VCF files of loci in approximate linkage equilibrium for two common spruce budworm larval parasitoids: Apanteles fumiferanae (Hymenoptera: Braconidae) and Glypta fumiferanae (Hymenoptera: Ichneumonidae)

Periodic and spatially synchronous outbreaks of insect pests have dramatic consequences for boreal and sub-boreal forests. Within these multitrophic systems, parasitoids can be stabilizing agents by dispersing toward patches containing higher host density (the so-called birdfeeder effect). However, we know little about the dispersal abilities of parasitoids in continuous forested landscapes, limiting our understanding of the spatiotemporal dynamics of host–parasitoid systems, and constraining our ability to predict forest resilience in the context of global changes. In this study, we investigate the spatial genetic structure and spatial variation in genetic diversity of two important species of spruce budworm larval parasitoids during outbreaks: Apanteles fumiferanae Viereck (Braconidae) and Glypta fumiferanae (Viereck) (Ichneumonidae). Using parasitoids sampled in 2014 from 26 and 29 locations across a study area of 350,000 km2, we identified 1,012 and 992 neutral SNP loci for A. fumiferanae (N = 279 individuals) and G. fumiferanae (N = 382), respectively. Using DAPC, PCA, AMOVA, and IBD analyses, we found evidence for panmixia and high genetic connectivity for both species, matching the previously described genetic structure of the spruce budworm within the same context, suggesting similar effective dispersal during outbreaks and high parasitoid population densities between outbreaks. We also found a significant negative relationship between genetic diversity and latitude for A. fumiferanae but not for G. fumiferanae, suggesting that northern range limits may vary by species within the spruce budworm parasitoid community. These spatial dynamics should be considered when predicting future insect outbreak severities in boreal landscapes.We provide two VCF files of loci in approximate linkage equilibrium: one for Apanteles fumiferanae (Apanteles_fumiferanae_279inds_1056loci_LD_filtered.vcf) and one for Glypta fumiferanae (Glypta_fumiferanae_382inds_1044loci_LD_filtered.vcf). We also provide the associated metadata from Table 1 of the article (outbreak areas, sites numbers, and latitude and longitude) in two text files (Apanteles_fumiferanae_Metadata.txt and Glypta_fumiferanae_Metadata.txt). Funding provided by: Fonds Québécois de la Recherche sur la Nature et les TechnologiesCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100003150Award Number: Team grant: 174142Spruce budworm larvae were collected from late May to mid-July 2014 from 29 sites located within and around six distinct outbreak areas in the province of Québec, Canada. Larvae were reared individually on artificial diet until egression of Apanteles fumiferanae and Glypta fumiferanae larvae. Adult parasitoids were then stored individually in dry conditions at -80°C prior to DNA extraction. Because both A. fumiferanae and G. fumiferanae are haplodiploid, total genomic DNA was extracted only from females (1–23 females per species per site). Genotyping-by-sequencing libraries were constructed using the protocol of Poland, Brown, Sorrells, and Jannink (2012), and then single-end sequenced (100 bp reads) using an Illumina HiSeq2000. Because there is no reference genome available for A. fumiferanae and G. fumiferanae, we used the TASSEL 3.0 UNEAK (Universal Network-Enabled Analysis Kit) pipeline for single-nucleotide-polymorphism (SNP) calling (Lu et al., 2013). Only sequences with five or more reads over all individuals were kept, creating a master list of sequences which serves as a 'pseudo reference genome'. During alignment, a 0.03 error tolerance rate. SNP loci with a minor allele frequency 25% of individuals and individuals with > 10% missing loci were pruned out of the data matrix using TASSEL 5. We then created a pruned dataset of SNP loci for each species by removing SNPs in high linkage disequilibrium (correlation coefficient ≥ 0.3) were dropped to remove highly correlated variants