Whole genome prediction method based on multi-omics integration

Through the multi-omics integrated genome prediction method, multiple prediction models are constructed and cross-validated training is carried out, which solves the problems of low prediction accuracy of complex traits and poor multi-omics integration in the existing technology, and improves the phenotype prediction accuracy of rice hybrids.

CN120108492AActive Publication Date: 2025-06-06YANGZHOU UNIV

Patent Information

Application Number
CN202510165371.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-14
Publication Date
2025-06-06
Estimated Expiration
2045-02-14

AI Technical Summary

Technical Problem

Existing genome-wide selection methods are difficult to capture complex interactions between genes and their downstream regulation, resulting in low prediction accuracy of complex traits and the multi-omics integration method has not improved or reduced the prediction power of some traits.

Method used

A genome-wide prediction method based on multiomics integration was adopted, and the prediction model of genome-phenotype, metabolomics and transcriptome data was obtained, and the prediction matrix was preprocessed and integrated, and the genome-phenotype, metabolomics-phenotype, transcriptome-phenotype, and transcriptome-phenotype were constructed. The model was trained using the ten-fold cross-validation training model, and the prediction matrix was finally input into the linear regression model for prediction.

Benefits of technology

It improves the prediction accuracy of rice hybrid phenotype, improves the comprehensive effect of multi-omic predictive power, and exceeds the limitations of single-omic predictive power.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120108492A_ABST
    Figure CN120108492A_ABST
Patent Text Reader

Abstract

The invention belongs to the field of plant breeding, and provides a whole genome prediction method based on multi-omics integration. Comprising the steps of genome data acquisition, metabolome data acquisition, transcriptome data acquisition, phenotype data acquisition, data preprocessing, hybrid data speculation, training set and test set division, prediction model construction, prediction model training, linear regression model training and phenotype value prediction. According to the method, the genome-phenotype prediction model, the metabolome-phenotype prediction model, the transcriptome-phenotype prediction model and the linear regression model are constructed, and ten-fold cross validation is used for training the models, so that the multi-omics predictive power of the models is improved, and the rice hybrid phenotype prediction precision is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of plant breeding, and in particular to a whole genome prediction method based on multi-omics integration. Background Art

[0002] Whole genome selection is abbreviated as GS. GS breeding uses the association between genetic markers and phenotypes across the whole genome to construct statistical models and then predict individuals with unknown phenotypes. Compared with traditional plant breeding methods, it can improve predictive power and genetic gain. However, even with complete gene sequences, GS still finds it difficult to capture the complex interactions between genes and their downstream regulation, resulting in a bottleneck in the accuracy of complex trait prediction. Currently, low-cost and high-throughput molecular technologies are developing rapidly, which can accurately quantify metabolites and transcripts. Since the metabolome and transcriptome are between the genotype and phenotype information flows and contain rich site interaction information, integrating these data is a way to break through the GS prediction bottleneck.

[0003] Existing multi-omics integration methods, such as directly merging the genotype matrix, metabolite matrix, and transcript matrix into a large matrix by column, which can be input into the model for training and phenotype prediction, have made some progress. However, for some traits, the predictive power has not been improved and is only equivalent to or even lower than that of a single omics. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the object of the present invention is to provide a whole genome prediction method based on multi-omics integration to improve the prediction accuracy of rice hybrid phenotypes.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A genome-wide prediction method based on multi-omics integration, including:

[0007] Obtain parental genomic data, parental metabolomic data, parental transcriptomic data, and hybrid phenotypic data;

[0008] Filtering out SNPs with a minimum allele frequency lower than 0.05 and a missing rate greater than 0.2 in the parental genome data to obtain preprocessed genome data, and performing standardization processing on the parental metabolome data and the parental transcriptome data, respectively, to obtain preprocessed metabolome data and preprocessed transcriptome data;

[0009] Inferring a hybrid genome matrix, a hybrid metabolome matrix, and a hybrid transcriptome matrix respectively according to the pre-processed genome data, the pre-processed metabolome data, and the pre-processed transcriptome data;

[0010] The hybrid phenotypic data, the hybrid genome matrix, the hybrid metabolome matrix, and the hybrid transcriptome matrix are integrated and randomly divided into a training set and a test set;

[0011] The pre-constructed genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model are trained using the training set according to ten-fold cross validation to obtain a training prediction matrix, and the trained genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model are used to predict the test set to obtain a test prediction matrix;

[0012] The training prediction matrix is ​​input as a feature into a linear regression model for training, and the linear regression model is predicted according to the test prediction matrix to obtain a final hybrid phenotype prediction value.

