Phenotypic and Genetic Structure Support Gene Flow Generating Gene Tree Discordances in an Amazonian Floodplain Endemic Species.

5 years ago

Phenotypic and Genetic Structure Support Gene Flow Generating Gene Tree Discordances in an Amazonian Floodplain Endemic Species.

Michael J Hickerson, Cristina Y Miyaki, Alexandre Aleixo, Fabio R Amaral, Lucas E Araujo-Silva, Camila C Ribas, Erik Choueri, Gregory Thom

Before populations become independent evolutionary lineages, the effects of micro evolutionary processes tend to generate complex scenarios of diversification that may affect phylogenetic reconstruction. Not accounting for gene flow in species tree estimates can directly impact topology, effective population sizes and branch lengths and the resulting estimation errors are still poorly understood in wild populations. In this study we used an integrative approach, including sequence capture of Ultra Conserved Elements (UCEs), mtDNA Sanger sequencing and morphological data to investigate species limits and phylogenetic relationships in face of gene flow in an Amazonian endemic species (Myrmoborus lugubris: Aves). We used commonly implemented species tree and model-based approaches to understand the potential effects of gene flow in phylogenetic reconstructions. The genetic structure observed was congruent with the four recognized subspecies of M. lugubris. Morphological and UCEs data supported the presence of a wide hybrid zone between M. l. femininus from the Madeira river and M. l. lugubris from the Middle and lower Amazon river, which were recovered as sister taxa by species tree methods. When fitting gene flow into simulated demographic models with different topologies, the best-fit model indicated these two taxa as non sister lineages, a finding that is in agreement with the results of mitochondrial and morphological analyses. Our results demonstrated that failing to account for gene flow when estimating phylogenies at shallow divergence levels can generate topological uncertainty, which can nevertheless be statistically well supported, and that model testing approaches using simulated data can be useful tools to test alternative phylogenetic hypotheses.

Publisher URL: http://doi.org/10.1093/sysbio/syy004

DOI: 10.1093/sysbio/syy004