The medicinal plant genomes are more complex than animal genomes, so the process of sequencing the medicinal plant genomes has been hindered, and it entered a period of rapid development from 2016. This may be due to the decline in sequencing price and the development of long-read sequencing technologies. The status of medicinal plant genomes articles reported each year is shown in Figure 2A. In 2020, the number of published medicinal plant genomes has reached up to 53. In 2021, 33 medicinal plant genome articles have been published until June 4th, and the total article number is inferred to be more than 60. As more and more medicinal plant genomes have been revealed, several plants have been sequenced twice or multiple times for genomes. Among these repeatedly sequenced medicinal plant genomes, some are because of sequencing at the same time, some are due to improved level and quality, and some are genomes of different varieties from the same species. Among those 53 medicinal plant genomes reported in 2020, 18 genomes were reported repeatedly, accounting for 34%. This tells us that sequencing technology is continuously developing and progressing, bringing us to a completer and more accurate genome. Take Panax notoginseng (Chinese name: Sanqi) as an example, five versions of its genomes have been reported, the first two versions published in 2017 were sequenced by the next-generation sequencing (NGS) technology of Illumina platform (Chen et al., 2017; Zhang D. et al., 2017), and the recent three versions published in 2020 and 2021 were sequenced by the third-generation sequencing technologies of Pacific Biosciences (PacBio) and Oxford Nanopore (ONT) (Fan G. et al., 2020; Jiang et al., 2021; Yang et al., 2021b). The latest two versions of the genome were assembled to the chromosome level, the length of the assembled sequences was hundreds of times longer than the first two versions, and the accuracy and credibility of annotation have also been greatly improved. The statistical results of detailed information about medicinal plant genomes were shown in Table 1.

With the widespread application of NGS technology, genome sequencing of medicinal plants has become more feasible due to the greatly reduced cost and time required to complete the project. According to the whole genome sequence, the basic information of biology and biomedical functions can be well understood.

We have made statistics on the medicinal plant genome articles over the years, and have a basic understanding of the general characteristics of the reported medicinal plant genomes. The comparison of size and repetitiveness ratio of these published medicinal plant genomes and their evolution relationship is shown in Figure 2B. Among them, the genomes of five medicinal plants are much larger than other medicinal plants, they are Allium sativum, Paeonia suffruticosa, Aloe vera, Taxus wallichiana, and Ginkgo biloba. In the plants whose genomes have been sequenced, there are 123 angiosperms (including 12 monocots, 105 eudicots, and 6 magnoliids), two gymnosperms, and one lycopodiophyta plant. The simplified phylogeny of the major clades of sequenced medicinal plants is also shown in Figure 2B. Angiosperms account for the vast majority of sequenced medicinal plants, and eudicots make up the majority of angiosperms. Genome size has a positive correlation with the ratio of repetitive elements, when the genome size is larger, the proportion of repetitive elements also tends to be correspondingly larger. Most of the genome size is concentrated within 4 Gb, and the repetitiveness ratio sequences are concentrated between 30 and 90%.

It has been said that plant genome reports were formulaic and lack biology significance, their descriptions mainly included the assembly, protein-coding genes, repeats, evolution analysis, some aspects of biology, usually with a focus on transcription factors and active compounds biosynthesis pathway (Michael and Jackson, 2013). According to these published medicinal plant genomes, most of them have not yet been used to solve specific application problems, such as discovering new medicinal mechanisms, cultivating new resistant varieties, explaining evolutionary events, and so on. But the assembly of the genomes provides us with the guarantee of the database. Once we need the support of genetic information, the genome is the solid foundation and reference.

The start of genomics is from the early 1990s, and automated sequencing methods using dideoxy chain termination with fluorescent molecules developed, which is known as Sanger sequencing. The effectiveness of the Sanger platform for large eukaryotic genomes was first reported in 2000 for Drosophila melanogaster, ushering in a new era of genomics (Adams et al., 2000). This method was also applied in plant biology, like sequencing ESTs in Arabidopsis thaliana (Newman et al., 1994), and then sequencing the whole genome of various plants, like Oryza sativa (Yu et al., 2002), Populus trichocarpa (Tuskan et al., 2006), Carica papaya (Ming et al., 2008) and Brachypodium distachyon (International Brachypodium Initiative, 2010). However, there are still gaps and errors in the assembly of these genomes, so they are not completely “finished,” because the process of “finishing” needs inspection and experimental resolution of inconsistencies, and it’s a time-consuming, tough, and expensive work (Hamilton and Robin Buell, 2012). In the germination stage of the development process about medicinal plant genome sequencing, considering this and cost, the Sanger sequencing method is only used to sequence the genome of major economic crops that are also regarded as medicinal plants, like Ricinus communis, to provide references and templates for subsequent research.

