An information-based identification method for alleles

Abstract

The present invention relates to the technical field of gene data processing, and proposes an information-based identification method for alleles, including: obtaining a base sequence of a gene to be identified and a comparison gene base sequence; using a sequence matching algorithm to obtain a gene similarity score coefficient between the base sequence of the gene to be identified and the comparison gene base sequence; obtaining a sequence matching score coefficient and a gene difference feedback coefficient according to the matching characteristics of the bases between the secondary sequence to be identified and the comparison base secondary sequence; obtaining a gene matching excellent index between the matching base sequence to be identified and the comparison matching base sequence according to the gene difference feedback coefficient; obtaining an allele feature vector according to the value of the gene matching excellent index and the allele category association index; using a clustering algorithm to obtain a classification result of the comparison gene base sequence; and obtaining an identification result of the allele according to the classification result. The present invention avoids the problem of low accuracy in allele identification caused by direct matching of bases at the same site of two gene sequences.

Description

Technical Field

[0001] The present invention relates to the technical field of gene data processing, and in particular to an information-based identification method for alleles. Background Art

[0002] In genetics, a pair of genes with different morphologies that control the same trait at the same position on homologous chromosomes are called alleles, which can lead to genetic differences between individuals at specific loci. The information-based identification method of alleles refers to the method of identifying and distinguishing different alleles by calculating and analyzing bioinformatics data. It is widely used in the fields of genotyping and population genetics. The identification of alleles can provide important information on individual genetic characteristics, which helps to take better medical care, timely disease prevention and more complete scientific research. For example, in the medical field, it has been found that many genetic diseases are caused by mutations in specific alleles. The identification of alleles can help diagnose genetic diseases, and doctors can use this to accurately assess the risk of patients getting sick.

[0003] In genetics, genes can mutate. Mutation refers to changes in gene structure caused by the replacement, addition or deletion of gene base pairs. Mutation often produces new alleles, and the base sequence differences between these alleles are often relatively small. Traditional clustering algorithms can cluster known genes with high similarity to the gene to be identified based on the large similarity between alleles, such as the K-means algorithm and the K-shape algorithm. Among them, the K-means algorithm is simple and easy to implement, but it requires setting the initial number of clusters; the hierarchical clustering algorithm has the problem of determining in advance how to split the dendrogram to generate the final cluster; the K-shape algorithm can directly cluster the sequence, but the inconsistent length of the allele base sequence requires additional gene base sequence processing. Summary of the invention

[0004] The present invention provides an information-based identification method for alleles to solve the problem of low identification accuracy caused by the addition and deletion of bases due to gene mutation when directly comparing the bases at the same site of two gene sequences. The technical solution adopted is as follows:

[0005] An embodiment of the present invention provides an information-based identification method for alleles, the method comprising the following steps:

[0006] Respectively obtaining the base sequence of the gene to be identified and the base sequence of the comparison gene;

[0007] The Needleman-Wunsch algorithm is used to obtain the gene similarity score coefficient between the matching base sequence to be identified and each aligned matching base sequence;

[0008] The matching base sequence to be identified and each aligned matching base sequence are divided into a secondary sequence to be identified and an aligned base secondary sequence of the same length in a uniform segmentation manner; a sequence matching score coefficient between the secondary sequence to be identified and the aligned base secondary sequence is obtained according to the matching characteristics of the bases between the secondary sequence to be identified and the aligned base secondary sequence; a gene difference feedback coefficient between the matching base sequence to be identified and each aligned matching base sequence is obtained according to the sequence matching score coefficient between the secondary sequences between the matching base sequence to be identified and each aligned matching base sequence;

[0009] Obtaining the gene matching excellence index between the to-be-identified matching base sequence and each of the compared matching base sequences according to the gene difference feedback coefficient between the to-be-identified matching base sequence and all the compared matching base sequences; obtaining the allele category association index between the to-be-identified matching base sequence and each of the compared matching base sequences according to the gene matching excellence index and the allele group of each of the compared matching base sequences; obtaining the allele feature vector of each compared gene base sequence according to the value of the gene matching excellence index corresponding to each of the compared matching base sequences;

[0010] A clustering algorithm is used to obtain classification results of all compared gene base sequences based on the allele feature vectors; and an identification result of the allele of the gene base sequence to be identified is obtained according to the classification results.

