Method for judging insertion event of unknown exogenous fragment into transgene based on targeted sequencing

Abstract

The invention discloses a method for judging an unknown exogenous fragment insertion transgenic event based on targeted sequencing, and relates to the field of transgenic technology judgment. According to the invention, the insertion position of a transgenic event can be accurately identified, and even if an exogenous gene is inserted into a repetitive sequence region, transgenic and non-transgenic sequences can be effectively distinguished; the sequencing data after quality control is classified and counted, and a machine learning model is combined to optimize a proportion fluctuation range, so that the accuracy of judging the transgenic event is remarkably improved; the introduction of the Bayesian optimization method can further improve the accuracy and reliability of the model for judging the property of the transgenic event, and avoids the deviation and error in the traditional method.

Description

Technical Field

[0001] The present invention relates to the field of transgenic technology identification, and in particular to a method for identifying transgenic events in which unknown exogenous fragments are inserted based on targeted sequencing. Background Art

[0002] Transgenic technology has the characteristics of strong targeting, high breeding efficiency and short cycle, and has become one of the important means of plant molecular breeding. Determining whether the transgenic event is homozygous or heterozygous is of great significance for ensuring the genetic stability of transgenic plants, improving breeding efficiency, ensuring the consistency of trait expression, conducting accurate application research, and meeting regulatory and safety requirements.

[0003] During the growth and development of crops, they are often affected by insect pests, weeds and diseases. The current prevention and control measures mainly use a large amount of chemical pesticides and herbicides, which not only seriously affects the ecological environment and biodiversity, increases production costs and labor intensity, but also increases the chance of human poisoning. The most economical and effective method is to explore and use insect-resistant, herbicide-resistant and disease-resistant genes for variety breeding. In the breeding process, homozygous transgenic plants are more genetically stable because they contain exogenous genes on both homologous chromosomes. They can be directly used in production or further breeding work. In the commercialization and regulatory approval process of transgenic products, homozygous plants may also be more easily approved. Therefore, it is of great significance to distinguish pure heterozygous from transgenic events.

[0004] Most of the existing methods for judging the insertion of transgenic events need to rely on the specific insertion position of the exogenous sequence on the reference genome as an entry point, but once the exogenous sequence is inserted into the repetitive fragment, the capture efficiency and accuracy will be greatly reduced. The present invention can just cope with this situation, and can judge the vast majority of materials without relying on the insertion position information on the reference genome. In addition, the prior art also has certain bottlenecks for the analysis of sequencing data. For the transformant sequence of genetically modified crops, the comparison algorithm mainly identifies the transgenic event by comparing the matching of the exogenous sequence with the reference genome, but when the exogenous gene is inserted into a complex repetitive sequence region, multiple comparisons often occur, further exacerbating the ambiguity of the insertion position of the exogenous gene. Summary of the invention

[0005] The present invention proposes a method for identifying an unknown exogenous fragment inserted into a transgenic event based on targeted sequencing, aiming to provide a solution to the inaccurate identification of the transgenic event due to unclear insertion position.

[0006] Among them, the method for identifying the unknown exogenous fragment insertion into the transgenic event based on targeted sequencing includes the following steps: S1. Confirming the intersection position of the exogenous sequence and the reference genome sequence in the transformant sequence of the target transgenic event through the alignment model, wherein the position is the breakpoint; S2. Perform next-generation sequencing on the samples to be tested to obtain the original clean reads, and align them with the transformant sequences to generate a BAM file; perform quality control on the alignment information in the BAM file to ensure that the mapping quality of the alignment is ≥40 and that the alignment only includes one CIGAR operation; S3. According to the alignment results after quality control, reads were classified into three types: Ref_reads, Cross_reads, and GMO_reads; the three types of reads were counted, and the counting results were corrected by logarithmic transformation with base 10; S4. Calculate the ratio between reads of different categories and optimize the ratio fluctuation range through machine learning to determine the nature of the material.