[0013] Preferably, it also includes:

[0014] Calculating the coefficient of determination between the final predicted phenotypic value and the observed phenotypic value of the test set to obtain an evaluation index;

[0015] The mean of the evaluation index of several trainings is obtained to obtain the final prediction power of the model.

[0016] Preferably, the ratio of the training set to the test set is 9:1.

[0017] Preferably, the hybrid genome matrix is ​​expressed as:

[0018]

[0019] The expression of the hybrid metabolome matrix is:

[0020]

[0021] The expression of the hybrid transcriptome matrix is:

[0022]

[0023] Wherein, G_m, M_m, T_m are the hybrid paternal genome matrix, the hybrid paternal metabolome matrix, and the hybrid paternal transcriptome matrix, respectively; G_f, M_f, T_f are the hybrid maternal genome matrix, the hybrid maternal metabolome matrix, and the hybrid maternal transcriptome matrix, respectively. Preferably, the model expressions of the genome-phenotype prediction model, the metabolome-phenotype prediction model, and the transcriptome-phenotype prediction model are:

[0024] y train =1μ+Oα+ε;

[0025] in, i,j=1,2,...,...n;y train is the hybrid phenotype observation value vector of the training set; μ is the fixed effect; O is any one of the hybrid genome matrix, the hybrid metabolome matrix and the hybrid transcriptome matrix in the training set; ε is the residual vector; h is the bandwidth parameter; p is any one of the number of SNPs, the number of metabolites and the number of transcripts; α is any one of the genomic random effect, the metabolome random effect and the transcriptome random effect; K is any one of the genomic Gaussian kernel, the metabolome Gaussian kernel and the transcriptome Gaussian kernel; is any one of the genomic variance, metabolomic variance, and transcriptomic variance; D ij is the Euclidean distance between the i-th sample and the j-th sample of the hybrid in the training set; n is the number of hybrids in the training set.

[0026] Preferably, the expression of the training prediction matrix is:

[0027]

[0028] The expression of the test prediction matrix is:

[0029]

[0030] in, and Respectively represent the predicted values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the training set; and They respectively represent the means of 10 prediction values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the test set; k represents taking the k-th fold data as the test data of the model when training the model using cross-validation.

[0031] Preferably, the linear regression model is expressed as:

[0032]

[0033] Among them, q∈{1,2,3}; is the ith phenotypic observation of the hybrid based on the training set; b is the intercept term; w 1 、w 2 、w 3 are the first regression coefficient, the second regression coefficient, and the third regression coefficient respectively; ε is the error term; I A is the indicative function; for The i-th element in the first column.

[0034] The present invention discloses the following technical effects:

[0035] The present invention provides a whole genome prediction method based on multi-omics integration, which solves the problem that the prior art only improves the prediction power of a single omic by constructing a genome-phenotype prediction model, a metabolome-phenotype prediction model, a transcriptome-phenotype prediction model and a linear regression model and training the model using ten-fold cross validation, thereby achieving an improvement in the multi-omics prediction power of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1 A schematic diagram of a whole genome prediction process based on multi-omics integration provided in an embodiment of the present invention;

[0038] Figure 2 Box plots of the prediction power of nine prediction models provided in the embodiments of the present invention for four traits of rice hybrids. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] The purpose of the present invention is to provide a whole genome prediction method based on multi-omics integration to improve the prediction accuracy of rice hybrid phenotypes.

