The phenotypic spaces of Arabidopsis thaliana F2s vs. wild accessions. The results shown correspond to one of the 10 hypervolume calculations. Left: Trait hypervolumes based on the first three principal component analysis (PCA) dimensions (with opaque dots representing observed data points, and semitransparent small dots representing uniformly distributed random points generated by the algorithm used for hypervolume computation). Hypervolumes with size equivalent to the mean of the 100 hypervolumes generated by resampling per group (F2s vs. accessions) are shown. Right: Trait variation within the phenotypic spaces of A. thaliana F2s (blue) and wild accessions (red) considering the first three PCA dimensions of trait hypervolumes. Hypervolume centroids, observed data points, and uniformly distributed random points generated by the algorithm used for hypervolume computation are depicted by large circles, opaque dots, and semitransparent small dots, respectively. Hypervolumes with size equivalent to the mean of the 100 hypervolumes generated by resampling per group (F2s vs. accessions) are shown. LA, leaf area; LDMC, leaf dry matter content; LNC, leaf nitrogen content; SLA, specific leaf area.
Why do we observe some plant #phenotypes but not others? This study explores the multivariate #PhenotypicSpace of a widely distributed #plant species to reveal how interplays between population history & natural selection shape phenotypic diversity in #Arabidopsis @plosbiology.org 🧪 plos.io/43Z8IWi