Construction method and application of SNP (Single Nucleotide Polymorphism) liquid phase chip for forest tree breeding

Abstract

The invention discloses a construction method and application of an SNP (Single Nucleotide Polymorphism) liquid phase chip for forest tree breeding, and belongs to the technical field of forest tree molecular breeding. The method comprises the steps that based on whole genome re-sequencing data of multiple germplasm resources of a target species, a high-quality candidate background site pool is obtained through primary screening of a first set of screening parameters, probe design feasibility and genome space distribution optimization are conducted on the high-quality candidate background site pool, and a uniformly-distributed and probe-feasible background site set is formed; secondly, based on a whole genome association analysis result, screening out a functional site set directly associated with breeding traits; and finally, combining the two types of site sets, and carrying out probe design and synthesis to prepare the liquid phase chip. The invention also provides the chip constructed by the method and application of the chip in forest genome selective breeding. According to the invention, the design limitation of a traditional chip is broken through, a special chip with high density and high breeding value can be systematically constructed, and the breeding efficiency is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of plant bioinformatics technology, and more specifically to a method for constructing and applying SNP liquid-phase chips for forest tree breeding. Background Technology

[0002] The improvement of Chinese fir began in the 1950s, and its genetic improvement work ranks among the top in China's forest tree genetic improvement. After decades of genetic improvement, Chinese fir has entered the high-generation improvement stage. Currently, the fourth-generation seed orchard of Chinese fir in Fujian Province has been completed, and it is about to enter a new stage of fifth-generation breeding. Traditional forest tree breeding is characterized by long breeding cycles, low phenotypic selection efficiency, time-consuming and labor-intensive progeny testing, and long time required for the selection of new varieties, which seriously restricts the breeding of superior varieties of Chinese fir and the development of plantations.

[0003] Genome-wide selection breeding is a novel plant breeding method based on the association analysis of molecular markers and sample phenotypic values, model construction, and their applications. It is an early selection breeding technique based on genotype selection. By constructing a GS breeding technology system for Chinese fir, the breeding process of Chinese fir can be accelerated. With the development of sequencing technology, obtaining high-throughput, high-quality SNPs is one of the prerequisites for implementing GS molecular breeding of Chinese fir. However, as a coniferous species, Chinese fir has a huge and complex genome, making the application of whole-genome sequencing to detect variant sites costly and difficult to implement. SNP chips developed using targeted sequencing genotype detection (GBTS) technology offer higher resolution, lower cost, and greater flexibility. They can be widely used in population genetic diversity analysis, population structure, QTL (quantitative trait loci) mapping and genetic map construction, GWAS, and the mining of economically important trait genes.

[0004] However, current SNP chip designs in the forestry field generally suffer from the following problems:

[0005] 1) Most microarray designs focus only on constructing genetic maps or analyzing population structure, and the selected loci are mostly neutral background loci evenly distributed throughout the genome. Although these loci can provide basic diversity information, they are not directly related to important economic traits of Chinese fir such as growth, material properties, and adaptability, which limits their efficiency and application value in marker-assisted selection and genome prediction.

[0006] 2) Although some studies may have discovered some trait-related loci through genome-wide association analysis, there is a lack of a systematic approach in the current technology to effectively integrate large-scale, high-quality background loci with rigorously validated functional loci to build an "integrated" solution that both meets the needs of basic genetics research and directly serves breeding decisions.

[0007] 3) Existing chip design processes fail to fully consider the special needs of breeding application scenarios, such as ensuring the accuracy, reliability and cost-effectiveness of genotyping in highly heterogeneous breeding populations.

[0008] Therefore, there is an urgent need in this field for an innovative SNP chip construction method to break through the limitations of traditional neutral site selection and to develop a dedicated chip with both high-density genome coverage and high breeding application value through an optimized, breeding-oriented design strategy, so as to truly promote the molecular design breeding process of Chinese fir and other forest tree species. Summary of the Invention

[0009] To address the problems of limited site selection strategies, systematic missing functional sites, and lack of breeding guidance in existing forest tree SNP microarray designs, this invention aims to provide an innovative method for constructing SNP liquid-phase microarrays. This method systematically integrates genomic background information and trait functional information to construct a dedicated microarray with both high-density genome coverage and high breeding application value, overcoming the low efficiency of traditional microarrays in molecular breeding.

