SNP datasets from invasive Bombus terrestris on Tasmania, Australia
Data files
Nov 09, 2023 version files 137.69 MB
Abstract
Invasive species are predicted to adjust their morphological, physiological, and life-history traits to adapt to their non-native environments. Although a loss of genetic variation during invasion may restrict local adaptation, introduced species often thrive in novel environments. Despite being founded by just a few individuals, the bumblebee Bombus terrestris (Hymenoptera: Apidae) has successfully spread across the island of Tasmania (Australia) in less than 30 years, becoming abundant and competitive with native pollinators. We use RADseq to investigate what neutral and adaptive genetic processes associated with environmental and morphological variation allow B. terrestris to thrive as an invasive species in Tasmania. Across 15 sites, we found high gene flow with low genetic diversity, significant isolation-by-distance, and spatial variation in effective migration rates. A longitudinal band of restricted migration was evident across the mid-central region of Tasmania, corresponding to sites with high elevation, pastural land, low wind speeds and low precipitation seasonality. Tajima’s D indicated a recent population expansion for central sites extending from the south to the north of the island. Significant selection signatures were found for loci in relation to precipitation, wind speed, and wing loading. Candidate loci were annotated to genes with functions related to cuticle water retention and insect flight muscle stability. Understanding how a genetically impoverished invasive bumblebee has rapidly adapted to a novel island environment provides further understanding about the evolutionary processes that determine successful insect invasions and the potential for invasive hymenopteran pollinators to spread globally.
README: SNP datasets
https://doi.org/10.5061/dryad.8931zcrxc
The three SNP datasets are VCF files and consists of 160 Bombus terrestris samples.
Description of the data and file structure
Dataset 1 (TAS_bombus_dataset1.vcf.gz) consists of 35,836 SNPs and was generated for neutral genetic clustering analyses. Setting in populations in STACKS were -R 0.80, --write-single-snp, --hwe 0.05 and --min-mac 2. The dataset was also pruned for SNPs in linkage-disequilibrium using the --indep-pairwise 50 (window size), 5 (step size), and 0.2 (r^2 threshold) option in Plink v.1.9.
Dataset 2 (TAS_bombus_dataset2.vcf.gz) (90,396 SNPs) was filtered identically to dataset 1 except it included all SNPs per locus in order to reconstruct the full haplotype information. This dataset was used for fineRADstructure analysis.
Dataset 3 (TAS_bombus_dataset3.vcf.gz) (33,832 SNPs) was generated for selection analyses (-R 0.80, --write-single-snp, --min-maf 0.05 and --min-mac 2), whereby loci in linkage disequilibrium were not pruned, but a single SNP per RAD tag was retained.
For all three datasets full-siblings has been removed (--relatedness2 filter in VCFtools). Individuals with high missing data and low read depth were removed (populations: -R 0.5, --hwe 0.05). Data quality was checked using the --missing-indv and --site-mean-depth options in VCFtools.
Methods
Wild B. terrestris workers were sampled from 16 locations across the Australian continental island state of Tasmania and subsequently extracted for DNA (DNeasy Blood and Tissue extraction kit (Qiagen) and sent for RADseq at to Floragenex, Inc. (Portland, OR, USA). Samples were individually sequenced on an Illumina HighSeq 4000, generating 150bp paired-end reads with an average depth of coverage between 12–112x per sample.