Skip to main content

Genomics of hybrid parallel origin in Aquilegia ecalcarata

Abstract

Background

The parallel evolution of similar traits or species provides strong evidence for the role of natural selection in evolution. Traits or species that evolved repeatedly can be driven by separate de novo mutations or interspecific gene flow. Although parallel evolution has been reported in many studies, documented cases of parallel evolution caused by gene flow are scarce by comparison. Aquilegia ecalcarata and A. kansuensis belong to the genus of Aquilegia, and are the closest related sister species. Mutiple origins of A. ecalcarata have been reported in previous studies, but whether they have been driven by separate de novo mutations or gene flow remains unclear.

Results

In this study, We conducted genomic analysis from 158 individuals of two repeatedly evolving pairs of A. ecalcarata and A. kansuensis. All samples were divided into two distinct clades with obvious geographical distribution based on phylogeny and population structure. Demographic modeling revealed that the origin of the A. ecalcarata in the Eastern of China was caused by gene flow, and the Eastern A. ecalcarata occurred following introgression from Western A. ecalcarata population. Analysis of Treemix and D-statistic also revealed that a strong signal of gene flow was detected from Western A. ecalcarata to Eastern A. ecalcarata. Genetic divergence and selective sweep analyses inferred parallel regions of genomic divergence and identified many candidate genes associated with ecologically adaptive divergence between species pair. Comparative analysis of parallel diverged regions and gene introgression confirms that gene flow contributed to the parallel evolution of A. ecalcarata.

Conclusions

Our results further confirmed the multiple origins of A. ecalcarata and highlighted the roles of gene flow. These findings provide new evidence for parallel origin after hybridization as well as insights into the ecological adaptation mechanisms underlying the parallel origins of species.

Peer Review reports

Introduction

Independent populations that colonize similar environments and evolve similar traits provide strong evidence for the deterministic role of natural selection in evolution. The resulting pattern has been called “parallel” or “convergent” [1,2,3,4,5]. Replication occurs at different levels of biological tissue, including genes, pathways, networks, univariate and multivariate phenotypes, ecological traits, and biological communities, and may lead to replicated evolution of species or ecotypes [6,7,8,9]. Indeed, the genetic mechanisms underlying parallel evolution are often unclear in many studies of repeated evolution. Nevertheless, the traits that evolved repeatedly in parallel are often assumed to have arisen independently through separate de novo mutations (narrow-sense definition of parallel evolution), but such mutations could also have been recruited from shared ancestral polymorphisms or interspecific gene flow [10,11,12]. For example, widespread parallel evolution in sticklebacks was due to repeated fixation of ectodysplasin alleles [13], and Pundamilia cichlid species appeared after hybridization [14, 15].

With the advent of population genomic data, it is now possible to detect genomic regions putatively underlying recent convergent adaptations. Introgression hybridization has been proposed as an essential source of adaptive genetic variation [16]. Soria-Carrasco et al. [17] found that 17% of the single-nucleotide polymorphisms (SNPs) in the genome of Timema cristinae in California occurred between two or more pairs of parallel ecotypes and that 0.01% of SNPs were affected by natural selection according to a field experiment. Meier et al. [15] used genomic analyses to study the parallel ecological speciation of blue and red-backed Pundamilia cichlid species in Lake Victoria. Their findings revealed that a subset of the most strongly diverged regions in older species pairs also diverged in younger pairs, and these shared diverged regions exhibited parallel differences in allele frequency.

Parallel origins of Aquilegia ecalcarata have been documented by previous studies [18, 19]. A. ecalcarata has been divided into the Eastern clade and Western clade in China based on population genomic data. The Eastern clade includes A. ecalcarata and A. kansuensis, and the Western clade includes A. ecalcarata, A. rockii, and A. kansuensis. The genetic introgression from Western A. ecalcarata has contributed to the emergence of the A. rockii phenotype with straight and short nectar spur [19]. The main morphological differences between A. ecalcarata and A. kansuensis include the size of the flower organs and the presence or absence of nectar spurs, A. kansuensis has nectar spurs and the nectar spurs have played a key role in the floral isolation between A. ecalcarata and A. kansuensis [19]. The multiple origin of A. ecalcarata are adapted to a stony environment that differs from that of their closest relatives, indicating a habitat shift may have driven new adaptations [18]. Ballerini et al. [20] found that the POPOVICH plays a critically important role in nectar spur development and has recently been shown to encodes a C2H2 zinc-finger transcription factor. It has been reported that POPOVICH plays a central role in regulating cell proliferation in the Aquilegia petal during the early phase of spur development [20]. The POPOVICH gene is located on linkage chromosome3_27454200–27,455,760 in the A. coerulea ‘Goldsmith’ v3.1 reference genome (https://phytozome.jgi.doe.gov) corresponding to linkage chromosome3_26779918–26,781,011 of the A. oxysepala var. kansuensis v1.0 genome assembly [21]. Geng et al. [19] analyzed the shared alleles of the POPOVICH gene in A. ecalcarata lineages of different origins and found that the spurless trait might have a single origin. However, incomplete lineage sorting precluded determination of whether the different A. ecalcarata lineages originated through hybridization. Thus, whether the different origins of A. ecalcarata lineages were driven by multiple independent mutational or introgression hybridization remain unclear.

Here, we conducted a genomic analysis of 158 individuals of two repeatedly evolving pairs (Eastern species pair and Western species pair) of A. ecalcarata and A. kansuensis. Using population genetic structure, model simulations, treemix and ABBA-BABA tests, genetic differentiation, GO enrichment analysis and patterns of isolation by distance (IBD), determine the origin pattern of Eastern and Western A. ecalcarata, detect the important role of gene flow in parallel evolution, and infer the genetic basis of multiple origin of A. ecalcarata.

Materials and methods

Sample collection, sequencing, variant calling and filtering

A total of 158 individuals were selected for genomic analysis, populations information was shown in Table S1. The first author identifed and deposited voucher specimens in the herbarium of Shaanxi Normal University (SNNU) (Table S1). All specimens were collected from the wild. Eastern species pair (including 40 A. kansuensis and 55 A. ecalcarata) and Western species pair (including 23 A. kansuensis and 40 A. ecalcarata). DNA was extracted from dried leaves using a Genomic DNA extraction kit. Libraries were generated using the Illumina Next Ultra DNA Library Prep Kit (TIANGEN, BEIJING, China). Sequencing libraries were sequenced on the Illumina HiSeq 4000 platform (San Diego, CA, USA) to generate 150 bp paired reads with the sequencing depth of 20×. All obtained DNA reads have been deposited in the NCBI database under BioProject: PRJNA690975, and individual number information was shown in Table S1. High quality clean reads were mapped to A. yangii reference genome [21] using bwa v0.7.15 with default options [22]. Alignment results and marked duplicate reads were sorted using samtools v1.3.1 [23], and duplicate reads were removed using samtools v1.3.1. Variants were called using samtools v1.3.1 and filtered using VCFtools v0.1.13 [24]. The specific commands and parameters used were as follows: Samtools calling (multisample): samtools mpileup -b bam.list ‐C 50 ‐q 25 ‐f ‐‐output ‐v ‐u ‐t DP ‐t AD ‐t SP ‐e ‐h ‐L ‐o –p. Vcftools filtering: vcftools –vcf ‐‐minQ 25 –min‐meanDP 5 ‐‐max‐‐meanDP 30 ‐‐maf 0.02 ‐‐max‐missing 0.5 –out.

