Next Article in Journal
Chromatographic Fingerprinting of the Old World Lupins Seed Alkaloids: A Supplemental Tool in Species Discrimination
Previous Article in Journal
Exogenous p-Coumaric Acid Improves Salvia hispanica L. Seedling Shoot Growth
Previous Article in Special Issue
Structure–Function Analysis Reveals the Singularity of Plant Mitochondrial DNA Replication Components: A Mosaic and Redundant System

Plants 2019, 8(12), 547;

New Insights on Lilium Phylogeny Based on a Comparative Phylogenomic Study Using Complete Plastome Sequences
Institute of Agricultural Science and Technology, Chungbuk National University, Chungbuk, 28644, Korea
Department of Horticulture, Kyungpook National University, Daegu, 41566, Korea
Department of Forest Science, Chungbuk National University, Chungbuk, 28644, Korea
Author to whom correspondence should be addressed.
Received: 6 November 2019 / Accepted: 22 November 2019 / Published: 27 November 2019


The genus Lilium L. is widely distributed in the cold and temperate regions of the Northern Hemisphere and is one of the most valuable plant groups in the world. Regarding the classification of the genus Lilium, Comber’s sectional classification, based on the natural characteristics, has been primarily used to recognize species and circumscribe the sections within the genus. Although molecular phylogenetic approaches have been attempted using different markers to elucidate their phylogenetic relationships, there still are unresolved clades within the genus. In this study, we constructed the species tree for the genus using 28 Lilium species plastomes, including three currently determined species (L. candidum, L. formosanum, and L. leichtlinii var. maximowiczii). We also sought to verify Comber’s classification and to evaluate all loci for phylogenetic molecular markers. Based on the results, the genus was divided into two major lineages, group A and B, consisting of eastern Asia + Europe species and Hengduan Mountains + North America species, respectively. Sectional relationships revealed that the ancestor Martagon diverged from Sinomartagon species and that Pseudolirium and Leucolirion are polyphyletic. Out of all loci in that Lilium plastome, ycf1, trnF-ndhJ, and trnT-psbD regions are suggested as evaluated markers with high coincidence with the species tree. We also discussed the biogeographical diversification and long-distance dispersal event of the genus.
Lilium; phylogenomics; plastome; molecular markers; gene tree; species tree

1. Introduction

The genus Lilium L. is the type genus of Liliaceae that consists of approximately 100 species spread throughout the cold and temperate regions of the Northern Hemisphere [1,2]. Lilium species are economically important because of their ornamental features in horticulture as cut flowers and as potted and garden plants [3]. In addition, the flowers and bulbs of cultivated Lilium species are used for food and medicine [4].
The classification of the genus Lilium had been built based on flower shape before Comber’s classification [5]. As a result, the sectional (or subgeneral) boundaries in the genus frequently changed, and many species were referred to different sections according to different classification systems [6,7,8]. In contrast to the previous sectional concept based on flower shape, Comber [5] suggested the use of new natural characteristics, including characteristics of leaves, bulbs, and stems to classify the genus Lilium and divided it into seven sections. Although Comber’s classification was revised by De Jong [9] based on previous papers published up to that time, Comber’s classification has been primarily used to date to recognize Lilium species and circumscribe the sections within the phylogeny of Lilium.
Nishikawa et al. [10] constructed the phylogeny of 55 Lilium species using the internal transcribed spacer (ITS) region to clarify their phylogenetic relationships and found that most species formed a clade according to the classification based on the morphological features. Nonetheless, it is difficult to follow their results directly because most of the branch lengths were very short and not significantly supported. In addition, the phylogenies using the same marker with more samples indicated that most of the sections, except for Martagon, were polyphyletic [11,12,13]. Such discordance between classifications based on morphological characteristics and molecular phylogeny has also been found in other genera. For example, the subgenera of the genus Cymbidium based on morphological characteristics [14] was not found to be monophyletic using ITS [15] and ITS + matK [16,17], and a number of branches in the phylogenetic trees were collapsed in strict consensus trees. It is likely that gene trees based on molecular markers were often insufficient for investigating species trees because of certain evolutionary events, such as incomplete lineage sorting and horizontal gene transfer, or because convergent morphological evolution has occurred in certain lineages. To ensure the phylogenetic accuracy, increased taxa sampling has been preferred for a long time [18,19,20]. However, it has been suggested that the number of genes may be a more important determinant than the taxon number [21,22,23], and the rapid development of next-generation sequencing techniques has led us to the phylogenomic era, which provides numerous data to resolve ambiguous relationships.
Plants contain three genomes: a nuclear genome, a mitochondrial genome, and a plastid genome (plastome). The nuclear genomes [24] and mitochondrial genomes [23] in flowering plants substantially vary in length, whereas the plastomes maintain consistent lengths and typical structures for a long time [25], except for those of some specific lineages [26,27,28]. In addition, the typical plastome structure allows for next-generation sequencing (NGS) data to be assembled easily. As a result, the number of deposited plastome sequences is 10 times that of mitochondrial genome sequences in the NCBI genome database. The highly conservative nature of the plastome structure makes it possible for intergenic spaces to be used as molecular markers, as well as genes [29,30]. Using this feature, many researchers have attempted to increase phylogenetic accuracy for unresolved taxa using whole plastome sequences [31,32,33] and have also suggested new combinations of molecular markers for specific taxa [34,35,36,37,38]. Two studies on the phylogenomics of Lilium using plastome sequences have been published [4,39] to date. The phylogenetic relationships among the branches were well resolved with high support. However, taxon samplings were restricted because of the inability to use the data from both papers, which were published in the same year. Therefore, it was difficult to verify Comber’s classification based on well-resolved phylogeny. Du et al. [4] suggested molecular markers for the phylogeny of Lilium based on nucleotide diversity; however, the issues of polytomies and low supporting values still remained. Apparently, there was no comparison between species trees and gene trees to suggest molecular markers for the phylogenetic analysis. A gene tree can be easily acquired from each gene; however, it can conflict with the species tree [40] because not all genes have evolved in the same manner in the same lineage. Therefore, it is necessary to evaluate the reliability of any constructed species trees and gene trees and suggest molecular markers for the phylogeny of certain lineages.
In this study, three plastomes in Lilium were newly sequenced (1) to verify Comber’s classification and (2) to evaluate all loci of the Lilium plastome for phylogenetic molecular markers. Two taxa, Lilium formosanum Wallace and Lilium candidum L., were sampled to investigate the monophyly of Leucolirion and to determine the phylogenetic position of Liriotypus. Lilium leichtlinii var. maximowiczii (Regel) Baker was added because it was found to be closely or distantly related to Lilium lancifolium Thunb. in the previous studies [10,11,12,13]. The phylogenies of the genus Lilium that were built based on different methodologies using plastome sequences were constructed to determine a more accurate species tree. Based on this result, gene trees were compared to the species tree of the genus Lilium to evaluate which gene trees were similar to the species tree in a topology, while reflecting the generic relationships in Liliaceae. Additionally, based on the species tree generated in this study, the evolution of genomic characteristics in the genus Lilium was also discussed.