[0007] Furthermore, in step S3, the specific types of classifying reads into three types of Ref_reads, Cross_reads and GMO_reads are as follows: Reads that fully align to parts of the reference genome; Reads that partially align to the reference genome and partially align to the exogenous sequence, i.e., reads that span the breakpoints; Reads that fully align with the exogenous sequence portion.

[0008] Furthermore, in step S3, the specific classification process of classifying reads into three types of Ref_reads, Cross_reads and GMO_reads is as follows: When the CIGAR string only includes matching operations and is completely aligned to the reference genome, the read is Ref_reads; When the CIGAR string includes matching and insertion / deletion operations, and the two aligned regions are from the reference genome and the exogenous sequence respectively, the read is Cross_reads; When the CIGAR string only includes matching operations and is completely aligned to the exogenous sequence, the read is GMO_reads.

[0009] Furthermore, the step S4 specifically includes the following sub-steps: S401. According to the classification results of reads, the number of reads in each category is calculated to obtain the ratio between reads in different categories, and the ratio between reads in different categories is used as the feature of model input; S402. Taking maximizing classification accuracy as the optimization goal; S403. Define a hyperparameter space for Bayesian optimization; S404. Constructing an objective function, and adjusting the classification accuracy by selecting different ratio thresholds and fluctuation ranges; S405. Use the Gaussian process as a proxy model to predict the behavior of the objective function and guide the next evaluation point based on the current evaluation result; use the expected improvement as the acquisition function to select an optimal hyperparameter combination that can improve the model performance; update the proxy model through iterations and use the acquisition function to guide Bayesian optimization to find the optimal hyperparameters; after each iteration, the objective function returns the evaluation result; S406. Obtain the optimal parameter combination through Bayesian optimization, and use the optimized proportional relationship to determine the transgenic properties of the material.

[0010] Furthermore, in step S402, the optimization target is optimized by an optimization function, the output of which is the accuracy of the classifier, and the optimized proportional relationship is used to distinguish transgenic events.

[0011] Furthermore, the accuracy calculation process is specifically based on the actual number of transgenic reads. , the true number of non-GM reads , the number of non-GM reads incorrectly identified as GM and the number of transgenic reads incorrectly identified as non-transgenic The confusion matrix composed of is calculated, namely: ; Among them, the represents the calculation result of accuracy, Represents the actual number of transgenic reads. Represents the true number of non-transgenic reads. represents the number of non-GM reads that were incorrectly identified as GM. Indicates the number of transgenic reads that were incorrectly identified as non-transgenic.

[0012] Furthermore, in step S404, the objective function is constructed according to the accuracy, that is: ; Among them, the Indicates that the hyperparameter configuration is The objective function is and ) represent the hyperparameter configurations respectively. The true number of transgenic reads, the true number of non-transgenic reads, the number of non-transgenic reads mistakenly judged as transgenic, and the number of transgenic reads mistakenly judged as non-transgenic, the They represent the threshold of the proportion of reads that are completely aligned with the reference genome, the threshold of the proportion of reads that cross the breakpoints, and the threshold of the proportion of reads that are completely aligned with the exogenous sequence.

[0013] Furthermore, the , , and The calculation process is expressed as follows: ; ; ; ; Among them, the The threshold value of the ratio of reads that fully align the exogenous sequence is used to reflect the insertion amount of the exogenous gene sequence in the transgenic event. Indicates the number of reads related to the qualified exogenous sequence within a specific interval, that is, the number of reads that are completely aligned with the exogenous sequence. Represents the total number of all reads. Indicates the ratio of reads that are completely aligned with the reference genome to the total reads. Indicates the number of reads that are completely aligned with the reference genome sequence, that is, the reads that completely match the reference genome. Indicates the ratio threshold of reads crossing the breakpoint, that is, the ratio of reads crossing the breakpoint to the total reads. Indicates the number of reads that cross the breakpoint. Indicates the ratio threshold of reads spanning breakpoints, that is, the ratio of reads spanning breakpoints to total reads.