Phylogenetic inference

We converted the vcf file into a fasta file using a perl script. The script handled the loci as follows: replacing heterozygous loci with AC = > M, CA = > M, AG = > R, GA = > R,AT = > W, TA = > W, CG = > S, GC = > S, CT = > Y, TC = > Y, TG = > K, and GT = > K, and all non-variant sites were removed. We regarded A. yabeana (The name of the three individuals: NM0101, BJ0101 and HA0101, Table S1) as outgroup and constructed the maximum likelihood (ML) tree using IQ-TREE v2.0.3 [25] under the GTR model [26] with 1,000 bootstrap replicates [27]. The phylogenetic tree was edited and modified using FigTree v1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/).

Genetic structure

The population genetic structure in our samples was inferred using ADMIXTURE v.1.4.0 [28]. The postulated number of ancestral populations (K) was set from 2 to 8, and 10-fold cross-validation (--cv = 10) was performed. We selected the most likely K based on the minimum cross-validation (CV) error. Principal components analysis (PCA), a dimensionality-reduction method, was conducted to further assess the population structure. First, we converted the format using VCFtools v0.1.16 and PLINK v1.9 [29]. Then, SNPs were filtered with the parameters as “--indep-pairwise 50 5 0.5”. PCA was accomplished on all SNPs using smartpca program from EIGENSOFT v6.1.4 [30] with default parameters.

Demographic history

To discriminate among alternative evolutionary scenarios for the origin of the A. ecalcarata and A. kansuensis, we used fastsimcoal2 v2.6 [31] to conduct model simulations. First, in order to assess whether speciation occurred in a period of geographical isolation or in the face of gene flow, we conducted model simulations for both Eastern and Western species pair. To test whether the divergence of A. ecalcarata and A. kansuensis was accompanied gene flow, we compared four demographic models for the Eastern and Western pairs: (1) no gene flow; (2) secondary contact (only recent gene flow); (3) only early gene flow; (4) constant gene flow. The change time in gene flow was estimated as a model parameter (Table S2). Next, we constructed alternative demographic models for the evolution of Eastern and Western species pair combined. We compared six topologically different demographic models: (1) a single origin model, wherein Eastern A. kansuensis first diverged from ancestral group, followed by Western A. kansuensis, Eastern and Western A. ecalcarata eventually diverged, allowing for recent gene flow between species pairs and between the same species in Eastern and Western populations; (2) a single origin model, wherein Western A. kansuensis first diverged from ancestral group, followed by Eastern A. kansuensis, Eastern and Western A. ecalcarata eventually diverged, allowing for recent gene flow between species pairs and between the same species in Eastern and Western populations; (3) a single origin between A. kansuensis and A. ecalcarata with subsequent independent colonization of the West and the East by both species and interspecific gene flow and conspecific gene flow between species in Eastern and Western populations; (4) a parallel origin with two independent evolution events into A. kansuensis and A. ecalcarata, wherein one species pair occurs in the East, and the other one in the West, allowing for subsequent gene flow between species pairs and between the same species in Eastern and Western populations; (5) a hybrid parallel origin (paralle origin after hybridization) model, wherein the Western species pair is derived from a hybrid ancestor, allowing for recent gene flow between species pairs and between the same species in Eastern and Western populations; and (6) a hybrid parallel origin model, wherein the Eastern species pair is derived from a hybrid ancestor, allowing for recent gene flow between species pairs andas well as between the same species in Eastern and Western populations. The estimated generation time were set to 2 year and mutation rata were set to 7e−9 per base pair per generation based on the rate of Arabidopsis thaliana [32]. Alternative models of historical events were fitted to the joint site frequency spectra data, and two-dimensional joint SFS (2D-SFS) was constructed from posterior probabilities of sample allele frequencies using easySFS.py (https://github.com/isaacovercast/easySFS). Each model was run 50 times with 100,000 simulations to calculate composite likelihood and 40 expectation-conditional maximization (ECM) cycles. The best model was identified using the maximum likelihoods value distributions and Akaike’s information criterion (AIC) [31]. Finally, we calculated 95% confidence intervals of demographic parameters estimated from 100 bootstrap replicates by simulating SFS from the maximum composite likelihood estimates and re-estimating parameters. To further verify the results of fastsimcoal2, we used Migrate-n software to infer gene flow between the Eastern and Western A. ecalcarata [33]. Six models were used to infer different patterns of gene flow: (1) Western A. ecalcarata had a past gene flow to Eastern A. ecalcarata, (2) Eastern A. ecalcarata had a past gene flow to Western A. ecalcarata, (3) there was bidirectional gene flow between Eastern and Western A. ecalcarata, (4) there was no past gene flow between Eastern and Western A. ecalcarata, (5) Eastern A. ecalcarata had a past gene flow to the ancestral population of Western group, and (6) Western A. ecalcarata had a past gene flow to the ancestral population of Eastern group. The specific parameters used were: ML analysis strategy, 10 short chains (totaling 10,000 trees) and 3 long chains (totaling 500,000 trees), burn-in of the initial 100,000 trees, adaptive heating scheme (heating = ADAPTIVE), four temperature intervals of 1, 1.2, 1.5, and 3, with other settings using default parameters. The best model was determined using the maximum likelihood value.

We executed pairwise sequentially Markovian coalescent (PSMC) modeling [22] to estimate historical changes in Ne (effective population size) through periods based on each species. The Ne was also calculated using SMC + + v. 1.15.2. The mutation rate was 1.4E-8 per site per year, and the one generation was 2 years [19].

Detection of gene flow

Treemix v1.13 [34] was used to infer the direction and strength of gene flow. The number of migration edges (m parameter) of Treemix algorithm was set from 2 to 5, and per m parameter was iterated 5 times. The default “Evanno”-like method and various linear model in R package OptM [35] were used to determine the best m value. To further confirm the genetic introgression, ABBA-BABA analysis (D-statistic) was conducted with Dsuite v0.3 [36]. The D-statistic [37] is designed for a 4-taxon fixed phylogeny of P1, P2, P3, and outgroup (O). When there is no gene introgression between O, P1, P2 and P3, the D-statistic would be 0. If, however, there is gene introgression between P2 and P3 or P1 and P3, the D-statistic would be positive or negative. The vcf file was used directly as the input file, and SNPs were filtered using parameters “--maf 0.05 --geno 0.2”. The Dtrios procedure was then used to identify gene introgression at a significant level of p < 0.01. In addition, we used fd statistics to more accurately analyze the gene flow between different originated A. ecalcarata, and we used 5 kb window to analyze fd statistics [38].

