Donor ratio analysis method for mixed DNA samples based on STR typing and residual matrix analysis

Abstract

The present invention discloses a method for analyzing the proportion of donors in mixed DNA samples based on STR typing and residual matrix analysis. First, the minimum number of donors in the sample is calculated by the maximum allele counting method, and then the genotype distribution of the sample is fitted by an optimization algorithm. The difference between the predicted peak height and the actual measured peak height is compared to accurately estimate the proportion of each donor in the mixed DNA sample. Experimental results show that the present invention can effectively reduce the estimation error existing in traditional methods when processing multi-allele mixed DNA samples, especially showing high accuracy in complex sample analysis. By establishing a linear model and combining it with an optimization algorithm, the present invention not only provides high-precision technical support for DNA analysis in forensic science, but also provides a practical solution for the quantitative analysis of mixed DNA samples in other fields.

Description

Technical Field

[0001] The present invention relates to the field of biological detection technology, and in particular to a mixed DNA sample donor ratio analysis method based on STR typing and residual matrix analysis. Background Art

[0002] Mixed DNA samples, composed of DNA samples from multiple donors, are common in forensic identification and genetic analysis. Existing analysis of mixed DNA samples relies on the analyst's experience and understanding of DNA data. Analysis is typically performed on markers such as STRs (short tandem repeats), with the contributions of different donors determined based on the patterns of each marker's presence in the mixed DNA sample (e.g., repeat number, size differences, etc.). Specifically, using methods such as cluster maps and genetic marker analysis plots (such as spectrograms or electropherograms), technicians can manually analyze the intensity and frequency of each marker in the sample to preliminarily separate the DNA from different individuals. For experienced technicians, intuitively assessing signal intensity, heterozygosity, and homology at a given location can help them make decisions without the assistance of advanced software. This manual method is typically used for simpler mixed DNA sample analyses, but manual separation is often inefficient when dealing with complex, multi-source samples, and the results are easily influenced by personal experience and subjective judgment. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the present invention aims to provide a mixed DNA sample donor ratio analysis method based on STR typing and residual matrix analysis.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The method for analyzing the proportion of mixed DNA sample donors based on STR typing and residual matrix analysis includes the following steps:

[0006] S1. Determine the number of donors: After obtaining the mixed DNA sample, perform PCR amplification and capillary electrophoresis in sequence to obtain the size and fluorescence intensity of each DNA fragment. Based on this, a map is further generated and genotype analysis is performed to obtain the allele and peak height information of each site. The maximum value of the allele number of each site is taken as the maximum allele number, and the maximum allele number is determined according to the maximum allele number.

[0007] S2. Calculate the peak height ratio: For each locus, calculate the ratio of the peak height of each allele in the locus to the total peak height of all alleles at the locus, thereby obtaining the peak height ratio of each allele at the locus;

[0008] S3. Generate residual matrix:

[0009] The mixing ratio of donor 1 is defined as a variable Mx, and the initial value, end value and step size of the variable Mx are defined to obtain all values ​​of the variable Mx;

[0010] At a locus, for each value of the variable Mx, a corresponding allele mixing ratio matrix is ​​generated. Each row in the allele mixing ratio matrix corresponds to a possible genotype combination of two donors at the locus, and each column represents the allele mixing ratio calculated for each allele at the locus according to the mixing ratio Mx of donor 1.

[0011] At a locus, the following calculations are performed on the allele mixing ratio matrix corresponding to each value of the variable Mx and the corresponding matrix columns are generated:

[0012] First, the updated value of each column in the i-th row of the allele mixing ratio matrix is ​​calculated as follows:

[0013] updated_m(i)=(column_m[i]-x) 2

[0014] Where i∈[1,2,…,N], N represents the total number of rows in the allele mixing ratio matrix, column_m[i] represents the value of the i-th row and m-th column in the allele mixing ratio matrix, and x represents the peak height ratio of the allele corresponding to the m-th column at the corresponding site;

[0015] Then calculate the value of the i-th row in the matrix column, that is, the residual sum of squares RSS, as follows:

[0016] RSS[i]=updated_1(i)+updated_2(i)+,…,+updated_M(i)

