Details of the record

TitleRate Variation Among Nuclear Genes and the Age of Polyploidy in Gossypium
AuthorsDavid S. Senchina, Ines Alvarez, Richard C. Cronn, Bao Liu, Junkang Rong, Richard D. Noyes, Andrew H. Paterson, Rod A. Wing, Thea A. Wilkins, and Jonathan F. Wendel
PublicationMol Biol Evol 20(4): 633-643
AbstractMolecular sequence data are widely employed as molecular clocks to address questions of the absolute and relative ages of various divergence events. A number of methodological and biological phenomena may affect the accuracy of molecular clock-based estimates (Gaut 1998; Hillis, Moritz, and Mable 1996; Sanderson 1998; Soltis et al. 2002). Major issues include rate heterogeneity caused by various evolutionary factors (e.g., generation time, germ-line processes, metabolic rate); difficulties in interpreting fossil or biogeographic data against which a clock might be calibrated; use of nonindependent lineages for calibration; and the inappropriate usage of calibrations on lineages for which they were not intended. One of the most significant factors is rate variation among genes, even at putatively neutral sites. Early surveys based on a dozen or fewer genes reported only a twofold to threefold variation in synonymous substitution rates (Wolfe, Sharp, and Li 1989; Gaut 1998), but recent studies reveal higher levels of intergenic rate variation (Kusumi et al. 2002; Tiffin and Hahn, 2002; Zhang, Vision, and Gaut 2002). Because rates of nuclear gene evolution within an organism vary so widely, sampling only one or a few genes in molecular clock applications can lead to high variance in divergence time estimates when only a single clock calibration is employed. A reasonable strategy to minimize this problem would be to sample multiple genes so that an average rate may be estimated. This is not to claim that other important sources of error do not exist; in particular, clock calibration and lineage-specific effects remain vexing and, to a certain extent, intangible problems. However, minimizing the issue of intergenic rate variation would appear to be an achievable and worthwhile goal. In the present study, we applied this rationale to a model system from Gossypium (cotton), employing sequence variation at 48 nuclear genes. Gossypium L. contains 50 species whose phylogenetic relationships have been explored using multiple molecular data sets (Seelanan, Schnabel, and Wendel 1997; Small et al. 1998; Cronn et al. 2002b). Data indicate that shortly after its origin, Gossypium experienced rapid divergence (Cronn et al. 2002b), leading to modern monophyletic lineages that vary in chromosome size and interfertility (so-called

(Total records:175) Home | Search | Show all | Top

.: 2009 :: Barry Marler :: Plant Genome Mapping Laboratory :: University of Georgia