Comparative Analysis of the Chloroplast Genomes of Eight Species of the Genus Lirianthe Spach with Its Generic Delimitation Implications

Based on Sima and Lu’s system of the family Magnoliaceae, the genus Lirianthe Spach s. l. includes approximately 25 species, each with exceptional landscaping and horticultural or medical worth. Many of these plants are considered rare and are protected due to their endangered status. The limited knowledge of species within this genus and the absence of research on its chloroplast genome have greatly impeded studies on the relationship between its evolution and systematics. In this study, the chloroplast genomes of eight species from the genus Lirianthe were sequenced and analyzed, and their phylogenetic relationships with other genera of the family Magnoliaceae were also elucidated. The results showed that the chloroplast genome sizes of the eight Lirianthe species ranged from 159,548 to 159,833 bp. The genomes consisted of a large single-copy region, a small single-copy region, and a pair of inverted repeat sequences. The GC content was very similar across species. Gene annotation revealed that the chloroplast genomes contained 85 protein-coding genes, 37 tRNA genes, and 8 rRNA genes, totaling 130 genes. Codon usage analysis indicated that codon usage was highly conserved among the eight Lirianthe species. Repeat sequence analysis identified 42–49 microsatellite sequences, 16–18 tandem repeats, and 50 dispersed repeats, with microsatellite sequences being predominantly single-nucleotide repeats. DNA polymorphism analysis revealed 10 highly variable regions located in the large single-copy and small single-copy regions, among which rpl32-trnL, petA-psbJ, and trnH-psbA were the recommended candidate DNA barcodes for the genus Lirianthe species. The inverted repeat boundary regions show little variation between species and are generally conserved. The result of phylogenetic analysis confirmed that the genus Lirianthe s. l. is a monophyletic taxon and the most affinal to the genera, Talauma and Dugandiodendron, in Sima and Lu’s system and revealed that the genus Lirianthe s. s. is paraphyletic and the genus Talauma s. l. polyphyletic in Xia’s system, while Magnolia subsection Gwillimia is paraphyletic and subsection Blumiana polyphyletic in Figlar and Nooteboom’s system. Morphological studies found noticeable differences between Lirianthe species in aspects including leaf indumentum, stipule scars, floral orientation, tepal number, tepal texture, and fruit dehiscence. In summary, this study elucidated the chloroplast genome evolution within Lirianthe and laid a foundation for further systematic and taxonomic research on this genus.


