Joint complete interval mapping method for multiple parent-derived offspring groups

Abstract

The invention relates to a combined complete interval mapping method for multiple parent-derived offspring groups, which comprises the following steps of: obtaining genotype data of the multiple parent-derived offspring groups, and constructing an integrated linkage map of the genotype data; respectively constructing orthogonal indication variables of marked genotypes for each group, and establishing a linear relation model between phenotypes and marked genotypes; based on the established linear relation model between the phenotype and the genotype of the single population, selecting a significant marker in each population by using stepwise regression, and correcting the phenotype value by using the significant marker; hypothesis testing is established, joint test statistics are calculated through interval scanning, the QTL is positioned, the position and the genetic effect of the QTL are estimated, and complete interval mapping of multi-group joint QTL positioning is achieved. According to the method, combined complete interval mapping with background control can be carried out for the parent-derived offspring group, the problem that a combined QTL positioning method for a plurality of mapping groups is lacked at present is solved, and the accuracy of a gene positioning result is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of molecular biology and genetic breeding, and particularly relates to a joint complete interval mapping method for multiple parent-derived offspring populations. Background Art

[0002] The process of finding the location of quantitative trait loci (QTL) on chromosomes and estimating their genetic effects is called QTL mapping, which has become the main method for studying the genetic mechanism of complex traits. Common methods include additive dominant QTL mapping and epistatic QTL mapping. Based on the results of QTL mapping, genes can be finely mapped, or the located genes can be used as markers for assisted selection of traits (Wang Jiankang 2017). Commonly used QTL mapping methods include single marker mapping (SMA; Sax 1923), interval mapping (IM; Lander and Botstein 1989), composite interval mapping (CIM; Zeng 1994), multiple interval mapping (MIM; Kao et al. 1999), complete interval mapping (ICIM; Li et al. 2007; Zhang et al. 2008), and genome-wide composite interval mapping (GCIM; Wen et al. 2019). Commonly used QTL positioning tools include R / qtl (Broman et al. 2003), QTL Cartographer (Wang et al. 2005), MAPQTL (van Ooijen 2009), QTL IciMapping (Meng et al. 2015), etc. The efficacy of QTL detection is related to the size of the mapping population, the heritability of the trait, the QTL effect, the marker density, and whether there is linkage or interaction between QTLs (Wang et al. 2020).

[0003] In genetic research, in order to achieve specific research goals, it is often necessary to construct multiple genetic populations. It is usually necessary to analyze these populations together. For example, the six-generation joint analysis method in quantitative genetics can use six basic generations (P 1 , P 2 、F 1 、F 2 , B.C. 1 , B.C. 2 ) phenotypic data, and established an additive-dominant model for joint analysis, calculated the genetic variance and heritability of each generation, estimated the number of effective factors, and preliminarily analyzed the genetic effects, but could not carry out further genetic analysis (Wang Jiankang 2017). Pure line populations can accurately estimate additive effects, F 2Or other temporary populations can be used for dominance effect estimation and heterosis research. The comprehensive use of the two types of populations is conducive to obtaining accurate estimates of additive dominance effects. Multi-population joint analysis is of great significance for heterosis research, accurate estimation of genetic effects, and mining more excellent allele variations at important gene loci.

[0004] However, the current multi-population joint analysis usually first conducts genetic analysis in a single population, and then compares and integrates the analysis results, or merges them into a large population for association analysis. Different types of populations have different genetic structures, and the constructed linkage maps may have large differences in the number of linkage groups, the number of markers, the order of marker positions, and the average map distance (Liang Yongshu et al. 2007). When conducting comparative analysis of different populations, it is often necessary to construct a set of integrated maps to integrate the genetic information of multiple populations and increase the marker density (Galeano et al. 2011). The integration of QTL positioning results of different genetic populations is often achieved through QTL meta-analysis. Meta-analysis comprehensively utilizes the QTL information located in specific chromosome segments of multiple genetic populations to determine the QTL position (Arcade et al. 2004). However, when constructing an integrated map, it is impossible to fully utilize the recombination rate information between two markers in a single population. Meta-analysis cannot estimate the effect of consistent QTLs, nor can it reflect genetic information such as population size, trait heritability, and micro-effect QTLs that do not reach critical values ​​in a single population.

[0005] Therefore, this field urgently needs a method to directly use the original genotype and phenotypic information of each population to carry out joint gene positioning, fully explore the genetic information of each population, and improve the accuracy of QTL mapping results. Summary of the invention

[0006] The purpose of the present invention is to provide a multi-population joint complete interval mapping method for parent-derived offspring populations, to achieve joint QTL (Quantitative Trait Locus) positioning with background control in multiple mapping populations, and to improve the accuracy of genetic analysis.