[0041] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] Figure 1 A schematic diagram of a whole genome prediction process based on multi-omics integration provided in an embodiment of the present invention is shown in FIG. Figure 1 As shown, the present invention provides a whole genome prediction method (MOEGS) based on multi-omics integration, comprising:

[0043] Step 100: Obtaining parental genome data, parental metabolome data, parental transcriptome data, and hybrid phenotypic data;

[0044] Step 200: filtering out SNPs (single nucleotide polymorphisms) with a minimum allele frequency lower than 0.05 and a deletion rate greater than 0.2 in the parental genome data to obtain preprocessed genome data, and performing standardization processing on the parental metabolome data and the parental transcriptome data to obtain preprocessed metabolome data and preprocessed transcriptome data;

[0045] Step 300: inferring a hybrid genome matrix, a hybrid metabolome matrix, and a hybrid transcriptome matrix respectively according to the preprocessed genome data, the preprocessed metabolome data, and the preprocessed transcriptome data;

[0046] Step 400: Integrate the hybrid phenotypic data, the hybrid genome matrix, the hybrid metabolome matrix, and the hybrid transcriptome matrix and randomly divide them into a training set and a test set;

[0047] Step 500: training the pre-constructed genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model using the training set according to ten-fold cross validation to obtain a training prediction matrix, and predicting the test set using the trained genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model to obtain a test prediction matrix;

[0048] Step 600: Input the training prediction matrix as a feature into the linear regression model for training, and make predictions on the linear regression model according to the test prediction matrix to obtain the final hybrid phenotype prediction value.

[0049] Furthermore, it also includes:

[0050] Calculating the coefficient of determination between the final predicted phenotypic value and the observed phenotypic value of the test set to obtain an evaluation index;

[0051] The mean of the evaluation index of several trainings is obtained to obtain the final prediction power of the model.

[0052] Preferably, the ratio of the training set to the test set is 9:1.

[0053] Furthermore, the expression of the hybrid genome matrix is:

[0054]

[0055] The expression of the hybrid metabolome matrix is:

[0056]

[0057] The expression of the hybrid transcriptome matrix is:

[0058]

[0059] Among them, G_m, M_m, and T_m are the hybrid paternal genome matrix, hybrid paternal metabolome matrix, and hybrid paternal transcriptome matrix, respectively; G_f, M_f, and T_f are the hybrid maternal genome matrix, hybrid maternal metabolome matrix, and hybrid maternal transcriptome matrix, respectively.

[0060] Specifically, the model expressions of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model are:

[0061] y train =1μ+Oα+ε;

[0062] in, i,j=1,2,...,...n;y train is the hybrid phenotype observation value vector of the training set; μ is the fixed effect; O is any one of the hybrid genome matrix, the hybrid metabolome matrix and the hybrid transcriptome matrix in the training set; ε is the residual vector; h is the bandwidth parameter; p is any one of the number of SNPs, the number of metabolites and the number of transcripts; α is any one of the genomic random effect, the metabolome random effect and the transcriptome random effect; K is any one of the genomic Gaussian kernel, the metabolome Gaussian kernel and the transcriptome Gaussian kernel; is any one of the genomic variance, metabolomic variance, and transcriptomic variance; D ij is the Euclidean distance between the i-th sample and the j-th sample of the hybrid in the training set; n is the number of hybrids in the training set.

[0063] Furthermore, the expression of the training prediction matrix is:

[0064]

[0065] The expression of the test prediction matrix is:

[0066]

[0067] in, and Respectively represent the predicted values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the training set; and They respectively represent the means of 10 prediction values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the test set; k represents taking the k-th fold data as the test data of the model when training the model using cross-validation.

[0068] Specifically, the linear regression model is expressed as:

[0069]

[0070] Among them, q∈{1,2,3}; is the ith phenotypic observation of the hybrid based on the training set; b is the intercept term; w 1 、w 2 、w 3 are the first regression coefficient, the second regression coefficient, and the third regression coefficient respectively; ε is the error term; I A is the indicative function; for The i-th element in the first column.