Introduction
The family Magnoliaceae encompasses a group of plants that are highly valued for their ornamental qualities.The family involves about 300 species and ranges across tropical and temperate areas, mainly in Southeast Asia, North South America Central America, Southeast North America, including Mexico, and Antilles [1,2].In terms of classification, dividing Magnoliaceae into two subfamilies or two tribes, Liriodendroideae (or Liriodendreae) and Magnolioideae (or Magnolieae), has been widely accepted [3][4][5].Within the subfamily Liriodendroideae, there is no dispute that it contains only the single genus, Liriodendron Linnaeus.Currently, all research indicates that the classification controversies within the family Magnoliaceae are mainly focused on the delimitation and classification of genera within the subfamily Magnolioideae.The number of genera within the subfamily Magnolioideae has been divided into 16 by Xia [6], 15 by Law [4], 14 by Sima and Lu [7], 11 by Dandy [3], or even just 1 by Figlar and Nooteboom [8].Li [9] prefer the treatment of segregated or smaller genera in the family Magnoliaceae because that the treatment of the bigger genera cannot represent the later evolution process and levels in this family, and that the bigger genera are too inclusive to reflect the evolutional trends and the migratory routes mainly on the basis of morpho-geographical studies.Sima and Lu's Magnoliaceae system [7,10], established in 2012 based on DNA data and morphological characteristics, especially observations on living plants, consists of 14 monophyletic genera in their delimitation of the subfamily Magnolioideae, which has been strongly supported by an increasing number of DNA data analysis results and gradually accepted by many scholars [11][12][13][14].
The genus Lirianthe Spach s. l. in Sima and Lu's system is equivalent to Magnolia subgenus Magnolia section Gwillimia subsection Gwillimia Candolle plus the subsection Blumiana (Blume) Figlar & Nooteboom in Figlar and Nooteboom's system [15].It is a monophyletic group [7,10] and includes about 25 species mainly distributed from East Himalaya through South China to Southeast Asia [16].There are about 12 species of this genus in China, of which 5 are endemic to China [16,17].But this genus Lirianthe Spach s. s. in Xia's system excludes those species with the circumscissile mature carpels such as L. hodgsonii (J.D. Hooker & Thomson) Sima & S. G. Lu, and so on.Plants of the genus Lirianthe have very important practical value and are world-renowned ornamental plants, industrial timber sources, medicinal materials, and fragrance source species [18,19] Wu have large and fragrant flowers, making them popular choices for garden ornamentals.Additionally, their lush green foliage makes them ideal for greening up outdoor space.L. phanerophlebia (B.L. Chen) Sima is only known from Hekou, Jinping, and Maguan in the south-east of Yunnan province.Based on the results of the Global Trees Campaign field surveys in December 2005, it is estimated that the total wild population is less than 200 individuals.The biggest threat to the species is a decrease in habitat, with many suitable areas now replaced by banana plantations.Local awareness-raising is vital for this species, as is research into nursery techniques for its cultivation.L. henryi is a national second-level key protected wild plant from Yunnan province.It has been listed as an endangered plant by the international union for conservation of nature (IUCN).L. confusa Sima & W. N. Sima and L. brevisericea Sima & Hong Yu are two newly described species.
Earlier phylogenetics of Magnoliaceae plants based on DNA analysis employed from one to several nuclear or chloroplast DNA regions [20][21][22], but a recent systematic and phylogenetic study based on complete chloroplast genome (CPG) sequences of nine species of Lirianthe demonstrated that these nine species of Lirianthe form a monophyletic branch [23].This study shows the value of chloroplast genomic data for systematic or species classification research, but did not include L. henryi, L. phanerophlebia, L. confusa, and L. brevisericea.Here, the complete chloroplast genomes were sequenced and assembled for eight species of genus Lirianthe.Our goal was to (1) elucidate the genomic structure, gene content and genetic variability encoded in Lirianthe chloroplast DNA; (2) compare the chloroplast phylogenomics in the family Magnoliaceae using CPG sequences and protein coding sequence separately, with an emphasis on the generic delimitation for Lirianthe.The chloroplast sequence datasets established herein facilitate improved phylogenetic resolution and evolutionary inferences for these under-surveyed Asian endemics and their underexplored Lirianthe relatives.

The Sequence Coverage Analysis of Assembled Chloroplast Genome
The sequence coverage, also known as the sequencing depth, serves as a key indicator for determining the reliability and the quality of assembly results.The average sequence coverages of eight assembled plastomes ranged from 93.9× to 547.9×, evaluated by GetOrganelle.The raw sequencing reads were also aligned to the assembled plastomes using minimap2 and the coverage depth was calculated using samtools, which ranged from 105.79× to 1577.05×.The sequencing coverages are shown in Table 1.The specific coverage results of eight chloroplast assemblies, obtained through the minimaup2 + samtools methods, can be found in Table S1, and their graphical results are illustrated in Figure S1.