[0007] The technical solution for which the present invention seeks protection is as follows:

[0008] A joint complete interval mapping method for multiple parent-derived offspring populations comprises the following steps:

[0009] S1: obtaining genotype data of multiple parent-derived offspring populations; the genotype data of the multiple parent-derived offspring populations have a certain number of common molecular markers;

[0010] S2: constructing an integrated linkage map using the genotype data of the plurality of parent-derived offspring populations, and collecting and collating the phenotypic data of each population;

[0011] S3: construct orthogonal indicator variables of marker genotypes for each population, and establish a linear relationship model between phenotype and marker genotype;

[0012] S4: Based on the linear relationship model between the phenotype and marker genotype of a single population established in step S3, significant markers are selected in each population using stepwise regression, and the phenotypic values ​​are corrected using the significant markers, and interval mapping with background control for multiple populations is performed based on the complete interval mapping method;

[0013] S5: Establish hypothesis tests, calculate joint test statistics through interval scanning, locate QTLs, and estimate their positions and genetic effects, thereby achieving complete interval mapping for joint QTL positioning in multiple populations.

[0014] The plurality of parent-derived offspring populations may be two or more parent-derived offspring populations.

[0015] The population types of the parent-derived offspring populations may be identical or inconsistent, and the parents may be the same or different.

[0016] The parent-derived progeny population includes a hybrid F 2 or F 2:3 Derived populations, backcross (BC) populations, recombinant inbred lines (RIL) populations, and double haploid (DH) populations, etc.

[0017] The certain number of common molecular markers may be 10 or more common molecular markers, and an increase in the number of common molecular markers is conducive to the construction of an integrated linkage map.

[0018] In step S1, each population can be sequenced separately using the same sequencing platform to obtain common molecular markers among the populations.

[0019] In step S2, the integrated linkage map can be obtained using genetic analysis software that includes an integrated map construction function. The genetic analysis software can be QTL IciMapping, JoinMap, BioMercator, etc.

[0020] An integrated linkage map is a genetic linkage map constructed by integrating genetic information from different genetic populations.

[0021] In step S3, constructing orthogonal indicator variables of marker genotypes refers to defining additive effect indicator variables and dominant effect indicator variables.

[0022] Orthogonal indicator variables are a method used in statistical analysis to deal with categorical variables (also called nominal variables or discrete variables). The purpose of this method is to transform the categorical variables into a form that makes these variables independent or orthogonal in the model, thus avoiding the problem of multicollinearity.

[0023] In step S3 of the above method, if each locus in the hth population contains 2 genotypes, the linear relationship model is:

[0024]

[0025] If each seat in the hth population contains 3 genotypes, the linear relationship model is:

[0026]

[0027] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the integrated linkage map; m is the number of marker intervals;

[0028] y ih is the phenotypic value of the i-th individual in the h-th population;

[0029] b 0h is the mean of the linear relationship model between the phenotype and marker genotype of the hth population;

[0030] x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0031] x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0032] z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0033] z i(j+1)h is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0034] b jh 、cjh , b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect;

[0035] ε ih is the random error effect of the hth population and follows a normal distribution.

[0036] In step S4, computer programming or R language, SAS software, etc. can be used to perform stepwise regression based on the linear relationship model between the phenotype and marker genotype of the single population established in step S3, with the additive effect indicator variable x and the dominant effect indicator variable z as independent variables and the phenotypic value y as the dependent variable, so as to select significant markers.

[0037] Stepwise regression is a method for selecting independent variables in a linear regression model. The basic idea is to introduce variables one by one, provided that their partial regression sum of squares is empirically significant. At the same time, after each new variable is introduced, the old variables that have been selected for the regression model are tested one by one, and the variables that are considered insignificant after testing are deleted to ensure that every independent variable in the obtained independent variable subset is significant. This process goes through several steps until no new variables can be introduced. At this time, all independent variables in the regression model are significant to the dependent variable.

[0038] In step S4 of the above method, if each locus in the hth population contains two genotypes, the formula for correcting the phenotypic value is as follows:

[0039]

[0040] If each locus in the hth population contains three genotypes, the formula for correcting the phenotypic value is as follows:

[0041]

[0042] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the common molecular marker; k represents the number of the left end marker of the current scanning interval;

[0043] Δy ih is the phenotypic correction value of the i-th individual in the h-th population; y ih is the phenotypic value of the i-th individual in the h-th population;

[0044] Yes jh An estimated value of It is c jh An estimated value of Yes jh,(j+1)hAn estimated value of It is c jh,(j+1)h The estimated value of jh 、c jh , b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect;

[0045] x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0046] x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0047] z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0048] z i(j+1)h It is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population. When the genotype is the genotype of parent 1, the heterozygous genotype and the genotype of parent 2, the values ​​are 0, 1 and 0 respectively.