[0010] To achieve the above objectives, the technical solution of this application is as follows:

[0011] In a first aspect, the present invention provides a method for constructing an SNP liquid phase chip for forest tree breeding.

[0012] The core concept of this method lies in employing a "dual-channel screening" strategy. This involves using two independent yet complementary technical pathways to screen background loci for basic genetic analysis and functional loci directly serving breeding selection. These two pathways are then systematically integrated to construct a dedicated microarray that combines high-density coverage with high breeding value. The method specifically includes the following steps:

[0013] Step (1): Obtaining the background site set

[0014] Based on whole-genome resequencing data from multiple germplasm resources of the target forest tree species, SNP loci were initially screened according to the first set of screening parameters, aiming to obtain a high-quality pool of candidate background loci for population genetics analysis.

[0015] In a preferred embodiment, the first set of screening parameters includes: sequencing depth meeting specific upper and lower limits set based on the number of samples, a site deletion rate not exceeding 0.5, and a minimum allele frequency not less than 0.05. More preferably, the lower limit of the sequencing depth is set to be greater than the total number of resequencing samples, while the upper limit is set based on the statistical characteristics of the depth distribution of all resequencing data (e.g., based on the average depth and standard deviation) to effectively filter low-depth and abnormally high-depth regions, balancing data reliability and cost-effectiveness.

[0016] In another preferred embodiment, to further ensure the quality of the site and the accuracy of the genotyping, the first set of screening parameters also includes: passing the Hardy-Weinberg equilibrium test, with a significance p-value not higher than 1e-6; and requiring that the SNP site maintain a distance of at least 5 base pairs from the nearest insertion / deletion marker.

[0017] Step (2): Obtaining the set of functional sites

[0018] Based on genome-wide association analysis of one or more important phenotypic traits of the target forest tree species, the results of GWAS are screened according to a second set of screening parameters to obtain a set of functional loci directly associated with the breeding target traits.

[0019] In a preferred embodiment, the second set of screening parameters is: the significance p-value of the association between the SNP site and the trait in the GWAS analysis is no higher than 5e-8. This strict statistical threshold ensures the reliability of the screened functional sites.

[0020] Step (3): Optimization of background sites

[0021] The candidate background site pool obtained in step (1) is optimized for probe design feasibility and genomic spatial distribution to form a set of background sites that are evenly distributed on the reference genome and for which probe design is feasible.

[0022] In one specific implementation, the optimization includes dividing the reference genome into multiple non-overlapping windows, and within each window, selecting the optimal site based on probe design scoring (considering factors such as GC content, repetition rate, and specificity) as the representative background site for that window. This strategy ensures uniform coverage of microarray sites on the genome while maximizing the success rate of probe design.

[0023] Step (4): Chip integration and fabrication

[0024] The background site set obtained after optimization in step (3) is merged with the functional site set obtained in step (2). Based on the merged comprehensive site set, probe design and synthesis are performed, and finally, an SNP liquid phase chip is fabricated.

[0025] In some embodiments, to ensure the wide applicability and stability of the chip constructed by this method, the number of samples of the multiple germplasm resources used for resequencing in step (1) should not be less than 100, and they should come from at least 3 different geographical populations or genetic lineages to ensure that the input genetic diversity information is sufficiently representative.

[0026] This invention's method is particularly suitable for coniferous tree species with complex genomes and long breeding cycles. In a preferred embodiment, the target tree species is Chinese fir. The dedicated Chinese fir breeding chip constructed using the above method can significantly improve its molecular breeding efficiency.

[0027] In a second aspect, the present invention provides an SNP liquid phase chip constructed by any of the methods described in the first aspect above.

[0028] The unique feature of this chip lies in its probe set, which can simultaneously and specifically capture SNP loci combinations composed of the background locus set and the functional locus set obtained by the "dual-channel screening" strategy. This unique locus composition makes it not only a genotyping tool, but also an integrated breeding decision support platform.

[0029] Thirdly, the present invention provides a specific application of the SNP liquid phase chip described in the second aspect above in forest tree genetic breeding.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1) Through the original "dual-channel screening" strategy, the traditional diversity analysis tool is upgraded into an efficient breeding decision tool, which enables the chip to directly enrich loci that are strongly correlated with important economic traits.