[0017] M represents the total number of columns of the allele mixing ratio matrix;

[0018] Finally, the matrix columns corresponding to all values ​​of the variable Mx at a site are combined to obtain the residual matrix of the site; each row in the residual matrix corresponds to a genotype combination, each column corresponds to a value of the mixing ratio variable Mx, and each element represents the residual sum of squares under the corresponding genotype combination and mixing ratio;

[0019] S4. Based on the residual matrix corresponding to each site obtained in step S3, a mixing ratio-logarithmic residual relationship diagram corresponding to each site is drawn, where each curve in the mixing ratio-logarithmic residual relationship diagram represents a genotype combination, the abscissa is the mixing ratio, and the ordinate is the logarithmic residual calculated based on the residual sum of squares;

[0020] S5. After obtaining the admixture ratio-logarithmic residual relationship graph corresponding to all sites, the genotype combination and admixture ratio corresponding to the lowest point in the admixture ratio-logarithmic residual relationship graph is the most likely genotype combination and admixture ratio for the corresponding site;

[0021] S6. Calculate the average value of the mixing ratio corresponding to the most likely genotype combination of all sites, and perform genotype splitting to obtain the final analysis results.

[0022] Furthermore, in step S3, the initial value of the variable Mx is 0, the end value is 1, and the step size is 0.01.

[0023] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0024] The present invention also provides a computer device, comprising a processor and a memory, wherein the memory is used to store a computer program; and when the processor is used to execute the computer program, the above method is implemented.

[0025] The beneficial effects of the present invention are as follows: the present invention combines the maximum allele counting method and the residual sum of squares (RSS) optimization fitting technology, which can accurately determine the contribution ratio of each donor in a mixed DNA sample. First, the minimum number of donors in the sample is calculated by the maximum allele counting method, and the genotype distribution of the sample is further fitted by the optimization algorithm, and the difference between the predicted peak height and the actual measured peak height is compared to accurately estimate the proportion of each donor in the mixed DNA sample. Experimental results show that the present invention can effectively reduce the estimation error existing in traditional methods when processing multi-allelic mixed DNA samples, especially showing higher accuracy in complex sample analysis. By establishing a linear model and combining it with an optimization algorithm, the present invention not only provides high-precision technical support for DNA analysis in forensic science, but also provides a practical solution for the quantitative analysis of mixed DNA samples in other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is the STR typing pattern of sample 16.4 in Example 1 of the present invention;

[0027] Figure 2 Graph showing the relationship between the CSF1PO mixing ratio and the residual in Example 1 of the present invention;

[0028] Figure 3 Graph showing the relationship between the D3S1358 mixing ratio and the residual in Example 1 of the present invention;

[0029] Figure 4 Graph showing the relationship between the D18S51 mixing ratio and the residual in Example 1 of the present invention;

[0030] Figure 5 Graph showing the relationship between the D19S433 mixing ratio and the residual in Example 1 of the present invention;

[0031] Figure 6 Graph showing the relationship between the FGA mixing ratio and the residual in Example 1 of the present invention;

[0032] Figure 7 TH01 mixing ratio and residual error in Example 1 of the present invention;

[0033] Figure 8 This is the STR typing pattern of sample 12.8 in Example 2 of the present invention;

[0034] Figure 9 Graph showing the relationship between the CSF1PO mixing ratio and the residual in Example 2 of the present invention;

[0035] Figure 10 Graph showing the relationship between the D3S1358 mixing ratio and the residual in Example 2 of the present invention;

[0036] Figure 11 Graph showing the relationship between the D18S51 mixing ratio and the residual in Example 2 of the present invention;

[0037] Figure 12 Graph showing the relationship between the D19S433 mixing ratio and the residual in Example 2 of the present invention;

[0038] Figure 13 Graph showing the relationship between the FGA mixing ratio and the residual in Example 2 of the present invention;

[0039] Figure 14 1 is a graph showing the relationship between the TH01 mixing ratio and the residual in Example 2 of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings. It should be noted that this embodiment is based on the technical solution and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to this embodiment.

