Adams Lab: Current Research

Understanding the nature of phenotypic diversity follows two general themes in the Adams lab: 1) examining patterns of phenotypic diversity in empirical systems, and 2) developing analytical tools for examining patterns of phenotypic variation and diversity. Some recent projects related to these topics are briefly described below.

The primary focus of the Adams lab is to understand the evolution of phenotypic diversity. Much of this work is directed towards the evolution and ecology of salamanders, and what ecological and evolutionary forces are responsible for their phenotypic diversification. Major questions concern what role ecological pressures play in generating phenotypic diversity and in maintaining ecological communities. These questions are addressed at various spatial and temporal scales, and from both an ecological and evolutionary context. With this paradigm, niche use, food resources, behavioral traits, abiotic effects, and patterns of phenotypic variation are examined in concert to understand the form-function relationship, and its impacts on evolutionary diversification.

At the community-level, we link proximate ecological interactions to micro-evolutionary patterns of phenotypic diversification. Much of our work focuses on the effects of biotic interactions, but abiotic environmental effects are also examined. Through this work we have found that sympatric phenotypic divergence (character displacement) is a common evolutionary response to interspecific competition in Plethodon. For instance, in some communities, exploitative competition for food resources appears to have resulted in sympatric diversification (Adams, 2000; Adams and Rohlf, 2000). Similar ecomorphologial patterns within populations in cranial shape associated with food resource use (i.e. within-population trophic polymorphisms) suggest that adaptive diversification within habitats plays an important role in promoting phenotypic diversity (Maerz, Myers, and Adams, 2006). In other communities behavioral interference seems to dominate, but with similar evolutionary responses (Adams, 2004; Adams et al., 2007). However, this pattern is not uniformally the case. In certain communities, a combination of biotic and abiotic forces appear to regulate the distributions of salamanders and patterns of phenotpyic diversification (Arif, Adams, and Wicknick, 2007). Together, these studies describe a putative causal link between particular anatomical adaptations and ecological selective forces that shape patterns of diversification.

Examining these patterns at a broader spatial scale, we are addressing macroecological questions, such as whether the ecological forces at play in local communities influence not only phenotypic diversity, but also community composition and community structure. Indeed, our initial work suggests that community composition at a macroecological level is determined in part by such competitive-based 'community assembly rules' (Adams, 2007), implying a direct connection between microevolutionary process and macroecological community assembly. We are continuing this work to examine large-scale patterns of body size variation across communities, to determine the extent to which patterns can be identified (community-wide character displacement). Other current research examines the role that ecological forces play in generating larger scale patterns of community composition and phenotypic diversification. This work integrates community assembly models, ecological niche modeling (ENM), and microevolutionary patterns to determine the extent to which phenotypic patterns are impacted by community membership, and how community structure and species distributions affect, and are affected by, biotic and abiotic interactions.

At a broader temporal scale, we are using a phylogenetic perspective to determine the extent to which microevolutionary patterns translate into of macroevolutionary trends. Our initial work suggests that for some lineages of Plethodon, phenotypic diversification has proceeded consistently throughout the history of the group, however, considerable phenotypic homoplasy has resulted (Adams and Collyer, in prep). Thus, while diversification via competitive interactions may have been a prominant feature in these groups, it has not necessarily resulted in patterns of adaptive diversification. Other macoevolutionary work compares the relationship between species diversification and morphological diversification in tropical and temperate salamanders (in collaboration with J. Wiens), and examines broad-scale ecogeographic patterns of body size variation across Plethodon salamanders and amphibians in general (Adams and Church, in press).

Theoretical Methods for Understanding Phenotype Change

Ecology and evolutionary biology are largely concerned with patterns of change. Many interesting questions and disciplines in evolutionary ecology examine changes in an organism’s phenotype, and considerable effort has been devoted to documenting these patterns in nature. Examples of such disciplines include: phenotypic plasticity (change across environmental gradients), quantitative genetics (change across generations), ontogenetics (change during development), local adaptation and diversification (change across environments at a broader scale), and biomechanics (change during motion). In these studies, an organism’s phenotype is typically quantified using multivariate data, where numerous phenotypic attributes are examined simultaneously to estimate patterns of trait variation and covariation (e.g,. using factorial MANOVA). Remarkably, while considerable effort has been devoted to understanding the biological underpinnings of these disciplines, a general analytical framework for assessing phenotypic change is lacking. Additionally, such attributes of change are not fully described using the MANOVA: a significant interaction term inmplies that phenotypic change across environments is not cosistent across species, but does not describe how they differ!) Clearly, a more comprehensive approach is desired.

We are currently developing a general analytical framework for assessing patterns of phenotypic change. For the simple case of change across two 'states' (e.g., two environments), phenotypic change can be thought of as a vector, which has both a magnitude (the amount of change), and a direction in phenotype space (the manner in which the phenotypic covariance changes). Our approach statistically examines the magnitude and direction components separately, allowing for a more complete description of how phenotypic change differs between species or other groups (Collyer and Adams, 2007). One advantage of this approach however, is that it is general, and may be used to account for covariates, and other non-targeted sources of variation. For instance, in studies of character displacement, if the phenotypic traits of interest vary along an environmental gradient, patterns due to competitive interactions may be obscured. By including the gradient as a covariate, our approach can correctly identify character displacement along such clines (Adams and Collyer, 2007).

For phenotypic change across more than two states, the phenotypic change vector becomes a multivariate phenotypic change trajectory. These trajectories have a size (magnitude of change), an orientation (direction of change), and a shape (the manner in which the phenotype changes), all of which must be quantified and compared for a meaningful understanding of phenotypic change. We call our general approach for evaluating these patterns change: Procrustes Trajectory Analysis (Adams and Collyer, submitted). A recent application of this approach is found in comparing complex patterns of biomechanical motion, in which the change in posture during the motion generates distinct patterns of change through motion space (Adams and Cerney, 2007).

In our current theoretical research, we are examining the performance of PTA to specific biological problems, and comparing its performance to other approaches. Several ad-hoc approaches have been proposed for those specific biological problems (e.g., aspects of evolutionary diversification), but in many cases their performance has not been determined across a range of simulation conditions. We are comparing PTA to these approaches to determine the extent to which PTA is a general solution to many quantitative problems in ecology and evolutionary biology (e.g., Adams and Collyer submitted). Because many biological questions are fundamentally interested in patterns of change, the data addressing these questions exhibit the same general properties. As such, we posit that a single quantitative approach (PTA) is applicable for assessing these patterns, because any pattern of phenotypic change can be represented by a trajectory in multivariate phenotype space. As such, PTA provides the conceptual unification required for a general analytical framework for the analysis of phenotypic change data derived from any biological process.

---

Copyright © 2007, Iowa State University, all rights reserved. Send comments to dcadams@iastate.edu