A computational comparative study of the repetitive DNA in the genus Quercus L

Abstract

An overall picture comparing the repetitive components of the genomes of three Quercus species was obtained by genome skimming with Illumina sequence reads. Read sets of Q. lobata, Q. robur, and Q. suber species were subjected to hybrid clustering in order to assemble a repeatome of the Quercus genus and to annotate it. The repeatome was composed of 8573 clusters. The abundance of repeated sequences in the three species was assessed by mapping Illumina reads of each species onto the repeatome. The repetitive portion of the genome was similar among the three species. The most abundant repetitive sequences were long terminal repeat-retrotransposons. Copia elements were overrepresented when compared with Gypsy ones. The most abundant retrotransposon lineages were SIRE for the Copia superfamily and Ogre/TAT for the Gypsy superfamily. Some of the clusters belonging to these lineages showed different transpositional time profiles among the three species. Ribosomal DNAs accounted for 3.64–6.75% of the repetitive component. Satellite DNAs were much more abundant in Q. lobata (8.68% of the genome) than in the two other species. However, different satellite DNAs showed large variations in their abundances. Overall, the composition of the repetitive portion of the genome showed some differences among oak species, suggesting a possible role of repeats for Quercus species differentiation. In the cases of Q. lobata and Q. robur, both of which belong to the Quercus section of the Quercus genus, such differences may be related to the different geographical origins of the species.