2. Results

2.1. Newly Sequenced Plastomes of Three Lilium Species

Numbers of reads ranging from 35,028,408 (L. leichtlinii var. maximowiczii) to 89,224,524 (L. candidum) were used for raw data after removing less than 50 bp reads from each dataset for the three species (Table 1). After constructing a complete plastome sequence, the number of mapped reads to complete the plastome sequences ranged from 568,878 (L. leichtlinii var. maximowiczii) to 1,945,534 (L. formosanum). Consequently, the average coverage of each plastome sequence ranged from 520.1 (L. leichtlinii var. maximowiczii) to 1840.6 (L. formosanum). The plastomes of the three Lilium species (L. candidum, L. formosanum, and L. leichtlinii var. maximowiczii) were 152,101–152,653 bp in length with a large single copy (81,481–82,101 bp), a small single copy (17,524–17,644 bp), and two inverted repeats (26,488–26,514 bp) (Table S1).
The overall GC content was 37.0%, which is similar to those of other Lilium species. In total, 133 genes were annotated from each plastome with 85 protein-coding genes, 8 rRNA genes, 38 tRNA genes, and 2 partial genes (rps19 and ycf1). InfA was a pseudogene in all three plastomes, as well as other Lilium species, and cemA was pseudogenized in the plastomes of L. candidum and L. leichtlinii var. maximowiczii because of the copy number variation of the poly A-tract.

2.2. Nucleotide Diversity within Genera Lilium and Fritillaria

The total length of the aligned sequences of Lilium and Fritillaria plastomes was 159,458 bp. The maximum nucleotide diversities of Lilium and Fritillaria were 0.030 and 0.041, respectively (Figure 1). In total, the nucleotide diversity was lower in the inverted repeat (IR) region than in the large single copy (LSC) and small single copy (SSC) regions in both genera, as well as being inversely proportional to the GC content. The delta nucleotide diversity between the two genera fluctuated from 0.014 to −0.017. Four loci higher than 0.02 in Lilium plastomes and two loci higher than 0.025 in Fritillaria plastomes were caused by large deletions in certain species because we deleted the sites with missing data for at least one species. In addition, there were the loci with higher nucleotide diversity, particularly in Lilium plastome sequences, which were caused by small palindromic repeats and the IR expansion from IRa to SSC.

2.3. Phylogeny of Lilium in Liliaceae

The 28 phylogenetic trees constructed from the four datasets, four tools, and two models showed that all of the genera in Liliaceae were monophyletic (Figure 2). Amana and Erythronium in tribe Tulipeae were distinguished from Cardiocrinum, Fritillaria, and Lilium in tribe Lilieae. In tribe Lilieae, Cardiocrinum diverged first, and Lilium and Fritillaria were separated later. In contrast to the generic relationships within the tribe, the infrageneric relationships varied slightly among phylogenetic trees.
In the clades of Lilium in 28 phylogenies, different datasets affected the change in topology more than different tools and models; however, the major clades were highly conserved in all phylogenies (Figure S1). The genus was divided into two major lineages, group A and group B, consisting of eastern Asia + Europe species and Hengduan Mountains + North America species, respectively (Figure 3). Group A consisted of three clades and five independent lineages. Clade I comprised three Martagon species and L. sp. KHK_2014, except for the phylogeny created by ASTRAL with coding genes. Clade II consisted of L. amabile, L. lancifolium, and L. callosum, which belong to Sinomartagon. L. longiflorum, and L. formosanum of Leucolirion and L. brownii of Archelirion formed clade III. L. cernuum of Sinomartagon and L. candidum of Liriotypus were sister groups to Martagon and the rest of group A, respectively. The positions of L. leichtlinii var. maximowiczii were distantly related to L. lancifolium and L. amabile, which were morphologically close to L. leichtlinii var. maximowiczii.
Group B consisted of four clades and two independent lineages. Clade IV comprised species in Sinomartagon 5c. Clade V consisted of L. leucanthum of Leucolirion and L. henryi of Sinomartagon 5a. Clade VI comprised L. duchartrei and L. fargesii of Sinomartagon, and clade VII included L. washingtonianum, L. pardalinum, and L. superbum of Pseudolirium. L. distichum of Martagon and L. philadelphicum of Pseudolirium were sisters to clade IV and clade IV + V, respectively. Most of the branches, including seven clades in groups A and B, were strongly supported by bootstrap values of the maximum likelihood analysis, branch support values of ASTRAL, and posterior probabilities of Bayesian inference, but certain branches within clades had moderate support.
A consensus tree of 28 phylogenies was constructed using the 50% majority rule to infer a robust species tree of Lilium (Figure 2). Except for two polytomies, one in Lilium and another in Amana, the seven clades in Lilium, as described above, were maintained.

2.4. Comparing Gene Trees to Species Trees

Among the 189 total gene trees, only 22 gene trees showed that each genus was monophyletic, even though two trees built using trnS-trnG IGS and the trnV intron had different generic relationships compared with the other 20 gene trees (Figures S2–S5). Thirty-six clusters of similar trees from the 190 trees (189 gene trees + the consensus tree of 28 species trees) were generated using treespace [43], with five principal components and a cut-off height of 100 (Figure 4). Among them, 21 clusters consisted of more than three trees (Table 2), and the 10th cluster included a consensus tree of the 28 species trees and 21 gene trees. These 21 gene trees were identical to the gene trees wherein each genus was monophyletic, except for the ycf2 gene tree.

2.5. IR Expansion and Contraction in Lilium

Based on the topology of Lilium species using Bayesian inference with all loci and partition models, the movements of IR boundaries (LSC-IRb, IRb-SSC, and SSC-IRa) were investigated (Figure 5). The LSC-IRb boundaries of group A were identical to each other. In contrast to group A, IR expansions and contractions were found in group B, excluding L. distichum, L. washingtonianum, L. pardalinum, and L. superbum. Two IR-SSC boundaries were more diverse than LSC-IRb, although most IR expansions and contractions occurred in the SSC-IRa boundary. Overall, IR boundaries were highly conserved within clades, except clades IV and VI.

2.6. Insertions/Deletions in Lilium

In total, eight insertions/deletions (indels) longer than 50 bp occurred in over five species that were found throughout the whole plastome sequences. Five of them had distinguishable features between groups A and B in the Lilium phylogeny (Figure S5A), but others did not correspond to the phylogenetic relationships (Figure S5B).

3. Discussion

3.1. Evolution of Lilium Plastomes in Liliaceae

The plastome sequences of Lilium, including the three newly determined species in this study, are highly conserved in terms of gene content and order and genomic structure. Although the nucleotide diversities of the LSC and SSC regions were higher than that of the IR region, this is a common phenomenon across the angiosperms [25]. However, the nucleotide diversities of Lilium and Fritillaria were higher in the LSC and SSC regions but lower in the IR region as compared to that of Paris belonging to Melanthiaceae of Liliales [44]. Consequently, the IR regions in Liliaceae appear to be more stable by purifying selection, and the LSC and SSC regions were assumed to have been under relaxed purifying selective pressure compared to those of Melanthiaceae during their evolutionary process. In Liliaceae, the fluctuating delta nucleotide diversity between Lilium and Fritillaria also supported the idea that different loci of the plastome have undergone different selective regions in this lineage, except the overestimated nucleotide diversities owing to deletions and small inversions. These findings imply that the mutational dynamics with respect to plastome loci between Lilium and Fritillaria have been processed differently even though they are closely related taxa.

