Hybrid dynamics in a species group of swallowtail butterflies

Abstract

Hybrid zones provide unique natural laboratories for studying mechanisms of evolution. But identification and classification of hybrid individuals (F1, F2, backcross, etc.) can be complicated by real population changes over time as well as by use of different marker types, both of which challenge documentation of hybrid dynamics. Here, we use multiple genetic markers (mitochondrial DNA, microsatellites and genomewide single nucleotide polymorphisms) to re-examine population structure in a hybrid zone between two species of swallowtail butterflies in western Canada, Papilio machaon and P. zelicaon. Our aim was to test whether their hybrid dynamics remain the same as found 30 years ago using morphology and allozymes, and we compared different genetic data sets as well as alternative hybrid identification and classification methods. Overall, we found high differentiation between the two parental species, corroborating previous research from the 1980s. We identified fewer hybrid individuals in the main zone of hybridization in recent years, but this finding depended on the genetic markers considered. Comparison of methods with simulated data sets generated from our data showed that single nucleotide polymorphisms were more powerful than microsatellites for both hybrid identification and classification. Moreover, substantial variation among comparisons underlined the value of multiple markers and methods for documenting evolutionarily dynamic systems.

Mitochondrial phylogenomics, the origin of swallowtail butterflies, and the impact of the number of clocks in Bayesian molecular dating

“…They reach their greatest richness in tropical regions (Wallace, ; Braby et al ., ; Condamine et al ., ; Owens et al ., ), and also occur in temperate and cold climates (Omoto et al ., ; Katoh et al ., ; Nazari et al ., ; Michel et al ., ). Investigations of Papilionidae have contributed to fundamental insights into biogeography (Wallace, ), conservation biology (Collins & Morris, ), insect–plant interactions (Ehrlich & Raven, ; Berenbaum & Feeny, ), speciation (Boggs et al ., ; Dupuis & Sperling, ), and other evolutionary and ecological studies (Scriber et al ., ; Kunte, ; Condamine et al ., ; Kunte et al ., ). Addressing major questions from these fields requires a reliable and accurate phylogenetic framework for pinpointing significant phylogenetic events.…”

Gene flow and climate‐associated genetic variation in a vagile habitat specialist

“…We visited most known P. m. dodi occurrence locations in Canada and aimed to collect 5–10 individuals every 25–50 km along the Red Deer, South Saskatchewan, Old Man, and Milk Rivers, where suitable habitat is present (collection locations are visualized in Figure 1, inset h). As with previous studies (Dupuis et al., 2016, 2019; Dupuis & Sperling, 2015, 2016; Sperling, 1987, 1990), we attempted to locate and collect P. m. dodi between major river valleys (e.g., smaller eroding slopes with A. dracunculus and hilltopping features). However, consistent with our past work and historical records (e.g., Bird et al., 1995), we did not observe any individuals outside of known suitable habitat along major river valleys.…”

“…The aim of our study was to evaluate IBD, IBR, and IBE for a taxon with high habitat specificity, a discrete dispersal stage, and distribution across a variable environment. The Old World swallowtail butterfly ( Papilio machaon L.) species group has been the subject of considerable study in North America (e.g., Dupuis, Cullingham, Nielsen, & Sperling, 2019; Dupuis, Mori, & Sperling, 2016; Dupuis & Sperling, 2015, 2016; Sperling, 1987, 1990). One subspecies in particular, P. m. dodi McDunnough, 1939, is well suited for this investigation.…”

mvmapper: Interactive spatial mapping of genetic structures

“…Various visualization options for the human and swallowtail butterfly microsatellite data sets of Rosenberg et al. () (a and b) and Dupuis and Sperling () (c), respectively. (a) Population grouping vs. principal component 3, (b) latitude vs. principal component 3 and (c) principal component 2 vs. 1, coloured by COI clade and zoomed in to the Red Deer River valley in southeast Alberta, Canada [Colour figure can be viewed at wileyonlinelibrary.com]…”

“…Assessing patterns of genetic structure is one of the foundational challenges of population genetics (Pritchard, Stephens, & Donnelly, 2000;Slatkin, 1987;Verity & Nichols, 2016;Wright, 1949), and characterizing this structure across geographic space is one of the first steps in most population genetic studies. Such contextualization of genetic structure allows in-depth evolutionary investigations, such as characterizing dispersal and invasion pathways (Genton, Shykoff, & Giraud, 2005;Janes et al, 2014;Mori, Davis, & Evenden, 2016), assessing and prioritizing conservation efforts (Austin, Jelks, Tate, Johnson, & Jordan, 2011;Proshek et al, 2015;Zenboudji et al, 2016), quantifying hybridization (Chatfield, Kozak, Fitzpatrick, & Tucker, 2010;Dupuis & Sperling, 2016) and even utilizing genomic information to predict human origins (Das, Wexler, Pirooznia, & Elhaik, 2016;Elhaik et al, 2014;Flegontov et al, 2016). Some analyses explicitly incorporate spatial information in the assessment of population structure (e.g., TESS: Caye, Deist, Martins, Michel, and Francois (2016), BAPS: Cheng, Connor, Siren, Aanensen, and Corander (2013), GENELAND: Guillot, Mortier, and Estoup (2005), EEMS: Petkova, Novembre, and Stephens (2016), SCAT: Wasser et al (2004), sPCA: Jombart, Devillard, Dufour, and Pontier (2008)), and landscape genetics is a fast growing field of statistics combining population genetics and landscape ecology (Manel & Holderegger, 2013;Manel, Schwartz, Luikart, & Taberlet, 2003;Storfer et al, 2007).…”