After 2011, the NGS technology develops rapidly and occupies the position of the mainstream sequencing platform, becoming the preferred technology for sequencing the medicinal plant genomes. The widely and mainly used NGS platforms are Roche 454 platform and Illumina platform.

Roche 454 platform is the first commercially successful NGS system. This sequencing method uses a high-throughput pyrosequencing technology (Margulies et al., 2005). This platform utilizes emulsion PCR to detect the pyrophosphate released during nucleotide incorporation. In 2005, the read length of Roche 454 was only 100–150 bp with 20 Mb output data per run (Mardis, 2008). In 2008, the 454 GS FLX Titanium system appeared, with a reading length up to 700 bp and 0.7 G output data per run within 24 h. In late 2009, Roche simplified the library preparation and data processing and improved the output to 14 G per run (Liu et al., 2012). In 2012, the platform upgraded to the FLX+ and could generate 1 million reads, with a reading length up to 1,000 bp.

Illumina platform is a high-throughput technology of sequencing by synthesis using reversible dye terminators developed by Solexa and then purchased by Illumina in 2008 (Bentley et al., 2008). The mechanism of the Illumina platform is bridge PCR, which is different from the Roche/454 platform. The library DNA with fixed adaptors is denatured to single strands and linked on the flow cell, followed by bridge amplification to synthesize clusters of clonal DNA fragments. The library splices into single strands by linearization enzyme (Mardis, 2008), and then four kinds of fluorescently labeled nucleotides which have been modified with a terminator complement the template one base at a time, the signal is captured, then the terminator and fluorescent dye are cleaved, and a new round of synthesis repeats until coming up to the desired read length. In late 2011, the paired-end mode of the Hi-Seq2000 Illumina platform could generate more than 250 million reads total sequences of one lane.

Because the throughput of Hi-Seq 2000 is higher, the price is lower, and the application range is wider than Roche/454, the application of the Illumina platform in the medicinal plant genome sequencing occupies the mainstream position. The Illumina platform is widely applied for expression profiling, de novo sequencing, and re-sequencing in plant sequencing, like Thellungiella parvula (Dassanayake et al., 2011) and Arabidopsis thaliana (Cao et al., 2011). As more and more medicinal plant genomes have been reported, the medicinal plant genome sequencing has begun to enter the development stage, many large size medicinal plant genomes were successfully sequenced. However, another difficulty of plant genomes is the high repetition in the genome, so it is difficult to accurately assemble them by the NGS technologies.

The development of third-generation sequencing has overcome this problem. The most widely applied long-read sequencing platform is Single-Molecule Real-Time (SMRT) sequencing of Pacific Biosciences company. SMRT sequencing is run on cells, which have tiny wells called zero-mode waveguides (ZMWs). In each ZMWs, a DNA polymerase/template complex gets immobilized, and synthesizes a new DNA strand (Jiao and Schneeberger, 2017). Each incorporation generates a light pulse that can be recognized for differently labeled nucleotides (Eid et al., 2009). PacBio systems can sequence reads with an average size of about 20 kb and a maximum length of over 60 kb (Kim K. E. et al., 2014; Vanburen et al., 2015). Although the sequencing error rate of raw reads is up to 15%, self-correction by adequate coverage sequencing data (Chin et al., 2013) or correction with NGS data (Bashir et al., 2012; Koren et al., 2012) enables genome assemblies with the accuracy of over 99.999% simply by running bioinformatics analysis software (Chin et al., 2016). Besides the PacBio SMRT platform, there is also another long-read sequencing platform introduced by ONT Technologies, which provided access to their first sequencing system in 2014 (Quick et al., 2014; Deamer et al., 2016). Single DNA molecules are run through nanopores, and individual nucleotides create characteristic disruptions in them, which reveal the sequence of the nucleotides. The reads length and sequencing accuracy are similar with PacBio reads, and the longest reads can reach up to 200 kb. First, whole-genome assemblies using ONT data have reached N50 values of multiple hundred kb for fungal genomes, and bacterial genomes could be fully assembled with a nucleotide accuracy of over 99% (Goodwin et al., 2015; Loman et al., 2015).