[0049] In step S5 of the above method, the hypothesis test is:

[0050] If each locus in the hth population contains 2 genotypes, then H 0 :μ 1h =μ 2h (h=1,…,p), where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population;

[0051] If each locus in the hth population contains three genotypes, then H 0 :μ 1h =μ 2h =μ 3h (h=1,…,p), where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq, and qq in the hth population;

[0052] H 1: The QTL genotype values ​​in at least one population are not completely equal;

[0053] Among them, H 0 is the null hypothesis, H 1 is the alternative hypothesis, p represents the number of groups;

[0054] The genotype mean is the average of all individual genotype values ​​corresponding to the marker genotype.

[0055] The QTL genotype value refers to the average of all individual genotype values ​​corresponding to the genotype.

[0056] The joint test statistic is expressed as a joint LOD value; the joint LOD value is the sum of the individual LOD values ​​in all populations and is used to determine the presence of QTL; the LOD value of a single population (LOD h ) is the sum of the individual LOD values ​​in the population, which is used to judge the significance of QTL in a single population and reflect the difference in the performance of QTL among populations; the estimation of genetic effects is carried out separately in a single population.

[0057] LOD value refers to the logarithm of odds, which is a statistical indicator used in genetics to evaluate the linkage relationship between genes.

[0058] The calculation formulas for LOD values ​​and genetic effects are as follows:

[0059] If each locus in the hth population contains 2 genotypes, then H 1 The lower log-likelihood function is:

[0060]

[0061] If each locus in the hth population contains three genotypes, then H 1 The lower log-likelihood function is:

[0062]

[0063] H 0 The lower log-likelihood function is:

[0064]

[0065] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; p represents the number of populations; l represents the l-th marker genotype; n h represents the population size of the hth population;

[0066] L 1 For the alternative hypothesis H 1 The multi-group log-likelihood function under L 1hFor the alternative hypothesis H 1 The h-th group log-likelihood function under 0 The null hypothesis H 0 The multi-group log-likelihood function under 0h The null hypothesis H 0 The h-th group log-likelihood function under ;

[0067] If each locus in the hth population contains 2 genotypes, S hl (l=1,2,3,4) is the set of individuals of the lth labeled class in the hth population, π hl1 and π hl2 are the proportions of individuals with QTL genotypes QQ and qq in the lth marker class in the hth population, respectively;

[0068] If each locus in the hth population contains three genotypes, S hl (l=1,…,9) is the set of individuals of the lth labeled class in the hth population, π hl1 , π hl2 and π hl3 are the proportions of individuals with QTL genotypes QQ, Qq, and qq in the lth marker class in the hth population, respectively;

[0069] is the density function of the kth normal distribution in the hth population, k = 1, 2, 3; μ 0h is the mean genotype of all individuals in the hth population; Yes H 1 The variance of the normal distribution in the next h population; Yes H 0 The variance of the normal distribution in the next h population; Δy ih is the phenotypic correction value of the i-th individual in the h-th population;

[0070] The LOD value of a single group is: LOD h =L 1h -L 0h ;

[0071] The combined LOD values ​​are:

[0072] If each locus in the hth population contains two genotypes, then the additive effect (h=1,…,p), where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population;

[0073] If each locus in the hth population contains three genotypes, then the additive effect Dominance Effect (h=1,…,p), where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq and qq in the hth population.

[0074] Beneficial effects:

[0075] (1) The method provided by the present invention is based on the inclusive composite interval mapping (ICIM) method with background control, which can effectively control the influence of background genetic effects on the current mapping interval and improve the accuracy of QTL positioning.

[0076] (2) The method provided by the present invention comprehensively utilizes the genotype and phenotype data of multiple mapping populations. The position and effect of each QTL in the positioning results can be completely determined, and there will be no differences in QTL position and effect between populations, thereby better determining the co-localized QTLs in each population.

[0077] (3) The method provided by the present invention avoids the cumbersome steps of first mapping a single group, then comparing and integrating the positioning results of each group, and analyzing the results when processing multi-group data in the existing method, thereby improving the analysis efficiency.

[0078] (4) The method provided by the present invention makes full use of multi-population data, which is conducive to the mining of micro-effect QTLs, and the results obtained by analysis are richer.

[0079] In summary, the method provided by the present invention can carry out joint complete interval mapping with background control for parent-derived offspring populations, which solves the current problem of lack of joint QTL positioning methods for multiple mapping populations and improves the accuracy of gene positioning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 This is the LOD map of the joint QTL mapping of plant height traits of three wheat RIL populations in Experimental Example 1 of the present invention.