[0071] Specifically, q represents the indicative function I A When q=1

[0072]

[0073] Only and It is the predicted value of the hybrid phenotype based on the training set by the genome-phenotype prediction model and the metabolome-phenotype prediction model, and is therefore a genome-metabolome integration. When q = 2

[0074]

[0075] Only and It is the prediction value of the hybrid phenotype based on the training set by the genome-phenotype prediction model and the transcriptome-phenotype prediction model, and is therefore a genome-transcriptome integration. When q = 3

[0076]

[0077] and At the same time, the genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model have predicted the hybrid phenotype based on the training set, and are therefore a genome-metabolome-transcriptome integration.

[0078] Preferably, the genome-phenotype prediction model, the metabolome-phenotype prediction model, and the transcriptome-phenotype prediction model are trained based on the training set using ten-fold cross validation, and the test set is predicted to obtain a training prediction matrix and a test prediction matrix. The specific process is: the training set is randomly divided into 10 parts, and in each iteration, one of the parts is used as test data and the remaining 9 parts are used as training data, and then the parameters estimated by the training data are used to predict the hybrid phenotype value of the test data, and the hybrid phenotype value of the test set is predicted at the same time. After 10 rounds of iterations, three hybrid phenotype prediction value vectors based on the genome-phenotype prediction model, the metabolome-phenotype prediction model, and the transcriptome-phenotype prediction model on the training set are obtained, and three hybrid phenotype prediction value vectors based on the genome-phenotype prediction model, the metabolome-phenotype prediction model, and the transcriptome-phenotype prediction model on the test set are obtained, and finally the hybrid phenotype prediction value vectors based on the training set and the test set are respectively merged to obtain the final training prediction matrix and the test prediction matrix.

[0079] Optionally, the training prediction matrix is ​​input as a feature into a linear regression model for training, and the linear regression model is predicted based on the test prediction matrix to obtain a final hybrid phenotype prediction value. The specific process is: the training prediction matrix is ​​used as a new feature, and the hybrid phenotype observation value vector in the training set is input as a label into the linear regression model for training, and the trained linear regression model is used to predict on the test prediction matrix to obtain the final hybrid phenotype prediction value.

[0080] Specifically, the method provided in this embodiment is described in detail below:

[0081] 1) Data Acquisition

[0082] The genome, metabolome, transcriptome data of the parents and the phenotypic data of the hybrids are obtained, and the genome, metabolome and transcriptome of the hybrids are inferred based on the genome, metabolome and transcriptome of the parents. This example uses a set of publicly available rice data sets for experiments, including 210 parents, 278 hybrids, 1619 bins of the parents from 270820 SNPs, 1000 metabolites, 24994 gene expressions, and four yield-related traits: yield per plant, number of tillers, number of grains per panicle, and thousand-grain weight.

[0083] 2) Data preprocessing

[0084] The parental genome, metabolome, and transcriptome data were preprocessed. For the genome data, SNPs with a minimum allele frequency lower than 0.05 and a missing rate greater than 0.2 were filtered out, and for the metabolome and transcriptome data, they were standardized.

[0085] 3) Infer the genome, metabolome and transcriptome of hybrids

[0086] Assume that the hybrid paternal genome, metabolome and transcriptome matrices are G_m, M_m and T_m respectively, and the maternal genome, metabolome and transcriptome matrices are G_f, M_f and T_f respectively. Then the hybrid genome, metabolome and transcriptome matrices are:

[0087]

[0088] 4) Construction of omics-phenotype prediction model

[0089] Dataset division: The data set was randomly divided into 10 parts using ten-fold cross validation, one of which was used as the test set, and the remaining 9 were used as training sets. The genome-phenotype prediction model, metabolome-phenotype prediction model, and transcriptome-phenotype prediction model were constructed based on the training set. The model expression is as follows:

[0090] y train =1μ+Oα+ε

[0091] Among them, y train is the vector of hybrid phenotype observations based on the training set, and μ is the fixed effect.