3.2. Verification of Comber’s Sectional Classification

In this study, we reconstructed the phylogeny of 28 Lilium species using complete plastomes with four different datasets, four tools, and two models and compared these trees to determine the accurate species tree. There were slightly different topologies, but the major clades were coincident among the trees and were strongly supported by various branch support values (Figure S1). Before the discussion of the phylogeny of Lilium, we must evaluate the identification of plastome sequences of L. distichum (NC_029937) and L. sp. KHK_2014 (NC_027679), which were published by the same research group, because of their unreliable phylogenetic position in the genus. Du et al. [4] first raised a question regarding the phylogenetic position of L. distichum (NC_029937). The rbcL and matK sequences of L. distichum (NC_029937) were identical to those of L. speciosum (rbcL: AB034922.1; matK: AB030853, AB049526). On the contrary, atpB, rbcL, and ndhF of L. sp. KHK_2014 (NC_027679) were identical to those of L. distichum (atpB: KC796843, JX903928, KM085888; rbcL: JX903238, JN786059, JN417422, KP711933; ndhF: JX903509, KM085762). Consequently, it may be necessary to exclude these two plastomes or to consider L. sp. KHK_2014 as L. distichum to avoid improper conclusions for the phylogenomics of Lilium.
All phylogenetic trees using different datasets indicated that Lilium species could be divided into two major groups. These two groups were also distinguished by the mutational dynamics of the IR expansion/contraction and large indels (Figure 5 and Figure S5A).

3.2.1. The Phylogenetic Position of Martagon

Martagon consists of five species that are primarily distributed in northeastern Asia and Russia, except L. martagon, which ranges widely from central Europe to eastern Siberia. This section had been considered an early-diverging lineage in Lilium based on the morphological characteristics: hypogeal and delayed germination, whorled leaves, jointed scales, and heavy seeds [5,45]. In contrast to the morphological analyses, molecular phylogenetic analyses using ITS sequences showed that this section is a more recently derived lineage and is sister to some of the Sinomartagon + Leucolirion [10,11,13]. Based on these subgeneric relationships, Gao et al. [41] suggested that the ancestor of Martagon, Sinomartagon, and Leucolirion 6b had a distribution within the Hengduan Mountains before Martagon separated from Sinomartagon + Leucolirion 6b approximately 8.8 million years ago. However, based on the plastome sequences, Martagon is not a sister to Sinomartagon + Leucolirion 6b but forms a clade within Sinomartagon I (Figure 5). Additionally, the plastome structures of Lilium provide further support. The junctions between SCs and IRs of Martagon are identical to those of Sinomartagon I, whereas they differ from those of Sinomartagon 5c and Leucolirion 6b (Figure 5). These results imply that the ancestor of Martagon diverged from Sinomartagon species, i.e., the ancestor of L. cernuum in this study, and the divergence time was more recent than the expectation of Gao et al. [41]. In addition, four species within Sinomartagon I and three species within Martagon are commonly distributed in eastern Asia. As a result, the hypothesis that the origin of Martagon is the Hengduan Mountains [41] is controvertible.

3.2.2. The Polyphyly of Pseudolirium

Pseudolirium consists of all American lilies, including L. philadelphicum, which is a lectotype of the section [5]. Interestingly, when the data matrix involved L. philadelphicum, the phylogeny using ITS [10,11,13] showed the section was monophyletic, but using matK [46] revealed polyphyly for the section. Unfortunately, the phylogenetic position of the section using both markers was not resolved or was supported weekly. On the other hand, Kim et al. [47] suggested that L. philadelphicum seems to be distinguishable from other species in the section based on phylogenomics using plastome sequences. This relationship is well supported by the 28 species trees constructed in this study. In terms of DNA sequence mutations, a two base deletion in ITS was found specifically in Pseudolirium species, except for L. philadelphicum [10], and there was an obvious distinction between the IR-SC junctions of L. philadelphicum and other Pseudolirium species (Figure 5). Morphologically, subsection 2d, consisting of L. philadelphicum and L. catesbaei in Pseudolirium, is distinguished from other species by erect flowers and highly clawed perianth parts [3]. Consequently, a new circumscription of Pseudolirium should be considered to reflect the recent phylogenetic results.

3.2.3. The Polyphyly of Leucolirion

Leucolirion consists of eight species with scattered and sessile leaves and trumpet flowers [1]. The section is subdivided into two subsections based on bulb color: dark purple or brown for 6a and white for 6b [1]. In this study, the two subsections were distantly separated with strong support and this result was congruent with the previous phylogenetic studies [13,46]. In addition, L. henryi of Sinomartagon 5a and L. brownii of Archelirion formed a robust clade with Leucolirion 6a and 6b, respectively (Figure S1, Figure 5). Based on the phylogenetic and cytological studies, Du et al. [13] suggested that L. henryi and L. brownii should be classified into Leucolirion 6a and 6b, respectively. Consequently, our results provide further support for the modification of Leucolirion according to Du et al. [13].

3.2.4. The Position of Liriotypus in the Genus Lilium

Liriotypus comprises 20 species, including all European, Turkish, and Caucasian species, with the exception of Lilium martagon [5,48]. Among them, L. bulbiferum, having upright flowers, has been distinguished from the rest of the species within the section and forms a clade with the Sinomartagon species, including L. dauricum of Daurolirion based on the molecular phylogenetic analyses [13,48]. In this study, L. bulbiferum was placed far away from L. candidum, which is a lectotype of the section [5], agreeing with the results of previous studies. However, it forms a clade with Martagon + Sinomartagon I with strong support, although there are sampling gaps that prevent a concrete conclusion (Figure S1). Therefore, increased taxa sampling, particularly the members of Daurolirion, will help to resolve the discordance of the phylogenetic position of L. bulbiferum between this study and previous studies and offer the correct phylogenetic position for this species.
On the other hand, the phylogenetic position of Liriotypus, except L. bulbiferum, was irregular within the genus, although they formed a clade with the Sinomartagon 5c species-Nomocharis clade in previous studies [11,13]. On the contrary, L. candidum was an early-diverging taxon in group A, without alternative relationships with the rest of group A species in this study (Figure 3, Figure S1). Consequently, the newly suggested phylogenetic position of Liriotypus in this study is incongruent with that of previous phylogenetic studies using a few molecular markers. One possible explanation for this discordance is that the position of L. candidum does not represent Liriotypus in spite of its taxonomic importance by lectotype in the subsection. This species differs from other Liriotypus species based on rosette basal leaves and widely trumpet-shaped flowers [48,49], and the geographic circumscription of the sections in Lilium by Comber [5] collided with the molecular phylogeny of this study, i.e., Pseudolirium and Liriotypus. Another scenario is that the origin of Liriotypus came from early-diverging Sinomartagon (see additional details in the next section) and directly moved to Europe via the Caucasus during evolution.
In spite of low taxon sampling in Liriotypus, our result strongly supports the previous conclusions in which L. bulbiferum is separated from the rest of the species [48], and it suggests a new phylogenetic position of the section within the robust phylogenetic trees. However, because there still remains uncertainty as to whether the position of L. candidum belongs to the Liriotypus based on its unique morphological characteristics in the section, more sampling in the section will be needed to solve its position.