Genetic divergence and population selection

In order to study genetic divergence and population selection among species, genome scanning was used to analyze the genomic data, and all statistics were used a nonoverlapping 5Kb window. The relative divergence (Fst), the absolute divergence (Dxy) and the nucleotide polymorphism (π) were calculated using the Python script [38]. The population selection tests was performed through XP-CLR analysis [39] with nonoverlapping 5Kb window, linkage disequilibeium (LD) value is 0.95, and the minimum number of SNPs in each window is set to 200. Statistical analysis was conducted on the Eastern and Western species pairs, respectively. We standardized per-window Fst or Dxy in each group pair to a Z-score [40] based on the formula: Z-value = (window value − mean value)/standard deviation value. The windows with Z-FST ≥ 2 or Z-Dxy ≥ 2 from each species pair were treated as highly diverged regions (HDRs) [41]. Venn diagram was calculated and visualised using the jvenn online website (http://bioinfo.genotoul.fr/jvenn). Ggplot was used to visualize the boxplot, ggsignif is used to add significance markers, significance test was performed using t-test (https://0-doi-org.brum.beds.ac.uk/10.31234/osf.io/7awm6).

To further demonstrate the impact of gene introgression on genetic divergence in Eastern and Western specie, we then calculated and compared the genomic characteristics of the gene introgression regions, and we used “bedtools shuffle” to randomly generate 1000 genomic regions, each of which was the same size, so that these genomic regions were representative of the genome-wide range. We compared the Fst and Dxy between the introgression region and 1,000 randomly selected genomic regions with same size.

Gene Ontology (GO) annotation

We selected the target regions on the genome and bcftools v1.1 [23] was used to export the fasta sequence, which then imported into the OmicsBox software (http://manual.omicsbox.com). Blast analysis was used, followed by running InterproScan and merging InterproScan GOs to annotation. GO annotation was performed after GO mapping with default parameters. GO-weight was set to 5, e-value-hit-filter to 1e-3, and other parameters to default values.

Haplotype analysis and isolation by distance (IBD)

Haplotype were estimated with ape and pegas in R v3.4.3, and network were drawn through PopART v1.7 [42]. Then, pheatmap in R v3.4.3 was used to draw the haplotype heat map. Mantel test was performed to infer the effect of geographic distance on genetic divergence. PLINK v1.9 was used to calculate pairwise genetic distance of identity by state (IBS) for each population [29]. The geographic distance matrix among each population was calculated in accordance with latitude and longitude. Finally, the correlation between geographic distance and genetic distance was accomplished in R v3.4.3.

Results

Phylogenetic inference and genetic structure

The ML phylogenetic tree was performed based on all SNPs (Fig. 1b). The phylogenetic tree was divided into two distinct clades with high bootstrap support. According to the geographical distribution of A. kansuensis and A. ecalcarata (Fig. 1e), we further called the two large clades as the Western group and the Eastern group (corresponding to Western species pair and Eastern species pair). ADMIXTURE was performed to analyze population genetic structure, When K = 2, the Eastern group and Western group had specific genetic components. When K = 3, the populations in the Eastern group demonstrated two major genetic components. Meanwhile, the genetically mixed populations (HB03, GZ01 and CQ02) appeared, which were more pronounced at K = 4 (Fig. 1c and e). And the Western group simultaneously showed two genetic components corresponding to A. kansuensis (SC16, SC31, SC32, SC33 and SC34) and A. ecalcarata (QH11, SC01, SC04, SC05, SC06, SC08, SC13 and SC14). When K = 5, part of A. kansuensis in the Eastern group had a new genetic component. The results of PCA of all SNPs also reflected a population genetic structure (Fig. 1d). The first principal component (pc1; variance explained = 38.378%) clearly separated the Western group and the Eastern group. And the second principal component (pc2; variance explained = 27.856%) clearly separated A. kansuensis and A. ecalcarata in Western group.

Fig. 1
figure 1

Phylogenetic inference and genetic structure. (a) Floral morphology of two species pairs (Eastern species pair and Western species pair). (b) The Maximum Likelihood (ML) phylogenetic tree of 158 individuals of A. kansuensis and A. ecalcarata. (c) Results from ADMIXTURE with K = 2, 3, 4, 5 based on SNP data. (d) The first two dimensions of principal component analysis (PCA) based on SNP data. (e) Geographic distribution of the sampling locations. The colors represent ancestral genetic components (according to ADMIXTURE at K = 4)

Demographic history and gene flow

The highest likelihood distribution and the lowest AIC of Eastern and Western species pairs was both obtained for a model of divergence with recent gene flow (Fig. 2a; Figure S1 and S2). Among the six models for the origin of A. ecalcarata, the hybrid parallel origin model of the Eastern species pair derived from hybrid ancestors was the best model (highest likelihood distribution and lowest AIC) (Fig. 2c and S3; Table S3 and S4). Under this model, the first divergent event into A. kansuensis and A. ecalcarata in the West occurred at 2,174 kya (95% CI: 2,150-2,198 kya).

Fig. 2
figure 2

Demographic history and gene flow. (a) Four different models for inferring the gene flow patterns between species pair. (b) Comparing demographic models with different topologies (single, parallel and hybrid origin) with recent gene flow scenarios. A black arrow with ancestry proportion m indicates introgression event. Other black arrows indicate gene flow. (c) The demographic model with the best fit. Rectangles represent populations, whereas the numbers inside the rectangle indicates effective population size and the numbers corresponding to the dotted lines indicate the splitting times. (d) Three gene flow events inferred by Treemix. kan-w: Western A. kansuensis; eca-w: Western A. ecalcarata; kan-e: Eastern A. kansuensis; eca-e: Eastern A. ecalcarata. (e) and (f) PSMC and SMC + + estimations of the effective population size (Ne) for Eastern and Western species pairs. The time scale on the x-axis is calculated assuming a neutral mutation rate (µ = 1.4e-08 per site per year) and generation time (g = 2 years)

Subsequently, A. ecalcarata colonized the East and introgressed into the local A. kansuensis population (95% CI: 1,610-1,648 kya). Shortly after the introgression event, the admixed population was split into two population at 1,406 kya (95% CI: 1,394-1,418 kya). It was estimated that this introgression event was estimated to contribute a high proportion of the genetic variation (54%) of the ancestor of both Eastern species. Gene flow between the Western species pair was higher than that between the Eastern species pair, while recent gene flow was especially high from Western A. ecalcarata into Western A. kansuensis. To further verify the results of fastsimcoal2, Migrate-n software were used to infer gene flow between the Eastern and Western A. ecalcarata. The results indicate that Model 6 has the highest likelihood and probability (Table S5), supporting the hypothesis that Western A. ecalcarata had a past gene flow to the ancestral population of Eastern group.