[0081] Figure 2 The information of 11 QTLs identified by joint QTL mapping of plant height traits of three wheat RIL populations in Experimental Example 1 of the present invention; LOD 1 、LOD 2 、LOD 3 are the LOD values ​​of the single population in the ZC, DS and GZ populations, and LOD is the joint LOD value; Add 1 、Add 2 、Add 3 are the additive effects of QTLs in ZC, DS and GZ populations, respectively; PVE 1 、PVE2 、PVE 3 are the phenotypic variation explained by QTL in ZC, DS and GZ populations respectively; LeftCI and RightCI are the left and right ends of the confidence interval when the QTL drops by 1 LOD value.

[0082] Figure 3 are the positions, effects and genetic variances of the preset QTLs in the simulation model in Experimental Example 2 of the present invention; Q1, Q2, Q3, Q4, Q5 and Q6 are the preset 6 QTLs; Pop1, Pop2 and Pop3 are DH populations derived from 3 different parents.

[0083] Figure 4 It is the single population mapping, multi-population joint mapping efficacy and false positive rate of each QTL in Experimental Example 2 of the present invention; Q1, Q2, Q3, Q4, Q5 and Q6 are 6 preset QTLs; Pop1, Pop2, Pop3 represent single population mapping; CQM represents multi-population joint mapping; FDR is the false positive rate.

[0084] Figure 5 It is the single population mapping, multi-population joint mapping QTL position, estimated value and standard error of the effect of each QTL in Experimental Example 2 of the present invention; Q1, Q2, Q3, Q4, Q5 and Q6 are the preset 6 QTL; Pop1, Pop2, Pop3 represent single population mapping; CQM represents multi-population joint mapping; SE is the standard error of each estimated value; Add1, Add2, Add3 are the additive effects of QTL in the three populations, respectively. DETAILED DESCRIPTION

[0085] The present invention will be further described in detail below in conjunction with specific examples. It should be understood that the following examples are only for explanation and illustration and do not limit the scope of the present invention in any way. Unless otherwise defined, the meaning of all technical and scientific terms used herein is the same as that generally understood by those of ordinary skill in the art to which the present invention belongs.

[0086] This invention was funded by the Science and Technology Innovation 2030-Major Project (2023ZD0407501).

[0087] Example 1

[0088] The present invention provides a joint complete interval mapping method for multiple parent-derived offspring populations, comprising the following steps:

[0089] S1: obtaining genotype data of multiple parent-derived offspring populations; the genotype data of the multiple parent-derived offspring populations have a certain number of common molecular markers;

[0090] S2: using the genotype data of the plurality of parent-derived offspring populations, constructing an integrated linkage map with the aid of genetic analysis software, and collecting and collating the phenotypic data of each population;

[0091] S3: construct orthogonal indicator variables (additive effect indicator variables and dominant effect indicator variables) of marker genotypes for each population, and establish a linear relationship model between phenotype and marker genotype;

[0092] S4: Based on the linear relationship model between the phenotype and marker genotype of a single population established in step S3, significant markers are selected in each population using stepwise regression, and the phenotypic values ​​are corrected using the significant markers, and interval mapping with background control for multiple populations is performed based on the complete interval mapping method;

[0093] S5: Establish hypothesis tests, calculate joint test statistics through interval scanning, locate QTLs, and estimate their positions and genetic effects, thereby achieving complete interval mapping for joint QTL positioning in multiple populations.

[0094] In the above step S3, if each seat in the hth population contains 2 genotypes, the linear relationship model is:

[0095]

[0096] If each seat in the hth population contains 3 genotypes, the linear relationship model is:

[0097]

[0098] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the integrated linkage map; m is the number of marker intervals;

[0099] y ih is the phenotypic value of the i-th individual in the h-th population;

[0100] b 0h is the mean of the linear relationship model between the phenotype and marker genotype of the hth population;

[0101] x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0102] x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0103] z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0104] z i(j+1)h is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0105] b jh 、c jh , b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect;

[0106] ε ih is the random error effect of the hth population and follows a normal distribution.

[0107] In the above step S4, using computer programming or with the help of R language, SAS software, etc., based on the linear relationship model between the phenotype and marker genotype of the single population established in step S3, stepwise regression is performed with the additive effect indicator variable x and the dominant effect indicator variable z as independent variables and the phenotypic value y as the dependent variable to select significant markers.

[0108] If each locus in the hth population contains two genotypes, the formula for correcting the phenotypic value is as follows:

[0109]

[0110] If each locus in the hth population contains three genotypes, the formula for correcting the phenotypic value is as follows:

[0111]

[0112] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the common molecular marker; k represents the number of the left end marker of the current scanning interval;

[0113] Δy ih is the phenotypic correction value of the i-th individual in the h-th population; y ih is the phenotypic value of the i-th individual in the h-th population;

[0114] Yes jh An estimated value of It is c jh An estimated value of Yesjh,(j+1)h An estimated value of It is c jh,(j+1)h The estimated value of jh 、c jh , b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect;

[0115] x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0116] x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0117] z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively;

[0118] z i(j+1)h It is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population. When the genotype is the genotype of parent 1, the heterozygous genotype and the genotype of parent 2, the values ​​are 0, 1 and 0 respectively.