General Characteristics of Eight Chloroplast Genomes of Lirianthe Species
The sizes of the CPGs of eight Lirianthe species ranged from 159,548 bp to 159,833 bp, with significant variations between individual species (Table 2).The nucleotide compositions of the genomes were remarkably similar to each other, with 30.0%Adenine (A), 30.7%Thymine (T), 20.0%Cytosine (C), and 19.3% Guanine (G).This suggests a close evolutionary relationship among them.The CPGs were divided into a canonical quadripartite structure with a large single-copy region (LSC, 87,671~87,965 bp), a small single-copy region (SSC, 18,745~18,772 bp), and a pair of inverted-repeats regions (IRa/IRb, 26,538~26,560 bp). Figure 1 shows the chloroplast genome structure of the genus Lirianthe, taking L. confusa as an example.The GC content of the CPG, LSC, SSC, and IR regions were 39.27%~39.29%,37.97%~37.98%,34.32%~34.46%,and 43.16%~43.18%,respectively (Table 2).The number of annotated genes in the eight chloroplast genomes is the same, with a total of 130, including 85 protein-coding genes, 37 transfer RNA genes, and 8 ribosomal RNA genes.The gene names and classifications of the 85 protein-coding genes were shown in Table S2, and were classified into four categories for their different roles.The annotated CPG data have been submitted to the NCBI database.The specific accession numbers are shown in Table 2.
The relative synonymous codon usage (RSCU) values of eight Lirianthe CPGs are displayed in Table S3.
Perhaps due to the high genetic similarity, the RSCU values of the eight species are very close, and the RSCU analysis results of L. confusa (MT654129) are shown in Figure 2. Its protein-coding gene region contained 22,659 codons (including 78 terminal codons).Among these codons, Leu (Leucme, L) was the most abundant amino acid, with 2314 (10.21%), while Cys (Cysteine, C) was the least, with 262 (1.16%).In all the termination codons, AUU is the most frequently used amino acid.
Except for the stop codon, there are 30 codons with RSCU values greater than one, in which the codon ending with A/U was 94.80% and the codon ending with C/G was 5.20%, indicating that these codons tended to end in A/U.The codon usage of these Lirianthe species is notably conserved, and is consistent with earlier reports on the chloroplast genomes of many land plants [24].
The relative synonymous codon usage (RSCU) values of eight Lirianthe CPGs are displayed in Table S3.
Perhaps due to the high genetic similarity, the RSCU values of the eight species are very close, and the RSCU analysis results of L. confusa (MT654129) are shown in Figure 2. Its protein-coding gene region contained 22,659 codons (including 78 terminal codons).Among these codons, Leu (Leucme, L) was the most abundant amino acid, with 2314 (10.21%), while Cys (Cysteine, C) was the least, with 262 (1.16%).In all the termination codons, AUU is the most frequently used amino acid.

Repeat Identification
A total of 42~49 microsatellite sequence repeats (SSR) loci, 16~18 tandem repeats, and 50 dispersed repeats were detected in the eight Lirianthe CPGs (Table 3).The SSRs of eight species were all single nucleotides and dinucleotides, and single nucleotides were dominant, accounting for more than 90%; more than three nucleotide repeating units were not detected.Among the dispersed repeat sequences, palindromic repeats and forward repeats were the main ones, accounting for 50% and 30%, respectively.Except for the stop codon, there are 30 codons with RSCU values greater than one, in which the codon ending with A/U was 94.80% and the codon ending with C/G was 5.20%, indicating that these codons tended to end in A/U.The codon usage of these Lirianthe species is notably conserved, and is consistent with earlier reports on the chloroplast genomes of many land plants [24].

Repeat Identification
A total of 42~49 microsatellite sequence repeats (SSR) loci, 16~18 tandem repeats, and 50 dispersed repeats were detected in the eight Lirianthe CPGs (Table 3).The SSRs of eight species were all single nucleotides and dinucleotides, and single nucleotides were dominant, accounting for more than 90%; more than three nucleotide repeating units were not detected.Among the dispersed repeat sequences, palindromic repeats and forward repeats were the main ones, accounting for 50% and 30%, respectively.