The emergence of third-generation sequencing technology has made a great leap in sequencing read length and brought medicinal plant genome sequencing into a stage of rapid development. The strategy used in this stage is a combination of second- and third-generation sequencing technologies, which can ensure long read length, high throughput, and reasonable sequencing price at the same time. Medicinal plant genomes are large and have high-ratio repetitive elements, the frequently-used strategy is combining high coverage Illumina and low coverage PacBio SMRT or ONT data. Because third-generation sequencing can provide long-read sequences to increase the assembly accuracy and genome draft quality, but the price is relatively high, so Illumina platform is used to guarantee enough sequencing data. And this can make it possible to assemble large and complex medicinal plant genomes to the chromosome level. After these years of sequencing development, the medicinal plants not only can obtain draft genome relevant information and dig out target protein-coding genes, but also recognize the chromosome-level of the genome to discover the evolution, gene cluster’s function, repetitive elements effect, and so on.

We found that not only does the number of medicinal plant genomes sequenced continue to increase, but the number of medicinal plant genomes sequenced repeatedly is also increasing. Why? First of all, because the genomes of many medicinal plants have not been revealed yet, many teams are performing de novo sequencing of the genomes at the same time, and accordingly publish them at the same time. Then, with the continuous development of gene sequencing technology, we can obtain longer sequencing read lengths, so as to assemble more complete and accurate high-level genomes. Genomes assembled to the chromosome level are the current trend. The information that the genome gives us is no longer a contig or scaffold, but the chromosome and the position of a gene on the chromosome.

There are 25 medicinal plants with two reported genomes, three medicinal plants with three reported genomes, and one plant with five reported genomes. Representative medicinal plants include Momordica charantia (bitter gourd), Salvia miltiorrhiza (Danshen), Punica granatum (pomegranate), Panax notoginseng (Sanqi), Panax ginseng (Asian ginseng), etc. Bitter gourd and danshen have two reported versions of the genome. Bitter gourd completed the de novo assembly of the genome draft in 2017, as well as basic annotation and evolutionary analysis (Urasaki et al., 2017). In 2020, using PacBio long-read sequencing technology, the Momordica charantia genome was assembled to the chromosome level, and further investigate the genomic changes under domestication (Matsumura et al., 2020). The genome of Salvia miltiorrhiza was also assembled to eight chromosomes, the assembled genome size increased from 538 to 594.75 Mb, and the proportion of repetitive elements also increased from 54.44 to 64.84% (Xu et al., 2016; Song Z. et al., 2020). Punica granatum (pomegranate), which is a popular and nutritious fruit with medicinal properties, has three published genome versions (Qin et al., 2017; Yuan Z. et al., 2018; Luo et al., 2020). The third version of the genome is assembled to the chromosome level, and it is a high-quality genome map of the soft-seed pomegranate, which helps to clarify the genetic divergence between soft- and hard-seeded varieties and provides insights into the genetic diversity and population structure of pomegranates (Luo et al., 2020). Panax notoginseng (Sanqi) is a well-known TCM whose genome research is sought after by scientists, and a total of five versions have been reported. The three recent versions are assembled to the chromosome level (Fan G. et al., 2020; Jiang et al., 2021; Yang et al., 2021b), which are more complete than the previously available genome assemblies (Chen et al., 2017; Zhang D. et al., 2017), further reveal the biosynthesis pathways of ginsenosides and dencichine, as well as provide a resource for further exploration of the saponin biosynthesis, cultivation, and breeding of P. notoginseng. Panax ginseng (Asian ginseng), reputed as the king of medicinal herbs, belongs to the same genus Panax, which also has two versions of reported genomes (Xu et al., 2017; Kim et al., 2018). Both of these two genomes provide a comprehensive understanding for functional and evolutionary analysis as well as ginsenoside biosynthesis. Additionally, Kim et al. (2018) identified fatty acid desaturases that can increase freezing tolerance and chlorophyll a/b binding protein genes which enable efficient photosynthesis under low light. However, the read length of both genomes is not long enough by the current standards, and there is still space for further improvement in the integrity and accuracy of the ginseng genome.