[0119] In the above step S5, the hypothesis test is:

[0120] If each locus in the hth population contains 2 genotypes, H 0 :μ 1h =μ 2h (h=1,…,p), where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population;

[0121] If each locus in the hth population contains three genotypes, H 0 :μ 1h =μ 2h =μ 3h (h=1,…,p), where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq, and qq in the hth population;

[0122] H 1: The QTL genotype values ​​in at least one population are not completely equal;

[0123] Among them, H 0 is the null hypothesis, H 1 is the alternative hypothesis, p represents the number of groups;

[0124] The genotype mean is the average of all individual genotype values ​​corresponding to the marker genotype.

[0125] The QTL genotype value refers to the average of all individual genotype values ​​corresponding to the genotype.

[0126] The joint test statistic uses the joint LOD value, which is the sum of the individual LOD values ​​in all groups, while the LOD value of a single group (LOD h ) is the sum of the individual LOD values ​​in the population; the joint LOD value is used to determine the existence of QTL, while the single-population LOD value is used to determine the significance of QTL in a single population, reflecting the differences in the performance of QTL between populations; the estimation of genetic effects is performed separately in a single population.

[0127] LOD value refers to the logarithm of odds, which is a statistical indicator used in genetics to evaluate the linkage relationship between genes.

[0128] The calculation formulas for LOD values ​​and genetic effects are as follows:

[0129] If each locus in the hth population contains 2 genotypes, then H 1 The lower log-likelihood function is:

[0130]

[0131] If each locus in the hth population contains three genotypes, then H 1 The lower log-likelihood function is:

[0132]

[0133] H 0 The lower log-likelihood function is:

[0134]

[0135] Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; p represents the number of populations; l represents the l-th marker genotype; n h represents the population size of the hth population;

[0136] L 1 For the alternative hypothesis H 1 The multi-group log-likelihood function under L 1hFor the alternative hypothesis H 1 The h-th group log-likelihood function under 0 The null hypothesis H 0 The multi-group log-likelihood function under 0h The null hypothesis H 0 The h-th group log-likelihood function under ;

[0137] If each locus in the hth population contains 2 genotypes, S hl (l=1,2,3,4) is the set of individuals of the lth labeled class in the hth population, π hl1 and π hl2 are the proportions of individuals with QTL genotypes QQ and qq in the lth marker class in the hth population, respectively;

[0138] If each locus in the hth population contains three genotypes, S hl (l=1,…,9) is the set of individuals of the lth labeled class in the hth population, π hl1 , π hl2 and π hl3 are the proportions of individuals with QTL genotypes QQ, Qq, and qq in the lth marker class in the hth population, respectively;

[0139] is the density function of the kth normal distribution in the hth population, k = 1, 2, 3; μ 0h is the mean genotype of all individuals in the hth population; Yes H 1 The variance of the normal distribution in the next h population; Yes H 0 The variance of the normal distribution in the next h population; Δy ih is the phenotypic correction value of the i-th individual in the h-th population;

[0140] The LOD value of a single group is: LOD h =L 1h -L 0h ;

[0141] The combined LOD values ​​are:

[0142] If each locus in the hth population contains two genotypes, then the additive effect (h=1,…,p), where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population;

[0143] If each locus in the hth population contains three genotypes, then the additive effect Dominance Effect (h=1,…,p), where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq and qq in the hth population.

[0144] Experimental Example 1. Joint gene mapping experiment of three wheat RIL populations

[0145] Experimental purpose: The joint complete interval mapping method for multiple parent-derived offspring populations described in Example 1 (hereinafter referred to as the method of the present invention) was used to perform joint gene mapping on three wheat recombinant inbred lines (RIL) populations of the Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, to investigate the QTL (Quantitative Trait Locus) mapping effect of this method. The RIL population is a genetically stable pure line population obtained by continuous selfing after hybridization between two parents.

[0146] Experimental materials: The RIL population derived from Zhou 8425B and Chinese Spring, denoted as ZC, contains 245 RIL individual plants; the RIL population derived from Doumai and Shi 4185, denoted as DS, contains 275 RIL individual plants; the RIL population derived from Gaocheng 8901 and Zhoumai 16, denoted as GZ, contains 176 RIL individual plants.