[0092] 5) Omics-phenotype prediction model training

[0093] The above model is trained using 10-fold cross validation, and predictions are made on the test set. When the model is trained, a prediction matrix based on the training set is obtained:

[0094]

[0095] And a prediction matrix based on the test set:

[0096]

[0097] 6) Secondary training

[0098] The predicted values ​​on the training set are used as new features to input into the linear regression model for training, and predictions are made based on the predicted values ​​on the test set to obtain the final hybrid phenotype prediction values. The model expression is:

[0099]

[0100] Use the trained linear regression model to predict the phenotypic value of the test set, and finally predict the phenotypic value for:

[0101]

[0102] where q∈{1,2,3}, is the phenotypic prediction value of the sth sample of the hybrid based on the test set, s = 1, 2, ... m, m is the number of hybrids in the test set, is the estimated value of the intercept term of the linear regression model obtained by estimating the training set, and are the regression coefficients of linear regression estimated by the training set, I A is an indicative function. When q=1, it indicates genome-metabolome integration; when q=2, it indicates genome-transcriptome integration; and when q=3, it indicates genome-metabolome-transcriptome integration.

[0103] 7) Model Evaluation

[0104] Using the predicted phenotypic values ​​obtained by the linear regression model and the observed phenotype values ​​of the test set (y test ) between the coefficient of determination (R 2 ) as the evaluation index of the model. After each fold in the divided data set is trained as a test set, the R of 10 training runs can be obtained. 2 , 10 times R 2 The average R of a ten-fold cross validation can be obtained 2 In order to ensure the randomness and reliability of the results, the process was repeated 20 times and the R 2 The mean of the values ​​is taken as the final predictive power of the model.

[0105] refer to Figure 2, the four traits are yield per plant, number of tillers, number of grains per panicle, and 1000-grain weight; the nine prediction models are G, M, T, GM, GT, GMT, G+M, G+T, and G+M+T, which respectively represent genome prediction, metabolome prediction, transcriptome prediction, integrated genome-metabolome prediction, integrated genome-transcriptome prediction, integrated genome-metabolome-transcriptome prediction, genome-metabolome integrated prediction, genome-transcriptome integrated prediction, and genome-metabolome-transcriptome integrated prediction. In each box plot, different lowercase letters at the bottom indicate significant differences between different models. The above methods were used to predict yield per plant, number of tillers, number of grains per panicle, and 1000-grain weight of rice hybrids; the cross-validation results showed that compared with genome prediction, MOEGS significantly improved the predictive power of traits. The genome-metabolome integration improved the predictive power of yield per plant, number of tillers, number of grains per panicle and 1000-grain weight by 48.4%, 11.8%, 17.1% and 1.5%, respectively; the genome-transcriptome integration improved it by 33.3%, 12.1%, 16.0% and 2.5%; and the genome-metabolome-transcriptome integration improved it by 45.8%, 10.4%, 17.8% and 2.7%. In addition, the predictive power of MOEGS is higher than that of tandem integrated multi-omics; compared with genome-metabolome integration, genome-metabolome integration increased the predictive power of yield per plant, number of tillers and number of grains per panicle by 12.9%, 9.6% and 3.7%, respectively; compared with genome-transcriptome integration, genome-transcriptome integration increased the predictive power of yield per plant, number of tillers, number of grains per panicle and 1000-grain weight by 21.3%, 40.9%, 9.7% and 1.8%, respectively; compared with genome-metabolome-transcriptome integration, genome-metabolome-transcriptome integration increased the predictive power of yield per plant, number of tillers, number of grains per panicle and 1000-grain weight by 32.6%, 33.5%, 10.3% and 1.5%, respectively.

[0106] The beneficial effects of the present invention are as follows:

[0107] The present invention improves the multi-omics prediction ability of the model and the prediction accuracy of rice hybrid phenotype by constructing a genome-phenotype prediction model, a metabolome-phenotype prediction model, a transcriptome-phenotype prediction model and a linear regression model and training the model using ten-fold cross validation.