[0011] Preferably, the method of using the Needleman-Wunsch algorithm to obtain the gene similarity score coefficient between the matching base sequence to be identified and each aligned matching base sequence is:

[0012] The base sequence of the gene to be identified and each of the compared gene base sequences are used as inputs of the Needleman-Wunsch algorithm, and the optimal matching base sequence of the base sequence of the gene to be identified output by the Needleman-Wunsch algorithm is recorded as a matching base sequence to be identified; the optimal matching base sequence of each compared gene base sequence output by the Needleman-Wunsch algorithm is recorded as an compared matching base sequence;

[0013] The gene similarity score coefficient between the to-be-identified matching base sequence and each of the compared matching base sequences is obtained according to the base matching results between the to-be-identified matching base sequence and each of the compared matching base sequences.

[0014] Preferably, the method for obtaining the gene similarity score coefficient between the to-be-identified matching base sequence and each of the compared matching base sequences according to the base matching results between the to-be-identified matching base sequence and each of the compared matching base sequences is:

[0015] The matching scores of the bases in each order between the to-be-identified matching base sequence and each comparison matching base sequence that meet the preset conditions and do not meet the preset conditions are set to 1 and -1 respectively;

[0016] Determine the matching score weight of the bases in each order according to the number of blank bases in each order of the matching base sequence to be identified and each comparison matching base sequence;

[0017] The product of the matching score of the base in each order and the matching score weight is taken as the first cumulative factor, and the accumulation of the first cumulative factor in all orders of the matching base sequence to be identified and each compared matching base sequence is taken as the gene similarity score coefficient between the matching base sequence to be identified and each compared matching base sequence.

[0018] Preferably, the method for obtaining the sequence matching score coefficient between the secondary sequence to be identified and the secondary sequence of the aligned base according to the matching characteristics of the bases between the secondary sequence to be identified and the secondary sequence of the aligned base is:

[0019] The sum of the matching scores of the secondary sequence to be identified and the bases in all orders in each aligned base secondary sequence is taken as the sequence matching score coefficient between the secondary sequence to be identified and each aligned base secondary sequence.

[0020] Preferably, the method for obtaining the gene difference feedback coefficient between the to-be-identified matching base sequence and each aligned matching base sequence according to the sequence matching score coefficient between the secondary sequence between the to-be-identified matching base sequence and each aligned matching base sequence is:

[0021] A threshold segmentation algorithm is used to obtain a segmentation threshold of all the sequence matching score coefficients between the identification matching base sequence and each comparison matching base sequence; the comparison base secondary sequences whose sequence matching score coefficients are greater than the segmentation threshold are marked as similar sequences, and the comparison base secondary sequences whose sequence matching score coefficients are less than the segmentation threshold are marked as deviation sequences;

[0022] The set of each adjacent deviation sequence in all aligned base secondary sequences divided by each aligned matching base sequence is regarded as a gene deviation group;

[0023] The absolute value of the difference between the order value of the last deviation sequence in the current gene deviation group and the order value of the first deviation sequence in the next gene deviation group adjacent to the current gene deviation group is taken as the gene deviation distance of the current gene deviation group;

[0024] Among them, for the last gene deviation group in each aligned matching base sequence, the average of the gene deviation distances from the first gene deviation group to the penultimate gene deviation group in each aligned matching base sequence is taken as the gene deviation distance of the last gene deviation group;

[0025] The gene difference feedback coefficient between the matching base sequence to be identified and each matching base sequence is obtained according to the gene deviation distance of all gene deviation groups in each matching base sequence.

