Abstract Detail

Systematics Section/ASPT

QIU, YIN-LONG [1], TAYLOR, ALEXANDER B. [2], FINE, ISAAC [2].

A Temporal Evolutionary Framework for Land Plants.

A fossil-calibrated molecular dating study of land plants was conducted, sampling five genes (cp-atpB, cp-rbcL, mt-atp1, mt-nad5, and nu-18S) from 616 species representing >70% families of land plants and their green algal relatives. A RAxML analysis produced a phylogeny largely in agreement with the current consensus on land plant phylogeny. In a series of Bayesian statistical analyses of divergence time implemented in BEAUTi and BEAST, different topologies of angiosperm-gymnosperm relationships and various combinations of rate, substitution, and speciation models were explored. The overall results from these analyses were largely in agreement. Further, age estimates for many nodes matched the fossil record relatively well. One major node that showed significant disagreement with the fossil record was that of crown-group angiosperms, but different analyses produced widely variable estimates for this node. For nodes in which basal relationships were unresolved, e.g., basal monilophytes, basal angiosperms, and core eudicots, age estimates were highly variable as topology changed from analysis to analysis. Several limitations of the study were revealed during analysis. First, the sequence length for each species was too short. One indication was that in the dating analyses, confidence intervals for most nodes were too large, typically ranging from 15-20% of the estimated ages. In addition, bootstrap values in phylogenetic analyses were low for some nodes, particularly those that were controversial. Second, data from either the chloroplast or the mitochondrial genome were insufficient for detecting historical information content and genome-specific bias in phylogenetic or dating analyses. Similarly, there were not enough genes in each rate category to assess the effect of evolutionary rate on topology recovery or age estimation, though the analyses seemed to be robust to perturbation caused by rate variation, possibly because of extensive and dense taxon sampling. Third, only 24 fossil calibration points were used; some of these fossils had not been thoroughly evaluated for their phylogenetic placement. Three to four times this many calibration points would be needed for resampling different subsets to allow assessment of the quality of the land plant fossil record and its impact on molecular dating analyses. Nevertheless, these experimental analyses clearly demonstrated the value and feasibility of such a study and revealed areas for improvement in the next phase. 

1 - University Of Michigan, Department Of Ecology And Evolutionary Biology, 830 N University Ave., Ann Arbor, MI, 48109-1048, USA
2 - University of Michigan, Ecology and Evolutionary Biology, 830 N. University Ave., Ann Arbor, MI, 48109, United States

Keywords:
land plants
divergence times.

Presentation Type: Oral Paper:Papers for Sections
Session: 38
Location: Salmon/Snake/Boise Centre
Date: Wednesday, July 30th, 2014
Time: 10:45 AM
Number: 38003
Abstract ID:473
Candidate for Awards:None