[0014] Furthermore, in step S405, the expected improvement is used as the acquisition function, and the calculation process of the expected improvement of selecting an optimal hyperparameter combination that can improve the model performance is expressed as: ; Among them, the Indicates that the hyperparameter configuration is The expected improvement is Indicates the current optimal reads ratio in the Bayesian optimization process. Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the cumulative distribution function of the standard normal distribution, which is used to calculate the probability that the objective function value is lower than the current optimal value. Indicates that the hyperparameter configuration is The standard deviation of the prediction error is Represents the probability density function of the standard normal distribution and is used to calculate the probability density of the difference between the current best value and the predicted mean.

[0015] Furthermore, in step S405, the proxy model is specifically represented as: ; Among them, the Represents a proxy model, the represents the Gaussian process function, Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the kernel function.

[0016] The beneficial effects of the invention are: The present invention can accurately identify the insertion position of transgenic events, and can effectively distinguish transgenic and non-transgenic sequences even if the exogenous gene is inserted into the repetitive sequence region; by classifying and counting the sequencing data after quality control, and optimizing the proportion fluctuation range in combination with the machine learning model, the accuracy of determining transgenic events is significantly improved; the introduction of the Bayesian optimization method can further improve the accuracy and reliability of the model in judging the nature of transgenic events, avoiding the deviations and errors in traditional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A method flow chart of a method for identifying an unknown exogenous fragment insertion transgenic event based on targeted sequencing provided in an embodiment of the present invention; Figure 2 A flowchart of threshold verification through experiments provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0019] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention. It should be noted that relational terms such as the terms "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0021] Moreover, the terms "comprises," "comprising," or any other variation thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or machine that includes a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such process, method, article, or machine. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or machine that includes the element.

[0022] The features and performance of the present invention are further described in detail below in conjunction with the embodiments.

[0023] Among them, Figure 1 , a method for identifying an unknown exogenous fragment insertion transgenic event based on targeted sequencing, comprising the following steps: S1. Confirming the intersection position of the exogenous sequence and the reference genome sequence in the transformant sequence of the target transgenic event through the alignment model, wherein the position is the breakpoint; S2. Perform next-generation sequencing on the samples to be tested to obtain the original clean reads, and align them with the transformant sequences to generate a BAM file; perform quality control on the alignment information in the BAM file to ensure that the mapping quality of the alignment is ≥40 and that the alignment only includes one CIGAR operation; S3. According to the alignment results after quality control, reads were classified into three types: Ref_reads, Cross_reads, and GMO_reads; the three types of reads were counted, and the counting results were corrected by logarithmic transformation with base 10; S4. Calculate the ratio between reads of different categories, and optimize the ratio fluctuation range through experimental verification to determine the properties of the material.

[0024] Specifically, for step S1, the specific implementation principle is: through alignment, the intersection position of the exogenous sequence and the reference genome sequence in the transformant sequence of the target transgenic event is clarified, and this position is called the breakpoint. In this step, a variety of alignment tools (such as BWA, Bowtie, Minimap2, etc.) are used to align the clean reads with the reference genome and GMO.fa. The alignment results are fused through a weighted voting mechanism or a weighted average method to ensure the high accuracy of the alignment results, especially when dealing with complex regions (such as repetitive sequences and structural variation regions), using multiple tools to reduce the error impact of a single tool.

[0025] For step S2, the specific implementation principle is: perform next-generation sequencing on the sample to be tested to obtain the original cleanreads, and compare them with GMO.fa to generate a BAM file. Perform quality control on the alignment information in the BAM file to ensure that the mapping quality (MappingQuality) of the alignment is ≥40, and the alignment contains only one CIGAR operation. That is, the alignment is either a simple match without insertion or deletion, or a single insertion or deletion.

[0026] Furthermore, in step S3, the specific types of classifying reads into three types of Ref_reads, Cross_reads and GMO_reads are as follows: Reads that fully align to parts of the reference genome; Reads that partially align to the reference genome and partially align to the exogenous sequence, i.e., reads that span the breakpoints; Reads that fully align with the exogenous sequence portion.

