Exome capture from the spruce and pine giga‐genomes

Abstract

Sequence capture is a flexible tool for generating reduced representation libraries, particularly in species with massive genomes. We used an exome capture approach to sequence the gene space of two of the dominant species in Canadian boreal and montane forests - interior spruce (Picea glauca x engelmanii) and lodgepole pine (Pinus contorta). Transcriptome data generated with RNA-seq were coupled with draft genome sequences to design baits corresponding to 26 824 genes from pine and 28 649 genes from spruce. A total of 579 samples for spruce and 631 samples for pine were included, as well as two pine congeners and six spruce congeners. More than 50% of targeted regions were sequenced at >10× depth in each species, while ~12% captured near-target regions within 500 bp of a bait position were sequenced to a depth >10×. Much of our read data arose from off-target regions, which was likely due to the fragmented and incomplete nature of the draft genome assemblies. Capture in general was successful for the related species, suggesting that baits designed for a single species are likely to successfully capture sequences from congeners. From these data, we called approximately 10 million SNPs and INDELs in each species from coding regions, introns, untranslated and flanking regions, as well as from the intergenic space. Our study demonstrates the utility of sequence capture for resequencing in complex conifer genomes, suggests guidelines for improving capture efficiency and provides a rich resource of genetic variants for studies of selection and local adaptation in these species.

Using transcriptome sequencing and pooled exome capture to study local adaptation in the giga‐genome of Pinus cembra

“…The single‐copy gene set showed no enrichment in GO terms compared to the complete gene set of the transcriptome, indicating that the single‐copy genes represent an unbiased subset of the full gene set. The proportion of removed SNPs is very similar to that reported by Yeaman et al () in two North‐American conifer species, for which the SNP set (Suren et al, ) was reduced by an order of magnitude. The discrepancy in the proportion of removed SNPs versus contigs comes from the fact that paralogous contigs generally contain more SNPs, and the probability of detecting multi‐copy contigs increases with the number of SNPs.…”

“…In the absence of genomic resources, this information is best acquired by RNA‐Seq and subsequent transcriptome assembly. Exome capture has been successfully applied in nonmodel species with large genomes, some of them over 20 Gbp (e.g., Neves, Davis, Barbazuk, & Kirst, ; Suren et al, ; Yeaman et al, ), and is a promising approach for population‐based genome scans and association studies (Jones & Good, ). Disadvantages of using exome capture include the fact that noncoding regions are ignored (including introns that can interfere with hybridization), the labour‐intensive design of probes (but see Puritz & Lotterhos, ) and the large variation in capture efficiency among probes and samples (Bamshad et al, ; Neves et al, ).…”

Capturing variation in Lens (Fabaceae): Development and utility of an exome capture array for lentil

“…1) and include not only the exons but also introns, untranslated flanking regions, and some extent of intergenic space (Table 2). The amount of capture outside of the target regions depends on DNA fragment length, which was targeted to be 350–380 bp in this study, and this is consistent with previous studies (Henry et al., 2014; Suren et al., 2016). More than 86% of the sequence data are within 400 bp of the probe target regions.…”

Expressed exome capture sequencing: A method for cost‐effective exome sequencing for all organisms

“…In general, it is thought that capture probes that span exon boundaries will result in low coverage of these regions (Jones & Good, ) or that certain regions will not be covered at all (Neves, Davis, Barbazuk, & Kirst, ). Inclusion of too many boundaries may also lower overall capture performance by increasing off‐target capture (Suren et al., ). EecSeq exome probes are derived from mature RNA, so some of the probes will inevitably span exon boundaries.…”