[0026] Preferably, the method for obtaining the gene difference feedback coefficient between the to-be-identified matching base sequence and each aligned matching base sequence according to the gene deviation distance of all gene deviation groups in each aligned matching base sequence is:

[0027] The difference between the mean of the sequence matching score coefficients corresponding to all similar sequences in each aligned matching base sequence and the mean of the sequence matching score coefficients corresponding to all deviation sequences is taken as the first difference; the calculation result with the natural constant as the base and the first difference as the exponent is taken as the first multiplication factor;

[0028] The difference between the segmentation threshold and the sequence matching score coefficient corresponding to each deviation sequence in each gene deviation group is accumulated on each gene deviation group as the numerator, the gene deviation distance of each gene deviation group is used as the denominator, and the ratio of the numerator to the denominator is used as the group deviation coefficient of each gene deviation group; the sum of the group deviation coefficients of all gene deviation groups in each aligned matching base sequence is used as the second multiplication factor;

[0029] The gene difference feedback coefficient between the to-be-identified matching base sequence and each comparison matching base sequence consists of a first multiplication factor and a second multiplication factor, wherein the gene difference feedback coefficient is in direct proportion to the first multiplication factor and the second multiplication factor.

[0030] Preferably, the method for obtaining the gene matching excellence index between the to-be-identified matching base sequence and each of the aligned matching base sequences according to the gene difference feedback coefficient between the to-be-identified matching base sequence and all aligned matching base sequences is:

[0031] Obtaining the minimum value between the base sequence length of the gene to be identified and the base sequence length of each compared gene, and using the natural constant as the base and the absolute value of the difference between the minimum value and the length of the matching base sequence to be identified as the exponent as the second difference;

[0032] The sum of the opposite number of the second difference and the preset parameter is used as the numerator, the calculation result with the natural constant as the base and the absolute value of the length difference between each aligned matching base sequence as the exponent is used as the denominator, and the ratio of the numerator to the denominator is used as the gene sequence length matching coefficient between the matching base sequence to be identified and each aligned matching base sequence;

[0033] The difference between the gene similarity score coefficient corresponding to each aligned matching base sequence and the minimum value of the gene similarity score coefficient between the matching base sequence to be identified and all aligned matching base sequences is taken as the numerator, and the ratio of the numerator to the gene difference feedback coefficient corresponding to each aligned matching base sequence is taken as the first component factor;

[0034] The gene matching excellence index between the to-be-identified matching base sequence and each comparison matching base sequence is composed of two parts: the gene sequence length matching coefficient and the first component factor, wherein the gene matching excellence index is proportional to the gene sequence length matching coefficient and the first component factor.

[0035] Preferably, the method for obtaining the allele category association index between the to-be-identified matching base sequence and each aligned matching base sequence according to the gene matching excellence index and the allele group of each aligned matching base sequence is:

[0036] Obtaining the same category allele group of each aligned gene base sequence from the gene database; taking the ratio of the sum of the gene matching excellence indexes corresponding to the aligned matching base sequences corresponding to each aligned gene base sequence in the same category allele group to the number of all aligned gene base sequences in the same category allele group as the third component factor;

[0037] The sum of the gene matching excellence index corresponding to each aligned matching base sequence and the third component factor is used as the allele category association index between the matching base sequence to be identified and each aligned matching base sequence.

[0038] Preferably, the method for obtaining the allele feature vector of each aligned gene base sequence according to the value of the gene matching excellence index corresponding to each aligned matching base sequence is:

[0039] The homologous genome and non-homogeneous genome of each aligned gene base sequence are obtained from all aligned gene base sequences by random sampling;

[0040] Obtaining the average of the gene matching excellence indexes between each compared gene base sequence and all compared gene base sequences in its corresponding non-similar genome as the representative value of the non-similar genome, and taking the average of the representative values ​​of a preset number of the non-similar genomes as the first decision threshold;

[0041] Obtaining the average of the gene matching excellence indexes between each compared gene base sequence and all compared gene base sequences in its corresponding similar genome as the representative value of the similar genome, and taking the average of the representative values ​​of a preset number of the similar genomes as the second decision threshold;

[0042] Obtaining the allele confidence coefficient of each aligned matching base sequence according to the comparison results of the gene matching excellence index corresponding to each aligned matching base sequence and the first decision threshold and the second decision threshold;

[0043] The vector consisting of the gene matching excellence index, allele category association index and allele confidence coefficient corresponding to each aligned matching base sequence is used as the allele feature vector of each aligned gene base sequence corresponding to each aligned matching base sequence.