PSMC and SMC + + were also used to infer population demographic (Fig. 2e and f). The results of PSMC analysis indicated a decrease in the Ne (effective population size) of Eastern and Western species pairs before 2 kya (kilo year ago), and the impact is greatest in Western A. kansuensis. After that, different groups underwent varying degrees of population expansion. SMC + + analysis also showed that four groups experienced different degrees of population contraction and expansion ranging from 1 to 10 kya, and Western A. kansuensis was more affected than Western A. kansuensis, and the same pattern was also observed in A. ecalcarata. These results suggest that the western population may have been more severely affected by the Quaternary glacial period.

The potential intraspecific and interspecific gene flow were also examined with Treemix and D-statistic. The OptM determined the optimal migration model as m = 3, suggesting that three migration edges might have occurred (Fig. 2d and S4). A strong signal of gene flow was detected from Western A. ecalcarata to Eastern A. ecalcarata. Furthermore, two other relatively weak gene flow were inferred from Eastern A. kansuensis to Western A. kansuensis and outgroup. The result was confirmed by the remarkable D value in D-statistic (Fig. 3). Gene flow between Eastern A. ecalcarata and Western A. ecalcarata had occurred as well when Eastern A. kansuensis was P1 (Fig. 3a), while the D value (0.111) was remarkable (p = 1.644e-10), and f4 admixture ratio (f4-ratio) was 9.68% (Table 1). Another significant (p = 1.241e-08) gene flow between Western A. kansuensis and Eastern A. kansuensis (0.117) had occurred when Western A. ecalcarata was P1 (Fig. 3d), and f4 admixture ratio (f4-ratio) was 8.83% (Table 1).

Table 1 The results of D-statistic and f4-ratio among the four Trios
Fig. 3
figure 3

Genetic introgression inference. (a-d) Analysis of ABBA-BABA. A non-zero D statistic reflects an asymmetric pattern of allele sharing, implying gene flow. Z value and p value reflect the significance of the test. An absolute value of the Z score above 3 is often used as a critical value

Shared high differentiation regions between species pairs

To determine the parallelism and non-parallelism of genetic divergence, a sliding window was used to calculate the genetic divergence (Fst) among the species pairs (Fig. 4a), and windows with Z-Fst ≥ 2 or Z-Dxy ≥ 2 were identified as highly diverged regions (HDRs). This approach resulted in 2446 HDRs in Eastern species pair and 2061 HDRs in Western species pair based on Z-Fst ≥ 2, respectively. The Fst estimates for both species pairs were significantly higher in HDRs than in Non-HDRs (Figure S5a and S5b). By comparing the parts of the HDRs that overlap between Eastern species pair and Western species pair, we obtained 123 shared HDRs, accounting for 5.03% HDRs in Eastern species pair and 5.97% HDRs in Western species pair (Figure S6a). The sharing ratio of HDRs was significantly lower than that of Non-HDRs (Chi-square test, p-value = 2.2e-15). These results indicate that only a subset of the highly differentiated regions of the original species pair is also differentiated between the younger species, and most regions of the genome were non-parallel. The Fst estimates for both Eastern and Western species pairs were not significantly higher in shared HDRs than in remaining HDRs (Figure S7a and S7b). Nucleotide polymorphisms can substantially affect relative divergence (Fst), so we analyzed absolute sequence divergence (Dxy) (Fig. 4b). According to Z-Dxy ≥ 2, 1771 HDRs in Eastern species pair and 1741 HDRs in Western species pair were identified (Fig. 4d and e). The Dxy estimates for both species pairs were significantly higher in HDRs than in Non-HDRs (Figure S5c and S5d). By comparing the parts of the HDRs that overlap between Eastern species pair and Western species pair, we obtained 742 shared HDRs, accounting for 41.90% HDRs in Eastern species pair and 42.62% HDRs in Western species pair (Figure S6b). The sharing ratio of HDRs was significantly lower than that of Non-HDRs (Chi-square test, p-value = 2.2e-14). The Dxy estimates for both Eastern and Western species pairs were significantly higher in shared HDRs than in remaining HDRs (Figure S7c and S7d). These results indicate that the shared HDRs play an important role in the divergence of the species pair.

Analysis of selection pressure

The selection pressure was performed through XP-CLR and XP-EHH analysis, and the candidate HDRs located in the top 5% XP-CLR or top 5% XP-EHH windows were considered as a candidate positive selection region. According to Fst and XP-CLR, a total of 930 HDRs had experienced positive selection in the Eastern species pair, accounting for 38% of the HDRs, and 351 HDRs had experienced positive selection in the Western species pair, accounting for 17% (Figure S8). According to Dxy and XP-CLR, a total of 82 HDRs had experienced positive selection in the Eastern species pair, accounting for 5% of the HDRs, and 142 HDRs had experienced positive selection in the Western species pair, accounting for 8% (Figure S8). Upon GO enrichment analysis of the candidate positive selection regions, we found that both Eastern and Western species were enriched in the category of regulation of photoperiodism, flowering, regulation of flower development, leaf development, seed germination, response to cold, DNA repair, defense response to bacterium (Figure S8; Gene IDs are listed in Table S6-S9 ).

Fig. 4
figure 4

The genetic divergence and selection pressure among Eastern and Western species pairs. (a) The relative sequence divergence among species pairs. (b) The absolute sequence divergence among species pairs. (c) Selective sweeps analysis based on XP-CLR. (d) and (e) Venn diagram plots with overlapping windows among HDRs and top 5% XP-EHH. (f) Enrichment categories of candidate positive selection region based on XP-EHH in Eastern A. ecalcarata. (g) Enrichment categories of candidate positive selection region based on XP-EHH in Western A. ecalcarata

According to Fst and XP-EHH, a total of 430 HDRs had experienced positive selection in the Eastern A. ecalcarata, accounting for 18% of the HDRs, and 664 HDRs had experienced positive selection in the Western A. ecalcarata, accounting for 32% (Fig. 4). According to Dxy and XP-CLR, a total of 194 HDRs had experienced positive selection in the Eastern A. ecalcarata, accounting for 11% of the HDRs, and 284 HDRs had experienced positive selection in the Western A. ecalcarata, accounting for 16% (Fig. 4). Upon GO enrichment analysis of the candidate positive selection regions, we found that both Eastern and Western species were enriched in the category of response to water deprivation, response to hypoxia, regulation of photoperiodism, flowering, regulation of flower development, leaf development, seed germination, response to cold, DNA repair, defense response to bacterium and virus (Fig. 4; Gene IDs are listed in Table S10-S13).

This suggests that the photoperiodism, precipitation, temperature, DNA repair, flowering time, regulation of flower and leaf development, seed germination, and defensive response of bacterium and virus may be important driving factors behind the parallel evolution of eastern and western A. ecalcarata.

