Chromosome-scale assemblies of three Ormosia species: repetitive sequences distribution and structural rearrangement

Abstract

Background: The genus Ormosia belongs to the Fabaceae family; almost all Ormosia species are endemic to China, which is considered one of the centers of this genus. Thus, genomic studies on the genus are needed to better understand species evolution and ensure the conservation and utilization of these species. We performed a chromosome-scale assembly of O. purpureiflora and updated the chromosome-scale assemblies of O. emarginata and O. semicastrata for comparative genomics.

Findings: The genome assembly sizes of the 3 species ranged from 1.42 to 1.58 Gb, with O. purpureiflora being the largest. Repetitive sequences accounted for 74.0-76.3% of the genomes, and the predicted gene counts ranged from 50,517 to 55,061. Benchmarking Universal Single-Copy Orthologs (BUSCO) analysis indicated 97.0-98.4% genome completeness, whereas the long terminal repeat (LTR) assembly index values ranged from 13.66 to 17.56, meeting the "reference genome" quality standard. Gene completeness, assessed using BUSCO and OMArk, ranged from 95.1% to 96.3% and from 97.1% to 98.1%, respectively.Characterizing genome architectures further revealed that inversions were the main structural rearrangements in Ormosia. In numbers, density distributions of repetitive elements revealed the types of Helitron and terminal inverted repeat (TIR) elements and the types of Gypsy and unknown LTR retrotransposons (LTR-RTs) concentrated in different regions on the chromosomes, whereas Copia LTR-RTs were generally evenly distributed along the chromosomes in Ormosia.Compared with the sister species Lupinus albus, Ormosia species had lower numbers and percentages of resistance (R) genes and transcription factor genes. Genes related to alkaloid, terpene, and flavonoid biosynthesis were found to be duplicated through tandem or proximal duplications. Notably, some genes associated with growth and defense were absent in O. purpureiflora.By resequencing 153 genotypes (∼30 Gb of data per sample) from 6 O. purpureiflora (sub)populations, we identified 40,146 single nucleotide polymorphisms. Corresponding to its very small populations, O. purpureiflora exhibited low genetic diversity.

Conclusions: The Ormosia genome assemblies provide valuable resources for studying the evolution, conservation, and potential utility of both Ormosia and Fabaceae species.

Keywords: RNA-seq; SNP calling; comparative genomics; gene duplication; gene families; genetic diversity; outlier loci; population genetics; repeat-mediated chromosome architectures; structure variation.

Figure 1:

Photographs of Ormosia purpureiflora. (A) O. purpureiflora seeds. Seed sizes measured using a ruler are illustrated in the lower right panel. (B) O. purpureiflora flowers. (C) O. purpureiflora fruit in the distance showing a diseased state or insect invasion. (D) O. purpureiflora seeds showing invasion by worms/insects or a diseased state. (E) O. purpureiflora flowers in a diseased state or under insect invasion. (F) O. purpureiflora natural habitat; the arrow shows the sampled individual (a small shrub) used for genome assembly.

Figure 2:

Sampled (sub)populations and population genetics of Ormosia purpureiflora. (A) Map showing (sub)populations sampled for Ormosia purpureiflora. (B) Principal component analysis (PCA) results showing the first 3 PCs (PC1 versus PC2; PC1 versus PC3) for O. purpureiflora individuals sampled from different (sub)populations. (C) Admixture results representing data for K = 2–6 clusters.

Figure 3:

Ormosia genomes and comparative genomics. (A) Hi-C interaction heat maps (bin length, 100,000 bp) for the genome assemblies of three Ormosia species. (B) Circos plot showing the genome features (chromosome, repeat density in length proportions, repeat density in numbers, gene density, and syntenic blocks from outer to inner) across chromosomes of the genome assemblies of three Ormosia species. Repeat densities in each Circos plot was quantified by all repetitive elements. For O. purpureiflora, the Circos plot also includes SNP density results between the results of gene density and syntenic blocks. All densities were estimated using a 1 Mbp sliding window. (C) The inferred phylogenetic tree, divergence time, and contracted (–) and expanded (+) gene families in O. purpureiflora and other species. (D) The density distribution of synonymous nucleotide substitutions (Ks) in the whole genome duplication analysis for Ormosia species and their sister species, L. albus. (E) Syntenic blocks among Ormosia species and L. albus. (F) Intrachromosomal structural variations observed among the three Ormosia species.

Figure 4:

Smoothing lines for gene and repeat density distributions (bin size, 100,000 bp) along chromosomes in Ormosia species. The repeat densities were measured by the number of each repetitive element. Scatterplots for the gene density and repeat distribution are presented in Supplementary Fig. S4. Box sizes correspond to chromosome sizes in Ormosia species, whereas the black bars on the upper part of each box (chromosome) represent the hot structural rearrangement region in the chromosomes.