[0044] Preferably, the method of using a clustering algorithm to obtain the classification results of all aligned gene base sequences based on the allele feature vector is:

[0045] The allele feature vectors of all aligned gene base sequences are used as input of the clustering algorithm, and the clustering algorithm is used to obtain the classification results of the allele feature vectors of all aligned gene base sequences.

[0046] The beneficial effects of the present invention are as follows: the present invention adopts the Needleman-Wunsch algorithm to obtain the matching base sequence to be identified and the matching base sequence based on the base sequence of the gene to be identified and the comparison gene base sequence, respectively, and constructs a gene similarity score coefficient by analyzing the matching situation between the matching base sequence to be identified and the comparison matching base sequence, so as to reflect the similarity between the two base sequences, avoid directly comparing the bases at the same sites of the two gene sequences, and avoid the situation where the accuracy of the obtained similarity is low due to the addition and deletion of bases caused by gene mutation; secondly, the matching base sequence to be identified is used to compare the base sequence to be identified and the matching base sequence to be compared. Matching base sequences and comparing matching base sequences are divided into the form of secondary sequences to be identified and compared base secondary sequences. By analyzing the base matching characteristics between the secondary sequences to be identified and the compared base secondary sequences, a gene difference feedback coefficient is constructed to solve the problem of low accuracy of the similarity obtained above due to the possible existence of identical gene fragments in non-allelic sequences; and the obtained allele feature vector is used as the input of the clustering algorithm to avoid the problem of poor efficiency of the clustering algorithm in processing high-dimensional data of gene base sequences, complete the clustering of compared gene base sequences, and improve the efficiency of allele information identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] 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 or the description of the prior art 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.

[0048] Figure 1 A schematic diagram of a flow chart of an information-based identification method for alleles provided by one embodiment of the present invention;

[0049] Figure 2 A schematic diagram of sequence matching using the Needleman-Wunsch algorithm provided in one embodiment of the present invention;

[0050] Figure 3 The present invention provides a flowchart of an implementation of an information-based identification method for alleles according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] 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.

[0052] See also Figure 1 , which shows a flow chart of an information-based identification method for alleles provided by one embodiment of the present invention, the method comprising the following steps:

[0053] Step S001, respectively obtaining the base sequence of the gene to be identified and the base sequence of the compared gene.

[0054] The present invention aims to utilize the characteristics that the similarity between allele base sequences is relatively large, and the base sequences between non-allelic genes are usually different and relatively large to identify the alleles belonging to the gene to be identified in the gene database.Want to carry out the information identification of alleles, need to obtain the base sequence of the gene to be identified first.The acquisition process of the base sequence of the gene to be identified in the present invention is as follows: adopt polymerase chain reaction (PCR) technology to carry out gene amplification to the gene to be identified, obtain a large number of identical genes, secondly for the gene to be identified after amplification, by gene sequencing technology, such as Sanger sequencing technology, high-throughput sequencing technology, etc. carry out gene sequencing, adopt Sanger sequencing technology to obtain the base sequence W of the gene to be identified in the present invention, wherein the polymerase chain reaction technology and Sanger sequencing technology are known technologies, and the specific process is not repeated.Secondly, the base sequences of other genes belonging to homology with the gene to be identified are extracted from the gene database, such as genbank database, and the base sequence of each homologous gene is used as a comparison gene base sequence, and the set composed of all the comparison gene base sequences is recorded as homologous gene set S. It should be noted that implementers can choose appropriate gene sequencing technology and gene database according to the genes that need to be identified by information technology.

[0055] At this point, the base sequence of the gene to be identified and the aligned gene base sequence are obtained, which are used to subsequently obtain the matching base sequence to be identified and the aligned matching base sequence.

[0056] Step S002, using the Needleman-Wunsch algorithm to obtain the gene similarity score coefficient between the to-be-identified matching base sequence and each aligned matching base sequence, and obtaining the sequence matching score coefficient between the to-be-identified secondary sequence and the aligned base secondary sequence according to the base matching characteristics between the to-be-identified secondary sequence and the aligned base secondary sequence.

