabarcoding holds great promise as a simple and efficient tool to detect plant-insect interactions. However, the influence of both abiotic and biotic factors on eDNA deposition in plant-insect interactions is not well understood. It especially remains to be tested how much eDNA abundances reflect the frequency or intensity of interactions of insects and plants.

We experimentally analyzed the quantitative deposition of eDNA from three insect species (the southern green stink bug Nezara viridula, the mustard beetle Phaedon cochleariae and the fall armyworm Spodoptera frugiperda) on leaves of two host plants of varying attractiveness: savoy cabbage (Brassica oleracea var. sabauda, preferred host) and tomato (Solanum lycopersicum, less preferred host). We tested for the effects of insect taxonomy, host plant, and exposure time and abundances of interacting insects on the plant material.

Our data shows a clear quantitative ecological signal in eDNA deposition. Insect abundance and interaction time is reflected in the amount of deposited eDNA. Moreover, significantly more eDNA was deposited on cabbage, the preferred host plant. Besides these ecological drivers, a very strong taxonomic eDNA deposition bias between different insect taxa was observed.

eDNA detection is strongly influenced by the ecological interactions of the targeted taxa, highlighting the utility of the tool to specifically detect trophic interactions between plants and arthropods. However, taxonomic biases in eDNA deposition preclude comparative analyses of interactions at the community level.

https://doi.org/10.5061/dryad.4mw6m90jh

Scripts for making the figures and analyses from “The quantity of deposited environmental DNA in plant-insect interactions depends on taxon, abundance, and interaction time” plus the raw reads used for the study.

Description of the data and file structure

Sarah-ZBJ*fastq.gz files

Those are the raw reads are, available in several compressed FASTQ files. We processed them through the bioinformatic pipeline described in the methods of the manuscript.

Files used for the R analyses

data.csv

It is a comma-delimited CSV file, which can be opened with a spreadsheet software (e.g., Microsoft Excel), a text editor or R. It has basically all the data used for the analyses regarding the read count. It is split in the following columns:

species: The target species in the replicate
vegetable: The host plant used in the replicate
individuals: Number of target species’ individuals used in the replicate
hours: How long the individuals were left on the host plant
reads: How many reads of the target species were recovered
exp: Code that summarize the replicate: e.g., SBcF1 indicate Nezara viridula (which has the code SB) on cabbage (c), with few individuals (F) and that was left for 1 hour (1).

gc_all.xlsx

XLSX file, which can be opened with a spreadsheet software (e.g., Microsoft Excel), a text editor or R. It has the data with the GC content for the COI of the considered species. It is split in the following way:

species: The target species
G: Guanine percentage
C: Cytosine percentage
GC: GC content

Code/Software

The R scripts are present in the “.R” files. Specifically:

-“1_plotting.R” has the scripts for allowing to make the barplots and the boxplots of the analyses.

-“2_statistical_analyses_reads.R” has the scripts for making the ANOVA, Shapiro-Wilks and GLM performed in the study.

-“3_GC_analyses.R” has the scripts for the t-test regarding the GC content of the species.

The “.R” files can be opened with a text editor or straight with R.

For the analyses, we used R (version 4.3.1), the “base” package (version 4.3.1) and the “stats” package (version 4.3.1). For plotting and tables, we used R, the “gtsummary” package (version 1.7.2) and “tidyverse” (version 2.0.0).