Patterns and causes of incongruence between plastid and nuclear Senecioneae (Asteraceae) phylogenies

Abstract

One of the longstanding questions in phylogenetic systematics is how to address incongruence among phylogenies obtained from multiple markers and how to determine the causes. This study presents a detailed analysis of incongruent patterns between plastid and ITS/ETS phylogenies of Tribe Senecioneae (Asteraceae). This approach revealed widespread and strongly supported incongruence, which complicates conclusions about evolutionary relationships at all taxonomic levels. The patterns of incongruence that were resolved suggest that incomplete lineage sorting (ILS) and/or ancient hybridization are the most likely explanations. These phenomena are, however, extremely difficult to distinguish because they may result in similar phylogenetic patterns. We present a novel approach to evaluate whether ILS can be excluded as an explanation for incongruent patterns. This coalescence-based method uses molecular dating estimates of the duration of the putative ILS events to determine if invoking ILS as an explanation for incongruence would require unrealistically high effective population sizes. For four of the incongruent patterns identified within the Senecioneae, this approach indicates that ILS cannot be invoked to explain the observed incongruence. Alternatively, these patterns are more realistically explained by ancient hybridization events.

Analysis of plastid and nuclear DNA data in plant phylogenetics—evaluation and improvement

“…As an internationally famous and reputable journal that publishes research on plant molecular phylogenetics, Systematic Botany was considered to be representative to some extent of the present status of cpDNA and nrDNA methodology. Given that some statistical methods for distinguishing hybridization from incomplete lineage sorting have only been published relatively recently [16,27,29], we restricted our search to the period between 2011 and 2005. The articles investigated had to fulfill the following two criteria: (i) one of the goals of the paper was to reconstruct phylogenetic relationships among members of a group or to determine the systematic position of a taxon, and (ii) the paper employed both cpDNA and nrDNA data to construct phylogenetic trees for the same taxa.…”

“…However, it is sometimes extremely difficult to determine the cause of incongruence, particularly to distinguish hybridization from incomplete lineage sorting [28,29], which may result in similar phylogenetic patterns. Fortunately, several novel approaches have been developed for statistically distinguishing hybridization from incomplete lineage sorting [16,29].…”

“…Plastid (cp) and nuclear (nr) genomes are the most frequently used sources of genetic data for reconstructing the phylogeny of plant groups [10][11][12][13][14]. However, certain evolutionary events, such as gene duplication, hybridization, and lineage sorting of ancestral polymorphisms, may result in conflicting topologies based on data sets from these two genomes at all taxonomic levels [12,15,16]. It is well known that hybridization and lineage sorting of ancestral polymorphisms have occurred much more frequently in nature than previously envisioned [17,18].…”

Frequent colonization and little in situ speciation in Senecio in the tropical alpine‐like islands of eastern Africa

“…Substitution model and best partitioning scheme were inferred using the program PartitionFinder v.1.1.1 ( Lanfear et al, 2012 ). Phylogenetic relationships were reconstructed using nrDNA and ptDNA separately, accounting for incongruence between the ptDNA and nrDNA data described earlier ( Pelser et al, 2007( Pelser et al, , 2010( Pelser et al, , 2012Calvo et al, 2013 ). Indels were coded for the plastid data only (Appendix S1 of the online Supplement).…”

“…Senecio has a nearly cosmopolitan distribution with major diversity centers in SA and South America ( Bremer, 1994 ;Pelser et al, 2007 ). Earlier molecular analyses and ancestral area reconstructions have shown that the genus is likely to have originated in sub-Saharan Africa in the mid to late Miocene, from where it dispersed to other parts of Africa and to all other continents except Antarctica ( Pelser et al, 2007( Pelser et al, , 2010( Pelser et al, , 2012. More specifi cally, we recently traced the origin of the genus to the dry or winter rainfall regions of western southern Africa (WSA) ( Kandziora et al, 2016 ).…”