DNA Polymorphism
The polymorphism of the CPG nucleotide sequence is displayed graphically using the nucleotide diversity values (Pi) obtained by sliding window analysis, as shown in Figure 3.This allowed for a comprehensive visualization of the genetic variations among CPGs.The Pi value ranges from 0 to 0.012, with an average of 0.0014.The 10 intervals with the highest Pi values are shown in Figure 3, and are petA-psbJ, trnH-psbA, psbM-trnY, rpl32-trnL, ccsA-ndhD, accD-ycf4, ndhF, ycf1, matK-rps16, and trnN-ndhF, respectively.Moreover, these highly variable regions are in both the LSC and SSC regions.The petA-psbJ has the highest Pi value (0.012).Furthermore, only the LSC and SSC regions include these highly variable regions; this finding is also supported by the graphical results of sequence alignments (Figure S2).these highly variable regions are in both the LSC and SSC regions.The petA-psbJ has the highest Pi value (0.012).Furthermore, only the LSC and SSC regions include these highly variable regions; this finding is also supported by the graphical results of sequence alignments (Figure S2).

Contraction and Expansion of IR Region
The length of the IR region of the eight Lirianthe species varies between 26,538 and 26,560 bp (Table 2).To better understand the boundary and detect changes in genes, we employed IRscope software (https://irscope.shinyapps.io/irapp/) to visualize the chloroplast into four regions: LSC-IRb (JLB), IRb-SSC (JSB), SSC-IRa (JSA), and IRa-LSC (JLA).Although the differences among them are small, the IR boundary region does show some differences in Figure S3.

Contraction and Expansion of IR Region
The length of the IR region of the eight Lirianthe species varies between 26,538 and 26,560 bp (Table 2).To better understand the boundary and detect changes in genes, we employed IRscope software (https://irscope.shinyapps.io/irapp/) to visualize the chloroplast into four regions: LSC-IRb (JLB), IRb-SSC (JSB), SSC-IRa (JSA), and IRa-LSC (JLA).Although the differences among them are small, the IR boundary region does show some differences in Figure S3.

Phylogenetic Analysis
The phylogenetic tree was constructed for the 8 Lirianthe species and the other 36 Magnoliaceous species using maximum likelihood (ML) and Bayesian methods based on the complete chloroplast genome (CPG) data.The ML consensus tree (Figure 4) was congruent with the Bayesian one (Figure 5) in the phylogenetic topologies, showing that the 44 Magnoliaceae species were divided into 15 monophyletic groups (Aromadendron, Dugandiodendron, Houpoea, Kmeria, Lirianthe, Liriodendron, Magnolia, Manglietia, Metamagnolia, Michelia, Oyama, Pachylarnax, Paramagnolia, Talauma, Yulania), successfully supporting Sima and Lu's system [7].But the 15 species of the genus Lirianthe had been scattered on different branches and classified in the two different genera, Talauma and Lirianthe, in Xia's system [6] or in the two different subsections, Magnolia subsection Gwillimia and subsection Blumiana, in Figlar and Nooteboom's system [8].The statistical supports, bootstrap values (BP) and posterior probabilities (PP), varied slightly among some branches or clades in Figures 4 and 5, but overall were relatively high, indicating that the tree-building results from ML and Bayesian methods both achieved good results in this study.

Plant Morphology
The main morphological characteristics of the leaves, flowers, and fruits of the eight Lirianthe species are described in Table 4. Except for the midveins of the leaf blade of L. henryi and L. hodgsonii, which are adaxially prominent, the other six species are all impressed.Regarding the presence and texture of trichomes, L. brevisericea, L. confusa, L. delavayi and L. odoratissima have various types of indumenta on the abaxial leaf surface, including sericeous and tomentose hairs; L. henryi has scattered appressed trichomes on the abaxial surface; whereas L. coco, L. hodgsonii and L. phanerophlebia have glabrous leaves.Except for L. phanerophlebia, whose stipule scars only extend to 1/3-1/2 of the petiole length, all other species have stipule scars that reach the top of the petiole.In terms of floral orientation, except for L. brevisericea, L. delavayi, and L. hodgsonii which are erect, the flowers of the remaining species are pendulous.The tepals are mostly 9 in number, ranging from 8 to 12; they are usually white, with only L. delavayi sometimes pink or red.The presence, texture, sparseness, and color of indumentum on the gynoecia are key distinguishing characteristics.Except for the mature carpel of L. hodgsonii, which dehisces along a circumference, all other species dehisce along the dorsal suture.