[0027] Furthermore, as a preferred step S3 of the above embodiment, a "soft-clipping" operation (S) is introduced to process partially aligned reads close to the breakpoint. For example, 10S50M10S means that the front and back ends of the read do not match the reference genome at all, but the middle part matches the reference genome completely, and can be considered as part of the cross-breakpoint reads. Furthermore, in step S3, the specific classification process of classifying reads into three types of Ref_reads, Cross_reads and GMO_reads is as follows: When the CIGAR string only includes matching operations and is completely aligned to the reference genome, the read is Ref_reads; When the CIGAR string includes matching and insertion / deletion operations, and the two aligned regions are from the reference genome and the exogenous sequence respectively, the read is Cross_reads; When the CIGAR string only includes matching operations and is completely aligned to the exogenous sequence, the read is GMO_reads.

[0028] Furthermore, the process of threshold value verification through experiments preferred in the above embodiment is proposed. Specifically, the basic principle of its implementation is as follows: Figure 2 ,Right now: (1) Determine the insertion position of the exogenous sequence in the transformant sequence: By comparison, the intersection position of the exogenous sequence and the reference genome sequence in the transformant sequence of the target transgenic event (hereinafter referred to as GMO.fa) is determined, and this position is hereinafter referred to as the breakpoint.

[0029] (2) Obtain BAM file through sequencing: The samples to be tested were subjected to second-generation sequencing, and the clean reads without any filtering were compared with GMO.fa to obtain a BAM file. The information in the BAM file was quality controlled. The default mapping quality (Mapping Quality) was ≥40, and the alignment only contained one CIGAR operation, that is, the alignment was either a simple match without insertion or deletion, or a single insertion or deletion.

[0030] (3) Classify and count the reads that meet the requirements and pass quality control.

[0031] Specifically, all reads are divided into the following three types: Reads that are completely aligned to the reference sequence (Ref_reads); Part of it is aligned to the reference genome and part of it is aligned to the exogenous sequence, that is, the reads that cross the breakpoints (Cross_reads); The reads of the exogenous sequence part (GMO_reads) were completely aligned. The above three reads were counted, and considering the skewed distribution problem, the counting results were corrected using a logarithmic transformation with a base of 10 to reduce the impact of extreme values ​​on the results and the skewness of the data.

[0032] (4) Calculate the ratio between reads of different categories and optimize the ratio fluctuation range through model verification to determine the properties of the material.

[0033] The specific experimental verification threshold is set as follows: The target transgenic event X has two transformant sequences: RB (5' is the reference genome sequence, 3' is the exogenous sequence, and the total length is 774bp) and LB (5' is the exogenous sequence, 3' is the reference genome sequence, and the total length is 726bp). The breakpoint position of X on RB is 262, and the breakpoint position on LB is 256. 25 real samples (10 homozygous positive samples, with P as the sample name prefix; 10 heterozygous positive samples, with H as the sample name prefix; 5 negative control samples, with N as the sample name prefix) were selected, and the judgment threshold was set to the theoretical value ± 0.2 for the experiment; In the initial experiment, 8 positive heterozygous samples were misjudged as negative samples, with a correct judgment rate of 68%. Due to the deviation of the theoretical value, the threshold fluctuation range needs to be further adjusted. After continuous attempts, the final judgment threshold is as follows: (1) The ratio of Ref_reads to Cross_reads is 0.7-1.4 (including the boundary, closed interval) and is considered to be a homozygous negative transgenic sample; (2) The ratio of Ref_reads to Cross_reads is 1.4-2.1 (left open and right closed interval) and is considered to be a heterozygous positive transgenic sample; (3) The ratio of GMO_reads to Cross_reads is 0.7-1.4 (open interval) and is considered to be a homozygous positive transgenic sample. Finally, 25 real samples (10 homozygous positive samples, with P as the sample name prefix; 10 heterozygous positive samples, with H as the sample name prefix; 5 negative control samples, with N as the sample name prefix) were selected again for the experiment.

