Phylogenetics of Pinus Subsection Cembroides Engelm. (Pinaceae) Inferred from Low-Copy Nuclear Gene Sequences

Abstract

Abstract—Pinus subsection Cembroides comprises approximately 15 taxa distributed from the southwestern United States to south central Mexico. Despite previous phylogenetic studies based on morphology, nuclear ribosomal DNA, and plastid DNA, we still lack a robust phylogenetic hypothesis and clear delimitation for the closely-related species within the group. We studied the evolutionary relationships within subsection Cembroides and explored incomplete lineage sorting and reticulation using low-copy number nuclear genes. Concatenation and multispecies coalescent phylogenies were inferred from samples representing all taxa from subsection Cembroides and outgroups corresponding to the closely-related subsections Balfourianae, Nelsoniae, Gerardianae, and Krempfianae. The concatenation and coalescence-based trees mainly agreed with one another in recovering Pinus subsection Cembroides as monophyletic and in recovering similar relationships among species as in previous plastid DNA-based studies. Phylogenetic position and admixture analysis suggest that P. californiarum should be treated as a separate species from P. monophylla. Furthermore, our results support recognizing P. fallax as a species rather than as an infraspecific taxon of P. monophylla or P. edulis. The ASTRAL-III tree was consistent with the presence of very high levels of ILS in the group of pinyon pines with small cones. Analyses that account for both incomplete lineage sorting and reticulation identify some unexpected hybridization scenarios that were not reported in the literature.

Patterns of hybridization and cryptic introgression among one- and four-needled pinyon pines

“…Previous studies have either failed to include P. juarezensis in their analyses or found inconclusive results regarding its taxonomic validity ( Zavarin et al , 1980 ; Snajberk et al , 1982 ; Gernandt et al , 2001 ; Montes et al , 2019 ). In our study, we included several populations with the five-needled pinyon individuals, including some from Lanner’s original discovery ( Lanner, 1974 a ).…”

“…Hybridization in pines has been widely acknowledged ( Critchfield, 1975 , 1986 ; Willyard et al , 2009 ; Menon et al ., 2018 , 2020 ). The pinyon pine complex is an excellent system to study hybridization due to its lack of: strong reproductive isolating mechanisms ( Lanner, 1974 a ); taxonomic congruence ( Gernandt et al ., 2001 , 2003 ); and conclusive genetic studies ( Montes et al , 2019 ). Most pine species lack interspecific incompatibility mechanisms ( Critchfield, 1975 ) and have wind-mediated, long-distance pollen dispersals ( Williams, 2010 ) that can facilitate gene flow between allopatric groups (Wright, 1952).…”

“…Bailey, has more resin canals and occupies drier habitats in higher elevations than P. monophylla . In a recent study, hybridization between P. californiarum and P. quadrifolia was explored using a Hyb-seq method, yet no gene flow was detected ( Montes et al , 2019 ). One reason could be the lack of adequate sample size due to the taxonomic focus of the study which aimed to capture as many species as possible using only a few individuals.…”

“…One reason could be the lack of adequate sample size due to the taxonomic focus of the study which aimed to capture as many species as possible using only a few individuals. Montes et al (2019) stated themselves that increased sampling would be required to draw accurate conclusions about admixture events in this system. In our study, we will be testing whether P. juarezensis hybridizes with P. monophylla or with the recently proposed taxa P. californiarum.…”

Analysis of paralogs in target enrichment data pinpoints multiple ancient polyploidy events inAlchemillas.l. (Rosaceae)

“…Commonly used target enrichment assembly pipelines (e.g.,Faircloth 2016;Andermann et al 2018) use different criteria to flag assembled loci with putative paralogs that are later filtered or sorted prior to phylogenetic analysis. The most common approach for processing paralogs in target enrichment datasets is removing the entire loci that show any signal of potential paralogy (e.g.,Crowl et al 2017;Montes et al 2019;Bagley et al 2020). Other approaches either retain or remove contigs based on the distinction between putative allelic variation (flagged sequences form monophyletic groups) or putative paralogs (non-monophyletic) in combination with study-specific threshold or random selection (e.g.,Villaverde et al;Liu et al 2019; Stubbs et al 2019), or manual processing (e.g.,Garcia et al.…”