[0108] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0109] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A genome-wide prediction method based on multi-omics integration, characterized in that: include: Obtain parental genomic data, parental metabolomic data, parental transcriptomic data, and hybrid phenotypic data; Filtering out SNPs with a minimum allele frequency lower than 0.05 and a missing rate greater than 0.2 in the parental genome data to obtain preprocessed genome data, and performing standardization processing on the parental metabolome data and the parental transcriptome data, respectively, to obtain preprocessed metabolome data and preprocessed transcriptome data; Inferring a hybrid genome matrix, a hybrid metabolome matrix, and a hybrid transcriptome matrix respectively according to the pre-processed genome data, the pre-processed metabolome data, and the pre-processed transcriptome data; The hybrid phenotypic data, the hybrid genome matrix, the hybrid metabolome matrix, and the hybrid transcriptome matrix are integrated and randomly divided into a training set and a test set; The pre-constructed genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model are trained using the training set according to ten-fold cross validation to obtain a training prediction matrix, and the trained genome-phenotype prediction model, metabolome-phenotype prediction model and transcriptome-phenotype prediction model are used to predict the test set to obtain a test prediction matrix; The training prediction matrix is ​​input as a feature into a linear regression model for training, and the linear regression model is predicted according to the test prediction matrix to obtain a final hybrid phenotype prediction value.

2. A genome-wide prediction method based on multi-omics integration according to claim 1, characterized in that: Also includes: Calculating the coefficient of determination between the final predicted phenotypic value and the observed phenotypic value of the test set to obtain an evaluation index; The mean of the evaluation index of several trainings is obtained to obtain the final prediction power of the model.

3. A genome-wide prediction method based on multi-omics integration according to claim 1, characterized in that: The ratio of the training set to the test set is 9:

1.

4. The whole genome prediction method based on multi-omics integration according to claim 1, characterized in that: The expression of the hybrid genome matrix is: The expression of the hybrid metabolome matrix is: The expression of the hybrid transcriptome matrix is: Among them, G_m, M_m, and T_m are the hybrid paternal genome matrix, hybrid paternal metabolome matrix, and hybrid paternal transcriptome matrix, respectively; G_f, M_f, and T_f are the hybrid maternal genome matrix, hybrid maternal metabolome matrix, and hybrid maternal transcriptome matrix, respectively.

5. A genome-wide prediction method based on multi-omics integration according to claim 4, characterized in that: The model expressions of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model are: y train =1μ+Oα+ε; in, y train is the hybrid phenotype observation value vector of the training set; μ is the fixed effect; O is any one of the hybrid genome matrix, the hybrid metabolome matrix and the hybrid transcriptome matrix in the training set; ε is the residual vector; h is the bandwidth parameter; p is any one of the number of SNPs, the number of metabolites and the number of transcripts; α is any one of the genomic random effect, the metabolome random effect and the transcriptome random effect; K is any one of the genomic Gaussian kernel, the metabolome Gaussian kernel and the transcriptome Gaussian kernel; is any one of the genomic variance, metabolomic variance, and transcriptomic variance; D ij is the Euclidean distance between the i-th sample and the j-th sample of the hybrid in the training set; n is the number of hybrids in the training set.

6. A genome-wide prediction method based on multi-omics integration according to claim 5, characterized in that: The expression of the training prediction matrix is: The expression of the test prediction matrix is: in, and Respectively represent the predicted values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the training set; and They respectively represent the means of 10 prediction values ​​of the genome-phenotype prediction model, the metabolome-phenotype prediction model and the transcriptome-phenotype prediction model on the test set; k represents taking the k-th fold data as the test data of the model when training the model using cross-validation.

7. A genome-wide prediction method based on multi-omics integration according to claim 6, characterized in that: The linear regression model is expressed as: Among them, q∈{1,2,3}; ytrain i is the i-th phenotypic observation value of the hybrid based on the training set; b is the intercept term; w1, w2, w3 are the first regression coefficient, the second regression coefficient and the third regression coefficient respectively; ε is the error term; I A is the indicative function; for The i-th element in the first column.

Citation Information

Patent Citations

  • Hybrid seed prediction method based on Bayesian model integrating parent phenotypes

    CN113053459A

  • Multi-view GBLUP method for integrating multiple types of data for phenotype prediction

    CN117153247A

  • Maize hybrid phenotype prediction method integrating genome and rhizosphere microbiomics data

    CN117637023A

  • Methods and systems for genome analysis

    US20160092631A1

  • Predicting disease outcomes using machine learned models

    US20210366577A1

Cited By

  • New variety breeding method for improving yield of bitter gourds

    CN121694206A