[0034] Furthermore, as a preferred threshold setting method of the above embodiment, a method of replacing the experimental verification process by machine learning is proposed. Specifically, the experimental verification process in step S4 is improved to verification by machine learning, including the following steps: S401. According to the classification results of reads, the number of reads in each category is calculated to obtain the ratio between reads in different categories, and the ratio between reads in different categories is used as the feature of model input; S402. Taking maximizing classification accuracy as the optimization goal; S403. Define a hyperparameter space for Bayesian optimization; S404. Constructing an objective function, and adjusting the classification accuracy by selecting different ratio thresholds and fluctuation ranges; S405. Use the Gaussian process as a proxy model to predict the behavior of the objective function and guide the next evaluation point based on the current evaluation result; use the expected improvement as the acquisition function to select an optimal hyperparameter combination that can improve the model performance; update the proxy model through iterations and use the acquisition function to guide Bayesian optimization to find the optimal hyperparameters; after each iteration, the objective function returns the evaluation result; S406. Obtain the optimal parameter combination through Bayesian optimization, and use the optimized proportional relationship to determine the transgenic properties of the material.

[0035] Furthermore, in step S402, the optimization target is optimized by an optimization function, the output of which is the accuracy of the classifier, and the optimized proportional relationship is used to distinguish transgenic events.

[0036] Furthermore, the accuracy calculation process is specifically based on the actual number of transgenic reads. , the true number of non-GM reads , the number of non-GM reads incorrectly identified as GM and the number of transgenic reads incorrectly identified as non-transgenic The confusion matrix composed of is calculated, namely: ; Among them, the represents the calculation result of accuracy, Represents the actual number of transgenic reads. Represents the true number of non-transgenic reads. represents the number of non-GM reads that were incorrectly identified as GM. Indicates the number of transgenic reads that were incorrectly identified as non-transgenic.

[0037] Furthermore, in step S404, the objective function is constructed according to the accuracy, that is: ; Among them, the Indicates that the hyperparameter configuration is The objective function is and ) represent the hyperparameter configurations respectively. The true number of transgenic reads, the true number of non-transgenic reads, the number of non-transgenic reads mistakenly judged as transgenic, and the number of transgenic reads mistakenly judged as non-transgenic, the They represent the threshold of the proportion of reads that are completely aligned with the reference genome, the threshold of the proportion of reads that cross the breakpoints, and the threshold of the proportion of reads that are completely aligned with the exogenous sequence.

[0038] Furthermore, the , , and The calculation process is expressed as follows: ; ; ; ; Among them, the The threshold value of the ratio of reads that fully align the exogenous sequence is used to reflect the insertion amount of the exogenous gene sequence in the transgenic event. Indicates the number of reads related to the qualified exogenous sequence within a specific interval, that is, the number of reads that are completely aligned with the exogenous sequence. Represents the total number of all reads. Indicates the ratio of reads that are completely aligned with the reference genome to the total reads. Indicates the number of reads that are completely aligned with the reference genome sequence, that is, the reads that completely match the reference genome. Indicates the ratio threshold of reads crossing the breakpoint, that is, the ratio of reads crossing the breakpoint to the total reads. Indicates the number of reads that cross the breakpoint. Indicates the ratio threshold of reads spanning breakpoints, that is, the ratio of reads spanning breakpoints to total reads.

[0039] Furthermore, in step S405, the expected improvement is used as the acquisition function, and the calculation process of the expected improvement of selecting an optimal hyperparameter combination that can improve the model performance is expressed as: ; Among them, the Indicates that the hyperparameter configuration is The expected improvement is Indicates the current optimal reads ratio in the Bayesian optimization process. Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the cumulative distribution function of the standard normal distribution, which is used to calculate the probability that the objective function value is lower than the current optimal value. Indicates that the hyperparameter configuration is The standard deviation of the prediction error is Represents the probability density function of the standard normal distribution and is used to calculate the probability density of the difference between the current best value and the predicted mean.

[0040] Specifically, when Close to or less than ,and If is large, the objective function under the current configuration has greater uncertainty; if near ,and If it is smaller, it means that the current configuration may be very close to the optimal solution, and the optimization will gradually converge.

