Airborne DNA and spider webs outperform other eDNA sources for monitoring terrestrial vertebrates
Data files
Oct 14, 2025 version files 58.56 GB
-
bc1_12S.txt
6.10 KB
-
bc1_16S.txt
9.61 KB
-
bc2_12S.txt
3.77 KB
-
bc2_16S.txt
12.88 KB
-
bc3_12S.txt
7.35 KB
-
bc4_12S.txt
5.83 KB
-
Distance_matrix.csv
31.04 KB
-
eDNAFlow_LCA_results_12S-V5.zip
20.60 MB
-
eDNAFlow_LCA_results_16S_Reptile.zip
50.03 MB
-
NextSeq02_FTP01_S1_R1_001.fastq.gz
19.83 GB
-
NextSeq02_FTP01_S1_R2_001.fastq.gz
20.27 GB
-
NSRun03_FTP401_S1_L001_R1_001.fastq.gz
18.39 GB
-
README.md
4.33 KB
-
Substrate_comparison_metadata.csv
23.03 KB
Abstract
Understanding the strengths and limitations of different environmental DNA substrates is essential for optimizing terrestrial vertebrate surveys and monitoring. However, the performance of newly explored substrates (airborne eDNA, vegetation swabs, spiderwebs) compared to longstanding eDNA sources (water and soil) is uncertain. Using a metabarcoding approach, we assessed vertebrate eDNA diversity across seven substrates: three airborne DNA collection methods (a powered fan system, and two passive collection methods), spider webs, vegetation swabs, water, and soil at two sites. The highest species richness was detected by powered air samples and spider webs (powered air: 83 at Perth Zoo, 44 at Karakamia; spider webs: 62 and 40, respectively), with no significant difference in the community composition, suggesting they capture eDNA from similar sources however, all substrates contributed unique detections. Passive airborne DNA collection, though less efficient than powered devices, (mean species richness per sample: 14.8 vs. 5.8 at Perth Zoo; 6.9 vs. 2.7 at Karakamia), demonstrated great potential as their low cost and simplicity may enable increased replication or longer deployment times, potentially increasing the probability of species detection. Our direct comparison of terrestrial eDNA substrates shows that airborne DNA is a step forward and not just hype. However, substrate-specific biases were evident, with vegetation swabs favouring arboreal mammals while water was dominated by aquatic and semi-aquatic species, highlighting the influence of species ecology on DNA deposition. eDNA studies targeting terrestrial vertebrates must consider the heterogeneity of vertebrate DNA distribution across ecosystems and the need for careful selection of eDNA substrates.
Dataset DOI: 10.5061/dryad.6wwpzgnb8
Description of the data and file structure
Metabarcoding data generated from two assays and seven sample types collected: three different methods were used to collect eDNA from the air - powered air samplers, wind-powered samplers, and bare filter media as a passive collection method and soil, spider webs, swabbed vegetation surfaces and water samples. All samples were collected throughout both Karakamia Wildlife sanctuary and Perth Zoo.
Files and variables
File: bc1_16S.txt and bc2_16S.txt
Description: Text file containing the unique ID barcoded tags required to demultiplex raw 16S Reptile sequences
File: bc1_12S.txt, bc2_12S.txt, bc3_12S.txt and bc4_12S.txt
Description: Text file containing the unique ID barcoded tags required to demultiplex raw 12S-V5 sequences:
File: Substrate_comparison_metadata.csv
Description: Contains site and sample metadata
Variables
- sample_id: Unique sample identifier
- site_id: Unique sample site identifier
- Latitude: Sample site latitude
- Longitude: Sample site longitude
- Site: Greater site identifier (Karakamia, Perth Zoo)
- Habitat: Habitat type
- Sample_method: Substrate type sampled
File: NextSeq02_FTP01_S1_R1_001.fastq.gz
Description: compressed sequence file containing raw sequences from 16S Reptile paired-end illumina NextSeq run
File: NextSeq02_FTP01_S1_R2_001.fastq.gz
Description: compressed sequence file containing raw sequences from 16S Reptile paired-end illumina NextSeq run
File: NSRun03_FTP401_S1_L001_R1_001.fastq.gz
Description: compressed sequence file containing raw sequences from 12S-V5 single-end illumina NextSeq run
File: eDNAFlow_LCA_results_16S_Reptile.zip
Description: Filtered data produced after sequence analysis using the eDNAFlow bioinformatics pipeline and a custom lowest common ancestor (LCA) assignment script for the 16S Reptile assay. These data represent the final curated sequence outputs used in the associated publication.
This folder contains four subfolder containing the following files:
- 06_Unique_ZOTUs* – fasta files containing zero-radius operational taxonomic units (ZOTUs) generated after quality filtering and dereplication and .txt files containing associated ZOTU table.
- 07_blast* – Results of BLASTN searches of each ZOTU against the local GenBank nucleotide database.
- 08_lulu* – curtained_zotuTable.tab containing output after LULU post-clustering curation and lulu_zotu_map.tab listing retained and merged ZOTUs with similarity and co-occurrence information used to identify redundant OTUs.
- 09_taxonomyAssigned* – Final taxonomic assignments produced using custom LCA pipeline.
File: eDNAFlow_LCA_results_12S-V5.zip
Description: Filtered data produced after sequence analysis using the eDNAFlow bioinformatics pipeline and a custom lowest common ancestor (LCA) assignment script for the 12S-V5 assay. These data represent the final curated sequence outputs used in the associated publication.
This folder contains four subfolder containing the following files:
- 06_Unique_ZOTUs* – fasta files containing zero-radius operational taxonomic units (ZOTUs) generated after quality filtering and dereplication and .txt files containing associated ZOTU table.
- 07_blast* – Results of BLASTN searches of each ZOTU against the local GenBank nucleotide database.
- 08_lulu* – curated_zotuTable.tab containing output after LULU post-clustering curation and lulu_zotu_map.tab listing retained and merged ZOTUs with similarity and co-occurrence information used to identify redundant OTUs.
- 09_taxonomyAssigned* – Final taxonomic assignments produced using custom LCA pipeline.
File: Distance_matrix.csv
Description: Data matrix consisting of both occurrence (presence/absence) and distance to eDNA detection from Perth Zoo eDNA samples
Access information
Data was derived from the following sources:
- All sequencing data was generated on the illumina NextSeq 1000/2000 platform
- All sequencing analysis was conducted using eDNAFlow bioinformatics pipeline