16 October 2019 in New Phytologist
1 March 2018 in Applications in Plant Sciences
Premise of the Study: Polyploidy has profound evolutionary consequences for land plants. Despite the availability of large phylogenetic and chromosomal data sets, estimating the rates of polyploidy and chromosomal evolution across the tree of life remains a challenging, computationally complex problem. We introduce the R package chromploid, which allows scientists to perform inference of chromosomal evolution rates across large phylogenetic trees.Methods and Results: chromploid is an open‐source package in the R environment that calculates the likelihood function of models of chromosome evolution. Models of discrete character evolution can be customized using chromploid. We demonstrate the performance of the BiChroM model, testing for associations between rates of chromosome doubling (as a proxy for polyploidy) and a binary phenotypic character, within chromploid using simulations and empirical data from Solanum. In simulations, estimated chromosome‐doubling rates were unbiased and the variance decreased with larger trees, but distinguishing small differences in rates of chromosome doubling, even from large data sets, remains challenging. In the Solanum data set, a custom model of chromosome number evolution demonstrated higher rates of chromosome doubling in herbaceous species compared to woody.Conclusions: chromploid enables researchers to perform robust likelihood‐based inferences using complex models of chromosome number evolution across large phylogenies.
13 October 2023 in Evolution
Geological events such as mountain uplift affect how, when, and where species diversify, but measuring those effects is a longstanding challenge. Andean orogeny impacted the evolution of regional biota by creating barriers to gene flow, opening new habitats, and changing local climate. Bomarea (Alstroemeriaceae) are tropical plants with (often) small, isolated ranges; in total, Bomarea species occur from central Mexico to central Chile. This genus appears to have evolved rapidly and quite recently, and rapid radiations are often challenging to resolve with traditional phylogenetic inference. In this study, we apply phylogenomics—with hundreds of loci, gene-tree-based data curation, and a multispecies-coalescent approach—to infer the phylogeny of Bomarea. We use this phylogeny to untangle the potential drivers of diversification and biogeographic history. In particular, we test if Andean orogeny contributed to the diversification of Bomarea. We find that Bomarea originated in the central Andes during the mid-Miocene, then spread north, following the trajectory of mountain uplift. Furthermore, Andean lineages diversified faster than non-Andean relatives. Bomarea thus demonstrates that—at least in some cases—geological change rather than environmental stability has driven high species diversity in a tropical biodiversity hotspot. These results also demonstrate the utility (and danger) of genome-scale data for making macroevolutionary inferences.
16 November 2022 in Systematic Biology
Testing adaptive hypotheses about how continuous traits evolve in association with developmentally structured discrete traits, while accounting for the confounding influence of other, hidden, evolutionary forces, remains a challenge in evolutionary biology. For example, geophytes are herbaceous plants—with underground buds—that use underground storage organs (USOs) to survive extended periods of unfavorable conditions. Such plants have evolved multiple times independently across all major vascular plant lineages. Even within closely related lineages, however, geophytes show impressive variation in the morphological modifications and structures (i.e.,“types” of USOs) that allow them to survive underground. Despite the developmental and structural complexity of USOs, the prevailing hypothesis is that they represent convergent evolutionary “solutions” to a common ecological problem, though some recent research has drawn this conclusion into question. We extend existing phylogenetic comparative methods to test for links between the hierarchical discrete morphological traits associated with USOs and adaptation to environmental variables, using a phylogeny of 621 species in Liliales. We found that plants with different USO types do not differ in climatic niche more than expected by chance, with the exception of root morphology, where modified roots are associated with lower temperature seasonality. These findings suggest that root tubers may reflect adaptations to different climatic conditions than those represented by other types of USOs. Thus, the tissue type and developmental origin of the USO structure may influence the way it mediates ecological relationships, which draws into question the appropriateness of ascribing broad ecological patterns uniformly across geophytic taxa. This work provides a new framework for testing adaptive hypotheses and for linking ecological patterns across morphologically varying taxa while accounting for developmental (non-independent) relationships in morphological data. [Climatic niche evolution; geophytes; imperfect correspondence; macroevolution.].
25 December 2024 in New Phytologist
14 November 2019 in Frontiers in Ecology and Evolution
We investigate a class of information criteria based on the informational complexity criterion (ICC), which penalizes model fit based on the degree of dependency among parameters. In addition to existing forms of ICC, we develop a new complexity measure that uses the coefficient of variation matrix, a measure of parameter estimability, and a novel compound criterion that accounts for both the number of parameters and their informational complexity. We compared the performance of ICC and these variants to more traditionally used information criteria (i.e., AIC, AICc, BIC) in three different simulation experiments: simple linear models, nonlinear population abundance growth models, and nonlinear plant biomass growth models. Criterion performance was evaluated using the frequency of selecting the generating model, the frequency of selecting the model with the best predictive ability, and the frequency of selecting the model with the minimum Kullback-Leibler divergence. We found that the relative performance of each criterion depended on the model set, process variance, and sample size used. However, one of the compound criteria performed best on average across all conditions at identifying both the model used to generate the data and at identifying the best predictive model. This result is an important step forward in developing information criterion that select parsimonious models with interpretable and tranferrable parameters.