[0147] Experimental methods: Genotype data of the three wheat RIL populations were obtained by 90K chip sequencing, and a multi-population integrated linkage map was constructed using QTLIciMapping software. Phenotypic data of plant height (PH) traits were collected for many years and multiple points. The best linear unbiased prediction (BLUP) value of PH multi-year multi-point data was calculated, and the joint QTL mapping was carried out using the method of the present invention, and compared with the QTL mapping results of the single-population ICIM method reported by Qu et al. (2021). The scanning step size was set to 0.2 cM, the LOD critical value was set to 5.0, and the probability of the variable entering the model during stepwise regression was set to 0.001.

[0148] Experimental results: Figure 1 and Figure 2 As shown in FIG. 1 , the method of the present invention identified 11 plant height QTLs in the joint mapping of the three wheat RIL populations. Among them, there was one QTL on chromosomes 1B, 2B and 6D, two QTLs on chromosome 5A, and three QTLs on chromosomes 4B and 4D. Figure 2 As shown, qPH4D-1 had the highest LOD value of 19.07. Figure 2 Medium LOD 1 、LOD 2 、LOD 3are the LOD values ​​of the single population in the ZC, DS and GZ populations, and LOD is the joint LOD value. 1 、Add 2 、Add 3 are the additive effects of QTLs in ZC, DS and GZ populations, respectively. 1 、PVE 2 、PVE 3 are the sizes of phenotypic variation explained by QTL in ZC, DS and GZ populations, respectively. LeftCI and RightCI are the left and right end positions of the confidence interval when the QTL drops by 1 LOD value. If two QTLs with a distance less than 10 cM are considered to be overlapping QTLs, 6 of these 11 QTLs overlap with the single-population QTLs reported by Qu et al. (2021), namely qPH1B, qPH4B-2, qPH4D-1, qPH4D-3, qPH5A-2 and qPH6D. The method of the present invention also discovered 5 new QTLs, namely qPH2B, qPH4B-1, qPH4B-3, qPH4D-2 and qPH5A-1. Among these 5 QTLs, qPH4B-1 has the highest LOD value of 9.34, and its LOD values ​​in ZC and GZ populations are both over 4.5, with high significance, which can be used for further fine positioning research.

[0149] Experimental Example 2. Monte Carlo Simulation Population Joint Gene Mapping Experiment

[0150] Purpose of the experiment: Since it is difficult to fully determine the reliability of mapping results based on real populations, this experiment used the simulation breeding platform Blib for simulation to verify the accuracy of the joint complete interval mapping method for multiple parent-derived offspring populations described in Example 1 (hereinafter referred to as the method of the present invention) for multi-population joint QTL (Quantitative Trait Locus) mapping.

[0151] Experimental method: Assume that joint QTL mapping is carried out on doubled haploid (DH) populations derived from three different parents, and the population size is 200. During the simulation, it is assumed that there are 4 chromosomes, each with a length of 110 cM, and 12 markers are evenly distributed, that is, the total number of markers is 72, and the distance between each marker is 10 cM. ​​Six QTLs (Q1, Q2, Q3, Q4, Q5 and Q6) are preset, with 2 QTLs on chromosomes 1 and 2, and 1 QTL on chromosomes 3 and 4. Random numbers between -1 and 1 are assigned to the QTL additive effect. Figure 3As shown, Pop1, Pop2, and Pop3 represent three populations, and the additive effects of Q1 in the three populations are -0.73, 0.07, and -0.96, respectively; the additive effects of Q2 in the three populations are 0.87, -0.87, and 0.57, respectively; the additive effects of Q3 in the three populations are -0.07, -0.04, and 0.39, respectively; the additive effects of Q4 in the three populations are -0.58, 0.46, and -0.41, respectively; the additive effects of Q5 in the three populations are -0.69, -0.20, and -0.55, respectively; and the additive effects of Q6 in the three populations are 0.43, 0.81, and 0.16, respectively. The simulation was repeated 1000 times. The method of the present invention is used to carry out multi-population joint QTL mapping for the three populations simulated each time, the scanning step is set to 1.0cM, the LOD critical value is 5.0, and the probability of the variable entering the model during stepwise regression is set to 0.001. The support interval length is 10cM (that is, if the detected QTL falls within the range of 5cM around the preset QTL, it is regarded as a true positive, otherwise it is regarded as a false positive). The detection efficacy of each QTL and the false positive rate of the model are calculated. The detection efficacy is the number of times the QTL is correctly detected divided by the number of simulations, and the false positive rate is the number of false positive QTLs divided by the total number of QTLs detected. For ease of comparison, single-population QTL mapping is also performed on the three populations simulated each time, and the mapping method is single-population ICIM. Except that the LOD critical value is set to 2.5, other mapping parameters are the same as multi-population joint QTL mapping.

