Phylogenetic incongruence between “gene trees” and “species trees” has been widely acknowledged in phylogenetic research.Conflicts may emerge from several processes including paralogy, hybridization, and incomplete lineage sorting.
Then we investigated how phylogenetic conflict bears on taxonomic classification within the family as well as on inferences on biogeographical history, floral evolution, and measures of phylogenetic diversity (PD).
• Our study demonstrated that phylogenetic conflict not only affects the inference of organismal relationships but also impacts our understanding of biogeographical history, morphological evolution, and phylogenetic diversity.
In plant phylogenetic studies, plastid and nuclear ribosome DNA (18S-5.8S-28S), particularly the internal transcribed spacer (ITS), probably represent the most frequently used markers for phylogenetic inference (e.g., Qiu et al., 1999; Moore et al., 2007, 2010; Qiu et al., 2010).
The advantages of plastid and ribosome DNA are clear and repeatedly demonstrated in previous studies.
These include the presence of numerous duplications of homogenized copies that facilitate amplifications using polymerase chain reactions (PCR), easy design of universal primers on conserved regions at both ends, and inclusion of both conserved and rapidly evolving regions that are applicable across a broad range of taxonomic levels (e.g., land plants, seed plants, angiosperms, orders, families down to the genus, species, and populations levels; Chase et al., 1993; Soltis et al., 1999; Savolainen et al., 2000; Kong et al., 2002).
Phylogenetic hypotheses arising from these markers have been used in diverse studies, encompassing phylogeography and historical biogeography, DNA barcoding and analyses of species diversification through time, as well as quantifying phylogenetic diversity (PD) to inform conservation and phylogenetic community ecology approaches (Soltis et al., 2006; Forest et al., 2007; Antonelli and Sanmartín, 2011; Li et al., 2011b; Liu et al., 2014).In general, longer DNA sequences improve the resolution of phylogenetic relationships.Consequently, data from different loci are usually combined as total evidence in phylogenetic reconstruction.However, different loci, especially those that are genetically unlinked or from different genomes (e.g., plastid sequences vs.nuclear ribosome regions), could carry discordant phylogenetic information and yield conflicting phylogenetic trees (e.g., Pamilo and Nei, 1988; Page and Charleston, 1997; Sang et al., 1997; Calviño et al., 2008). These include paralogy, hybridization, incomplete lineage sorting of ancestral polymorphisms and horizontal gene transfer (HGT).Conflict may also result from systematic, methodological (e.g., misspecification of models) and stochastic errors (e.g., during sequencing and alignment), factors that are difficult to rule out (Pamilo and Nei, 1988; Page and Charleston, 1997; Keeling and Palmer, 2008; Smith, 2011; Knox, 2014).