Species
Leaves Flowers Fruits

Plant Morphology
The main morphological characteristics of the leaves, flowers, and fruits of the eight Lirianthe species are described in Table 4. Except for the midveins of the leaf blade of L. henryi and L. hodgsonii, which are adaxially prominent, the other six species are all impressed.Regarding the presence and texture of trichomes, L. brevisericea, L. confusa, L. delavayi and L. odoratissima have various types of indumenta on the abaxial leaf surface, including sericeous and tomentose hairs; L. henryi has scattered appressed trichomes on the abaxial surface; whereas L. coco, L. hodgsonii and L. phanerophlebia have glabrous leaves.Except for L. phanerophlebia, whose stipule scars only extend to 1/3-1/2 of the petiole length, all other species have stipule scars that reach the top of the petiole.In terms of floral orientation, except for L. brevisericea, L. delavayi, and L. hodgsonii which are erect, the flowers of the remaining species are pendulous.The tepals are mostly 9 in number, ranging from 8 to 12; they are usually white, with only L. delavayi sometimes pink or red.The presence, texture, sparseness, and color of indumentum on the gynoecia are key distinguishing characteristics.Except for the mature carpel of L. hodgsonii, which dehisces along a circumference, all other species dehisce along the dorsal suture.Mature carpels dehiscent along dorsal sutures.

