Search

Long terminal repeat retrotransposons (LTR-RTs) contribute a large fraction of many sequenced plant genomes and play important roles in genomic diversity and phenotypic variations. LTR-RTs are abundantly distributed in plant genomes, facilitating the development of markers based on LTR-RTs for a variety of genotyping purposes. Whole-genome analysis of LTR-RTs was performed in Cleistogenes songorica.A total of 299,079 LTR-RTs were identified and classified as Gypsytype, Copiatype, or other type. LTR-RTs were widely distributed in the genome, enriched in the heterochromatic region of the chromosome, and negatively correlated with gene distribution. However, approximately one-fifth of genes were still interrupted by LTR-RTs, and these genes are annotated. Furthermore, four types of primer pairs (PPs) were designed, namely, retrotransposon-based insertion polymorphisms, inter-retrotransposon amplified polymorphisms, insertion site–based polymorphisms, and retrotransposon-microsatellite amplified polymorphisms. A total of 350 PPs were screened in 23 accessions of the genus Cleistogenes, of which 80 PPs showed polymorphism, and 72 PPs showed transferability among Gramineae and non-Gramineae species. In addition, a comparative analysis of homologous LTR-RTs was performed with other related grasses. Taken together, the study will serve as a valuable resource for genotyping applications for C. songoricaand related grasses.

Melilotus albusMedik. is a legume species with potential to be used as a forage or conserved fodder in Northwest China. Breeding M. albusas a forage crop is focused on simultaneously improving forage yield and reducing coumarin (Cou) content. The goal of the present study was to estimate genotypic variation for several key agronomic traits and to evaluate the impact of single‐trait selection and Smith–Hazel‐index‐based multitrait selection for herbage dry weight (DW), plant height (PH), stem number (SN), and leaf area (LA). A quantitative genetic analysis was conducted on trait measurements from 25 half‐sib (HS) families evaluated at Yuzhong and Linze, in China. Significant (P< 0.05) genotypic variation among the HS families for all traits was observed. There was also significant (P <0.05) genotype × environment interaction for all traits except for LA. The predicted genetic gain, based on estimated genotypic variance, for single‐trait selection varied from 2.3 to 30.3% per selection cycle. Pattern analysis of the HS family × multitrait best linear unbiased prediction matrix provided a graphical summary of the association among the traits, as well as the four HS family groups generated from clustering. Some members within Group 1 could be crossed to generate a breeding population to be used in recurrent selection for M. albuscultivar development in Northwest China. The applicability of a Smith–Hazel selection index to increase expression of the traits DW, PH, and SN and reduce LA, indirectly reducing Cou content, was confirmed. This index enabled identification of families for increasing key biomass traits and decreasing Cou production.