Genome-wide analysis of butterfly bush (Buddleja alternifolia) in three uplands provides insights into biogeography, demography and speciation

Ma, Yong-Peng; Liao, Rong-Li; Zhang, Ren-Gang; Yun, Quan-Zheng; Olmstead, Richard G.; Chau, John H.; Milne, Richard I.; Van de Peer, Yves; Sun, Wei-Bang

Genome-wide analysis of butterfly bush (Buddleja alternifolia) in three uplands provides insights into biogeography, demography and speciation

Ma, Yong-Peng; Liao, Rong-Li; Zhang, Ren-Gang; Yun, Quan-Zheng; Olmstead, Richard G.; Chau, John H.; Milne, Richard I.; Van de Peer, Yves; Sun, Wei-Bang

URI: https://repository.up.ac.za/handle/2263/87293

Date: 2021-11

Abstract:

Understanding processes that generate and maintain large disjunctions within plant species can provide valuable insights into plant diversity and speciation. The butterfly bush Buddleja alternifolia has an unusual disjunct distribution, occurring in the Himalaya, Hengduan Mountains (HDM) and the Loess Plateau (LP) in China. We generated a high-quality, chromosome-level genome assembly of B. alternifolia, the first within the family Scrophulariaceae. Whole-genome re-sequencing data from 48 populations plus morphological and petal colour reflectance data covering its full distribution range were collected. Three distinct genetic lineages of B. alternifolia were uncovered, corresponding to Himalayan, HDM and LP populations, with the last also differentiated morphologically and phenologically, indicating occurrence of allopatric speciation likely to be facilitated by geographic isolation and divergent adaptation to distinct ecological niches. Moreover, speciation with gene flow between populations from either side of a mountain barrier could be under way within LP. The current disjunctions within B. alternifolia might result from vicariance of a once widespread distribution, followed by several past contraction and expansion events, possibly linked to climate fluctuations promoted by the Kunlun–Yellow river tectonic movement. Several adaptive genes are likely to be either uniformly or diversely selected among regions, providing a footprint of local adaptations. These findings provide new insights into plant biogeography, adaptation and different processes of allopatric speciation.

Description:

Supplementary Material 1: Dataset S1 Morphological measurement and floral colour reflectance data for populations of Buddleja alternifola. Fig. S1 Phylogenetic trees inferred by ASTRAL- and ML-based approaches. Fig. S2 Patterns of linkage disequilibrium (LD). Fig. S3 Models 1–3, during the process of divergence among the three linkages, no gene flows with no changes in effective population size and (Model 1); with changes in effective population sizes starting from the divergence of TB (TDIV1), as well as SC and HT (TDIV2, Model 2); with changes in effective population sizes starting from TDIV1. Fig. S4 The phylogenomic tree used for time assignment of divergence for ancestral area reconstruction using representative samples of B. alternifolia and three species in the genus are currently available with re-sequencing data. Fig. S5 Cross-validation (CV) error and marginal likelihood values for different model K. Fig. S6 Reconstructing the phylogenomic relationships for 46 species of Buddleja using single-copy genes. Methods S1 Site ancestral state estimation. Methods S2 Estimating mutation rate of B. alternifolia. Methods S3 Reconstructing the phylogenomic relationships for 46 species of Buddleja using single-copy genes.

Supplementary Material 2: Notes S1 Reproducibility of analyses for BEAST and r8s files.

Supplementary Material 3: Table S1 Statistics of all assemblies. Table S2 Basic information with regards to genomes of 17 plants that were used to gene family analysis and the phylogenetic tree construction. Table S3 A matrix information on geographic distances among populations. Table S4 Environmental parameters used for assessment of ecological niche differentiation in B. alternifolia. Table S5 Geographical coordinate of B. alternifolia. Table S6 WGS-PacBio sequencing statistics. Table S7 WGS Illumina sequencing statistics. Table S8 HiC sequencing statistics. Table S9 Repeat annotations of the Buddleja alternifolia genome assembly. Table S10 Gene annotation statistics of the Buddleja alternifolia assembly. Table S11 Functional annotation of predicted genes in the Buddleja alternifolia genome. Table S12 Summary of the gene family analyses. Table S13 Basic information on location and genome mapping characteristics of all sampled individuals. Table S14 Summary of SNP annotations. Table S15 Global pairwise Fst between areas at the whole-genome level. Table S16 Pairwise Fst between areas in the divided nine subgroups of the whole genome, that is, eight in the gene region and one in the intergene region. Table S17 Results of nine models used in the fastsimocal analysis. Table S18 Basic parameters of three models compared in BioGeoBears, that is, Dec and Divalike based on dispersal-vicariance analysis, and Bayarea based on Bayesian inference of historical biogeography for discrete areas, with and without the founder-event speciation ‘J’ parameter. Table S19 Results of IBD and IBA analysis using simple and partial Mantel tests. Table S20 Shared genes detected by both approaches, red colour font indicating the shared genes of a significant overrepresentation with a specific GO term (P < 0.05). Table S21 Annotation of genes with significant GO terms (P < 0.05) detected by both approaches. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Show full item record