Discussion
The chloroplast genome is a superior option to the nuclear genome for studying nucleotide diversity and reconstructing the phylogeny of related species due to its smaller genome size, lower nucleotide substitution rate, uniparental inheritance, and haploid characteristics [25].The development of next-generation sequencing technology and bioinfor-matics analysis methods has significantly reduced the cost of obtaining genome sequences.Therefore, chloroplast genome-scale data are seeing increasing use.As a result, inferring evolutionary relationships at higher taxonomic levels [26], even at the interspecific or intraspecific level in some species, such as ginseng authentication by 18 species-specific markers [27], authentication of ginseng cultivars by 17 polymorphic sites [28], differentiation of hazelnut cultivars by a combination of markers of two InDels and one SNP [29], markers were mostly developed from whole-chloroplast genomes in these studies.The chloroplast genome assembly and annotation, as well as comparative analysis, are the basis and prerequisite for the above work.
We had previously completed the assembly and annotation of chloroplasts of L. hodgsonii [23] and L. coco [30].Here, we newly assembled the chloroplast genomes of six other Lirianthe species.The CPGs of these plants were found to possess a similar size and structure, all demonstrating a typical quadripartite pattern.There is a known connection between the phylogenetic position and the total GC content.Early differentiated lineages, like those found in the Magnoliaceae, have a greater GC content [31].Compared to the median GC content of 35% in most angiosperms, plants of the genus Houpoea N. H. Xia & C. Y. Wu have a higher GC content of around 39.2% [32], which is almost identical to the GC content of 39.3% in the Lirianthe species in this study.The GC content in the IR region of all Lirianthe species is notably high (43.2%),and alignment is increasingly conducted using chloroplast genome-scale data, with similar findings in other plants like Carthamus (43.2%) [33] and Cypripedium (42.7%) [34], potentially due to the inclusion of four highly conserved rRNA genes with high GC content (Figure 1).
SSR markers have been widely used in population genetics research due to their reliability and high variability.In this study, a small amount of SSR markers were detected, with about 170 in Houpoea species [32] and 246 in Bougainvillea spectabilis [35], while our findings indicated only 42~49 SSRs.The primary explanation is the use of different search parameters, as Carthamus species can only recognize 36-40 SSRs [33] when the same parameters (1-10 2-6 3-5 4-5 5-5 6-5) are applied.The expansion and contraction of IR regions at the junctions of LSC and SSC have significant impacts, including the creation of pseudogenes and alterations in genome size and evolutionary rate [36].There was an absence of substantial progress even at lower taxonomic levels.Species identification and the detection of structural changes in IR boundaries in this study suggest that Magnoliaceae may have remained relatively primitive and conservative in nature.
The analysis of CPG DNA polymorphism is a proven approach to detecting mutation hotspots, which can function as distinct DNA barcodes.Two of the ten genes or DNA segments with the highest assessment of nucleotide diversity detected in this study, rpl32-trnL and petA-psbJ, are also present in the genus Houpoea [32], which also belongs to the Magnoliaceae family.trnH-psbA is one of the early recommended plant DNA barcodes [37] and was also the second most polymorphic fragment in this study.rpl32-trnL, petA-psbJ, and trnH-psbA are good candidate barcodes for Lirianthe species based on the results of comparative analysis of DNA polymorphisms.These potentially highly variable chloroplast barcodes will greatly enhance our ability to identify and protect rare and endangered species within the genus Lirianthe.Resolution of phylogenetic relationships and reconstruction of evolutionary history have become feasible through comparison of chloroplast genomes and phylogenetic analysis.
The genus Lirianthe s. s. is one of the genera reinstated by Xia and Wu for the Magnolia subsection Gwillimia when several genera were recognized in the family Magnoliaceae.It has been accepted in many recent publications from China and Vietnam [38].CPG has proved to be a powerful tool in unraveling the evolutionary associations among land plants [26,39,40].Based on Sima and Lu's system [7], 15 species of the genus Lirianthe s. l. and 29 species of other genera of Magnoliaceae were selected, and a phylogenetic tree was drawn using a ML method with chloroplast whole-genome sequences (Figure 4).Among them, all 15 species of the genus Lirianthe s. l. gathered together and formed an independent branch.This confirmed that the genus Lirianthe s. l. is a monophyletic taxon in Sima and Lu's system.But the 15 species of the genus Lirianthe s. l had been scattered on different branches and classified in the two different genera, Talauma and Lirianthe, in Xia's system [6] or in the two different subsections, Magnolia subsection Gwillimia, and subsection Blumiana in Figlar and Nooteboom's system [8].These revealed that the genus Lirianthe s. s. is paraphyletic and the genus Talauma s. l. polyphyletic in Xia's system, while the Magnolia subsection Gwillimia is paraphyletic and the subsection Blumiana is polyphyletic in Figlar and Nooteboom's system.Furthermore, the 15 Lirianthe species show a crossover mixed pattern in Xia's system.The four species marked with green shading demonstrate the results of this cross-mixing classification in Figure 4.The processing results of Lirianthe plants indicate that the plant range of Lirianthe delimited in Sima and Lu's system is reasonable and scientific, and chloroplast genome evidence supports these taxonomic treatments.
However, as mentioned in the introduction, there are too many generic delimitation problems within the tribe Magnolieae or the subfamily Magnolioideae.In 2004, Figlar and Nooteboom, based on the latest available data on DNA and morphology, degraded the tribe Magnolieae or the subfamily Magnolioideae into the genus Magnolia Linnaeus, reduced all of its former segregated genera to Magnolia, and reconstructed a complex infrageneric system of the biggest genus, Magnolia s.l., in the family Magnoliaceae, including 3 subgenera, 12 sections, and 13 subsections.Thus, it appeared that this treatment had solved the generic delimitation problems within the tribe Magnolieae or the subfamily Magnolioideae.In essence, this just changed the problems from generic delimitation to infrageneric delimitation and the delimitation problems remained unsolved.In order to better unravel the generic delimitation problems above, Xia [6] and Sima & Lu [7] published a new system for the family Magnoliaceae in 2012.A total of 17 genera are recognized in Xia's system [6] and a total of 15 genera are recognized in Sima and Lu's system [7].The system by Sima and Lu, based on the data on DNA and the observations of morphological characters, especially in living plants, is strongly supported by recent DNA work [21,23,30,32,41] and has been accepted by many scholars [11][12][13][14].In order to better reflect the evolutionary steps and levels of Magnoliaceous plants, as well as the evolutionary trends and migration routes based primarily on morphological-geographical studies, it is a more feasible and useful approach to define smaller independent genera.Overall, the results of this study can provide valuable sequence information for the molecular systematics of Lirianthe in Magnoliaceae and offer a theoretical basis for the utilization and conservation of Lirianthe germplasm resources.