[0041] This example provides a method for analyzing the proportion of mixed DNA sample donors based on STR typing and residual matrix analysis, which is implemented based on maximum allele counting, heterozygote balance analysis, and residual sum of squares optimization fitting. It includes the following steps:

[0042] S1. Determine the number of donors.

[0043] After obtaining the mixed DNA sample, PCR amplification and capillary electrophoresis are performed in sequence to obtain the size and fluorescence intensity information of each DNA fragment. Based on this, a map is further generated and genotype analysis is performed to obtain the allele and peak height information of each site; the maximum value of the number of alleles at each site is taken as the maximum allele number, and the minimum number of donors is calculated based on the maximum allele number, thereby determining the minimum number of donors contained in the mixed DNA sample.

[0044] Take sample 16.4 as an example (standard products 9947A and 9948 are mixed in a ratio of 16:4). Figure 1 As shown, the CSF1PO, D3S1358, TH01, D19S433, D18S51, and FGA loci all have three alleles. It's known that one donor has two alleles at one locus, and the number of alleles at multiple loci is greater than two, indicating that there was more than one donor. 3 divided by 2, rounded up, is 2. Preliminary analysis indicates that the number of donors in the mixed DNA sample is two. However, the mixing ratio and the genotypes of the two donors at each locus cannot be determined.

[0045] The allele and peak height information at each locus obtained based on sample 16.4 is shown in Table 1.

[0046] Table 1 Allele peak height information of samples

[0047]

[0048]

[0049] S2. Calculate the peak height ratio. For each locus, sort each allele in the locus by molecular weight from smallest to largest, and calculate the ratio of the peak height of each allele to the total peak height of all alleles at that locus to obtain the peak height ratio of each allele at that locus.

[0050] Taking CSF1PO in Table 1 as an example, allele 10, allele 11, and allele 12 were reordered in ascending order based on molecular weight and named a, b, and c, respectively. The proportion of the peak height of each allele in the total peak height was calculated, as shown in Table 2.

[0051] Table 2 Allele peak height information of samples

[0052]

[0053] S3. Generate a residual matrix.

[0054] The mixing ratio of donor 1 is defined as a variable Mx, with an initial value of 0, an end value of 1, and a step size of 0.01. That is, Mx∈[0.0,0.01,0.02,…,1.0]. At a locus, a corresponding allele mixing ratio matrix is ​​generated for each value of the variable Mx. Each row in the allele mixing ratio matrix corresponds to a possible genotype combination of two donors at that locus, and each column represents the allele mixing ratio calculated for each allele at that locus according to the mixing ratio of donor 1, Mx.

[0055] Continuing with CSF1PO in Table 1 as an example, the allele mixing ratio matrix is ​​shown in Table 3.

[0056] Table 3 Generate allele mixing ratio matrix

[0057]

[0058] At a locus, the following calculations are performed on the allele mixing ratio matrix corresponding to each value of the variable Mx and the corresponding matrix columns are generated:

[0059] First, the updated value of each column in the i-th row of the allele mixing ratio matrix is ​​calculated as follows:

[0060] updated_m(i)=(column_m[i]-x) 2

[0061] Wherein, i∈[1,2,…,N], N represents the total number of rows of the allele mixing ratio matrix, column_m[i] represents the value of the i-th row and m-th column in the allele mixing ratio matrix, and x represents the peak height ratio of the allele corresponding to the m-th column at the corresponding site (calculated in step S2).

[0062] Then calculate the value of the i-th row in the matrix column, that is, the residual sum of squares RSS, as follows:

[0063] RSS[i]=updated_1(i)+updated_2(i)+,…,+updated_M(i)

[0064] M represents the total number of columns of the allele mixing ratio matrix;

[0065] Finally, the matrix columns corresponding to all values ​​of the variable Mx at a site are combined to obtain the residual matrix of the site; each row in the residual matrix corresponds to a genotype combination, each column corresponds to a value of the mixing ratio variable Mx, and each element represents the residual sum of squares under the corresponding genotype combination and mixing ratio.