3.2.5. Biogeographic History of the Genus Lilium

The geographical origin of the tribe Lilieae has been speculated to be the Himalayas + Hengduan Mountains and multiple intercontinental dispersal events for Lilieae were suggested, as well as Lilium [41,50]. In this study, we also found long distant dispersals and simpler than the previous speculation. Gao et al. [41] suggested a long-distance dispersal model of Lilium based on ITS and matK phylogenies, in which there were movements among four regions: Hengduan Mountains, eastern Asia, Europe, including Caucasus, and Northern America. However, from the results in this study, it was suggested that there were two main long dispersals in eastern Asia—Europe in group A and Hengduan Mountains—North America in group B (Figure 3).
To explain these distributions, it was supposed that Lilium comprised “Sinomartagon and its derivatives.” Sinomartagon comprises approximately 30 species [11] with epigeal and immediate germination (except L. henryi), scattered leaves, an entire bulb, and Turk’s cap flowers [5], and it has a distribution from the Hengduan Mountains to eastern Asia. In molecular phylogenetic trees, including those in this study, this section has polyphyly regardless of the types of marker or taxon samplings [12,13,41,46]. Therefore, if the manner for distant dispersals was paved during glacial periods or pollinators traveled a great distance at the beginning of the Lilium diversification, certain populations of different species in the Sinomartagon could have moved together. Therefore, the adaptations to new circumstances may have led to new populations that morphologically converged. Consequently, populations within new circumstances had similar morphological characteristics and belonged to the same section by morphological classification. This may cause the discordance between classifications based on morphological characteristics and molecular phylogeny. Regarding the basal lineages of group A and B, the “Sinomartagon and its derivatives” hypothesis is unclear because of our low taxon sampling or extinction of ancient Sinomartagon species. This hypothesis may further confuse the evolutionary history of the genus because (1) nobody has placed Sinomartagon as a basal lineage in the Lilium based on the morphological characteristics, and (2) most phylogenies of Lilium constructed using ITS are not consistent with the present results. In addition, inheritance from single parents, such as a plastome, makes it difficult to detect hybridization events. However, the phylogenetic tree generated in this study was the first robust phylogenetic tree with more than 20 samples. There is no doubt that the most important action for the discussion of phylogenetic relationships or divergence times of certain lineage is the construction of accurate species trees without polytomy and poor support. Therefore, we cannot rule out this “Sinomartagon and its derivatives” hypothesis based on the well-resolved phylogenetic tree.

3.3. Molecular Markers for the Phylogeny of Lilium and its Relatives

ITS has been used as a valuable molecular marker for the phylogeny of Lilium [10,11,13,41] but the phylogenetic relationships among sections or species were incongruent or were weakly supported. To overcome this problem, we constructed the phylogeny of Lilium using whole plastome sequences in this study. However, the production and manipulation of NGS data also require significant time and cost, as well as a higher-level technique than the Sanger sequencing method.
To evaluate which gene trees were similar to the species tree in terms of topology, while reflecting the generic relationship in Liliaceae, 189 gene trees were constructed, and only 20 were found to be consistent with the results presented in previous phylogenetic studies [51,52]. Among them, the ycf1, trnF-ndhJ, and trnT-psbD regions were coincident with highly variable regions of the Lilium plastome, as suggested by Du et al. [4].
This could serve as an alternative choice of NGS-based phylogenomics in Lilium when we use insufficient conditioned samples, such as very low concentration, fragmentation, or extraction from very old specimens that are difficult for use in the preparation of an NGS library.

4. Materials and Methods

4.1. Plant Materials and DNA Extraction

The bulbs of L. leichtlinii var. maximowiczii (Wooriseed, Korea) and the seeds of L. formosanum (Wageningen University, Netherlands) and L. candidum (Royal horticultural society Lily group, UK) were germinated on media at Kyungpook National University of Korea. Genomic DNA was extracted from young fresh leaves using a DNeasy plant mini kit (Qiagen, Valencia, CA, USA), following the manufacturer’s protocol.

4.2. Sequencing, Assembly, and Annotation

Genomic DNA was sequenced using the HiSeq 2500 instrument (Illumina, San Diego, CA, USA). The following assembling procedures were implemented using Geneious 10.2.5 [53]. Both ends of raw reads were trimmed with more than a 1% chance of an error per base. Reads exceeding 50 bp in length were extracted and used as raw reads after this step. Raw reads were mapped to the plastome sequence of Lilium pardalinum [47] with medium-low sensitivity. Reads were aligned to the reference then de novo assembled with zero mismatches and gaps among the reads to generate contigs. Raw reads were realigned to the contigs with zero mismatches and gaps among the reads for up to 100 iterations. The generated contigs were concatenated into a circular form using de novo assembled circularizing contigs with matching ends. Finally, the raw reads were mapped to the complete plastome sequence with zero mismatches and gaps among the reads to verify the coverage depths through the genome, because of the fact that many plastome-like sequences distributed in the mitochondrial genome and nuclear genome have relatively low coverage depths compared to that of the plastome.
All of the genes in the three plastome sequences were annotated and compared with those of L. pardalinum using Geneious annotation with 90% similarity, then re-checked separately using BLASTP [54] and tRNAscan-SE [55].

4.3. Sequence Diversity and GC Content Analyses

Twenty Fritillaria and 28 Lilium plastome sequences were aligned by MAFFT [56], then the AT-rich regions were realigned by MUSCLE [57] to increase the alignment accuracy at these regions. The alignment sequences were loaded in R ver. 3.5.1 [58], and the nucleotide diversities of the two genera were analyzed using sliding window analysis (window size = 600 bp, step size = 200 bp) by deleting the sites including at least one missing data point for all sequences. In addition, the GC content was also calculated to compare the relationship between nucleotide diversity and GC content.

4.4. Phylogenetic Analysis

Fifty-five plastome sequences from the genera Lilium, Fritillaria, Cardiocrinum, Amana, and Erythronium were downloaded from GenBank to construct a phylogeny for Lilium. We extracted 78 coding sequences (CDSs), 90 intergenic spaces (IGSs) longer than 100 bp, and 21 intron regions (two regions from clpP, trnK, and ycf3) from each plastome (Table S2).
All of the extracted sequences were aligned using MAFFT [55] according to the region (Table S3) and merged into four datasets as follows: All_loci (including CDSs, IGSs, and introns), CDS_loci, IGS_loci, and Intron_loci. The models for each partition for the four datasets were estimated using PartitionFinder 2 [59], ModelFinder [60], or Jmodeltest 2 [61] for different phylogenetic analyses. The phylogenetic trees were constructed using the RAxML Black Box with 1000 bootstraps [62] in the CIPRES gateway [63], MrBayes with ngen = 10,000,000, samplefreq = 1000, and burninfrac = 0.25 [64], or IQ-TREE with 1000 bootstraps [65]. In total, 189 gene trees were constructed using IQ-TREE with ModelFinder, and these were used to estimate the species trees by ASTRAL [66,67,68]. All of the details for phylogenies are summarized in Table S2. A consensus tree of 28 phylogenies was constructed by majority rule.