The genes related to medicinal plant growth and development, disease resistance, important genetic traits, and germplasm characters which are the important functional genes in medicinal plants, taking advantage of genome annotation information, discovering good genes, using genetic engineering methods to break the reproductive isolation, and cultivating the new species with excellent agronomic characters and high content of active ingredients, so that it can lay the foundation for the large-amount extraction of active ingredients and extensive clinical application. By combing transcriptome and resequencing of individual species within or between species, the large-scale molecular markers can be identified rapidly and accurately, and genetic linkage study of molecular markers and qualified characters can also be accelerated, the phenotypes of medicinal plants and the relationship of physical characteristics and genotypes are discovered quickly so that efficiency of breeding are improved obviously.

The study of Scutellaria baicalensis (Huangqin) genome sequencing revealed that a specialized metabolic pathway for the synthesis of 40-deoxyflavonebioactives evolved in the genus Scutellaria and found that the gene encoding a specific cinnamate coenzyme A ligase likely obtained its new function following recent mutations and that four genes encoding enzymes in the 40-deoxyflavone pathway are present as tandem repeats in the genome of Huangqin. Further analysis discovered that gene duplications, segmental duplication, gene amplification, and point mutations coupled to gene neo- and subfunctionalizations were involved in the evolution of 40-deoxyflavone synthesis in Scutellaria. These results not only provide significant insight into the evolution of specific flavone biosynthetic pathways in the mint family Lamiaceae but also facilitate the development of tools for enhancing bioactive productivity by molecular breeding in plants (Zhao Q. et al., 2019).

Whole-genome sequencing cannot only elucidate the biosynthesis pathways of natural products but also give insight into their evolution. The evolution will bring the whole genome change, like WGD and whole-genome triplication (WGT), to adapt to the environment alteration and explain the characters of plants. We summarized the WGD and WGT events of some representative species reported in the medicinal plant genome articles, and these situations are shown in Figure 5. These WGD and WGT events are summarized and introduced into three types of plants, which are eudicots, monocots, and magnoliids.

In the eudicots part, we select five representative branches to demonstrate the situation. The representative medicinal plants of Araliaceae and Apiaceae are clustered together. P. ginseng, P. notoginseng, and E. senticosus belong to Araliaceae, and P. notoginseng is diploid, while P. ginseng and E. senticosus are tetraploid. Two rounds of WGD were discovered in these Araliaceae plants, the first round occurred around 29.6 Mya, P. ginseng, and E. senticosus both had the second round of WGD, which were found almost 2.2 Mya in P. ginseng and 13 Mya in E. senticosus, respectively. Additionally, these recent WGDs were discovered to contribute to the ability of P. ginseng to overwinter and E. senticosus to adapt to cold environment, enabling them to live and spread broadly through the cold area (Kim et al., 2018; Jiang et al., 2021; Yang et al., 2021a). These two rounds of WGD occurred in the family Araliaceae after divergence with the Apiaceae, which may be one of the reasons why its genome was bigger than other medicinal plants. In the D. carota and A. graveolens that belonged to Apiaceae, one shared WGD occurred in about 43 Mya, and one recent WGD only existed in A. graveolens in approximately 1.9 Mya, and this duplication contributed to the expansion of terpene synthase gene families (Song X. et al., 2020). The second branch in the eudicots part includes six plants belonging to Lamiales, one shared WGD (almost 60.7 Mya) was identified in S. baicalensis, S. barbata, S. miltiorrhiza, and S. indicum, which might be responsible for chromosomal expansion and rearrangement (Xu et al., 2020a), and two rounds of WGD were found in S. splendens and L. angustifolia, which could result in the gene families expansion related to terpenoid biosynthesis (Li et al., 2021). In P. cuspidatum, it experienced current lineage-specific WGD at 6.6 Mya after the divergence with F. tataricum from the ancestor, and it shared the ancient and common WGD with F. tataricum at 65 Mya (Zhang Y. et al., 2019), after this WGD, the genome of F. tataricum experienced dramatic chromosomal rearrangements, resulting in very fragmented intra-genome collinear blocks (Zhang L. et al., 2017). There is also a WGT event identified and reported in the medicinal plant genome articles. T. wilfordii was found to have a WGT event in approximately 21 Mya, which enabled it to cope better with and adapt to the markedly changed environment, and the duplication of the triptolide biosynthesis genes were almost generated by this WGT event, suggesting this WGT event was important to the evolution of triptolide biosynthesis (Tu et al., 2020).