Gene flow in highly diverged regions

Determining whether the phenotypic similarity between two similar taxa stems from hybridization is a major challenge because the magnitude of gene flow can vary among regions in the genome. We detected gene flow in the parallel diverged regions of the Eastern and Western species pairs. According to D-statistical analysis, the trio (kan_east, eca_east), eca_west) revealed gene flow between Eastern A. ecalcarata and Western A. ecalcarata had occurred (Fig. 3a), and f4 admixture ratio (f4-ratio) was 9.68% (Table 1). So we further localized introgressed regions by calculating fd statistics, which have been proven to be more useful to assist in locating introgressed loci in small genomic regions compared with the D-statistics. Introgressed regions were defined as the top fd windows that summed to the genomic proportion estimated from the f4-ratio (9.68%), and 1657 introgression windows were identified (Fig. 5a). According to Fst, 242 HDRs in Eastern species pairs overlap with the introgression windows, which accounted for 10% of the HDRs (2446 HDRs), and 111 HDRs in Western species pairs overlaps with the introgression windows, which accounted for 5% of the HDRs (2061 HDRs) (Fig. 5b). 24% (29/123) shared HDRs overlap with the introgression windows between Eastern and Western species pairs (Fig. 5c). According to Dxy, 72 HDRs in Eastern species pairs overlap with the introgression windows, which accounted for 4% of the HDRs (1771 HDRs) in the Eastern species pairs, and 82 HDRs in Western species pairs overlaps with the.

Fig. 5
figure 5

The influence of gene flow on the divergence of species pairs. (a) Manhattan plot showing the fd values across genome. Dark spots indicate the locations of the candidate introgressed regions. (b) Venn diagram plots with overlapping windows between HDRs and Introgression. (c) Venn diagram plots with overlapping windows between shared HDRs and Introgression. (d) Diagram showing the comparison of genetic divergence (Fst and Dxy) between introgressed regions (solid line) and the distribution resulting from the 1,000 randomizations (dashed line: average across randomizations). (e) The Fst distribution for both species pairs were significantly higher in the HDRs overlap with the introgression windows than in remaining HDRs

introgression windows, which accounted for 5% of the HDRs (1741 HDRs) in the Western species pairs (Fig. 5b). 5% (29/123) shared HDRs overlap with the introgression windows between Eastern and Western species pairs (Fig. 5c). The proportion of both shared HDRs by Fst and Dxy overlap with the introgression windows is significantly higher than that of the remaining HDRs (Chi-square test, p-value = 2.2e-16; Chi-square test, p-value = 0.0028). These results indicate that gene introgression plays a very important role in the parallel evolution of Eastern and Western A. ecalcarata.

To further demonstrate the impact of gene introgression on genetic divergence in Eastern and Western species, we compared the genetic divergence coefficients between the introgression window and the 1000 random sampling windows. The Fst and Dxy estimates of introgression windows for both species pairs were significantly higher than at random (Fig. 5d). The Fst estimates for both species pairs were significantly higher in the HDRs overlap with the introgression windows than in remaining HDRs (Fig. 5e). Both model analysis and gene flow detection showed that there was a significant gene flow from western to eastern species, so these results reflect the significant effect of the gene flow from Western A. ecalcarata to Eastern A. ecalcarata on the origin and divergence of the Eastern species pair.

Gene flow and function in shared positive selection regions

Based on the XP-CLR and Fst, Eastern and Western species share 8 candidate positive selection HDRs, four of which have higher gene flow, including two genes. Based on the XP-CLR and Dxy, Eastern and Western species share 4 candidate positive selection HDRs, all of which have higher gene flow, including 3 genes. By conducting GO analysis on these genes, the category of proton-transporting ATP synthase activity, G-protein coupled receptor activity, protein binding were enriched. (GO Terms and Gene IDs are listed in Table S14). Based on the XP-EHH, there are no shared HDRs with higher gene flow. In order to further clarify the functions of these gene IDs, we conducted gene identification, and found that the gene corresponding to Chr1_20.1976 is named PIA2, which has a function of response to high light intensity [43]. The phylogenetic tree and heat map show that the PIA2 gene has higher similarity between Eastern and Western A. ecalcarata (Fig. 6a and b).

we also analyzed the haplotypes of POPOVICH. A total of 12 haplotypes were generated from 158 individuals (Fig. 6b). Haplotype network and heatmap cluster tree showed that POPOVICH did not differentiate between A. kansuensis and A. ecalcarata, and there was no pattern of two major branches in the East and West (Fig. 6c and d). The haplotype network revealed that A. kansuensis and A. ecalcarata shared many haplotypes. Moreover, The window chromosome3_26779918–26,781,011 in which POPOVICH was located did not belong to the HDRs, and the fd values of this window did not rank in the top 9.68% of the entire genome.

Fig. 6
figure 6

Phylogeny and haplotype analysis of PIA2 and POPVICH genes. (a) The ML phylogenetic tree of PIA2 gene in all individuals of A. ecalcarate and A. kansuensis, Jones-Taylor-Thornton model. (b) Heatmap analysis of PIA2 gene. Each row represents a genomic position for all accession, and the column represents a individual. (c) Median-joining network of POPVICH gene haplotypes. The areas of the circles are proportional to the number of individuals. (d) Heatmap analysis of POPVICH gene. Each row represents a genomic position for all accession, and the column represents a individual

Isolation by distance (IBD)

In the Western species pair, a Mantel test revealed that genetic distance and geographic distance were significantly correlated (i.e., isolation by distance) between A. ecalcarata and A. kansuensis. A significant pattern of IBD was also detected between A. ecalcarata and A. kansuensis in the Eastern species pair. We also combined the Eastern and Western groups for IBD analysis, and the results still showed a significant relationship (Table S15). These findings suggested that geographical isolation contributes to the genetic divergence between A. ecalcarata and A. kansuensis. The value of the Mantel statistic in the Western species pair was higher than that in the Eastern species pair, indicating that geographical isolation had a greater effect on genetic differentiation in the Western species pair than in the Eastern species pair.