4.5. Comparison of Gene Trees

The generic relationships of each gene tree were compared to those presented in previous phylogenetic studies of Liliaceae [51,52]. Clusters of similar gene trees were identified using the Kendall Colijn metric [69] in treespace [43] with five principal components and a cut-off distance of 100. Phylogeny using selected markers was constructed by IQ-TREE, with 1000 bootstraps for comparison to the consensus tree constructed by 28 phylogenies.

4.6. R Packages for Manipulation of Phylogenies

APE [70], Biostrings [71], dplyr [72], ggplot2 [73], ggtree [74,75], gridExtra [76], pegas [77], phytools [78], tidytree [79], and treespace [43] were used in this study.

Supplementary Materials

The following are available online at, Figure S1: A clade of Lilium species extracted from 28 species trees, Figure S2: Gene trees constructed using 78 coding regions. Coloured line refers to each genus, Figure S3: Gene trees constructed using 90 intergenic spacers. Coloured line refers to each genus, Figure S4: Gene trees were constructed using 21 introns. Coloured line refers to each genus, Figure S5: Large indertions/deletions occurring at least five species with longer than 50 bp. Red line distinguishes between two groups, Table S1: Summary of three plastome sequences, Table S2: List of estimated species trees according to different options, Table S3: 189 loci used for phylogenetic analyses in this paper.

Author Contributions

Performed research design, writing of the manuscript, lab work, and data analysis, H.T.K.; collected samples and edited the manuscript, K.-B.L.; contributed to the writing and editing of the manuscript, J.S.K.


This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Project No. NRF-2017R1D1A1A02018573/2016R1D1A1B04932913).


We thank Ji-Hyang Park for growing the plants from seeds and bulbs.

Conflicts of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.