[0057] The base sequence of the i-th aligned gene in the homologous gene set S is recorded as S i , if the base sequence of the gene to be identified W and the base sequence of the comparison gene S i is a pair of alleles, then the base sequences W, S i There is a high base matching similarity between them, but if the above alleles are caused by gene mutation, there may be a phenomenon such as deletion or addition of the base sequence at a certain site, which causes the above two base sequences to satisfy the allele relationship.

[0058] The present invention adopts the Needleman-Wunsch algorithm to compare the base sequence W of the gene to be identified and the base sequence S of the gene to be compared. i Perform an overall comparison, the algorithm input is the base sequence W of the gene to be identified and the base sequence S of the gene to be comparedi The output is the optimal matching base sequence for each input, and the optimal matching base sequence of the gene base sequence to be identified is recorded as the matching base sequence to be identified w i , the alignment gene base sequence S output by the Needleman-Wunsch algorithm i The best matching base sequence is recorded as the matching base sequence s i , set the blank bases or spaces in the optimal matching base sequence to N, such as Figure 2 As shown, it should be noted that the matching base sequence to be identified is w i Matching base sequence s i The lengths of are equal, and the length is the number of bases in each matching base sequence. The Needleman-Wunsch algorithm is a well-known technology, and the specific process will not be repeated here.

[0059] In bioinformatics, the addition or deletion of bases has a great influence on the genes that control biological traits, such as the production of amino acids and proteins. Although the replacement of gene bases means that the bases are changed, the synthesized amino acids and proteins are synthesized normally. Therefore, in gene mutations, the addition or deletion of bases often has a greater impact on the control of traits and gene identification than the replacement of bases.

[0060] Based on the above analysis, the gene similarity score coefficient is constructed here to characterize the matching degree between the matching base sequence to be identified and each sequential base in each comparison matching base sequence, and to calculate the matching base sequence to be identified w i Matching base sequence s i The gene similarity score coefficient P(w i ,s i ):

[0061]

[0062] In the formula, Gnp(w i,k ,s i,k ) is the base w i,k 、s i,k The base matching score between i,k 、s i,k are the matching base sequences to be identified, w i Matching base sequence s i The kth base in i,k =s i,k ≠N means w i,k 、s i,k are equal and none of them are blank bases;

[0063] pw i,kis the matching score weight of the base matching score, Fd() is the prior condition between them, and Fd(w i,k ,s i,k )=Fa means w i,k 、s i,k None of them is base N; Fd(w i,k ,s i,k )=Fb means w i,k 、s i,k One of them is base N; Fd(w i,k ,s i,k )=Fc means w i,k 、s i,k All are base N;

[0064] P(w i ,s i ) is the matching base sequence to be identified w i Matching base sequence s i The gene similarity score coefficient between them, K is the matching base sequence w to be identified i The number of bases in .

[0065] Among them, the matching base sequence to be identified is w i Matching base sequence s i When the bases at the same position in the sequence are the same and not N, Gnp(w i,k ,s i,k ) is set to 1; the more the number of additions or deletions of the two bases in the kth order in the gene mutation, the greater the impact on the base sequence matching, and the matching score weight pw i,k The larger the value of is, the more likely the matching base sequence w is to be identified. i Matching base sequence s i The higher the similarity, the closer the base sequence W of the gene to be identified and the base sequence S of the gene to be compared i The higher the similarity score is after being processed by the Needleman-Wunsch algorithm.

[0066] In bioinformatics, different genes may have the same gene sequence fragments. This is because a gene may produce more than one protein. There are two genes that are not alleles. However, since both genes can produce the same protein, similar gene fragments may be matched together after the above steps. In addition, since there are actually only fixed types of gene bases and the degree of variation between bases is limited, other gene fragments may have different functions, such as different amino acids and proteins produced, although the base types at the same site of the two genes may be the same. At this time, the two genes may also have a high gene similarity score.

