Effects of taxon sampling and tree reconstruction methods on phylodiversity metrics. Issue 17 (21st August 2019)

• Record Type: Journal Article
• Authors: Jantzen, Johanna R.; Whitten, William M.; Neubig, Kurt M.; Majure, Lucas C.; Soltis, Douglas E.; Soltis, Pamela S.
• Is Part Of: Ecology and evolution. Volume 9:Issue 17(2019)
• Page Start: 9479
• Page End: 9499
• Publication Date: 2019-08-21
• Subjects: community phylogenetics -- phylodiversity -- phylogenetic diversity -- phylogeny reconstruction methods -- taxon sampling; Ecology -- Periodicals; Evolution -- Periodicals; 577.05
• DOI: 10.1002/ece3.5425
• Languages: English
• ISSNs: 2045-7758

Abstract

Abstract: The amount and patterns of phylodiversity in a community are often used to draw inferences about the local and historical factors affecting community assembly and can be used to prioritize communities and locations for conservation. Because measures of phylodiversity are based on the topology and branch lengths of phylogenetic trees, which are affected by the number and diversity of taxa in the tree, these analyses may be sensitive to changes in taxon sampling and tree reconstruction methods. To investigate the effects of taxon sampling and tree reconstruction methods on measures of phylodiversity, we investigated the community phylogenetics of the Ordway‐Swisher Biological Station (Florida), which is home to over 600 species of vascular plants. We studied the effects of (a) the number of taxa included in the regional phylogeny; (b) random versus targeted sampling of species to assemble the regional species pool; (c) including only species from specific clades rather than broad sampling; (d) using trees reconstructed directly for the taxa under study compared to trees pruned from a larger reconstructed tree; and (e) using phylograms compared to chronograms. We found that including more taxa in a study increases the likelihood of observing significantly nonrandom phylogenetic patterns. However, there were no consistent trends in the phylodiversity patterns based on random taxon sampling compared to targeted sampling, or within individual clades compared to the complete dataset. Using pruned and reconstructed phylogenies resulted in similar patterns of phylodiversity, while chronograms in some cases led to significantly different results from phylograms. The methods commonly used in community phylogenetic studies can significantly impact the results, potentially influencing both inferences of community assembly and conservation decisions. We highlight the need for both careful selection of methods in community phylogenetic studies and appropriate interpretation of results, depending on the specific questions to be addressed. Abstract : Measures of phylogenetic diversity are calculated based on the topology and branch lengths of phylogenetic trees. Using an empirical dataset of 14 communities in North Central Florida, we found that the taxon sampling strategy and the type of phylogeny used to calculate phylogenetic diversity can significantly influence the patterns of phylogenetic diversity that are observed. These different methodologies can result in different conclusions, so methods for community phylogenetic studies should be chosen with care.