Empirical Research: The central goal of my empirical research is to discover, document, and explain the diversity of amphibians, including their taxonomic, genomic, biochemical, behavioral, anatomical, ecological, and biogeographic diversity. A general theme of my research is the combination of evidence from multiple sources to test phylogenetic hypotheses. Since 1994 I have had an active, hypothesis-driven field program in Colombia , with each expedition designed to address a specific set of problems. My empirical studies are also hypothesis-driven, often resulting in the discovery of new species, novel morphological structures, complex behaviors, or previously unrecognized clades—all of which provide insight into the phylogenetic diversification of amphibians and suggest new directions for research.

In a collaborative study led by Darrel Frost, me, and Julian Faivovich and funded primarily by NASA, we targeted 532 terminals, representing the global diversity of amphibians and appropriate outgroup taxa, for DNA sequences (ca. 4,500 bases from three mitochondrial and five nuclear genes) and morphology in one of the largest phylogenetic studies of any vertebrate group to date. The findings of this study (Frost et al., 2006) result in a fundamental reshaping of our understanding of amphibian relationships and a new taxonomy of living amphibians.

For over a decade I have investigated the diversity of dart-poison frogs (Dendrobatoidea). My NSF-funded doctoral research into their phylogeny brought together evidence from DNA sequences multiple mitochondrial and nuclear, morphology, alkaloid profiles, and behavior (174 characters) for a large sample of dendrobatid species (Grant et al., 2006). I am particularly interested in the evolution of dendrobatid toxicity, which includes over 450 lipophilic alkaloids, many with extensive pharmacological applications. These alkaloids are not synthesized by the frogs but are instead accumulated from the diet, which leads to questions about the evolutionary ecology of this system. I have used my phylogenetic results to design feeding experiments (with Ron Gagliardo of the Atlanta Botanical Gardens and Ralph Saporito of Florida International University) to test hypothesis about alkaloid uptake, and these results will allow us to initiate comparative genomics studies into its genetic basis. I have also analyzed the evolution of the complex behaviors of dendrobatids, including parental tadpole transportation and provision of nutritive oocytes.

Theoretical Research: A defensible theoretical foundation is required to interpret empirical observations and understand the phylogenetic diversification of amphibians, and my research aims to strengthen that foundation. The challenges faced by historical sciences, which aim to make ideographic inferences involving unique things and events, differ from those of the fields that have traditionally attracted the attention of philosophers (e.g., physics), which aim to make nomothetic inferences involving universal laws and class concepts. Specifically, my goal is to develop a consistent, rational, and objective basis for phylogenetic knowledge claims (e.g., Kluge & Grant, 2006). I am primarily interested in cognitive values and their implementation conceptually, operationally, and empirically, but research does not occur in a social vacuum, and socialogical factors also affect the course of science. Therefore, I am also interested in how the scientific community, especially funding agencies, may be structured such that sociological pressures support, and do not impede, implementation of cognitive values.

Analytical Research: In order to make empirical inferences, theoretically defined concepts must be operationalized. The analytical challenges faced by evolutionary informatics continue to increase as our ability to obtain genomic data rapidly outstrips our ability to analyze them thoroughly. Finding both the optimal topology for a given alignment and the optimal alignment for a given topology are among the most difficult computational problems known (both are NP-complete). Likewise, although automated laboratory procedures are helpful, assurance of quality control becomes more difficult as the magnitude of studies increases. The potential for errors to accumulate in large studies involving multiple gene fragments (and data files) for hundreds (and soon thousands) of terminals is enormous, and efficient error detection is an important bioinformatics problem that requires creative solutions. I am interested in both software and hardware to address these problems. I collaborate closely with Ward Wheeler in testing heuristic algorithms and designing search strategies using the AMNH 560-processor mixed Pentium III 1 Ghz and 500 Mhz cluster and 256-processor Pentium 4 Xeon 2.8 Ghz cluster, and we recently acquired an 8-processor Opteron 846 2.2 Ghz 128 GB RAM 64-bit computer for memory-intensive analyses.

Last Revised 2/2/2007 © 2006-2007 Taran Grant, All Rights Reserved