From paralogy to hybridization: Investigating causes of underlying phylogenomic discordance using the complex genusPackera(Senecioneae; Asteraceae)

Abstract

Premise of the study: Underlying discordance in phylogenomic studies is becoming more common, and the answer is not as simple as adding more data. Biological processes such as polyploidy, hybridization, and incomplete lineage sorting are main contributors to these issues and must be considered when generating phylogenies. Otherwise, interpretations of evolutionary relationships could be misleading. Methods: To obtain a better understanding of potential gene flow and its effect on phylogenetic trees, we investigated the causes and consequences of nuclear discordance using the genusPackerato understand how they influence the phylogenetic patterns seen in this complex group. To do this, we compared the topology and support values ofPackeraphylogenies resulting from various paralog selection or pruning methods. We then investigated whether pruning the paralogs instead of performing a selection process affected the topology and support of our phylogeny. To investigate hybridization and its effect on species relationships in our tree, we used likelihood methods to infer phylogenetic networks to find any evidence of gene flow among species lineages in this complicated genus. Key results: We found that performing different paralog selection or pruning methods does impact our understanding of the evolutionary relationships withinPackera, and that addressing these paralogs with more rigorous methods than the typical pipeline increases concordance within the resulting phylogenies. Additionally, investigating reticulation events within highly discordant clades showed that ancestral hybridization and reticulation events are common throughoutPackera. Conclusions: Investigating underlying biological processes by testing various methods can provide further insight into complex species relationships and levels of discordance within phylogenomic studies.

Compositae‐ParaLoss‐1272: A complementary sunflower‐specific probe set reduces paralogs in phylogenomic analyses of complex systems

“…Similar conclusions have been found in other groups (e.g., Sessa et al, 2012 ; Vargas et al, 2017 ; Morales‐Briones et al, 2018 ). Interestingly, a recent study in Packera showed that low support or discordant clades may be the result of ancient reticulation events in Packera 's history (Moore‐Pollard and Mandel, 2023b ), ultimately influencing the relationships and support within the species trees. We hypothesize that using Comp‐ParaLoss‐1272 will not only directly reduce issues associated with polyploidy, but also reduce issues from hybridization even if not addressed directly.…”

“…While removing paralogs from a data set may alleviate issues associated with ortholog determination in phylogenomic studies, it is important to note that paralogs are still reflective of the true evolutionary history of genes within some groups, including Packera . For example, hybridization and polyploidy are common in Packera , with around 40% of all Packera members exhibiting polyploidy (Trock, 1999 ; Moore‐Pollard and Mandel, 2023a , 2023b ), and thus paralogs are expected in the data set as it reflects the true evolutionary history of the group. Therefore, removing paralogs can remove full gene histories, impacting the ability to accurately model processes like reticulation and polyploidy.…”