  1. Haw, S.G.; Liang, S.-Y. The Lilies of China: The Genera Lilium, Cardiocrinum, Nomocharis and Notholirion; Timber Press: Portland, OR, USA, 1986. [Google Scholar]
  2. McRae, E.A.; Austin-Mcrae, E.; MacRae, E. Lilies: A Guide for Growers and Collectors; Timber Press: Portland, OR, USA, 1998; Volume 105. [Google Scholar]
  3. Lim, K.-B.; van Tuyl, J.M.L. Flower Breeding and Genetics; Springer: Berlin, Germany, 2007; pp. 517–537. [Google Scholar]
  4. Du, Y.-P.; Bi, Y.; Yang, F.-P.; Zhang, M.-F.; Chen, X.-Q.; Xue, J.; Zhang, X.-H. Complete chloroplast genome sequences of Lilium: Insights into evolutionary dynamics and phylogenetic analyses. Sci. Rep. 2017, 7, 5751. [Google Scholar] [CrossRef] [PubMed]
  5. Comber, H.F. A New Classification of the Genus Lilium; Royal Horticultural Society: London, UK, 1949; Volume 13, pp. 85–105. [Google Scholar]
  6. Endlicher, S. Genera plantarum. Vindobonae. Apud Fr. Beck Universitatis Bibliopolam 1836. Available online: (accessed on 25 November 2019).
  7. Reichenbach, H. Flora Germanica Excursoria.–Leipzig. 1830. Available online: (accessed on 25 November 2019).
  8. Baker, J. The Gardeners’ Chronicle and and agricultural gazette. 1871. Available online: (accessed on 27 November 2019).
  9. De Jong, P. Some notes on the evolution of lilies. Lily Yearb. N. Am. Lily Soc. 1974, 27, 23–28. [Google Scholar]
  10. Nishikawa, T.; Okazaki, K.; Uchino, T.; Arakawa, K.; Nagamine, T. A molecular phylogeny of Lilium in the internal transcribed spacer region of nuclear ribosomal DNA. J. Mol. Evol. 1999, 49, 238–249. [Google Scholar] [CrossRef] [PubMed]
  11. Nishikawa, T.; Okazaki, K.; Arakawa, K.; Nagamine, T. Phylogenetic analysis of section Sinomartagon in genus Lilium using sequences of the internal transcribed spacer region in nuclear ribosomal DNA. Breed. Sci. 2001, 51, 39–46. [Google Scholar] [CrossRef]
  12. Lee, C.S.; Kim, S.-C.; Yeau, S.H.; Lee, N.S. Major lineages of the genus Lilium (Liliaceae) based on nrDNA ITS sequences, with special emphasis on the Korean species. J. Plant Biol. 2011, 54, 159–171. [Google Scholar] [CrossRef]
  13. Du, Y.P.; He, H.B.; Wang, Z.X.; Li, S.; Wei, C.; Yuan, X.N.; Cui, Q.; Jia, G.X. Molecular phylogeny and genetic variation in the genus Lilium native to China based on the internal transcribed spacer sequences of nuclear ribosomal DNA. J. Plant Res. 2014, 127, 249–263. [Google Scholar] [CrossRef]
  14. Du Puy, D.; Cribb, P. The Genus Cymbidium; Royal Botanic Gardens: Sydney, Australia, 2007. [Google Scholar]
  15. Sharma, S.K.; Dkhar, J.; Kumaria, S.; Tandon, P.; Rao, S.R. Assessment of phylogenetic inter-relationships in the genus Cymbidium (Orchidaceae) based on internal transcribed spacer region of rDNA. Gene 2012, 495, 10–15. [Google Scholar] [CrossRef]
  16. van den Berg, C.; Ryan, A.; Cribb, P.J.; Chase, M.W. Molecular phylogenetics of Cymbidium (Orchidaceae: Maxillarieae): Sequence data from internal transcribed spacers (ITS) of nuclear ribosomal DNA and plastid matK. Lindleyana 2002, 17, 102–111. [Google Scholar]
  17. Yukawa, T.; Miyoshi, K.; Yokoyama, J. Molecular phylogeny and character evolution of Cymbidium (Orchidaceae). Bull. Natl. Sci. Mus. Tokyo B 2002, 28, 129–139. [Google Scholar]
  18. Zwickl, D.J.; Hillis, D.M. Increased taxon sampling greatly reduces phylogenetic error. Syst. Biol. 2002, 51, 588–598. [Google Scholar] [CrossRef]
  19. Graybeal, A. Is it better to add taxa or characters to a difficult phylogenetic problem? Syst. Biol. 1998, 47, 9–17. [Google Scholar] [CrossRef] [PubMed]
  20. Heath, T.A.; Hedtke, S.M.; Hillis, D.M. Taxon sampling and the accuracy of phylogenetic analyses. J. Syst. Evol. 2008, 46, 239–257. [Google Scholar] [CrossRef]
  21. Rokas, A.; Carroll, S.B. More genes or more taxa? The relative contribution of gene number and taxon number to phylogenetic accuracy. Mol. Biol. Evol. 2005, 22, 1337–1344. [Google Scholar] [CrossRef]
  22. Rosenberg, M.S.; Kumar, S. Incomplete taxon sampling is not a problem for phylogenetic inference. Proc. Natl. Acad. Sci. USA 2001, 98, 10751–10756. [Google Scholar] [CrossRef] [PubMed]
  23. Mower, J.P.; Sloan, D.B.; Alverson, A.J. Plant mitochondrial genome diversity: The genomics revolution. In Plant Genome Diversity Volume 1; Springer: Berlin, Germany, 2012; pp. 123–144. [Google Scholar]
  24. Gregory, T.R.; Nicol, J.A.; Tamm, H.; Kullman, B.; Kullman, K.; Leitch, I.J.; Murray, B.G.; Kapraun, D.F.; Greilhuber, J.; Bennett, M.D. Eukaryotic genome size databases. Nucleic Acids Res. 2006, 35, D332–D338. [Google Scholar] [CrossRef] [PubMed]
  25. Ruhlman, T.A.; Jansen, R.K. The plastid genomes of flowering plants. Methods Mol. Biol. 2014, 1132, 3–38. [Google Scholar] [CrossRef]
  26. Weng, M.L.; Blazier, J.C.; Govindu, M.; Jansen, R.K. Reconstruction of the ancestral plastid genome in Geraniaceae reveals a correlation between genome rearrangements, repeats, and nucleotide substitution rates. Mol. Biol. Evol. 2014, 31, 645–659. [Google Scholar] [CrossRef]
  27. Guisinger, M.M.; Kuehl, J.V.; Boore, J.L.; Jansen, R.K. Extreme reconfiguration of plastid genomes in the angiosperm family Geraniaceae: Rearrangements, repeats, and codon usage. Mol. Biol. Evol. 2011, 28, 583–600. [Google Scholar] [CrossRef]
  28. Westwood, J.H.; Yoder, J.I.; Timko, M.P.; dePamphilis, C.W. The evolution of parasitism in plants. Trends Plant Sci. 2010, 15, 227–235. [Google Scholar] [CrossRef]
  29. Shaw, J.; Lickey, E.B.; Schilling, E.E.; Small, R.L. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: The tortoise and the hare III. Am. J. Bot. 2007, 94, 275–288. [Google Scholar] [CrossRef]
  30. Shaw, J.; Lickey, E.B.; Beck, J.T.; Farmer, S.B.; Liu, W.; Miller, J.; Siripun, K.C.; Winder, C.T.; Schilling, E.E.; Small, R.L. The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Am. J. Bot. 2005, 92, 142–166. [Google Scholar] [CrossRef] [PubMed]
  31. Yang, J.B.; Tang, M.; Li, H.T.; Zhang, Z.R.; Li, D.Z. Complete chloroplast genome of the genus Cymbidium: Lights into the species identification, phylogenetic implications and population genetic analyses. BMC Evol. Biol. 2013, 13, 84. [Google Scholar] [CrossRef]
  32. Zhang, Y.J.; Ma, P.F.; Li, D.Z. High-throughput sequencing of six bamboo chloroplast genomes: Phylogenetic implications for temperate woody bamboos (Poaceae: Bambusoideae). PLoS ONE 2011, 6, e20596. [Google Scholar] [CrossRef] [PubMed]
  33. Wysocki, W.P.; Clark, L.G.; Attigala, L.; Ruiz-Sanchez, E.; Duvall, M.R. Evolution of the bamboos (Bambusoideae; Poaceae): A full plastome phylogenomic analysis. BMC Evol. Biol. 2015, 15, 50. [Google Scholar] [CrossRef] [PubMed]
  34. Dong, W.; Liu, J.; Yu, J.; Wang, L.; Zhou, S. Highly variable chloroplast markers for evaluating plant phylogeny at low taxonomic levels and for DNA barcoding. PLoS ONE 2012, 7, e35071. [Google Scholar] [CrossRef] [PubMed]
  35. Sarkinen, T.; George, M. Predicting plastid marker variation: Can complete plastid genomes from closely related species help? PLoS ONE 2013, 8, e82266. [Google Scholar] [CrossRef] [PubMed]
