Unable to connect to database - 17:11:52 Unable to connect to database - 17:11:52 SQL Statement is null or not a SELECT - 17:11:52 SQL Statement is null or not a DELETE - 17:11:52 Botany 2006 - Abstract Search
Unable to connect to database - 17:11:52 Unable to connect to database - 17:11:52 SQL Statement is null or not a SELECT - 17:11:52

Abstract Detail


Bringing Together the Living and Dead: Integrating Extant and Fossil Biodiversity in Evolutionary Studies

Clarke, Julia [1], Bell, Charles [2].

Creating faux fossils and sequence fragments: the impact of the structure of missing data in morphological and molecular datasets on phylogenetic accuracy.

THE goal of reconstructing a complete “Tree of Life” ultimately rests on the inclusion of extinct taxa as well as extant taxa for which not all gene histories will be known or are knowable (i.e., entire genes non-present). In this study we considered empirical, as well as simulated data, to investigate the effects of missing data using a variety of estimation procedures. Simulations were designed to reflect a more “realistic” representation of fossil, and otherwise incomplete, data when included in a phylogenetic study. Questions approached through simulations included whether small, topologically-restricted blocks or sets of character data, ones typically recovered together in fragmentary fossil specimens and in commonly studied segments of gene sequence data, behave less independently than random subsamples. Further, if they do, what if any are the implications for phylogenetic accuracy. More realistic simulation of missing data, the way that it commonly exists in the fossil record and for incomplete molecular sequence data, investigates if there is any difference in its expected effect on phylogeny estimation. These results were compared to previous work using simulations that removed (made missing) randomly selected characters or “classes” character data (e.g., molecular, morphological) to approach the impact of missing data. We also investigated whether the subsampled data, these faux fossils and sequence fragments, behaved differently if they were representing deeply nested tip taxa or basal divergences within clades. Findings from these analyses will be discussed. Several recommendations of how we might go about including specifically fossil data in larger phylogenetic analyses in light of our results will also be presented.


Log in to add this item to your schedule

1 - North Carolina State University/NCMNS, Marine Earth and Atmospheric Sciences, Campus Box 8208, Raleigh, North Carolina, 27608, USA
2 - Xavier University of Louisiana, Biology, 1 Drexel Drive, New Orleans, Louisiana, 70125, USA

Keywords:
fossil
Missing data
phylogeny
simulations.

Presentation Type: Symposium or Colloquium Presentation
Session: TBA
Location: /
Date: Thursday, January 1st, 1970
Time: TBA
Abstract ID:865


Copyright © 2000-2006, Botanical Society of America. All rights