Discussion

Parallel origin of Aquilegia ecalcarata after hybridization

Systems with either parallel evolution or parallel maintenance of species differences are thus both useful for studying the processes underlying natural selection in the evolution of species or traits [12, 14]. It is crucial to study the underlying evolutionary mechanisms of parallel evolution, including natural selection, de novo mutations, gene flow, drift and standing genetic variation [14, 44]. However, tests discriminating single and multiple origins of ecotypes or species in the face of persistent gene flow are often lacking. Based on population level sampling, Huang et al. [45] found that A. ecalcarata is not monophyletic. But due to lack of resolution, the origin of A. ecalcarata and its phylogenetic relationships with related species remain unclear. Geng et al. [19] used genomic data to analyze the origin of A. ecalcarata, and the results indicated that A. ecalcarata could be divided into different groups; however, whether the different groups of A. ecalcarata were derived from independent parallel origins or genetic introgression remains unclear. In this study, demographic model showed that the genetic differentiation first occurred between A. ecalcarata and A. kansuensis in the Western group. Then A. ecalcarata in the Western group colonized the East and hybridization with the ancestral population in Eastern A. kansuensis. Western A. ecalcarata contributed 54% of the genetic components to the hybrid progeny, while Eastern ancestral population contributed 46% to the hybrid progeny, which is similar to the genetic contribution ratio of many hybrid cases, and the parents is close to 50% [14, 46, 47]. Shortly after the hybridization event, the admixed population was split into A. ecalcarata and A. kansuensis in the East. Analysis of Treemix and D-statistic also revealed that a strong signal of gene flow was detected from Western A. ecalcarata to Eastern A. ecalcarata. Thus, the origin of Western A. ecalcarata preceded the hybrid origin of Eastern A. ecalcarata. This pattern of parallel origin of A. ecalcarata species after hybridization is consistent with the hybrid parallel origin model of Pundamilia cichlid species [14, 16]. Parallel origin in the narrow sense emphasizes the independence of the evolutionary history of the two independently evolving pairs [48], and some well-known cases have been documented in sticklebacks [49], stick insects [17], wildflowers [50] and wild rice [51]. Hybrid parallel origin emphasizes the key role of introgression or hybridization in the divergence of repetitive evolving pairs of ecotypes or species. However, hybrid parallel origin events has been rarely reported. The results of our study provide new evidence for hybrid parallel origin of ecotypes or species.

On the other hand, the results from TreeMix and D-statistic conflict with the “best” model in Fig. 2c. The introgression from Western A. ecalcarata to the ancestry of Eastern species pair, so the migration edge to point to the internal branch instead of the tip branch of Eastern A. ecalcarata. similarly, the D value in Fig. 3a should not be significant because both the Eastern A. ecalcarata and A. kansuensis inherited the same introgression that occurred in ancestry. The reason for this conflict is that both TreeMix and D values measured the results of gene flow, reflecting only the results of past hybridization. The ancestral population in the Eastern population with spur crossed with Western A. ecalcarata to produce the Eastern A. ecalcarata and A. kansuensis. However, due to the adaptability of the Eastern population to the local climate, the Eastern A. ecalcarata and A. kansuensis will continue to backcross with eastern ancestral population to adapt to the local environment, and the phenotype of the Eastern A. kansuensis and Eastern ancestors with spur are more similar, the phenotype and habitat of the eastern and western A. ecalcarata are more similar, so there was more subsequent gene flow from the Western A. ecalcarata lineage to the Eastern A. ecalcarata lineage. Therefore, Eastern A. ecalcarata and A. kansuensis are unlikely to inherit the same infiltration that occurred in the ancestors. The simulation of the model is a simulation of past gene flow events, highlighting the process rather than the result, so it is different from the results of TreeMix and D-statistic.

Ecological adaptation and genetic mechanism underlying the hybrid parallel origin of A. Ecalcarata

Natural selection is an important driving force for parallel evolution [2, 3]. Huang et al. [18] described that A. ecalcarata and A. kansuensis have different habitats. A. kansuensis grows in fertile soil under low altitude forests, while A. ecalcarata grows on stony beaches with poor soil at high altitude, habitat shift may be an important driving factor in the multiple origins of A. ecalcarata. Our analysis of genetic divergence also revealed that environmentally-related pathways such as response to water deprivation, response to hypoxia, regulation of photoperiodism, flowering, regulation of flower development, leaf development, seed germination, response to cold, DNA repair are important drivers of divergence among eastern and western species pair. Secondly, the PIA2 gene is located in HDR shared by both Eastern and Western species pair with higher gene flow, and responds to high light intensity. Therefore, differences of the climate factors might have contributed to the divergence between A. ecalcarata and A. kansuensis and the parallel origins of A. ecalcarata. However, additional work is needed to verify these speculations.

The spurless trait is a novel phenotype of A. ecalcarata that has contributed to the divergence between A. ecalcarata and A. kansuensis [19]. Therefore, the key gene controlling the nectar spur is likely highly divergent between A. ecalcarata and A. kansuensis. Ballerini et al. [20] found that the C2H2 transcription factor POPOVICH plays a key role in spur formation, but the window in which the POPOVICH gene was located (chromosome3_26779918–26,781,011) was not one of the HDRs (top 10% Fst windows) in the Eastern and Western species pairs. Phylogenetic tree showed that POPOVICH did not differentiate between A. kansuensis and A. ecalcarata. Therefore, the POPOVICH gene might not be the key candidate gene underlying divergence in the nectar spur in our species pairs. However, the extent to which POPOVICH’s expression level varies within our species remains unclear and is a question that needs to be explored in the future.

Non-parallelism of the genomic differentiation between the independently evolving species pairs

Analysis of genomic differentiation revealed that 123 HDRs identified by Fst and 742 HDRs identified by Dxy were shared between Eastern and Western species pairs, the number of non-shared HDRs was 2323 (identified by Fst) and 1029 (identified by Dxy) in Eastern species pair, and the number of non-shared HDRs was 1938 (identified by Fst) and 999 (identified by Dxy) in Western species pair, which indicated that most of the HDRs in the genome were non-parallel. Thus, the genetic differentiation in the Eastern species pair was not restricted to a subset of the genomic differences that characterize the Western species pair; instead, genomic differentiation likely included several new regions in the Eastern species pair, which reflects the independence of the divergence among species pairs. Many factors might contribute to explaining the observed non-parallelism among species pairs. First, the differences in the divergence of phenotypic traits between Eastern and Western species pairs might explain non-parallelism, as previous studies have shown that the direction of differentiation in the three floral traits in Eastern and Western species pairs differs [19]. Second, demographic history of Eastern and Western species pairs might lead to differences in the divergence process. Because the sudden and large decrease in population size due to the bottleneck may affect population divergence, thereby increasing the possibility of fixing mildly deleterious and effectively neutral mutations [52,53,54]. Our demographic history indicated that both the Eastern and Western species pairs have experienced varying degrees of population contraction. Thus, the Eastern and Western species pair might have experienced a bottleneck at some point in its evolutionary history, increasing the probability of the fixation of mildly deleterious and effectively neutral mutations.

Conclusions

Our study supports the gene flow contributed to the parallel evolution of A. ecalcarata. The results of gene flow test reflect the significant effect of the gene introgression from Western A. ecalcarata to Eastern A. ecalcarata on the origin and divergence of the Eastern species pair. These findings provide new evidence for parallel origin after hybridization as well as insights into the mechanisms underlying the parallel origins of species. In the next study, we will still need to conduct field experiments and molecular biology experiments to explore the ecological adaptation and genetic mechanism of the repeated origin of A. ecalcarata.

Data availability

The genomic data support the finding of this study have been desposited in the GenBank database under BioProject: PRJNA690975.

Abbreviations

AIC:

Akaike information criterion

Fst:

Genetic differentiation index

Dxy:

absolute sequence divergence index

HDRs:

highly diverged regions

2D-SFS:

two-dimensional joint SFS

ECM:

expectation-conditional maximization

IBS:

identity by state

IBD:

Isolation by distance

References

  1. Arendt J, Reznick D. Convergence and parallelism reconsidered: what have we learned about the genetics of adaptation? Trends Ecol Evol. 2008;23(1):26–32.

