A long-read phased assembly of the highly heterozygous fig (Ficus carica L.) genome reveals epigenetic patterns.

Assembly of non-model plant species is challenging due to high DNA heterozygosity and repeat content. Here, we report our work to build a high-quality genome reference using single-molecule, real-time sequencing for one of the oldest known domesticated species: the fig (Ficus carica L.). Using long-read sequencing data performed better than combining long and short-reads. The assembled fig genome is ~333 Mb of which 80% have been anchored to 13 chromosomes. The genome has an estimated repeat content of 37% and contains 37,840 protein-coding genes. Genome-wide analysis of N6-methyladenine and N4-methylcytosine have revealed high methylation levels in both genes and transposable elements and a prevalence of methylated over un-methylated genes. The high quality reference genome along with the characterization of methylation profiles of two unconventional DNA modifications provide an important resource not only for fig breeding but also for investigating the relationship between specific epigenetic changes and the phenotype of plants.