Ancestral morphological polymorphisms can shape the landscape of phylogenetic discordance in fossil clades
Abstract
One of the most transformative advances in molecular systematics over the past several decades has been our understanding of how population processes like incomplete lineage sorting (ILS) give rise to phylogenetic discordance. Fossil clades, which are known almost exclusively from morphology, have benefited little from these developments. An emerging body of work has shown how processes related to those that cause discordance at the molecular level can also shape patterns at the phenotypic level. Conflicting signals can arise when phenotypic polymorphisms are maintained in an ancestral taxon sort among descendant taxa or persist beyond speciation. Here, I explore how ancestral polymorphisms contribute to the landscape of phylogenetic discordance in Pleistocene Homo and late-Cretaceous Micraster. I found rampant discordance stemming from the sorting and persistence of ancestral polymorphisms in both clades. This suggests that mechanisms related to those that drive gene-tree discordance may also shape morphological patterns. Simulations show that explicitly modelling the processes that give rise to this discordance can dramatically improve reconstructions. Further studying how biological processes like ILS generate discordance in phenotypic characters and incorporating them into phylogenetic models can improve tree reconstruction and generate novel biological insights in fossil clades.
Dataset DOI: 10.5061/dryad.kd51c5bmr
Description of the data and file structure
Two empirical analyses and many simulation replicate datasets are included within the dryad.tgz.
Files and variables
File: homo/homo.best.tre
Description: Tree topology used for empirical analyses of Pleistocene Homo taxa.
File: homo/homo.poly_format.samp.fa
Description: Character matrix used for Homo empirical analysis.
File: micraster/best_mfc2.tre
Description: Tree topology used for empirical analyses of* Micraster* clade.
File: micraster/micrasteridae-character_matrix.fa
Description: Character matrix used for Micraster empirical analysis.
File: simulations/sim_fastas_10
Description: 100 simulated datasets for assessing treatment of polymorphisms. Each dataset contained 10 characters.
File: simulations/sim_fastas_25
Description: 100 simulated datasets for assessing treatment of polymorphisms. Each dataset contained 25 characters.
File: simulations/sim_fastas_50
Description: 100 simulated datasets for assessing treatment of polymorphisms. Each dataset contained 50 characters.
File: simulations/monomorph_fastas_10
Description: 100 simulated datasets for assessing tree reconstruction accuracy when polymorphic characters are randomly "resolved" to represent only a single character state. Each dataset contained 10 characters.
File: simulations/monomorph_fastas_25
Description: 100 simulated datasets for assessing tree reconstruction accuracy when polymorphic characters are randomly "resolved" to represent only a single character state. Each dataset contained 25 characters.
File: simulations/monomorph_fastas_50
Description: 100 simulated datasets for assessing tree reconstruction accuracy when polymorphic characters are randomly "resolved" to represent only a single character state. Each dataset contained 50 characters.
Access information
Data was derived from the following sources:
- Homo character matrix derived from 10.7934/P2451
- Micraster character matrix and phylogeny borrowed from 10.1093/evolut/qpaf177