Phylogenetic Character Reconstruction

J S Albert, University of Louisiana, Lafayette, LA, USA © 2007 Elsevier Inc. All rights reserved.

1.03.1 Introduction to Character State Reconstruction and Evolution 42

1.03.2 Basic Concepts 43 Homology: Similarity Due to Common Ancestry 43 Homoplasy: Convergence, Parallelism, and Reversal 43 Character State Polarity 43 Character or Trait Data 44 Adaptation 44 Phylogenetic Trees 45

1.03.3 Methods 47 Parsimony Optimization of Discrete Traits 47 Binary and Multistate Characters 47 Squared-Change and Linear Parsimony 47 Maximum Likelihood and Bayesian Optimization 47 Which Optimization Approach to Use? 48 Correlative Comparative Methods 49

1.03.4 Limitations of Methods 50

1.03.5 Conclusions 51

Glossary adaptation anagenesis character polarity character state reconstruction clade cladogenesis cladogram comparative method continuous trait

A feature or phenotype or trait that evolved to serve a particular function or purpose.

The origin of evolutionary novelties within a species lineage by changes in gene allele frequencies by the processes of natural selection and/or neutral genetic drift. The temporal direction of change between alternative (primitive and derived) states of a character. The process of estimating the ancestral or primitive condition of a character at a given node (branching point) in a phylogenetic tree. A complete branch of the tree of life. A monophyletic group. The origin of daughter species by the splitting of ancestral species; may or may not occur under the influence of natural selection.

A branching tree-shaped diagram used to summarize comparative (interspecific) data on phenotypes or gene sequences. In contrast to a phylogeny, a cladogram has no time dimension. The study of differences between species.

A quantitatively defined feature with no easily distinguished boundaries between phenotypes (e.g., size, cell counts, and gene expression levels).

convergence Similarity of structure or function due to independent evolution from different ancestral conditions. discrete trait A qualitatively defined feature with only a few distinct phenotypes (e.g., polymorphism; presence vs. absence). homology Similarity of structure or function due to phylogeny (common ancestry). homoplasy Similarity of structure or function due to convergence, parallelism or reversal. monophyletic A systematic category that includes an ancestor and all of its descendants; a complete branch of the tree of life; a 'natural' taxon; a clade. node An internal branching point in a phylogenetic tree. optimization Methods for estimating ancestral trait values on a tree. Commonly used optimization criteria are: maximum parsimony (MP) which minimizes the amount of trait change, and maximum likelihood (ML) which maximizes the likelihood of a trait at a node given likelihood values for trait evolution. parallelism Similarity of structure or function due to independent evolution from a common ancestral condition. paraphyletic A systematic category that includes an ancestor and some but not all of its descendents (e.g., 'invertebrates', 'agnathans', 'fish', and 'reptiles' (sans birds)).

parsimony phenotypic evolution phylogenetic character phylogenetic systematics phylogenetic tree phylogeny polyphyletic reversal synapomorphy taxon trait evolution

A principle of scientific inquiry that one should not increase, beyond what is necessary, the number of entities required to explain anything.

Change in the developmental program descendents inherit from their ancestors.

A homologous feature or phenotype or trait of an organism or group of organisms.

A method for reconstructing evolutionary trees in which taxa are grouped exclusively on the presence of shared derived features. Genealogical map of interrelationships among species, with a measure of relative or absolute time on one axis. Also called a tree of life or a phylogeny.

The evolutionary history of a species or group of species that results from anagenesis and cladogenesis. A systematic category that includes taxa from multiple phylogenetic origins (e.g., 'homeothermia' consisting of birds and mammals). Change from a derived character state back to a more primitive state; an atavism. Includes evolutionary losses (e.g., snakes which have 'lost' their paired limbs). A shared, derived character used as a hypothesis of homology. A species or monophyletic group of species (plural taxa). The sequence of changes of a feature or phenotype on a phylogeny.

1.03.1 Introduction to Character State Reconstruction and Evolution

Comparisons among the features of living organisms have played a prominent role in the biological sciences at least since the time of Aristotle. The comparative approach takes advantage of the enormous diversity of organismal form and function to study basic biological processes of physiology, embryology, neurology, and behavior. This approach has given rise to the widespread use of certain species as model systems, based on what has become known as the August Krogh Principle: ''For many problems there is an animal on which it can be most conveniently studied'' (Krebs, 1975).

From an evolutionary perspective, interspecific (between species) comparisons allow for the systematic study of organismal design. Rensch (1959) conceived of phylogeny as being composed of two distinct sets of processes: anagenesis, the origin of phenotypic novelties within an evolving species lineage (from the Greek ana = up + genesis = origin), and cladogenesis, the origin of new species from lineage splitting (speciation) (from the Greek clado = branch). Anagenetic changes arise within a population by the forces of natural selection and genetic drift. Cladogenesis may or may not arise from these population-level processes, and in fact many (or perhaps most?) species on Earth are thought to have their origins from geographical (allopatric) speciation under the influence of landscape and geological processes (Mayr, 1963; Coyne and Orr, 1989).

Because species descend from common ancestors in a hierarchical fashion (i.e., from a branching, tree-like process of speciation) closely related species tend to resemble each other more than they do more distantly related species. Patterns in the diversification of phenotypes have therefore been described as mosaic evolution, in which different species inherit distinct combinations of traits depending on the position of that species in the tree of life (McKinney and McNamara, 1990). Under this view, character evolution is regarded as a process of historical transformation from a primitive to a derived state, and study of this process necessarily presumes knowledge of primitive or ancestral conditions. In other words, because character evolution is perceived as trait change on a tree, it is necessary to estimate 'ancestral trait values'.

Direct observations of ancient phenotypes may be taken from fossils, which provide unique information on entirely extinct groups of organisms, and are usually associated with stratigraphic information pertaining to relative and absolute geological ages (Benton, 1993). Nonetheless, the fossil record has many well-known shortcomings, including the famously incomplete levels of preservation, and usually very limited information about the nature of soft tissues such as nerves and brains (but see Edinger, 1941; Stensio, 1963). Paleontological information on ancient physiological and behavioral traits is even more scanty (but see Jerison, 1976; MacLeod and Rose, 1993; Rogers, 2005).

Recent years have seen great advances in the formulation of comparative methods to estimate or infer ancestral phenotypes from extant (living) species (Garland et al, 1992, 1999; Martins, 2000). These methods use patterns in the mosaic of traits present among species in the context of an explicit hypothesis of interrelationships. These methods also address new topics, such as whether rates of phenotypic evolution have differed among lineages (clades), the circumstances in which a phenotype first evolved, the selective and developmental mechanisms underlying the origin of new pheno-types, and the evolutionary lability of phenotypes (Albert et al., 1998; Blomberg et al., 2003; Blackledge and Gillespie, 2004).

In this article, I summarize the major recent developments in phylogenetically based methods of studying character evolution, with the goals of explaining both the strengths and weaknesses of alternative methods. Most of the empirical examples cited are among animals with the most complex central nervous systems (e.g., vertebrates) in which neurological and behavioral evolution has been (arguably) most extensively studied. A major goal of this article is to highlight some of the most exciting new developments in the study of character evolution now being explored in this fascinating area of comparative neurobiology.

Was this article helpful?

0 0

Post a comment