Stick Insect Genomes Reveal Natural Selection’s Role in Parallel Speciation

See allHide authors and affiliations

Science  16 May 2014:
Vol. 344, Issue 6185, pp. 738-742
DOI: 10.1126/science.1252136

You are currently viewing the figures only.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

  1. Fig. 1 Population structure and whole-genome divergence between host-plant ecotypes of T. cristinae.

    (A) A maximum likelihood tree from whole-genome data from individuals from T. cristinae populations (rooted with other species of Timema not shown). Black circles depict nodes with >80% bootstrap support (100 replicates). Populations LA, HVA, MRA, and R12A were collected from Adenostoma. Populations PRC, HVC, MRC, and R12C were collected from Ceanothus. The geographic locations of the populations are depicted in fig. S1. (B) Patterns of genetic divergence (FST) along the genome. LG indicates linkage group. Colors denote whether SNPs occur in moderately (black) or highly (red/orange) divergent genetic regions, delimited with a HMM approach (weakly differentiated regions in blue are too sparse to see). Results are shown at three different scales. (Top) FST estimates at all ~4.4 million SNPs for each of the four population pairs. (Middle) Genomic divergence for two scaffolds (scaffolds 1271 on LG7 and 2230 on LG6 from HVA × HVC). (Bottom) Magnified view of genomic divergence for one parallel HMM divergence region that harbored a SNP showing parallel allele frequency divergence between hosts upon transplantation (scaffold 556 on LG4 from R12A × R12C; see also Fig. 3). [Credit: courtesy of R. Ribas.]

  2. Fig. 2 Quantification of parallel and nonparallel divergence across population pairs.

    (A) The majority of divergent SNPs (those above the 90th quantile of the empirical FST distribution) were so only in a single comparison, indicative of largely nonparallel genetic divergence across the genome (pie chart; see table S1 for full results, including those from the 99th quantile). Nonetheless, a significant enrichment of parallelism was observed (bar plots, all P < 0.0001, permutation tests of observed versus expected numbers of parallel divergent SNPs). (B) High FST in two population pairs. (C) High FST in three population pairs. (D) High FST in four population pairs. [Credit: courtesy of R. Ribas.]

  3. Fig. 3 Allele frequency changes across the genome in a field transplant experiment.

    (A) The experimental design, where individuals from population LA were transplanted to a nearby location of five paired blocks that contained a single plant individual of each host species. (B) An empirical quantile plot of the mean divergence in allele frequencies between hosts for all SNPs. (C) SNPs showing the largest parallel divergence between hosts (>99.5th quantile) for individual blocks (orange, greater change on Adenostoma; blue, greater change on Ceanothus). (D) SNPs showing the largest parallel divergence between hosts (>99.5th quantile) averaged across blocks. (E) SNPs showing parallel divergence between hosts in the experiment were in parallel HMM divergence regions between natural population pairs more often than expected by chance (P < 0.05, randomization test). [Credit: courtesy of R. Ribas.]

Stay Connected to Science