  36. Ku, C.; Hu, J.M.; Kuo, C.H. Complete plastid genome sequence of the basal asterid Ardisia polysticta Miq. and comparative analyses of asterid plastid genomes. PLoS ONE 2013, 8, e62548. [Google Scholar] [CrossRef] [PubMed]
  37. Ku, C.; Chung, W.C.; Chen, L.L.; Kuo, C.H. The complete plastid genome sequence of madagascar periwinkle Catharanthus roseus (L.) G. Don: Plastid genome evolution, molecular marker identification, and phylogenetic implications in asterids. PLoS ONE 2013, 8, e68518. [Google Scholar] [CrossRef]
  38. Kim, H.T.; Kim, J.S.; Lee, Y.M.; Mun, J.-H.; Kim, J.-H. Molecular markers for phylogenetic applications derived from comparative plastome analysis of Prunus. J. Syst. Evol. 2019, 57, 15–22. [Google Scholar] [CrossRef]
  39. Kim, J.-H.; Lee, S.-I.; Kim, B.-R.; Choi, I.-Y.; Ryser, P.; Kim, N.-S. Chloroplast genomes of Lilium lancifolium, L. amabile, L. callosum, and L. philadelphicum: Molecular characterization and their use in phylogenetic analysis in the genus Lilium and other allied genera in the order Liliales. PLoS ONE 2017, 12, e0186788. [Google Scholar] [CrossRef]
  40. Maddison, W.P. Gene trees in species trees. Syst. Biol. 1997, 46, 523–536. [Google Scholar] [CrossRef]
  41. Gao, Y.D.; Harris, A.J.; Zhou, S.D.; He, X.J. Evolutionary events in Lilium (including Nomocharis, Liliaceae) are temporally correlated with orogenies of the Q-T plateau and the Hengduan Mountains. Mol. Phylogenet. Evol. 2013, 68, 443–460. [Google Scholar] [CrossRef] [PubMed]
  42. Yu, G. A tidy tool for phylogenetic tree data manipulation [R package tidytree version 0.1. 9]. Compr. R Arch. Netw. 2019. Available online: (accessed on 25 November 2019).
  43. Jombart, T.; Kendall, M.; Almagro-Garcia, J.; Colijn, C. Treespace: Statistical exploration of landscapes of phylogenetic trees. Mol. Ecol. Resour. 2017, 17, 1385–1392. [Google Scholar] [CrossRef] [PubMed]
  44. Song, Y.; Wang, S.; Ding, Y.; Xu, J.; Li, M.F.; Zhu, S.; Chen, N. Chloroplast genomic resource of Paris for species discrimination. Sci. Rep. 2017, 7, 3427. [Google Scholar] [CrossRef] [PubMed]
  45. Lighty, R. The lilies of Korea. Lily Yearb. RHS 1969, 31, 31–39. [Google Scholar]
  46. Hayashi, K.; Kawano, S. Molecular systematics of Lilium and allied genera (Liliaceae): Phylogenetic relationships among Lilium and related genera based on the rbcL and matK gene sequence data. Plant Species Biol. 2000, 15, 73–93. [Google Scholar] [CrossRef]
  47. Kim, H.T.; Zale, P.J.; Lim, K.-B. Complete plastome sequence of Lilium pardalinum Kellogg (Liliaceae). Mitochondrial DNA Part B 2018, 3, 478–479. [Google Scholar] [CrossRef]
  48. İkinci, N.; Oberprieler, C.; Güner, A. On the origin of European lilies: Phylogenetic analysis of Lilium section Liriotypus (Liliaceae) using sequences of the nuclear ribosomal transcribed spacers. Willdenowia 2006, 36, 647–656. [Google Scholar] [CrossRef]
  49. Muratović, E.; Hidalgo, O.; Garnatje, T.; Siljak-Yakovlev, S. Molecular phylogeny and genome size in European lilies (Genus Lilium, Liliaceae). Adv. Sci. Lett. 2010, 3, 180–189. [Google Scholar] [CrossRef]
  50. Patterson, T.B.; Givnish, T.J. Phylogeny, concerted convergence, and phylogenetic niche conservatism in the core Liliales: Insights from rbcL and ndhF sequence data. Evolution 2002, 56, 233–252. [Google Scholar] [CrossRef]
  51. Rønsted, N.; Law, S.; Thornton, H.; Fay, M.F.; Chase, M.W. Molecular phylogenetic evidence for the monophyly of Fritillaria and Lilium (Liliaceae; Liliales) and the infrageneric classification of Fritillaria. Mol. Phylogenet. Evol. 2005, 35, 509–527. [Google Scholar] [CrossRef] [PubMed]
  52. Fay, M.F.; Chase, M.W.; Rønsted, N.; Devey, D.S.; Pillon, Y.; Pires, J.C.; Peterson, G.; Seberg, O.; Davis, J.I. Phylogenetics of Lilliales. Aliso J. Syst. Evol. Bot. 2006, 22, 559–565. [Google Scholar] [CrossRef]
  53. Kearse, M.; Moir, R.; Wilson, A.; Stones-Havas, S.; Cheung, M.; Sturrock, S.; Buxton, S.; Cooper, A.; Markowitz, S.; Duran, C.; et al. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 2012, 28, 1647–1649. [Google Scholar] [CrossRef] [PubMed]
  54. Altschul, S.F.; Gish, W.; Miller, W.; Myers, E.W.; Lipman, D.J. Basic local alignment search tool. J. Mol. Biol. 1990, 215, 403–410. [Google Scholar] [CrossRef]
  55. Lowe, T.M.; Eddy, S.R. tRNAscan-SE: A program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 1997, 25, 955–964. [Google Scholar] [CrossRef]
  56. Katoh, K.; Misawa, K.; Kuma, K.; Miyata, T. MAFFT: A novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002, 30, 3059–3066. [Google Scholar] [CrossRef]
  57. Edgar, R.C. MUSCLE: A multiple sequence alignment method with reduced time and space complexity. BMC Bioinform. 2004, 5, 113. [Google Scholar] [CrossRef]
  58. Team, R.C. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2018. [Google Scholar]
  59. Lanfear, R.; Frandsen, P.B.; Wright, A.M.; Senfeld, T.; Calcott, B. PartitionFinder 2: New methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol. Biol. Evol. 2016, 34, 772–773. [Google Scholar] [CrossRef]
  60. Kalyaanamoorthy, S.; Minh, B.Q.; Wong, T.K.; von Haeseler, A.; Jermiin, L.S. ModelFinder: Fast model selection for accurate phylogenetic estimates. Nat. Methods 2017, 14, 587. [Google Scholar] [CrossRef]
  61. Darriba, D.; Taboada, G.L.; Doallo, R.; Posada, D. jModelTest 2: More models, new heuristics and parallel computing. Nat. Methods 2012, 9, 772. [Google Scholar] [CrossRef]
  62. Stamatakis, A. RAxML version 8: A tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014, 30, 1312–1313. [Google Scholar] [CrossRef] [PubMed]
  63. Miller, M.A.; Pfeiffer, W.; Schwartz, T. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In Proceedings of the 2010 Gateway Computing Environments Workshop (GCE), New Orleans, LA, USA, 14 November 2010; pp. 1–8. [Google Scholar]
  64. Ronquist, F.; Teslenko, M.; van der Mark, P.; Ayres, D.L.; Darling, A.; Hohna, S.; Larget, B.; Liu, L.; Suchard, M.A.; Huelsenbeck, J.P. MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Syst. Biol. 2012, 61, 539–542. [Google Scholar] [CrossRef]
  65. Nguyen, L.-T.; Schmidt, H.A.; von Haeseler, A.; Minh, B.Q. IQ-TREE: A fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 2014, 32, 268–274. [Google Scholar] [CrossRef]
  66. Zhang, C.; Rabiee, M.; Sayyari, E.; Mirarab, S. ASTRAL-III: Polynomial time species tree reconstruction from partially resolved gene trees. BMC Bioinform. 2018, 19, 153. [Google Scholar] [CrossRef]
  67. Mirarab, S.; Warnow, T. ASTRAL-II: Coalescent-based species tree estimation with many hundreds of taxa and thousands of genes. Bioinformatics 2015, 31, i44–i52. [Google Scholar] [CrossRef]
  68. Mirarab, S.; Reaz, R.; Bayzid, M.S.; Zimmermann, T.; Swenson, M.S.; Warnow, T. ASTRAL: Genome-scale coalescent-based species tree estimation. Bioinformatics 2014, 30, i541–i548. [Google Scholar] [CrossRef]
  69. Kendall, M.; Colijn, C. A tree metric using structure and length to capture distinct phylogenetic signals. arXiv 2015, arXiv:preprint/1507.05211. [Google Scholar]
  70. Paradis, E.; Claude, J.; Strimmer, K. APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 2004, 20, 289–290. [Google Scholar] [CrossRef]
