Abstract Detail

Phylogenomic Perspectives on Reproductive Isolation and Introgression

Rifkin, Joanna [1], Rausher, Mark [2], Liao, Irene [3].

Selection, introgression, and floral traits in selfing and outcrossing morning glories.

The highly selfing morning glory Ipomoea lacunosa and its mixed-mating sister species I. cordatotriloba differ in floral and life-history traits associated with self-pollination. A Qst-Fst analysis of the two species across their ranges is consistent with selection shaping these differences in floral traits (in review). With a multivariate analysis, we found evidence consistent with selection on the suite of traits, and univariate analyses were consistent with direct selection on corolla and nectar traits but not others. These species hybridize in sympatry, and in expressed sequence demonstrate variable introgression rates across the genome (Rifkin et al. 2019). Loci that are resistant to introgression also show signatures of natural selection in a McDonald-Kreitman test.
This suggests that the loci resistant to introgression may include loci underlying the selected floral traits. To address this question, we use QTL mapping to identify genetic regions associated with floral traits and determine whether loci that strongly resisted introgression and showed signatures of selection overlap with QTLs underlying those floral traits that diverged in response to selection.
To more precisely identify the extent to which different regions of the genome permit and resist introgression, we combine this with an analysis of introgression from whole-genome sequencing of the two species across their range. In this analysis, we compare introgression rates in the regions of QTLs underlying sites that showed evidence of direct selection in the Qst-Fst analysis (flower size and shape; nectar volume and concentration) with those that did not (inflorescence structure, pollen number, life history). If selection against immigrant floral phenotypes is responsible for the regions of reduced introgression we observed, then we expect that regions underlying traits not showing evidence of direct selection should show greater introgression than those that do. If introgression is not reduced compared to the rest of the genome in regions underlying putatively selected traits, then regions of reduced introgression may reflect sites selected for other reasons.

1 - University of Toronto, Ecology and Evolutionary Biology, 25 Willcocks Street, Toronto, ON, M5S 3B2, Canada
2 - Duke University, Biology, French Family Science Center, 24 Science Drive, Durham, NC, 27708, USA
3 - Duke University, Biology, 130 Science Drive, Room 137, Biological Sciences Building, Durham, NC, 27708, United States

floral divergence
floral traits
mating system

Presentation Type: Symposium Presentation
Number: 0007
Abstract ID:724
Candidate for Awards:Margaret Menzel Award

Copyright © 2000-2019, Botanical Society of America. All rights reserved