Repeat annotation
Repetitive elements, including tandem repeats and transposable elements (TEs), were predicted in the O. kokonorica genome based on a combination of homology-based and de novo approaches. Tandem repeats were annotated using Tandem Repeats Finder v. 4.09 (Benson, 1999). TEs were identified using RepeatMasker v. 4.1.055 and RepeatProteinMask (Tarailo-Graovac & Chen, 2009) with Repbase v. 22.11 (Jurka et al., 2005) as the query library. Next, RepeatModeler v. 1.0.1.0 (Tarailo-Graovac & Chen, 2009) was used to construct a de novorepeat library for the identification of TEs that were not found in the Repbase library. We used LTR_FINDER (Xu & Wang, 2007) to search the full-length long terminal repeat (LTR) retrotransposons in O. kokonorica , C. songorica . We extracted the 5′ and 3′ LTR sequences based on the LTR_FINDER annotation results, which were subsequently aligned using MUSCLE v. 3.8.31 (Edgar, 2004). The genetic distance (K) between the 5′ and 3′ LTR sequences was calculated using DNADIST, a program within PHYLIP v. 3.696 (Felsenstein, 2004). The insert time (T) of each LTR-RT was calculated using the formula: T = K/2r, where r is the nucleotide substitution rate estimated by BASEML (see below).