  71. Pagès, H.; Aboyoun, P.; Gentleman, R.; DebRoy, S. Biostrings: Efficient manipulation of biological strings. R Package Version 2018, 2. Available online: (accessed on 25 November 2019). [Google Scholar]
  72. Wickham, H.; Francois, R.; Henry, L.; Müller, K. Dplyr: A grammar of data manipulation. R Package Version 2018, 3. Available online: (accessed on 25 November 2019). [Google Scholar]
  73. Wickham, H. Ggplot2: Elegant Graphics for Data Analysis; Springer: Berlin, Germany, 2016. [Google Scholar]
  74. Yu, G.; Smith, D.K.; Zhu, H.; Guan, Y.; Lam, T.T.Y. Ggtree: An R package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol. Evol. 2017, 8, 28–36. [Google Scholar] [CrossRef]
  75. Yu, G.; Lam, T.T.-Y.; Zhu, H.; Guan, Y. Two methods for mapping and visualizing associated data on phylogeny using ggtree. Mol. Biol. Evol. 2018, 35, 3041–3043. [Google Scholar] [CrossRef]
  76. Auguie, B. GridExtra: Functions in Grid graphics. R Package Version 0.9 2012, 1. Available online: (accessed on 25 November 2019). [Google Scholar]
  77. Paradis, E. Pegas: An R package for population genetics with an integrated–modular approach. Bioinformatics 2010, 26, 419–420. [Google Scholar] [CrossRef]
  78. Revell, L.J. Phytools: An R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 2012, 3, 217–223. [Google Scholar] [CrossRef]
  79. Xinqi, C.; Songyun, L.; Jiemei, X.; Tamura, M. Liliaceae. Flora China 2000, 24, 73–263. [Google Scholar]
Figure 1. Nucleotide diversity (π) and GC content throughout the plastome sequence according to sliding window analysis (window size = 600 bp, step size = 200 bp). The red line and blue dashed line refer to π and GC content, respectively. The vertical dashed grey lines refer to the approximate boundaries of the plastome structure.
Figure 1. Nucleotide diversity (π) and GC content throughout the plastome sequence according to sliding window analysis (window size = 600 bp, step size = 200 bp). The red line and blue dashed line refer to π and GC content, respectively. The vertical dashed grey lines refer to the approximate boundaries of the plastome structure.
Plants 08 00547 g001
Figure 2. Phylogenetic trees based on four tools (ASTRAL, IQ-TREE, RAxML, and MrBayes) using four datasets (genes, introns, intergenic spacers, and all regions) and two different models (partition model and non-partition model of the dataset). The colored line refers to each genus.
Figure 2. Phylogenetic trees based on four tools (ASTRAL, IQ-TREE, RAxML, and MrBayes) using four datasets (genes, introns, intergenic spacers, and all regions) and two different models (partition model and non-partition model of the dataset). The colored line refers to each genus.
Plants 08 00547 g002
Figure 3. Consensus tree of 28 phylogenies based on four tools (ASTRAL, IQ-TREE, RAxML, and MrBayes) using four datasets (genes, introns, intergenic spacers, and all regions) and two different models (partition model and non-partition model of the dataset). The colored line refers to each genus. The distribution areas of clades are based on Gao et al. [41] and Xinqi et al. [42].
Figure 3. Consensus tree of 28 phylogenies based on four tools (ASTRAL, IQ-TREE, RAxML, and MrBayes) using four datasets (genes, introns, intergenic spacers, and all regions) and two different models (partition model and non-partition model of the dataset). The colored line refers to each genus. The distribution areas of clades are based on Gao et al. [41] and Xinqi et al. [42].
Plants 08 00547 g003
Figure 4. Hierarchical cluster analysis for 190 trees, including 189 gene trees and a consensus tree. The horizontal dashed line refers to the cut-off value for cluster trees.
Figure 4. Hierarchical cluster analysis for 190 trees, including 189 gene trees and a consensus tree. The horizontal dashed line refers to the cut-off value for cluster trees.
Plants 08 00547 g004
Figure 5. Inverted repeat (IR) expansion and contraction in Lilium. Pale grey refers to bases and colors on the bases represent disagreements with the consensus sequence. The red, blue, brown, and green blocks below bases stand for the large single copy (LSC) region, IRb region, small single copy (SSC) region, and IRa region, respectively. The tree on the left is constructed using Bayesian inference with all loci and partition models.
Figure 5. Inverted repeat (IR) expansion and contraction in Lilium. Pale grey refers to bases and colors on the bases represent disagreements with the consensus sequence. The red, blue, brown, and green blocks below bases stand for the large single copy (LSC) region, IRb region, small single copy (SSC) region, and IRa region, respectively. The tree on the left is constructed using Bayesian inference with all loci and partition models.
Plants 08 00547 g005
Table 1. Summary of genome assembly.
Table 1. Summary of genome assembly.
TaxonNo. of Raw Reads (≥50 bp)No. of Mapped ReadsAverage CoverageSRA a Accession
Lilium candidum89,224,5241,187,0281134.7SRR7617960, SRR7617961
Lilium leichtlinii var. maximowiczii35,028,408568,878520.1SRR7617965
Lilium formosanum80,985,6061,945,5341840.6SRR7617963, SRR7617964
a Sequence Read Archive.
Table 2. Information on 21 clustered trees using treespace, with five principal components and 100 cut-off distance.
Table 2. Information on 21 clustered trees using treespace, with five principal components and 100 cut-off distance.
ClusterClustered TreesNo. of Trees
1accD; ndhG; psbB; rpl20; rpl32; rpl33; rps19; accD_psaI; psaJ_rpl33; trnS_trnG; ycf4_cemA; trnK_intron212
2atpA; ccsA; ndhA; ndhH; atpF_atpH; petA_psbJ; rpoB_trnC; rps15_ycf1; rps16_trnQ; trnE_trnT; trnT_trnL; rpoC1_intron12
3atpE; rps15; ndhH_rps15; ndhJ_ndhK; trnW_trnP5
4atpF; petG; psbH; psbT; ndhG_ndhI; petB_petD; psbA_trnK; trnN_ycf18
5atpH; ndhC; psaI; psbE; psbF; psbJ; rpl2; rpl16; rpl23; ccsA_ndhD; petL_petG; trnA_rrn23; trnV_rrn1613
6atpI; ndhD_psaC; ndhF_rpl32; rpl14_rpl16; rps19_trnH; rrn5_trnR; trnS_rps47
7cemA; petB; rbcL; rpl22; rps3; rps18; ndhC_trnV; rps11_rpl36; rps14_psaB; trnK_rps16; trnR_trnN11
8clpP; rps11; trnI_intron3
9matK; ndhB; ndhI; rpl14; rps7; rps14; psbE_petL; psbH_petB; psbJ_psbL; rps4_trnT; trnG_trnfM; trnL_ndhB; trnL_trnF; clpP_intron1; petB_intron; trnG_intron16
10ndhD; ndhF; ndhK; petA; psaA; rpoB; rpoC1; rpoC2; ycf1; ycf2; atpH_atpI; trnF_ndhJ; trnQ_psbK; trnS_psbZ; trnT_psbD; ycf3_trnS; atpF_intron; ndhA_intron; petD_intron; rpl16_intron; rps16_intron; Consensus tree of species trees22
11ndhE; ndhJ; psbC; rpoA; rps16; clpP_psbB; rps2_rpoC2; ycf2_trnL8
12petN; psbL; psbN; rps12_intron4
13psaB; psbA; rps8; psbM_trnD; rrn16_trnI; trnP_psaJ6
14psaC; psaJ; psbM; ndhB_rps7; psbN_psbH; rpl23_trnI6
15psbD; psaA_ycf3; rps12_trnV; trnV_trnM; ndhB_intron; rpl2_intron6
16psbK; psbZ; rps12; psbC_trnS; rrn4.5_rrn5; trnA_intron6
17rps2; rps4; trnV_intron3
18ycf4; petG_trnW; petN_psbM; trnC_petN; ycf3_intron25
19atpI_rps2; cemA_petA; ndhE_ndhG; rpl32_trnL; rpl33_rps18; rpl36_rps8; trnD_trnY7
20psbI_trnS; rps8_rpl14; trnR_atpA3
21rpl16_rps3; trnfM_rps14; trnM_atpE3
Back to TopTop