Assembly and Functional Annotation of the Salix purpurea L. Chloroplast Genome

Assembly and functional annotation of Salix purpurea L. chloroplast genome provides a valuable resource for phylogenetics, genomics, and shrub willow feedstock breeding efforts. A subset of paired-end NGS reads were filtered and batch assembled into contigs de novo using Velvet and ABySS (mer=19..51,2; cov=5x), as well as mapped to related chloroplast genomes with Bowtie2, extracted and iteratively re-assembled. Consensus sequences were aligned by MAAFT, ambiguities resolved using GapCloser, and then manually annotated using DOGMA. Transcriptomes were assembled de novo, mapped to consensus gene models (id=95, len_cov=80) and verified by a BLASTN/X query. Transcript copy number variation was analyzed in egdeR and summarized using hierarchical clustering to define modules of co-expression within an intraspecific S. purpurea F₁ family and between eight tissue-types. The quadripartite structure is an estimated 156,405bp in length consisting of an 85 Kb Large Single Copy region, and 16 Kb Small Single Copy region, separated by a pair of 27 Kb inverted repeats. The genome encodes 93 mRNA, 29 tRNA (37 tRNA-Aragorn), and 8 rRNA. Of the 53 predicted RNA-editing sites (qual=99) within 20 transcripts, 58% reside in five NADH subunits. The number of editing sites positively correlates with transcript abundance; thus preferentially edited across tissue-type. The most significant differentially expressed transcripts (p <10E-5; logFC >5) between parent and hybrid were PS II, Cytb₆f, and NADH subunits. Comparative analyses in the Salicaceae will offer insight into the contrasting patterns of compensatory posttranscriptional editing and find use as a valuable genomic tool for divergence estimates.