Author
 |
LAMBOY W F - CORNELL UNIVERSITY |
|
Submitted to: Systematic Botany
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 1/31/1994 Publication Date: N/A Citation: N/A Interpretive Summary: Reconstruction of the evolutionary history (phylogeny) of groups of organisms is the most active area of systematics research today. The method which is most often used for determining the evolutionary history of a group is called maximum parsimony. Using known or measurable characteristics of a group of organisms, this method attempts to explain the group's evolution by finding an evolutionary tree for the organisms that requires the minimum number of evolutionary steps. Although the method is extremely popular, its accuracy has never been adequately tested. This research utilizes computer simulation to determine what biological factors affect the accuracy of maximum parsimony and how well it does at finding the true evolutionary tree. It was found that extinctions dramatically decrease the accuracy of the evolutionary relationships that are determined for the remaining organisms (those that did not become extinct). Also, this work shows that the faster evolution proceeds in a group of organisms, the more difficult it is for maximum parsimony to trace the true evolutionary history of the group. In addition, it is shown that if the same characteristics arise more than once during the evolution of the group, it is very difficult to use those characteristics to reconstruct the group's history. Finally, for the types of real data that are typically used in the determination of the evolutionary history of organisms, the results show that maximum parsimony finds the true evolutionary tree only about 10% of the time. Technical Abstract: The accuracy of maximum parsimony for phylogeny reconstruction using morphological data has never been adequately tested. In this study 128,000 known phylogenies of eight species and their associated discrete unordered character state data sets were generated by computer simulation, and all equally parsimonious trees were found using the program Hennig86. The accuracy of maximum parsimony was judged primarily by the percent of time the strict consensus tree was found to be identical to the true tree. Parallelisms, reversals, phyletic changes, and extinctions all significantly decrease maximum parsimony's ability to find the true phylogenetic tree. In addition, the method is less accurate when there are fewer characters, more character states per character, and more character state changes per speciation event. For simulated data sets that have properties similar to those of real data sets, maximum parsimony finds the true phylogenetic tree on average only about 10% of the time. Caution is advised, then, in using phylogenies reconstructed from morphological data to make inferences about biological processes or patterns that rely on knowledge of the true phylogeny. Additionally, the question is raised whether the great emphasis in systematics on phylogeny reconstruction using maximum parsimony and other cladistic methods is misplaced. |