[0067] Based on the above analysis, the present invention will identify the matching base sequence w i Matching base sequence s i Perform uniform segmentation processing, taking m as the length of a secondary sequence, and divide the base sequence to be identified into w i , compare and match base sequences s i Divide into n secondary sequences to be identified and n secondary sequences for comparison bases, and record the jth secondary sequence to be identified and the jth secondary sequence for comparison bases as d j 、b j . Based on the matching base sequence w to be identified i , compare and match base sequences s i The base matching score of each base in the sequence matching score coefficient between each secondary sequence to be identified and the secondary sequence of the comparison base is obtained, and d j 、b j The sequence matching score coefficient F(d j ,b j ):

[0068]

[0069] In the formula, F(d j ,b j ) is d j 、b j The sequence matching score coefficient between j (c) and (b) j (c) are the cth base in the secondary sequence to be identified and the secondary sequence of the comparison base, respectively. Gnp(d j (c),b j (c)) is base d j (c) and (b) j (c) is the base matching score between the two sequences, where m is the length of the secondary sequence and the empirical value of m is 15.

[0070] At this point, the sequence matching score coefficient corresponding to each aligned base secondary sequence in each aligned base sequence is obtained, which is used to subsequently calculate the gene matching excellence index corresponding to each aligned base sequence.

[0071] Step S003, obtaining the gene difference feedback coefficient corresponding to each aligned base sequence according to the sequence matching score coefficient corresponding to all aligned base secondary sequences in each aligned base sequence, and obtaining the gene matching excellence index based on the gene difference feedback coefficient.

[0072] According to the above steps, the matching base sequences w to be identified are obtained respectively. i Each secondary sequence to be identified is compared with the matching base sequence s iThe sequence matching score coefficient between each aligned base secondary sequence in . Further, the segmentation threshold of all the sequence matching score coefficients is obtained by using the Otsu threshold algorithm, and then the aligned base secondary sequences whose sequence matching score coefficients are greater than the segmentation threshold are marked as similar sequences, and the aligned base secondary sequences whose sequence matching score coefficients are less than the segmentation threshold are marked as deviation sequences. The Otsu threshold algorithm is a well-known technology, and the specific process is not repeated here.

[0073] The set of each deviation sequence and its adjacent deviation sequence is regarded as a gene deviation group. Each gene deviation group contains at least one deviation sequence. i The number of gene deviation groups in is recorded as n1. For any gene deviation group, the absolute value of the difference between the order value of the last deviation sequence in each gene deviation group and the order value of the first deviation sequence in the next gene deviation group adjacent to each gene deviation group is taken as the gene deviation distance of the current gene deviation group. It should be noted that for the alignment matching base sequence s i The last gene deviation group in the , will align the matching base sequence s i The average of the gene deviation distances from the first gene deviation group to the n1-1th gene deviation group is taken as the gene deviation distance of the n1th gene deviation group.

[0074] Based on the above analysis, the gene difference feedback coefficient between the matching base sequence to be identified and each matching base sequence is obtained according to the gene deviation distance of all gene deviation groups in each matching base sequence, and the matching base sequence to be identified w is calculated. i Matching base sequence s i The gene difference feedback coefficient V(w i ,s i ):

[0075]

[0076] Where D f (s i ) is the group deviation coefficient of the f-th gene deviation group, and f is the alignment matching base sequence s i The f-th gene deviation group in, m1 is the number of deviation sequences in the f-th gene deviation group, D f is the gene deviation distance of the f-th gene deviation group, y is the segmentation threshold of the sequence matching score coefficient, F(b g ) is the sequence matching score coefficient corresponding to the g-th deviation sequence in the f-th gene deviation group;

[0077] V(w i ,s i ) is the matching base sequence to be identified w iMatching base sequence s i The genetic difference feedback coefficient between are the aligned matching base sequences s i The mean of the sequence matching score coefficients corresponding to all similar sequences and all deviation sequences in , n1 is the comparison matching base sequence s i The number of genetically biased groups.