[0032] 2) The synergistic effect of background sites and functional sites maximizes the use of limited chip capacity, which not only meets the needs of basic research, but also greatly improves the accuracy and efficiency of breeding selection.

[0033] 3) In view of the highly heterogeneous nature of forest tree germplasm resources, the screening parameters were optimized (such as allowing a higher missing rate) to ensure high coverage and typing accuracy of the chip in real breeding populations.

[0034] 4) The chip constructed based on this method can solve the key technical needs of multiple breeding links, from gene mining to kinship correction and genome prediction, in one stop, providing a powerful platform tool for molecular design breeding of forest trees.

[0035] 5) The chip constructed in this invention provides an optimized data foundation for genomic selection breeding. The results of Examples 5 and 6 show that the genomic selection model constructed based on the genotype data of this chip performs well in terms of genetic parameter estimation and trait prediction. This is due, on the one hand, to the high-quality, high-density, and uniformly distributed background loci obtained through the first set of parameters, which provides a reliable guarantee for accurately estimating inter-individual kinship and correcting population structure in the model; on the other hand, to the functional loci significantly associated with traits obtained through the second set of parameters, which directly contribute to the prediction of target breeding values. Therefore, the locus combination determined by the "dual-channel screening" strategy of this invention combines the stability required for population genetics analysis with the directionality required for breeding selection, making it particularly suitable for modern breeding practices such as genomic selection, which have high requirements for both genotyping data quality and breeding guidance. Attached Figure Description

[0036] Figure 1 Distribution and density map of 50K SNPs in the Cunninghamia lanceolata genome using liquid-phase microarray;

[0037] Figure 2 A graph showing the consistency rate of repeated samples;

[0038] Figure 3 The results of genome-wide association analysis of diameter at breast height (DBH) trait in Chinese fir are based on a 50K liquid-phase gene chip.

[0039] Figure 4 This is a G-matrix heatmap of 20 Chinese fir samples based on a 50K liquid-phase gene chip. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention is further described below with reference to specific embodiments. Unless otherwise described in detail, the technical means used in the following embodiments are all conventional means well known to those skilled in the art. Alternatively, they may be carried out according to the kit and product instructions. Unless otherwise specified, the materials and reagents used in the following embodiments are commercially available.

[0041] Those skilled in the art will understand that the core of this application lies in providing an optimized and repeatable method for constructing SNP chips, rather than being limited to specific values. The specific parameters described below are optimal embodiments derived from extensive experimental verification, and the underlying technical principles enable those skilled in the art to apply them to different project scenarios.

[0042] The application and effects of the above principles will be further explained below through specific examples.

[0043] Example 1: Preparation of 50K cedar wood liquid phase chip

[0044] In this embodiment, the 548 Chinese fir germplasm resources were all derived from self-sequencing data. This material comprises 548 samples from 14 provenance sites (see Table 1).

[0045] Table 1. Distribution of resequencing samples from 548 Chinese fir specimens

[0046]

[0047] 1. SNP screening was performed on the resequencing results of *Cunninghamia lanceolata* using the following parameters: sequencing depth DP > 548 && DP < 2930, miss rate: 0.5, minimum allele frequency (MAF) > 0.05, HWE: 1e-6, and 5 bp away from an indel. Probe design was performed on the initial screened background sites, and sites that did not fit the design were removed (mainly based on GC content and site repetition rate). Then, a sliding window selection was performed on the background sites based on the designed panel size. Subsequently, the species reference genome was evenly divided into 49683 windows. Within each window, the feasibility of the probe design was evaluated using a sliding window, and the site with the best design score was selected as the representative background site for that window. The final number of background sites was 49683. The reference genome was divided into 50K segments.

[0048] 2. To better establish the association between SNP loci and phenotypes, genome-wide association analysis was performed on diameter at breast height (DBH), timber density, and bioclimatic traits. Loci with a threshold of 5e-8 or higher were selected based on the Manhattan plot, resulting in 3671 functional loci. After comprehensively considering the GC content of the loci and their overall repetition rate in the reference genome, 715 functional loci were ultimately adopted.

[0049] 3. Using 49,683 background sites and 715 functional sites, subsequent probes were synthesized to prepare the chip.

