Efficient de-novo Assembly of a Large Genome for Constructing a Reference Genome of the Pepper (Capsicum annuum L.)

As advances in next generation sequencing (NGS) technologies, overwhelming amount of genome sequencing data were generated to make a whole genome. Although lots of assemblers have been developed and computing performances have been much more powerful than previous conditions, completion of whole genome sequencing is still one of great challenge. Pepper (Capsicum annuum), which has a large genome (3.5 GB), is an important vegetable crop used as condiments for hot and spicy food. To construct reference genome of pepper, we generated 632.2 GB (182x) of whole genome shotgun (WGS) sequences of C. annuum CM334 using Illumina GA2 and Hiseq 2000. Raw data filtering processes removing duplicate reads and sequencing error using in-house pipeline were carried to get an accurate assembled data and 233 Gb (66.6x) sequences were remained. For genome assembly and scaffolding, velvet (EBI), NGScell (CLCbio), SOAPdenovo (BGI), and SSPACE (BaseClear) which were widely available and representative assemblers were used. The best version of assembly was selected among assembled results and gaps were filled using a gapcloser (BGI). The total length of assembled scaffolds was 3.09 GB (88.2% of whole genome) and the total number of scaffolds was 211,174 with 285,138 bp of N50. In addition to the WGS sequencing, 2,382 of pepper BAC clones were selected and sequenced by 454 or Solexa. Current objective of pepper genome project is increasing sizes of scaffolds and integration of WGS and BAC assembly is carried to generated unified pepper genome except 35 of completed BAC contigs already used for validation of WGS assembly. Taken together, it will be reference sequences for understanding and dissecting pepper genome.