Assembling the Genome Sequence of the Plastids and Mitochondria of the White Spruce (Picea glauca)

The SMarTForests project has published a draft genome sequence of the 20 gigabase white spruce genome, seven times the size of the human genome, which was sequenced using the Illumina HiSeq and MiSeq sequencing platforms (Birol et al. 2013). Within this large dataset, sequence reads originating from the organellar genomes are expected. Whereas one copy of the diploid nuclear genome is found in each cell, hundreds of organelles are present, and thus hundreds of copies of the organellar genome sequences. This abundance results in an overrepresentation of the organellar genome sequences in the whole genome sequencing data. We can therefore separate the sequencing reads that originate from the organellar genomes from the nuclear genome sequence by identifying reads that are highly abundant. Once the organellar reads are isolated, they can be assembled de novo to produce the genome sequence of the mitochondria and plastids.

Contigs of the draft white spruce genome that have unusually high depth of coverage were identified and separated from the remainder of the assembly. These high-depth contigs are composed of the organellar genomes as well as high copy number repeats. The sequencing reads were mapped to the entire draft white spruce genome, and reads that mapped to these contigs with high depth of coverage were separated. These reads were assembled using ABySS. The resulting contigs were aligned to the Norway spruce (Picea abies) complete plastid genome (NC_021456) and putative mitochondrial sequences using BWA-MEM to identify putative white spruce mitochondrial and plastid sequences.