Sample DNA Extraction and Sequencing
We dried and preserved the collected fresh leaves of 8 species (L.brevisericea, L. coco, L. confuse, L. delavayi, L. henryi, L. hodgsonii, L. odoratissima, L. We extracted the total genomic DNA from the samples using DNA Plantzol Reagent (Invitrogen, Carlsbad, CA, USA), and sent the DNA to the BGI Group (Shenzhen, China) for library construction and sequencing on an HiSeq 2500 platform (Illumina, Hayward, CA, USA).The 2 × 150 bp paired-end reads with an insert size of 400 bp were created and the resulting DNA sequences were obtained after removing reads with low quality and adapter contamination.

Chloroplast Genome Assembly and Annotation
The raw sequencing reads were assembled by the programs GetOrganelle v1.7.7.0 using default parameters [42].The genome annotation was conducted by using CP-GAVAS2 [43], with manual correction using Geneious software (version 2020.0.5) [44].The chloroplast genome coverage analysis was evaluated in two ways.The first was based on the assembly log from GetOrganelle.In the second evaluation, the raw reads were mapped to the assembled genomes using minimap2, and then samtools was used to count the read coverage at each position of the genome.

Comparitive Analysis of Chloroplast Genomes
To check the polymorphism of DNA, the CPGs sequences of 8 Lirianthe species were aligned using MAFFT v7.490 [47], then sliding window analysis was performed on the alignment result using DnaSP v5.10.1 [48] with the window length and step size both set to 500 bp.The regions with high Pi values were considered as mutational hotspots or candidate DNA barcodes.The contraction and expansion of IR region, namely LSC/IRb/SSC/IRa junctions, can be visualized in the online program IRscope (https://irscope.shinyapps.io/irapp/)[49].

Phylogenetic Analysis
To compare the phylogenetic positions of the genus Lirianthe in the family Magnoliaceae, 36 chloroplast genome sequences of other Lirianthe species and other genera in Magnoliaceae were downloaded from the NCBI database.All 44 species of plants from the Magnoliaceae family covered in this study and their accession number of the chloroplast genome are listed in Table S4.The phylogenetic trees were built using the complete chloroplast genome (CPG) DNA sequences of 44 species by maximum likelihood (ML) and Bayesian methods.The CPG DNA was aligned using MAFFT [47] in Geneious [44] with default settings.The ML analysis was run by RAxML 8.2.11 program [50] under the GTR + GAMMA model with 1000 bootstrap replicates.The Bayesian analysis was run by MrBayes 3.2.6 program [51] under the GTR + GAMMA model with Markov Chain Monte Carlo (MCMC) settings of 2 runs of 1,000,000 generations with a burn-in of 250,000 generations with trees sampled every 1000 runs.The phylogenetic tree was visualized by iTOL v5 (https://itol.embl.de/)[52].