[0066] Taking Table 3 as an example, for each Mx value corresponding to the allele mixing ratio matrix, the following calculations are performed:

[0067] Calculate the updated value of column_A[i]:

[0068] updated_A(i)=(column_A[i]-0.47) 2

[0069] Calculate the updated value of column_B[i]:

[0070] updated_B(i)=(column_B[i]-0.15) 2

[0071] Calculate the updated value of matrix column column_C[i]:

[0072] updated_C(i)=(column_C[i]-0.38) 2

[0073] Then calculate the residual sum of squares RSS:

[0074] RSS[i]=updated_A(i)+updated_B(i)+updated_C(i)

[0075] For example, for Mx=0, the residual sums of squares for 12 genotype combinations can be calculated according to Table 3, thereby obtaining the matrix columns shown in Table 4.

[0076] Table 4, residual sum of squares of each gene habit combination when Mx=0

[0077]

[0078] The same calculation is performed for other Mx values, and the residual matrix can be finally combined as shown in Table 5.

[0079] Table 5

[0080]

[0081] It should be noted that the step size of the actual variable Mx is 0.01, and some matrix column data of Mx values ​​are omitted in Table 5.

[0082] S4, draw mixing ratio-logarithmic residual graph. Residual values ​​are more, from 0 to 1, step length is 0.01, one genotype has 100 residuals, a total of 12 genotypes, a total of 1200 residual data, for observation convenience, in the present embodiment, according to the residual matrix corresponding to each site obtained in step S3, draw the mixing ratio corresponding to each site-logarithmic residual relationship graph, each curve in the mixing ratio-logarithmic residual relationship graph represents a genotype combination, the abscissa is the mixing ratio, and the ordinate is the logarithmic residual calculated according to the residual sum of squares (the ordinate is the logarithm of the residual sum of squares with 10 as the base), and taking the logarithm is to compress the data and increase the contrast effect.

[0083] S5. After obtaining the mixing ratio-logarithmic residual relationship graph for each site, find the lowest point of the mixing ratio-logarithmic residual relationship graph. The genotype combination and mixing ratio corresponding to the lowest point are the most likely genotype combination and mixing ratio for the corresponding site.

[0084] Figure 2 The following is the relationship between the mixing ratio and the logarithmic residual corresponding to CSF1PO: Figure 2 As shown, the lowest points correspond to ab and ac, which are the most likely genotype combinations, and the corresponding mixing ratios are 0.27 (the mixing ratio value corresponding to the lowest point of the curve) and 0.73.

[0085] The mixing ratio-logarithmic residual relationship diagrams of the remaining sites D3S1358, D18S51, D19S433, FGA, and TH01 are shown as follows: Figure 3 、 4 , 5, 6, and 7.

[0086] like Figure 3 As shown, for locus D3S1358,cc, ab are the most likely genotype combinations, and the admixture ratios are 0.09 and 0.91.

[0087] like Figure 4 As shown, for the locus D18S51, bb, ac are the most likely genotype combinations, and the admixture ratios are 0.15 and 0.85.

[0088] like Figure 5 As shown, for locus D19S433, ab, bc are the most likely genotype combinations, and the mixing ratios are 0.27 and 0.73.

[0089] like Figure 6 As shown, for the loci FGA, bc, ab are the most likely genotype combinations, and the mixing ratios are 0.18 and 0.82.

[0090] like Figure 7As shown, for locus TH01, ac, bc are the most likely genotype combinations, and the admixture ratios are 0.19 and 0.81.

[0091] S6. Calculate the average value of the mixing ratio corresponding to the most likely genotype combination of all sites, and perform genotype splitting to obtain the final analysis results.

[0092] In this example, the average mixing ratios of CSF1PO, D3S1358, TH01, D19S433, D18S51, and FGA were calculated, resulting in an average mixing ratio of 0.19 for donor 1 and 0.81 for donor 2. The genotype splitting results are shown in Table 6.

[0093] Table 6 Average admixture ratio and allele splitting

[0094]

[0095] S7. Verification.

[0096] The known typing of 9947A and 9948 is shown in Table 7:

[0097] Table 7 Genotypes of standards

[0098]

[0099] As shown in Table 7, the correct typing at each locus and the splitting results shown in Table 6 are not accurate. However, the result of calculating the mixing ratio using the average value is 0.19:0.81, while the actual mixing ratio is 0.2:0.8, with an error of about 6.5%, which is relatively accurate.

