PacBio Single Molecule Sequencing to Improve the Norway Spruce Genome Annotation

The recent advances of massive parallel sequencing (RNA-seq) provide great opportunities for gene discovery with unprecedented breadth and depth. The first Norway spruce (Picea abies) genome annotation is well supported by RNA-seq data from a broad biological sample collection. However, correct splice alignment of short reads in a complex plant genome such as spruce remains a challenging task due to the presence of huge amounts of repetitive sequences, large gene family sizes and the presence of ‘gene-like’ fragments. On the other hand, single molecule sequencing instruments can generate multi Kb sequences with the potential to improve gene discovery without transcriptome assembly. The main drawback of single molecule sequencing however is the high error rate, which hinders current applications in genome sequencing projects.

Here, we demonstrate a proof-of-concept long read error correction workflow to improve the read quality in a timely fashion. High quality long reads allow the well developed long read splice aligner such as GMAP to precisely determine splice site  junctions. We also present the roadmap of the second version of the spruce genome annotation based on combining the strengths of RNAseq and Pacific Biosciences (PacBio) cDNA sequencing to improve our understanding of spruce genes. Splice variants can be unambiguously confirmed by more than one transcripts without the need of assembling short reads or based on short read abundance to infer isoforms. Long introns commonly seen in the spruce genes can be easily verified by long reads without laborious PCR confirmation. The full-length cDNA enrichment protocol allows us to determine transcription start sites and polyadenylation tails. Full transcripts provide direct evidence to support the RNA-scaffolded genome sequence in the first release of assembly. PacBio cDNA reads will be further used in the RNA scaffolding procedure to improve the second version of the spruce genome assembly.