[0050] The genomic coordinates and sequence information of the 50,398 SNP loci ultimately contained in the liquid-phase chip of Chinese fir constructed by the above method are disclosed in the following manner:

[0051] 1. Reference Sequence: The genomic data of *Cunninghamia lanceolata* has been submitted to NCBI GenBank, accession number PRJNA1390470. This data contains the genomic region sequences of all SNP sites in this invention, i.e., 100 bp above and below each site (URL: https: / / www.ncbi.nlm.nih.gov / sra / PRJNA1390470).

[0052] 2. SNP coordinate file: Submitted to Zenodo, with the permanent identifier DOI: 10.5281 / zenodo.17972273 (URL: https: / / doi.org / 10.5281 / zenodo.17972273). This file details the chromosomal location, reference allele, and variant alleles of all 50,398 SNPs. Those skilled in the art can access all site information of the liquid-phase chip in this application through the aforementioned accession number and DOI.

[0053] This chip, developed by Novogene, is a liquid-phase chip product specifically designed and optimized for breeding needs using liquid-phase chip technology. Based on the principle of thermodynamic stability, probe design comprehensively considers factors such as the complexity of the target species' genome, the location of the target site, and the GC content near the target site to ensure 100% capture of the target region. Probe preparation includes template synthesis and probe preparation. The synthesized DNA template (Oligo Pool) undergoes quality control and in vitro amplification experiments to finally prepare the finished DNA probe. The probe length range is 80-120 bp, with an overall average probe length of approximately 100 bp. Biotin-labeled NTPs and nucleotide analogs are used as synthetic raw materials in DNA probe preparation. Nucleotide analogs can increase the Tm value of probe-liquid binding, making the binding between the probe and the library more stable. Simultaneously, the biotin label on the synthesized probes allows them to bind to streptavidin-coated magnetic beads in subsequent experiments.

[0054] 4. SNP chip testing principle and process:

[0055] (1) The principle of chip detection is to capture target regions by probe hybridization in order to detect genetic and genome-related variation information. The sample DNA is randomly fragmented, then WGS library is constructed, target regions are captured by probe hybridization, streptomycin magnetic beads are used for adsorption, non-target molecules are eluted, purified and enriched, then PCR amplification is performed, and sequencing is performed to obtain the sequencing information of the target region of the sample.

[0056] (2) Library construction and capture: DNA samples are subjected to sonication or enzyme digestion to break down DNA, end repair and 3' end addition of "A", adapter ligation and purification, and Pre-PCR library amplification to obtain the library required for hybridization capture. The capture method is performed according to "NGP Hyb & Wash Kit v2.0". After library and probe hybridization, probe binding to magnetic beads, rinsing of non-specifically bound libraries, PCR amplification after capture, library quantification and quality control, sequencing is performed.

[0057] (3) Sequencing and Analysis: After the library concentration was detected using the NGP Qubit dsDNA HS Assay Kit reagent on a Qubit 4.0 Fluorometer, high-throughput sequencing was performed. The raw image data obtained from high-throughput sequencing (Illumina platform, MGI platform, etc.) was identified and converted into raw sequencing data in Fastq format by a specific program. Low-quality sequences, adapter sequences, and abnormal sequences such as polyG were removed from the raw sequencing data to obtain clean data. The clean data was compared, sorted, and deduplicated with the Chinese fir reference genome using BWA software to obtain a bam file. The bam file was then analyzed for variant sites using BCFTOOLS software to finally obtain the genotyping results of the target locus.

[0058] Example 2: Genotyping and Evaluation of 50K Liquid-Phase Chip in Chinese Fir

[0059] To test the detection performance of the 50K liquid phase chip on Chinese fir samples, 22 Chinese fir leaf samples were selected for testing and evaluation.

[0060] The detection system employs a precise localization sequencing and genotyping technology based on liquid-phase capture of target genomic sequences. Synthetic specific probes are used to capture and enrich multiple different target sequences located at different genomic positions using liquid-phase hybridization. Then, high-throughput sequencing is performed on the captured and enriched target regions to obtain the genotypes of all SNP loci within the target regions. Two replicates are included in the 22 samples, designated as technical replicates.

[0061] The results of sequencing and data analysis are shown in Table 2. Figure 1 and Figure 2 .