[0100] Example 2

[0101] This embodiment provides another application example of the method of Example 1 using another sample 12.8 (standard products 9947A and 9948 are mixed in a ratio of 12:8), including the following steps:

[0102] S1. Determine the number of donors. After obtaining a mixed DNA sample, perform PCR amplification and capillary electrophoresis sequentially to obtain information on the size and fluorescence intensity of each DNA fragment. Based on this information, a map is generated and genotype analysis is performed to obtain allele and peak height information for each locus. The maximum number of alleles at each locus is used as the maximum allele number. Based on the maximum allele number, the minimum number of donors is calculated, thereby determining the minimum number of donors contained in the mixed DNA sample.

[0103] like Figure 8As shown, the CSF1PO, D3S1358, TH01, D19S433, D18S51, and FGA loci all have three alleles. It's known that one donor has two alleles at one locus, and the number of alleles at multiple loci is greater than two, indicating that there was more than one donor. 3 divided by 2, rounded up, is 2. Preliminary analysis indicates that the number of donors in the mixed DNA sample is two. However, the mixing ratio and the genotypes of the two donors at each locus cannot be determined.

[0104] The allele heights at each site in the mixed DNA sample are derived. The specific information is shown in Table 8:

[0105] Table 8 Allele peak height information of samples

[0106]

[0107]

[0108] S2. Calculate the peak height ratio. Using CSF1PO as an example, reorder allele 10, allele 11, and allele 12 based on molecular weight from smallest to largest and name them a, b, and c, respectively. Calculate the proportion of the peak height of each allele in the total peak height. This is shown in Table 9.

[0109] Table 9 Allele peak height information of samples

[0110]

[0111] S3. Generate a residual matrix. The method used is the same as step S3 of Example 1. Taking CSF1PO as an example, the obtained residual matrix is ​​shown in Table 10.

[0112] Table 10 Residual table under different Mx

[0113]

[0114]

[0115] It should be noted that the step size of the actual variable Mx is 0.01, and some column data of Mx values ​​are omitted in Table 10.

[0116] S4. Draw the relationship between the mixing ratio and the logarithmic residual.

[0117] The process of drawing the mixing ratio-logarithmic residual relationship diagram is the same as that in Example 1.

[0118] S5. Obtain the mixing ratio-logarithmic residual relationship graph for all loci CSF1PO, D3S1358, TH01, D19S433, D18S51, and FGA, and find the lowest point of the mixing ratio-logarithmic residual relationship graph. The genotype combination and mixing ratio corresponding to the lowest point are the most likely genotype combination and mixing ratio for the corresponding locus.

[0119] In this embodiment, the relationship between the mixing ratio and logarithmic residual of CSF1PO is shown in FIG. Figure 9 Among them, the lowest point corresponds to aa, bc, which is the most likely genotype combination, and the corresponding mixing ratio is 0.46 and 0.54.

[0120] The mixing ratio-logarithmic residual relationship diagrams of D3S1358, D18S51, D19S433, FGA, and TH01 are shown in Figure 2. Figure 10 、 11 , 12, 13, and 14.

[0121] like Figure 10 As shown, bc and ab are the most likely genotype combinations, and the mixing ratios are 0.32 and 0.68.

[0122] like Figure 11 As shown, ab and ac are the most likely genotype combinations, and the mixing ratios are 0.46 and 0.54.

[0123] like Figure 12 As shown, ab and bc are the most likely genotype combinations, and the mixing ratios are 0.46 and 0.54.

[0124] like Figure 13 As shown, bc and ab are the most likely genotype combinations, and the mixing ratios are 0.37 and 0.63.

[0125] like Figure 14 As shown, ac and bc are the most likely genotype combinations, and the mixing ratios are 0.36 and 0.64.

[0126] S6. Calculate the average value of the mixing ratios corresponding to the most likely genotype combinations at all sites, and perform genotype splitting to obtain the final analysis results. In this embodiment, the results are shown in Table 11.

