Western redcedar single nucleotide polymorphism (SNP) genotyping data for genomic selection and population genetics

Western redcedar (Thuja plicata) Single Nucleotide Polymorphism (SNP) data in Variant Call Format (VCF) for genomic selection training and target populations, genomic selection parents, and self-fertilized (selfing) lines, comprising 4,833 trees. Targeted sequencing-based genotyping was done by Capture-Seq methodology at Rapid Genomics (Neves est al. 2013). A set of 57,000 probes as designed for initial marker discovery, from which a panel of 20,858 probes was selected for genotyping. A set of transcriptomes (Shalev et al. 2018) (PRJNA704616) was aligned to the reference genome to identify SNPs. Candidate probes (120 nt) were initially designed in silico and 57,000 selected by removing candidates with poor base composition for hybridization (GC content 0.6, high G content >0.2 and long homopolymers >7), followed by removing probes aligning to more than one position on the reference genome (≥90% identity and length). The 57,000 probes represent 14,517 scaffolds (average 3.9 probes/scaffold), with 37,275 targeting at least one SNP and 19,725 mapping to intergenic regions not containing pre-identified SNPs. A set of 128 individuals were selected to validate the 57,000 probe panel and associated polymorphisms. Genomic DNA (0.5 ug) was fragmented (mean size 300 bp), followed by repair of ends, phosphorylation, adenylation, ligation of Illumina compatible adapters containing 8bp indexes and 5’ T-overhang, and 10 cycles PCR amplification with universal primers to produce sequencing-ready libraries. Libraries were quantified using PicoGreen. Libraries from 16 samples were pooled, hybridized to the 120 nt RNA probes following Agilent’s SureSelect Target Enrichment System (Agilent Technologies) and sequenced on an Illumina HiSeq X machine with paired-end 150bp cycle for an average sequencing depth per sample of 15X. Sequence data were aligned to the reference genome with BWA-MEM (http://arxiv.org/abs/1303.3997) and sets of four samples were combined to increase sequencing depth for identifying markers. Putative SNPs were identified using Freebayes (http://arxiv.org/abs/1207.3907) in 150bp on either side of the 57,000 probes and filtered probes that had more than 17 SNPs per 420 bp target region (150bp + 120bp + 150bp). The sequencing depth of the probes was used to select the final set of 20,885 probes, removing probes on both sides of the distribution (low and high sequencing depth), for Capture-Seq on the remainder of the samples.