    Article  PubMed  Google Scholar 

  2. Clarke B. The contribution of ecological genetics to evolutionary theory: detecting the direct effects of natural selection on particular polymorphic loci. Genetics. 1975;79(Suppl):101–13.

    PubMed  Google Scholar 

  3. Langerhans RB, DeWitt TJ. Shared and unique features of evolutionary diversification. Am Nat. 2004;164(3):335–49.

    Article  PubMed  Google Scholar 

  4. Losos JB. Convergence, adaptation, and constraint. Evolution. 2011;65(7):1827–40.

    Article  PubMed  Google Scholar 

  5. Wake DB, Wake MH, Specht CD. Homoplasy: from detecting pattern to determining process and mechanism of evolution. Sci (New York N Y). 2011;331(6020):1032–5.

    Article  CAS  Google Scholar 

  6. Ostrowski EA, Woods RJ, Lenski RE. The genetic basis of parallel and divergent phenotypic responses in evolving populations of Escherichia coli. Proc Biol Sci. 2008;275(1632):277–84.

    CAS  PubMed  Google Scholar 

  7. Tenaillon O, Rodríguez-Verdugo A, Gaut RL, McDonald P, Bennett AF, Long AD, Gaut BS. The molecular diversity of adaptive convergence. Volume 335. New York, N.Y.): Science; 2012. pp. 457–61. 6067.

    Google Scholar 

  8. Woods R, Schneider D, Winkworth CL, Riley MA, Lenski RE. Tests of parallel molecular evolution in a long-term experiment with Escherichia coli. Proc Natl Acad Sci USA. 2006;103(24):9107–12.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. James ME, Brodribb T, Wright IJ, Rieseberg LH, Ortiz-Barrientos D. Replicated evolution in plants. Annu Rev Plant Biol. 2023;74:697–725.

    Article  CAS  PubMed  Google Scholar 

  10. Schluter D, Conte GL. Genetics and ecological speciation. Proceedings of the National Academy of Sciencess of the United States of America. 2009; 106 Suppl, 9955–9962.

  11. Johannesson K, Panova M, Kemppainen P, André C, Rolán-Alvarez E, Butlin RK. Repeated evolution of reproductive isolation in a marine snail: unveiling mechanisms of speciation. Ser B Biol Sci. 2010;365(1547):1735–47. Philosophical transactions of the Royal Society of London.

    Article  Google Scholar 

  12. Faria R, Renaut S, Galindo J, Pinho C, Melo-Ferreira J, Melo M, Jones F, Salzburger W, Schluter D, Butlin R. Advances in ecological speciation: an integrative approach. Mol Ecol. 2014;23(3):513–21.

    Article  PubMed  Google Scholar 

  13. Colosimo PF, Hosemann KE, Balabhadra S, Jr Dickson M, Grimwood J, Schmutz J, Myers RM, Schluter D, Kingsley DM. Widespread parallel evolution in sticklebacks by repeated fixation of ectodysplasin alleles. Science. 2005;307:1928–33.

    Article  CAS  PubMed  Google Scholar 

  14. Meier JI, Sousa VC, Marques DA, Selz OM, Wagner CE, Excoffier L, Seehausen O. Demographic modelling with whole-genome data reveals parallel origin of similar Pundamilia cichlid species after hybridization. Mol Ecol. 2017;26(1):123–41.

    Article  CAS  PubMed  Google Scholar 

  15. Meier JI, Marques DA, Wagner CE, Excoffier L, Seehausen O. Genomics of parallel ecological speciation in Lake Victoria Cichlids. Mol Biol Evol. 2018;35(6):1489–506.

    Article  CAS  PubMed  Google Scholar 

  16. Abbott RJ. Plant speciation across environmental gradients and the occurrence and nature of hybrid zones. J Syst Evol. 2017;55:238–58.

    Article  Google Scholar 

  17. Soria-Carrasco V, Gompert Z, Comeault AA, Farkas TE, Parchman TL, Johnston JS, Buerkle CA, Feder JL, Bast J, Schwander T, Egan SP, Crespi BJ, Nosil P. Stick insect genomes reveal natural selection’s role in parallel speciation. Science. 2014;344(6185):738–42.

    Article  CAS  PubMed  Google Scholar 

  18. Huang L, Geng FD, Fan JJ, Zhai W, Xue C, Zhang XH, Ren Y, Kang JQ. Evidence for two types of Aquilegia ecalcarata and its implications for adaptation to new environments. Plant Divers. 2022;44(2):153–62.

    Article  PubMed  Google Scholar 

  19. Geng FD, Xie JH, Xue C, Sun L, Li JJ, Niu CY, Huang L, Zhang XH, Kang JQ, Kong HZ, Ren Y, Zhang JQ. Loss of innovative traits underlies multiple origins of Aquilegia Ecalcarata. J Syst Evol. 2022;60(6):1291–302.

    Article  Google Scholar 

  20. Ballerini ES, Min Y, Edwards MB, Kramer EM, Hodges SA. POPOVICH, encoding a C2H2 zinc-finger transcription factor, plays a central role in the development of a key innovation, floral nectar spurs, in Aquilegia. Proc Natl Acad Sci USA. 2020;117(36):22552–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Xie J, Zhao H, Li K, Zhang R, Jiang Y, Wang M, Guo X, Yu B, Kong H, Jiao Y, Xu G. A chromosome-scale reference genome of Aquilegia oxysepala var. Kansuensis. Hortic Res. 2020;7(1):113.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinf (Oxford England). 2009;25(14):1754–60.

    CAS  Google Scholar 

  23. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R. 1000 Genome Project Data Processing Subgroup. The sequence Alignment/Map format and SAMtools. Bioinf (Oxford England). 2009;25(16):2078–9.

    Google Scholar 

  24. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, Handsaker RE, Lunter G, Marth GT, Sherry ST, McVean G, Durbin R. 1000 Genomes Project Analysis Group. The variant call format and VCFtools. Bioinf (Oxford England). 2011;27(15):2156–8.

    CAS  Google Scholar 

  25. Nguyen LT, Schmidt HA, von Haeseler A, Minh BQ. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol. 2015;32(1):268–74.

    Article  CAS  PubMed  Google Scholar 

  26. Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods. 2017;14(6):587–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Hoang DT, Chernomor O, von Haeseler A, Minh BQ, Vinh LS. UFBoot2: improving the Ultrafast bootstrap approximation. Mol Biol Evol. 2018;35(2):518–22.