[0127] Table 11 Average admixture ratio and allele splitting

[0128]

[0129] S7. Verification.

[0130] The known typing of 9947A and 9948 is shown in Table 12.

[0131] Table 12 Genotypes of Standards

[0132]

[0133]

[0134] As shown in Table 12, the correct typing at each locus and the splitting results shown in Table 11 are highly accurate, with only CSF1PO having an error. The result of calculating the mixing ratio using the average value is 0.41:0.59, while the actual mixing ratio is 0.4:0.6, with an error of approximately 2.1%, which is relatively accurate.

[0135] Those skilled in the art can make various corresponding changes and modifications based on the above technical solutions and concepts, and all of these changes and modifications should be included in the scope of protection of the claims of the present invention.

Claims

1. A mixed DNA sample donor ratio analysis method based on STR typing and residual matrix analysis, characterized in that: The steps include: S1. Determine the number of donors: After obtaining the mixed DNA sample, perform PCR amplification and capillary electrophoresis in sequence to obtain the size and fluorescence intensity of each DNA fragment. Based on this, a map is further generated and genotype analysis is performed to obtain the allele and peak height information of each site. The maximum value of the allele number of each site is taken as the maximum allele number, and the maximum allele number is determined according to the maximum allele number. S2. Calculate the peak height ratio: For each locus, calculate the ratio of the peak height of each allele in the locus to the total peak height of all alleles at the locus, thereby obtaining the peak height ratio of each allele at the locus; S3. Generate residual matrix: The mixing ratio of donor 1 is defined as a variable Mx, and the initial value, end value and step size of the variable Mx are defined to obtain all values ​​of the variable Mx; At a locus, for each value of the variable Mx, a corresponding allele mixing ratio matrix is ​​generated. Each row in the allele mixing ratio matrix corresponds to a possible genotype combination of two donors at the locus, and each column represents the allele mixing ratio calculated for each allele at the locus according to the mixing ratio Mx of donor 1. At a locus, the following calculations are performed on the allele mixing ratio matrix corresponding to each value of the variable Mx and the corresponding matrix columns are generated: First, the updated value of each column in the i-th row of the allele mixing ratio matrix is ​​calculated as follows: updated_m(i)=(column_m[i]-x) 2 Where i∈[1,2,…,N], N represents the total number of rows in the allele mixing ratio matrix, column_m[i] represents the value of the i-th row and m-th column in the allele mixing ratio matrix, and x represents the peak height ratio of the allele corresponding to the m-th column at the corresponding site; Then calculate the value of the i-th row in the matrix column, that is, the residual sum of squares RSS, as follows: RSS[i]=updated_1(i)+updated_2(i)+,…,+updated_M(i) M represents the total number of columns of the allele mixing ratio matrix; Finally, the matrix columns corresponding to all values ​​of the variable Mx at a site are combined to obtain the residual matrix of the site; each row in the residual matrix corresponds to a genotype combination, each column corresponds to a value of the mixing ratio variable Mx, and each element represents the residual sum of squares under the corresponding genotype combination and mixing ratio; S4. Based on the residual matrix corresponding to each site obtained in step S3, a mixing ratio-logarithmic residual relationship diagram corresponding to each site is drawn, where each curve in the mixing ratio-logarithmic residual relationship diagram represents a genotype combination, the abscissa is the mixing ratio, and the ordinate is the logarithmic residual calculated based on the residual sum of squares; S5. After obtaining the admixture ratio-logarithmic residual relationship graph corresponding to all sites, the genotype combination and admixture ratio corresponding to the lowest point in the admixture ratio-logarithmic residual relationship graph is the most likely genotype combination and admixture ratio for the corresponding site; S6. Calculate the average value of the mixing ratio corresponding to the most likely genotype combination of all sites, and perform genotype splitting to obtain the final analysis results.

2. The method according to claim 1, characterized in that In step S3, the initial value of the variable Mx is 0, the end value is 1, and the step size is 0.

01.

3. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 2 is implemented.

4. A computer device, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program; and when the processor is used to execute the computer program, the method according to any one of claims 1 to 2 is implemented.

Images