In the monocots part, A. sativum and H. citrina are the representatives. A. sativum has undergone two rounds of WGD, suggesting WGD can be the important driving force of the proliferation of TEs and genome expansion in garlic (Sun et al., 2020). Otherwise, H. citrina experienced a recent WGD event at about 15.73 Mya, which was the main factor resulting in multiple copies of the orthologous genes (Qing et al., 2021). In the magnoliids part, C. salicifolius and P. nigrum are the representatives. Two rounds of ancient WGD were inferred in the C. salicifolious genome, one was shared by Calycanthaceae at ∼87 Mya after its divergence with Lauraceae, and the other was dating back to approximately 142 Mya in the ancestry of Magnoliales and Laurales (Lv Q. et al., 2020). Meanwhile, the P. nigrum genome was speculated to have a WGD event at ∼17.9 Mya, which brought genetic changes that were responsible for the particular biosynthesis of piperine (Hu et al., 2019).

Domestication is a complex evolutionary process, which is one of the most important technological innovations in human history, humans use plants to change their morphology and physiology traits, distinguishing them from wild ancestors, and ultimately giving rise to the current human cultures (Diamond, 2002; Hancock, 2005). Some of the domesticated plants are medicinal plants. The timing and geographical origins of domesticated traits, as well as the genes that lead to changes in traits, can be sent to find clues from genomic information (Purugganan and Fuller, 2009).

Coix is a widely cultivated grass crop with high nutritional and medicinal value, which has been domesticated as early as the Neolithic era. However, its genetic research and breeding were hampered by the lack of a sequenced genome. Two chromosome-level genomes of coix have been reported simultaneously, which belong to elite cultivar Beijing (Liu H. et al., 2020) and wild relative Coix aquatica Daheishan (Guo et al., 2020), respectively. They both find that hull thickness is an important domestication trait between the wild relatives and cultivars, and selection of papery hull from the stony hull in wild progenitors was a key step in coix domestication. Combining resequencing analysis and comparative analysis, several domesticated loci or genes (like loci in the ∼2 to 150 kb region upstream of ub3) and two major quantitative trait loci associated with hull thickness and color (Ccph1 and Ccph2), were discovered to be the potential identification loci for domestication. These findings will greatly facilitate and benefit the molecular breeding of coix and provide a powerful reference for the domestication and evolution of medicinal plants.

The active components of medicinal plants with complex and diverse structures are the material basis for their medicinal effect, and it’s also an important source of new drug discovery. However, many medicinal plant materials often face a series of problems in the process of development and utilization, for example, the growth of many medicinal materials is greatly affected by environmental factors; some rare herbs grow slowly and are difficult to grow by artificial cultivation; most of the active ingredients are low in content, complex in chemical structure and difficult in chemical synthesis; traditional methods of natural extraction or artificial chemical synthesis cannot meet the needs of scientific research and new drug development. Synthetic biology will be an effective way to resolve these problems.

As high-throughput sequencing technology for genome and transcriptome studies have developed rapidly, using bioinformatics method and functional genomics approach to screen and identify enzyme-coding genes on specific secondary biosynthesis pathway from a large number of the original species of medicinal plants, which will greatly accelerate the analysis process of secondary biosynthesis pathway and lay a solid foundation for herbal synthetic biology research of medicinal plants.

Tripterygium wilfordii genome is one of the typical examples. Because of the extremely low yield of triptolide extracted from T. wilfordii, its original plant cannot be grown on a large scale, and the current chemical synthesis route is limited to a yield of less than 1.64%. A more promising method to obtain more triptolide could be metabolic engineering, which can be realized via a synthetic biology strategy. However, it required elucidation of the triptolide biosynthesis pathway. Therefore, the sequencing of the T. wilfordii genome was completed, and cytochrome P450 TwCYP728B70 involved in triptolide biosynthesis was identified, accordingly, the triptolide content in the CYP728B70 overexpression line increased obviously (Tu et al., 2020). It’s important to make full use of genomic resources to reveal the biosynthesis pathways of active compounds in medicinal plants and use candidate genes in these pathways for the heterologous bioproduction under synthetic biology strategy.