Conclusions
By studying the chloroplast genome of eight Lirianthe species, we were able to clearly outline their characteristics and provide valuable insights into the divergence and phylogenetic evolution of Lirianthe species.It was evident from our analysis that the Lirianthe species exhibited a strong level of conservation in its genome structure, gene content, IR boundaries, repetitive elements, and codon usage.This implies that these species have maintained a steady genetic structure throughout their evolutionary history.The chloroplast genome is a robust tool for unraveling the phylogenetic connections among species within the genus Lirianthe, as shown by the high resolution of the ML tree.Our investigation of phylogenetic relationships has provided valuable insights into the connections between Lirianthe and other genera within family Magnoliaceae.The knowledge presented here enhances our understanding of how these species are related.By enriching the knowl-edge of the chloroplast genome of Lirianthe species, our study paves the way for future investigations into their classification and evolution.

Figure 1 .
Figure 1.The gene and structure diagram of the Lirianthe species chloroplast genomes, taking L. confusa (MT654129) as an example.

Figure 1 .
Figure 1.The gene and structure diagram of the Lirianthe species chloroplast genomes, taking L. confusa (MT654129) as an example.

Figure 3 .
Figure 3.The DNA polymorphism analysis of eight Lirianthe chloroplast genomes by sliding window method.Window length: 500 bp; step size: 500 bp.

Figure 3 .
Figure 3.The DNA polymorphism analysis of eight Lirianthe chloroplast genomes by sliding window method.Window length: 500 bp; step size: 500 bp.

Figure 4 .
Figure 4. Phylogenetic trees of the genus Lirianthe and other Magnoliaceous species based on wholechloroplast genome sequences using Maximum likelihood method based on 1000 replicates.Bootstrap percentages less than 100 are shown.Note: Gray, blue and orange bars indicate the results or belonging positions of the corresponding plants in Sima and Lu's [7], Xia's[6] and Figlar and Nooteboom's[8] systems, respectively.The green background color highlights the genus Lirianthe had been scattered on different branches and classified in the two different genera, Talauma and Lirianthe, in Xia's system[6]; The blue and yellow background colors highlight the genus Lirianthe had been scattered in the two different subsections, Magnolia subsection Gwillimia, and subsection Blumiana in Figlar and Nooteboom's system[8].

Figure 4 .
Figure 4. Phylogenetic trees of the genus Lirianthe and other Magnoliaceous species based on whole-chloroplast genome sequences using Maximum likelihood method based on 1000 replicates.Bootstrap percentages less than 100 are shown.Note: Gray, blue and orange bars indicate the results or belonging positions of the corresponding plants in Sima and Lu's [7], Xia's[6] and Figlar and Nooteboom's[8] systems, respectively.The green background color highlights the genus Lirianthe had been scattered on different branches and classified in the two different genera, Talauma and Lirianthe, in Xia's system[6]; The blue and yellow background colors highlight the genus Lirianthe had been scattered in the two different subsections, Magnolia subsection Gwillimia, and subsection Blumiana in Figlar and Nooteboom's system[8].

Figure 5 .
Figure 5. Phylogenetic trees of the genus Lirianthe and other Magnoliaceous species based on wholechloroplast genome sequences using Bayesian method.Numbers above branches are bootstrap support values based on 1,000,000 generations with a burnin proportion of 0.25.Posterior probabilities less than 1 are shown.Note: Gray, blue, orange bars and different background colors are same as Figure 4.

Figure 5 .
Figure 5. Phylogenetic trees of the genus Lirianthe and other Magnoliaceous species based on whole-chloroplast genome sequences using Bayesian method.Numbers above branches are bootstrap support values based on 1,000,000 generations with a burnin proportion of 0.25.Posterior probabilities less than 1 are shown.Note: Gray, blue, orange bars and different background colors are same as Figure 4.

Table 1 .
The assembly coverages of eight Lirianthe chloroplast genomes evaluated using two methods (average depth, ×).

Table 3 .
The type and number of repeat sequences in the chloroplast genomes of 8 Lirianthe species.

Table 3 .
The type and number of repeat sequences in the chloroplast genomes of 8 Lirianthe species.

Table 4 .
Morphological comparisons among eight species of the genus Lirianthe Spach.

Table 4 .
Morphological comparisons among eight species of the genus Lirianthe Spach.