    Article  CAS  PubMed  Google Scholar 

  28. Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009;19(9):1655–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, Sham PC. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Patterson N, Price AL, Reich D. Population structure and eigen analysis. PLoS Genet. 2006;2(12):e190.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Excoffier L, Dupanloup I, Huerta-Sánchez E, Sousa VC, Foll M. Robust demographic inference from genomic and SNP data. PLoS Genet. 2013;9(10):e1003905.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Ossowski S, Schneeberger K, Lucas-Lledó JI, Warthmann N, Clark RM, Shaw RG, Weigel D, Lynch M. The rate and molecular spectrum of spontaneous mutations in Arabidopsis thaliana. Volume 327. New York, N.Y.): Science; 2010. pp. 92–4. 5961.

    Google Scholar 

  33. Beerli P. Comparison of bayesian and maximum-likelihood inference of population genetic parameters. Bioinformatics. 2006;22(3):341–5.

    Article  CAS  PubMed  Google Scholar 

  34. Pickrell JK, Pritchard JK. Inference of population splits and mixtures from genome-wide allele frequency data. PLoS Genet. 2012;8(11):e1002967.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Fitak RR. OptM: estimating the optimal number of migration edges on population trees using Treemix. Biology Methods Protocols. 2021;6(1):bpab017.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Malinsky M, Matschiner M, Svardal H. Dsuite - fast D-statistics and related admixture evidence from VCF files. Mol Ecol Resour. 2021;21(2):584–95.

    Article  PubMed  Google Scholar 

  37. Durand EY, Patterson N, Reich D, Slatkin M. Testing for ancient admixture between closely related populations. Mol Biol Evol. 2011;28(8):2239–52.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Martin SH, Davey JW, Jiggins CD. Evaluating the use of ABBA-BABA statistics to locate introgressed loci. Mol Biol Evol. 2015;32(1):244–57.

    Article  CAS  PubMed  Google Scholar 

  39. Chen H, Patterson N, Reich D. Population differentiation as a test for selective sweeps. Genome Res. 2010;20(3):393–402.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Han F, Lamichhaney S, Grant BR, Grant PR, Andersson L, Webster MT. Gene flow, ancient polymorphism, and ecological adaptation shape the genomic landscape of divergence among Darwin’s finches. Genome Res. 2017;27:1004–15.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Ke F, Vasseur L, Yi H, Yang L, Wei X, Wang B, Kang M. Gene flow, linked selection, and divergent sorting of ancient polymorphism shape genomic divergence landscape in a group of edaphic specialists. Mol Ecol. 2022;31:104–18.

    Article  PubMed  Google Scholar 

  42. Leigh JW, Bryant D. POPART: full-feature software for haplotype network construction. Methods Ecol Evol. 2015;6:1110–6.

    Article  Google Scholar 

  43. Yoo JH, Cho MH, Lee SW, Bhoo SH. Phytochrome-interacting ankyrin repeat protein 2 modulates phytochrome A-mediated PIF3 phosphorylation in light signal transduction. J Biochem. 2016;4:243–9.

    Article  Google Scholar 

  44. Butlin RK, Saura M, Charrier G, Jackson B, André C, Caballero A, Coyne JA, Galindo J, Grahame JW, Hollander J, Kemppainen P, Martínez-Fernández M, Panova M, Quesada H, Johannesson K, Rolán-Alvarez E. Parallel evolution of local adaptation and reproductive isolation in the face of gene flow. Evolution. 2014;68(4):935–49.

    Article  PubMed  Google Scholar 

  45. Huang L, Geng FD, Fan JJ, Xue C, Zhang XY, Kang JQ, Zhang JQ, Ren Y. Genetic diversity and evolutionary history of four closely related Aquilegia species revealed by 10 nuclear gene fragments. J Syst Evol. 2018;56(2):129–38.

    Article  Google Scholar 

  46. Sangeet L, Fan H, Matthew TW, Leif A, Grant BR, Grant PR. Rapid hybrid speciation in Darwin’s finches. Science. 2017;359(6372):224–8.

    Google Scholar 

  47. Ru DF, Sun YS, Wang DL, Chen Y, Wang TJ, Hu QJ, Abbott RJ, Liu JQ. Population genomic analysis reveals that homoploid hybrid speciation can be a lengthy process. Mol Ecol. 2018;27:4875–87.

    Article  CAS  PubMed  Google Scholar 

  48. Schluter D, Nagel LM. Parallel speciation by natural selection. Am Nat. 1995;146(2):292–301.

    Article  Google Scholar 

  49. Rundle HD, Nagel L, Wenrick Boughman J, Schluter D. Natural selection and parallel speciation in sympatric sticklebacks. Volume 287. New York, N.Y.): Science; 2000. pp. 306–8. 5451.

    Google Scholar 

  50. Richards TJ, Walter GM, McGuigan K, Ortiz-Barrientos D. Divergent natural selection drives the evolution of reproductive isolation in an Australian wildflower. Evol Int J Org Evol. 2016;70(9):1993–2003.

    Article  Google Scholar 

  51. Cai Z, Zhou L, Ren NN, Xu X, Liu R, Huang L, Zheng XM, Meng QL, Du YS, Wang MX, Geng MF, Chen WL, Jing CY, Zou XH, Guo J, Chen CB, Zeng HZ, Liang YT, Wei XH, Guo YL, Ge S. Parallel speciation of Wild Rice Associated with Habitat shifts. Mol Biol Evol. 2019;36(5):875–89.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Barrick JE, Lenski RE. Genome dynamics during experimental evolution. Nat Rev Genet. 2013;14(12):827–39.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Elena SF, Lenski RE. Evolution experiments with microorganisms: the dynamics and genetic bases of adaptation. Nat Rev Genet. 2003;4(6):457–69.

    Article  CAS  PubMed  Google Scholar 

  54. Lang GI, Rice DP, Hickman MJ, Sodergren E, Weinstock GM, Botstein D, Desai MM. Pervasive genetic hitchhiking and clonal interference in forty evolving yeast populations. Nature. 2013;500(7464):571–4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank Cheng Xue, Wei Zhai, Xiao-Peng Chang, Xiao-Cheng Xing, Yuan-Zhen Zhang, and Yu-Qu Zhang for sample collection.

Funding

There was no Funding.

Author information

Authors and Affiliations

Authors

Contributions

F.-D.G designed the research. F.-D.G performed all analyses, F.-D.G. wrote the manuscript with the help of M.-F.L., M.-Q.L., L.-Z.W. and X.-D.Z. All authors have reviewed and approved the manuscript.

Corresponding author

Correspondence to Fang-Dong Geng.

Ethics declarations

Ethics approval and consent to participate

This study has been approved by the Chinese government and carried out with the laws of the People’s Republic of China. All participants had a license approval letter from the College of Life Sciences, Northwest University. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Geng, FD., Liu, MQ., Zhang, XD. et al. Genomics of hybrid parallel origin in Aquilegia ecalcarata. BMC Ecol Evo 24, 75 (2024). https://0-doi-org.brum.beds.ac.uk/10.1186/s12862-024-02266-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://0-doi-org.brum.beds.ac.uk/10.1186/s12862-024-02266-7

Keywords