[0041] Specifically, expected improvement (EI) is used to optimize the configuration of reads ratio in order to more accurately judge the nature of transgenic events. By calculating expected improvement (EI) in Bayesian optimization, the configuration that is most likely to bring improvement is selected from multiple candidate hyperparameters (such as reads ratio threshold), which can reduce the number of experiments and speed up the process of finding the optimal configuration.

[0042] Furthermore, in step S405, the proxy model is specifically represented as: ; Among them, the Represents a proxy model, the represents the Gaussian process function, Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the kernel function.

[0043] Specifically, a Gaussian process (GP) is a random process about an input space, which can be regarded as an infinite-dimensional vector, where each dimension represents a function value corresponding to an input point; illustratively, the kernel function selects the Mattern kernel, which has more flexibility in controlling smoothness compared to the commonly used Gaussian (RBF) kernel. The specific process is as follows: The basic form of the Mattern kernel is expressed as: ; Among them, the Represents two different input vectors, namely, feature vectors of different categories of reads, the representation, the represents the smoothness parameter of the Mattern kernel, The length scale parameter controls the range of influence between data points, Table verified by Mattern.

[0044] when When it approaches infinity, the Mattern kernel is close to the RBF kernel, indicating a very smooth function; when When , the Mattern kernel represents a discontinuous function. In this case, the kernel function has the weakest control over the smoothness between points and has a similar effect to the linear kernel function. when The larger it is, the larger the similarity range between data points is, that is, the distance between two points has less influence on the kernel function value; when The smaller it is, the smaller the similarity range between data points is, that is, the distance between two points has a greater impact on the kernel function value.

[0045] Furthermore, the covariance value is calculated for each pair of data points through the Mattern kernel, and the covariance value is used to reflect the similarity between reads of different categories.

[0046] Furthermore, the Gaussian process regression model is learned through the covariance matrix provided by the Mattern kernel, and the fluctuation of the proportion of reads of different categories is predicted based on the training data. On this basis, Bayesian optimization is used to adjust the hyperparameters in the model, optimize the fluctuation range of the proportion and improve the accuracy of material property judgment.

[0047] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing, characterized in that: The following steps are involved: S1. Confirming the intersection position of the exogenous sequence and the reference genome sequence in the transformant sequence of the target transgenic event through the alignment model, wherein the position is the breakpoint; S2. Perform next-generation sequencing on the sample to be tested to obtain the original clean reads, and align them with the transformant sequence to generate a BAM file; The alignment information in the BAM file was quality controlled to ensure that the mapping quality of the alignment was ≥40 and that the alignment included only one CIGAR operation; S3. According to the alignment results after quality control, reads were classified into three types: Ref_reads, Cross_reads, and GMO_reads; the three types of reads were counted, and the counting results were corrected by logarithmic transformation with base 10; S4. Calculate the ratio between reads of different categories and optimize the ratio fluctuation range through machine learning to determine the nature of the material.

2. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 1, characterized in that: In step S3, the reads are classified into three types: Ref_reads, Cross_reads and GMO_reads. The specific types are as follows: Reads that fully align to parts of the reference genome; Reads that partially align to the reference genome and partially align to the exogenous sequence, i.e., reads that span the breakpoints; Reads that fully align with the exogenous sequence.

3. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 1, characterized in that: In step S3, the specific classification process of classifying reads into three types of Ref_reads, Cross_reads and GMO_reads is as follows: When the CIGAR string only includes matching operations and is completely aligned to the reference genome, the read is Ref_reads; When the CIGAR string includes matching and insertion / deletion operations, and the two aligned regions are from the reference genome and the exogenous sequence respectively, the read is Cross_reads; When the CIGAR string only includes matching operations and is completely aligned to the exogenous sequence, the read is GMO_reads.

4. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 1, characterized in that: The step S4 specifically includes the following sub-steps: S401. According to the classification results of reads, the number of reads in each category is calculated to obtain the ratio between reads in different categories, and the ratio between reads in different categories is used as the feature of model input; S402. Taking maximizing classification accuracy as the optimization goal; S403. Define a hyperparameter space for Bayesian optimization; S404. Constructing an objective function, and adjusting the classification accuracy by selecting different ratio thresholds and fluctuation ranges; S405. Use the Gaussian process as a proxy model to predict the behavior of the objective function and guide the next evaluation point based on the current evaluation result; use the expected improvement as the acquisition function to select an optimal hyperparameter combination that can improve the model performance; By iteratively updating the proxy model, the acquisition function is used to guide Bayesian optimization to find the optimal hyperparameters; after each iteration, the objective function returns the evaluation result; S406. Obtain the optimal parameter combination through Bayesian optimization, and use the optimized proportional relationship to determine the transgenic properties of the material.

5. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 4, characterized in that: In step S402, the optimization target is optimized by an optimization function, the output of which is the accuracy of the classifier, and the optimized proportional relationship is used to distinguish transgenic events.

6. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 5, characterized in that: The accuracy calculation process is specifically based on the actual number of transgenic reads. , the true number of non-GM reads , the number of non-GM reads incorrectly identified as GM and the number of transgenic reads incorrectly identified as non-transgenic The confusion matrix composed of is calculated, namely: ； Among them, the represents the calculation result of accuracy, Represents the actual number of transgenic reads. Represents the true number of non-transgenic reads. represents the number of non-GM reads that were incorrectly identified as GM. Indicates the number of transgenic reads that were incorrectly identified as non-transgenic.

7. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 4, characterized in that: In step S404, the objective function is constructed according to the accuracy, that is: ； ； Among them, the Indicates that the hyperparameter configuration is The objective function is and ) represent the hyperparameter configurations respectively. The true number of transgenic reads, the true number of non-transgenic reads, the number of non-transgenic reads mistakenly judged as transgenic, and the number of transgenic reads mistakenly judged as non-transgenic, the They represent the threshold of the proportion of reads that are completely aligned with the reference genome, the threshold of the proportion of reads that cross the breakpoints, and the threshold of the proportion of reads that are completely aligned with the exogenous sequence.

8. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 7, characterized in that: Said , , and The calculation process is expressed as follows: ； ； ； ； Among them, the The threshold value of the ratio of reads that fully align the exogenous sequence is used to reflect the insertion amount of the exogenous gene sequence in the transgenic event. Indicates the number of reads related to the qualified exogenous sequence within a specific interval, that is, the number of reads that are completely aligned with the exogenous sequence. Represents the total number of all reads. Indicates the ratio of reads that are completely aligned with the reference genome to the total reads. Indicates the number of reads that are completely aligned with the reference genome sequence, that is, the reads that completely match the reference genome. Indicates the ratio threshold of reads crossing the breakpoint, that is, the ratio of reads crossing the breakpoint to the total reads. Indicates the number of reads that cross the breakpoint. Indicates the ratio threshold of reads spanning breakpoints, that is, the ratio of reads spanning breakpoints to total reads.

9. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 4, characterized in that: In step S405, the expected improvement is used as the acquisition function, and the calculation process of the expected improvement of selecting an optimal hyperparameter combination that can improve the model performance is expressed as: ； Among them, the Indicates that the hyperparameter configuration is The expected improvement is Indicates the current optimal reads ratio in the Bayesian optimization process. Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the cumulative distribution function of the standard normal distribution, which is used to calculate the probability that the objective function value is lower than the current optimal value. Indicates that the hyperparameter configuration is The standard deviation of the prediction error is Represents the probability density function of the standard normal distribution and is used to calculate the probability density of the difference between the current best value and the predicted mean.

10. The method for identifying unknown exogenous fragment insertion into transgenic events based on targeted sequencing according to claim 4, characterized in that: In step S405, the proxy model is specifically represented as: ； Among them, the Represents a proxy model, the represents the Gaussian process function, Indicates that the hyperparameter configuration is The classification accuracy of reads predicted by the model is Represents the kernel function.

Images