Geoherbs, controlled by genetic factors and affected by environmental conditions, are representative of high-quality medicinal materials. The utilization of sequencing technology and data can provide useful tools to elucidate the molecular mechanism of geoherbs. For the same medicinal plants in different areas, epigenomic studies of medicinal plants can be carried out to clarify the genetic variation of different production areas, especially the modification effect of different environments on the epigenome of medicinal material, including DNA methylation modification, small RNA sequencing analysis, chromatin immunoprecipitation analysis, and so on. In addition, microorganisms in soil are also important factors in the growth environment of geoherbs. Metagenomic analysis of soil microbial community can be sequenced to provide the basis for revealing the interaction between soil microorganisms and the growth of medicinal plants.

Recently, 545 genomes of ginkgo trees sampled from 51 populations across the world were sequenced to identify three refugia in China and detect multiple cycles of population expansion and reduction along with glacial admixture between relict populations in the southwestern and southern refugia, and multiple anthropogenic introductions of ginkgo were proved to occur from eastern China into different continents. This study provides insight into the evolutionary history of ginkgo and helps to provide protection and utilization way for its valuable genomic resources (Zhao Y. P. et al., 2019).

Because of the rapid development and progress of sequencing technology, more and more biosynthesis pathways of active ingredients from medicinal plants have been revealed. The early-stage was based on the mining from transcriptome data, and the later stage was based on the combined mining from genome and transcriptome data. Although transcriptome sequencing has so far occupied a major position in the research of biosynthesis pathways of medicinal ingredients, genome data can provide more important information, for example, it can reveal the evolution process of biosynthesis pathway genes, thereby efficiently synthesizing secondary metabolites with medicinal activity. In the opium poppy genome, a great discovery about a gene cluster including 15 genes was reported. Meanwhile, in its evolution process, the events like gene duplication, rearrangement, and fusion, could lead to the aggregation and co-expression of genes in the two metabolic pathways of noscapine and morphinan, so that it resulted in the formation of this supergene cluster, which could synergistically synthesize the medicinal ingredients in opium poppy (Guo et al., 2018). Therefore, the opium poppy genome helps to decipher the mystery of the synthesis of secondary metabolites. It is not only beneficial to the development of molecular plant breeding tools and cultivating new varieties, but also has great guiding significance for the selective improvement of the production of alkaloids with different efficacy in future artificial synthesis.

It also provides new ideas for the application of medicinal plant genomes. Through the evolution process, gene duplications and neofunctionalization can generate gene clusters, which may relate to specialized metabolites, and this phenomenon has already been observed in several model plants, like A. thaliana, Zea mays, and Solanum lycopersicum (Bharadwaj et al., 2021). In medicinal plants, we can refer to the research strategy of the opium poppy (Guo et al., 2018), which can help us understand the formation process of gene clusters related to medicinal active ingredients and improve their biosynthesis efficiency.

The continuous emergence of high-quality genomes has made the application of comparative genomics analysis more and more extensive and in-depth, and it is also a powerful tool for researchers to dig out biological problems and explain biological phenomena (Nobrega and Pennacchio, 2004). Comparative genomics, based on genome mapping and sequencing technology, are generally referred to as comparative analysis of the structural and functional gene regions of the genomes among multiple species or multiple individuals (populations) from one species. Specifically, it is to compare the similarities and differences in the structural characteristics, study the contraction and expansion of gene families, discover the differentiation time and evolution relationship, analyze the generation and evolution of new genes, etc.

One representative example of comparative genomics among different medicinal plant genome species can be Scutellaria baicalensis and Scutellaria barbata. The comparative genomic analysis of them showed the recent LTR may result in chromosomal rearrangement and expansion, and tandem duplication of paralogs after their speciation might contribute to the divergent evolution of flavonoid biosynthesis gene families, which provided a significant foundation for the evolution and chemodiversity studies in the Lamiaceae (Xu et al., 2020a).

Moreover, a representative of comparative genomics among different populations in the same species can be Forsythia suspense. Genome-wide comparative analysis was then conducted for the 15 natural populations across its current distribution range. The results revealed that candidate genes associated with local adaptation were functionally correlated with heterogeneous environmental factors, and supported the hypothesis that adaptive differentiation should be highly obvious in the genes of signal crosstalk between different environmental variables, which gave insights into the fundamental genetic mechanisms of the local adaptation to climatic gradients in plant species (Li L.-F. et al., 2020).