[0152] Experimental results: Whether it is single-population QTL mapping or multi-population combined QTL mapping, the detection power of QTL increases with the increase of QTL genetic variance (such as Figure 4 shown). Figure 4 The bar graphs of Pop1, Pop2, and Pop3 in the figure correspond to the QTL detection efficacy of single population mapping, CQM corresponds to the QTL detection efficacy of multi-population joint mapping, and FDR is the false positive rate. For example, Q3 and Q4 are linked, and the genetic variance of Q3 is smaller than that of Q4 in all three populations ( Figure 3 ), whether it is single population QTL mapping or multi-population combined QTL mapping, the detection efficiency of Q4 is higher than that of Q3 ( Figure 4 In addition, the QTL detection efficiency of multi-population joint QTL mapping is close to or higher than that of single-population QTL mapping ( Figure 4 ). For example, the single-population QTL mapping efficacy of Q4 in the three populations was 87.5%, 81.1% and 15.5%, respectively, while the multi-population joint QTL mapping efficacy was 96.9%. The false positive rate of multi-population joint QTL mapping was lower than that of single-population mapping. The false positive rates of single-population QTL mapping in the three populations were 15.55%, 14.90% and 18.90%, while the false positive rate of multi-population joint QTL mapping was only 14.05%.

[0153] When the QTL detection efficacy is similar, the estimated values ​​of the preset QTL position and effect by single-population QTL mapping and multi-population joint QTL mapping are also similar; however, when the single-population QTL mapping efficacy is too low, the estimated values ​​of the QTL position and effect by single-population QTL mapping are more biased (e.g. Figure 5 shown). Figure 5 SE is the standard error of each estimate; Pop1, Pop2, and Pop3 represent single-population mapping of three populations; CQM represents multi-population joint mapping; Add1, Add2, and Add3 are the additive effects of QTL in the three populations. For example, the true position of Q1 is at 25cM on chromosome 1. The single-population QTL mapping of Pop1 and Pop3 is similar to the multi-population joint QTL mapping. The estimated values ​​of the Q1 position of the three are 24.39, 24.44, and 24.38, respectively, and the standard errors of the estimated values ​​are 2.52, 2.12, and 2.21, respectively. The estimated values ​​and their standard errors are close. However, the detection power of Pop2 is only 0.3%, and the estimated value of Q1 position is 26.67, which has a large deviation, and the standard error is 4.71, which is also large. Therefore, compared with single-population QTL mapping, multi-population joint QTL mapping can obtain similar or more accurate estimates of QTL position and effect.

[0154] Based on the above results, it can be shown that the multi-population joint QTL mapping provided by the method of the present invention has higher QTL detection efficiency and lower false positive rate than single-population QTL mapping, and can obtain similar or more accurate estimates of QTL positions and effects, which is a more effective mapping method.

Claims

1. A joint complete interval mapping method for multiple parent-derived offspring populations, comprising the following steps: S1: obtaining genotype data of multiple parent-derived offspring populations; the genotype data of the multiple parent-derived offspring populations have a certain number of common molecular markers; S2: constructing an integrated linkage map using the genotype data of the plurality of parent-derived offspring populations, and collecting and collating the phenotypic data of each population; S3: construct orthogonal indicator variables of marker genotypes for each population, and establish a linear relationship model between phenotype and marker genotype; S4: Based on the linear relationship model between the phenotype and marker genotype of a single population established in step S3, significant markers are selected in each population using stepwise regression, and the phenotypic values ​​are corrected using the significant markers, and interval mapping with background control for multiple populations is performed based on the complete interval mapping method; S5: Establish hypothesis tests, calculate joint test statistics through interval scanning, locate QTLs, and estimate their positions and genetic effects, thereby achieving complete interval mapping for joint QTL positioning in multiple populations.

2. The method according to claim 1, It is characterized in that The parent-derived progeny population includes the hybrid F 2 or F 2:3 Derived populations, backcross populations, recombinant inbred line populations and doubled haploid populations.

3. The method according to claim 1, It is characterized in that The number of the common molecular markers is 10 or more.

4. The method according to claim 1, It is characterized in that The plurality of parent-derived offspring populations are sequenced separately using the same sequencing platform, thereby obtaining common molecular markers among the populations.

5. The method according to claim 1, It is characterized in that The integrated linkage map was constructed using genetic analysis software.

6. The method according to claim 1, It is characterized in that The orthogonal indicator variables of the marker genotypes include additive effect indicator variables and dominant effect indicator variables.

7. The method according to claim 1, It is characterized in that In step S3, If each seat in the hth population contains 2 genotypes, the linear relationship model is: If each seat in the hth population contains 3 genotypes, the linear relationship model is: Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the integrated linkage map; m is the number of marker intervals; y ih is the phenotypic value of the i-th individual in the h-th population; b 0h is the mean of the linear relationship model between the phenotype and marker genotype of the hth population; x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; z i(j+1)h is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes the values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; b jh 、c jh , b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect; ε ih is the random error effect of the hth population and follows a normal distribution.