[0062] Table 2. Chip detection results of 22 samples

[0063]

[0064] The average coverage of the target region for all samples was 98.14%, with coverage ranging from 97.66% to 98.58%. This indicates a well-designed chip capable of efficiently capturing the target region and covering almost the entire 50K probe set. Simultaneously, the sequencing depth was substantial, with an average depth of 93.16× for all samples, although significant differences existed between samples (ranging from 69.47× to 118.18×). Except for samples 7 (69.47×) and 8 (70.37×), the average depth of other samples exceeded 80×, with most above 90×. The average depth of 93.16× is far higher than the standard required for general variant detection, ensuring data reliability. Regarding coverage at different depth thresholds, the data showed good uniformity. Even at the required 30× coverage level, the coverage of all samples remained above 90%, with an average 30× coverage of 93.65%, ensuring high accuracy and reliability in variant detection. The technically replicated sample data showed a high degree of consistency, with a consistency rate between 99.22% and 99.25% and an average consistency rate of 99.235%, ensuring the reliability of genotype comparison analysis between different samples.

[0065] In summary, the 50K liquid phase chip for Chinese fir is stable and efficient, and can meet the stringent requirements of Chinese fir genome research.

[0066] Example 3: Genome-wide association analysis

[0067] To test the practicality of the 50K liquid phase chip in Cunninghamia lanceolata genetic research, a genome-wide association analysis (GWA) was performed on the diameter at breast height (DBH) of 887 Cunninghamia lanceolata accessions, and the results were saved. A Manhattan plot was generated from the GWA results using the qqman package in R. The results are shown below. Figure 3 As shown, a total of 3 SNP loci had p-values ​​exceeding the statistical threshold, indicating that these loci were significantly associated with the diameter at breast height (DBH) trait in Chinese fir. These results demonstrate that the 50K liquid-phase microarray for Chinese fir has high applicability in genetic research, facilitating the location and discovery of candidate genes.

[0068] Example 4: Correction of Phylogenetic Relationships in Chinese Fir

[0069] To test the applicability of the 50K liquid chromatography-mass spectrometry (LC-MS) chip in the identification and phylogenetic relationship study of Chinese fir (Cunninghamia lanceolata), 20 samples from one family were selected and tested using the 50K LC-MS chip. WGS library construction, target region capture based on probe hybridization, PCR amplification, and sequencing were performed to obtain sequencing information of the target regions of the samples, enabling genotyping. The G-matrix heatmap is shown in the image (see [link]). Figure 4Four samples were found to be unrelated to the other 16 samples. Among the other 16 samples, one pair was a full sibling, and the rest were half-siblings. Further comparison of these results with pedigree records revealed that the four unrelated samples were incorrectly classified into the same family. Analysis using a 50K liquid chromatography-mass spectrometry (LC-MS) chip accurately corrected this error, demonstrating that the 50K L-MS chip has high accuracy and resolution in identifying phylogenetic relationships in Chinese fir and can provide reliable data support for Chinese fir variety identification and phylogenetic research.

[0070] Example 5: Hereditary power estimation based on 50K cedar wood liquid-phase chip

[0071] To test the accuracy of 50K liquid-phase chip in estimating the heritability of growth traits in Chinese fir, genotyping was performed on 1724 samples from second-generation progeny stands of Chinese fir, and genetic data from these samples were collected. Combined with previously obtained phenotypic data, including key growth traits such as tree height and diameter at breast height (DBH), the genetic parameters of the statistical models ABLUP and GBLUP were compared. The results (see Table 3) show that among the four models constructed using genotype and pedigree data for tree height and DBH, respectively, the GBLUP model achieved the highest heritability estimate. Specifically, for DBH, the GBLUP-estimated heritability was 0.07 higher than the ABLUP-estimated heritability. This demonstrates that GBLUP can effectively capture more genetic variance. In the comparison of the Akaike Information Criterion (AIC), GBLUP achieved the minimum value for each trait, indicating that the GBLUP model fit was superior to ABLUP, proving the superiority of GBLUP. This demonstrates that the 50K liquid phase chip for Chinese fir exhibits high accuracy and reliability in heritability estimation, providing a powerful tool for genetic improvement of Chinese fir. It helps to more accurately screen individuals with superior traits, thereby improving the efficiency and effectiveness of Chinese fir breeding. Furthermore, it lays a solid foundation for subsequent work such as genome-wide association studies, kinship correction, and genomic selection prediction in Chinese fir.