8. The method according to claim 7, It is characterized in that In step S4, If each locus in the hth population contains two genotypes, the formula for correcting the phenotypic value is as follows: If each locus in the hth population contains three genotypes, the formula for correcting the phenotypic value is as follows: Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; j represents the j-th marker in the common molecular marker; k represents the number of the left end marker of the current scanning interval; Δy ih is the phenotypic correction value of the i-th individual in the h-th population; y ih is the phenotypic value of the i-th individual in the h-th population; Yes jh An estimated value of It is c jh An estimated value of Yes jh,(j+1)h An estimated value of It is c jh,(j+1)h The estimated value of jh 、c jh 、b jh,(j+1)h 、c jh,(j+1)h is the partial regression coefficient of the hth population phenotype on the jth marker, corresponding to the additive effect, dominant effect, additive-additive interaction effect, and dominant-dominant interaction effect; x ijh is the additive effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; x i(j+1)h is the additive effect indicator variable of the j+1th marker of the ith individual in the hth population, and takes values ​​of 1, 0, and -1 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; z ijh is the dominant effect indicator variable of the jth marker of the ith individual in the hth population, and takes values ​​of 0, 1, and 0 when the genotype is the genotype of parent 1, the heterozygous genotype, and the genotype of parent 2, respectively; z i(j+1)h It is the dominant effect indicator variable of the j+1th marker of the ith individual in the hth population. When the genotype is the genotype of parent 1, the heterozygous genotype and the genotype of parent 2, the values ​​are 0, 1 and 0 respectively.

9. The method according to claim 8, It is characterized in that The hypothesis test in step S5 is: If each locus in the hth population contains 2 genotypes, then H 0 :μ 1h =μ 2h (h=1,…,p), where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population; If each locus in the hth population contains three genotypes, then H 0 :μ 1h =μ 2h =μ 3h (h=1,…,p), where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq, and qq in the hth population; H 1 : The QTL genotype values ​​in at least one population are not completely equal; Among them, H 0 is the null hypothesis, H 1 is the alternative hypothesis, and p represents the number of groups.

10. The method according to claim 9, It is characterized in that The joint test statistic is expressed as a joint LOD value; the joint LOD value is the sum of the individual LOD values ​​in all populations, which is used to determine the existence of QTL; the LOD value of a single population is the sum of the individual LOD values ​​in the population, which is used to determine the significance of QTL in a single population; the estimation of genetic effects is performed separately in a single population; the calculation formulas for LOD values ​​and genetic effects are as follows: If each locus in the hth population contains 2 genotypes, then H 1 The lower log-likelihood function is: If each locus in the hth population contains three genotypes, then H 1 The lower log-likelihood function is: H 0 The lower log-likelihood function is: Where: i represents the i-th individual in a single population; h represents the h-th population in multiple parent-derived offspring populations; p represents the number of populations; l represents the l-th marker genotype; n h represents the population size of the hth population; L 1 For the alternative hypothesis H 1 The multi-group log-likelihood function under L 1h For the alternative hypothesis H 1 The h-th group log-likelihood function under 0 The null hypothesis H 0 The multi-group log-likelihood function under 0h The null hypothesis H 0 The h-th group log-likelihood function under ; If each locus in the hth population contains 2 genotypes, S hl (l=1,2,3,4) is the set of individuals of the lth labeled class in the hth population, π hl1 and π hl2 are the proportions of individuals with QTL genotypes QQ and qq in the lth marker class in the hth population, respectively; If each locus in the hth population contains three genotypes, S hl (l=1,…,9) is the set of individuals of the lth labeled class in the hth population, π hl1 , π hl2 and π hl3 are the proportions of individuals with QTL genotypes QQ, Qq, and qq in the lth marker class in the hth population, respectively; is the density function of the kth normal distribution in the hth population, k = 1, 2, 3; μ 0h is the mean genotype of all individuals in the hth population; Yes H 1 The variance of the normal distribution in the next h population; Yes H 0 The variance of the normal distribution in the next h population; Δy ih is the phenotypic correction value of the i-th individual in the h-th population; The LOD value of a single group is: LOD h =L 1h -L 0h ; The combined LOD values ​​are: If each locus in the hth population contains two genotypes, then the additive effect where μ 1h and μ 2h is the genotype mean of QTL genotypes QQ and qq in the hth population; If each locus in the hth population contains three genotypes, then the additive effect Dominance Effect where μ 1h , μ 2h and μ 3h is the genotype mean of QTL genotypes QQ, Qq and qq in the hth population.

Images