[0072] Table 3 Comparison of genetic parameters between GBLUP and ABLUP

[0073]

[0074] Example 6: Genomic selection prediction (GS) of Chinese fir growth traits

[0075] To test the accuracy of 50K liquid-phase chip in predicting the growth traits of Chinese fir, the statistical models ABLUP and GBLUP were used to predict and analyze the growth traits of the samples using the materials and methods of Example 5. The results (see Table 4) show that the predictive ability of the GBLUP model was tested using five-fold cross-validation, and compared with the traditional pedigree-based ABLUP model. In the same-generation prediction of second-generation offspring, the predictive ability PA of GBLUP for tree height was 0.14 and the accuracy ACC was 0.31, while the predictive ability PA of GBLUP for diameter at breast height was 0.12 and the accuracy ACC was 0.31. Further analysis revealed that although the prediction accuracy and predictive ability values ​​were not particularly high, compared with the traditional ABLUP method, using a 50K liquid-phase chip combined with the GBLUP model could more effectively capture the association information between genes and growth traits, thus improving the reliability of the prediction to some extent. Moreover, with further increases in sample size and continuous optimization of the model, such as incorporating more variables related to environmental factors or using more complex statistical models, the prediction accuracy and predictive ability are expected to be significantly improved. This indicates that the 50K liquid phase chip has great application potential and development space in predicting the growth traits of Chinese fir, and can provide strong technical support for the genetic improvement and efficient cultivation of Chinese fir.

[0076] Table 4 Comparison of Predictive Ability and Accuracy between GBLUP and ABLUP

[0077]

[0078] The above description is illustrative only and not restrictive of the present invention. Those skilled in the art will understand that many modifications, variations or equivalents can be made without departing from the spirit and scope defined by the appended claims, and all such modifications, variations or equivalents will fall within the protection scope of the present invention.

Claims

1. A method for constructing an SNP liquid phase chip for forest tree breeding, characterized in that, Includes the following steps: (1) Based on whole-genome resequencing data of multiple germplasm resources of the target forest tree species, in order to obtain a set of background loci for population genetic analysis, SNP loci are initially screened according to the first set of screening parameters to obtain a high-quality candidate background locus pool. (2) Based on genome-wide association analysis of the phenotypic traits of the target forest tree species, in order to obtain the set of functional loci directly associated with the breeding target traits, SNP loci that are significantly associated with the traits are screened from the GWAS results according to the second set of screening parameters to obtain the set of functional loci. (3) Optimize the probe design feasibility and genome spatial distribution of the candidate background site pool obtained in step (1) to form a set of background sites that are evenly distributed on the reference genome and have feasible probe design. (4) The background site set obtained in step (3) is merged with the functional site set obtained in step (2), and probes are designed and synthesized based on the merged site set to prepare liquid phase chips.

2. The method according to claim 1, characterized in that, The first set of screening parameters includes: sequencing depth with specific upper and lower limits based on the number of samples, site deletion rate not higher than 0.5, and minimum allele frequency not lower than 0.

05.

3. The method according to claim 2, characterized in that, The specific upper and lower limits of the sequencing depth are determined as follows: the lower limit is greater than the total number of resequencing samples, and the upper limit is set based on the statistical characteristics of the depth distribution of the resequencing data.

4. The method according to claim 1, characterized in that, The first set of screening parameters also includes: meeting the Hardy-Weinberg equilibrium test with a significance p-value not higher than 1e-6; and the distance between the SNP site and the nearest insertion / deletion marker is at least 5 base pairs.

5. The method according to claim 1, characterized in that, The second set of screening parameters is: the significance p-value of the association between SNP sites and traits in genome-wide association analysis is not higher than 5e-8.

6. The method according to claim 1, characterized in that, The optimization described in step (3) includes: filtering the candidate background site pool for probe design feasibility; dividing the reference genome into multiple windows, and selecting the optimal site within each window based on probe design score.

7. The method according to claim 1, characterized in that, The multiple germplasm resources used for resequencing in step (1) have a sample size of no less than 100 and come from at least 3 different geographical populations or genetic lineages.

8. The method according to claim 1, characterized in that, The target tree species is Chinese fir.

9. The SNP liquid-phase chip obtained by the method of any one of claims 1-8.

10. The application of the SNP liquid phase chip according to claim 9 in the selection breeding of forest tree genomes.

Images