[0078] Among them, the matching base sequence to be identified is w i Matching base sequence s i The lower the matching degree between the secondary sequences of the same order, the greater the difference between the matching score coefficient of the deviation sequence in the gene deviation group and the segmentation threshold, (yF(b g The larger the value of ), the smaller the interval between two adjacent gene deviation groups, and the smaller the distance between two adjacent gene deviation groups. f The smaller the value of D f (s i ) value, the larger the second multiplication factor The larger the value, the more likely the matching base sequence w i Matching base sequence s i The more secondary sequences there are with poor matching results and low similarity, The greater the difference between The larger the value of The larger the value of V(w i ,s i ) value is larger.

[0079] In bioinformatics, the lengths of genes are not always the same, and the lengths of two genes may differ greatly. For two genes with large differences, the similarity between the two genes during sequence matching is small. At the same time, if the two genes are a pair of alleles, the corresponding base sequences have a high similarity, and the lengths of the original gene sequences should not differ much.

[0080] Based on the above analysis, using the gene similarity score coefficient P(w i ,s i ) and the gene difference feedback coefficient V(w i ,s i ) Get the matching base sequence w to be identified i Matching base sequence s i The gene matching excellence index W(w i ,s i ):

[0081]

[0082] In the formula, A(wi ,s i ) is the matching base sequence to be identified w i Matching base sequence s i The gene sequence length coincidence coefficient between them, L1 and L2 are the base sequence W of the gene to be identified and the base sequence S of the gene to be compared, respectively. i length, K is the matching base sequence w to be identified i Length;

[0083] W(w i ,s i ) is the matching base sequence to be identified w i Matching base sequence s i The gene matching excellence index between i ,s i )、V(w i ,s i ) are the matching base sequences to be identified w i Matching base sequence s i The gene similarity score coefficient and gene difference feedback coefficient between min is the matching base sequence to be identified w i The minimum value of the gene similarity score coefficient between all aligned matching base sequences.

[0084] Among them, the base sequence of the gene to be identified W and the base sequence of the compared gene S i The greater the similarity, the smaller the difference between the length of the base sequence of the gene to be identified and the length of the base sequence of the compared gene. The smaller the value, the closer the base sequence W of the gene to be identified and the base sequence S of the gene to be compared i The more consistent the base lengths between the base sequences W of the gene to be identified and S of the comparison gene, the i The smaller the change in length after matching, the smaller the second difference The smaller the value, the greater the gene sequence length matching coefficient A(w i ,s i ) value is larger; the matching base sequence to be identified is w i Matching base sequence s i The higher the matching degree between the secondary sequences of the same order, the higher the P(w i ,s i ) value, V(w i ,s i ) is smaller, the first component factor The larger the value of i ,s i ) value is larger.

[0085] At this point, the gene matching excellence index corresponding to each aligned matching base sequence is obtained, which is used to subsequently construct the allele feature vector of each aligned gene base sequence.

[0086] Step S004, obtaining the allele category association index between the matching base sequence to be identified and each aligned matching base sequence according to the gene matching excellence index and the allele group of each aligned matching base sequence; obtaining the allele feature vector of each aligned gene base sequence according to the value of the gene matching excellence index corresponding to each aligned matching base sequence.

[0087] The purpose of the present invention is to identify the alleles of the base sequence W of the gene to be identified. It is impossible to directly determine which gene in the homologous gene set S is the allele of the base sequence W of the gene to be identified, but the alleles of the genes in the homologous gene set S are determined. If the base sequence W of the gene to be identified is an allele of a gene in the homologous gene set S, then the base sequence W of the gene to be identified belongs to the allele category of this gene. For example, the base sequence W of the gene to be identified is an allele of vola in the homologous gene set S, and the gene vola is also an allele of volb and volc. At this time, the base sequence W of the gene to be identified and vola, volb, and volc all belong to the vol allele category. If the base sequence W of the gene to be identified is similar to and consistent with the vola gene sequence, the gene consistency excellence index of the two is relatively large, and the gene consistency excellence index of the base sequence W of the gene to be identified and vola, volb, and volc is often relatively large.

[0088] Further, based on the comparison of gene base sequence S i Obtain the matching base sequence w to be identified for the same type of allele i Matching base sequence s i The allele class association index H(w i ,s i ):

[0089]

[0090] In the formula, H(w i ,s i ) is the matching base sequence to be identified w i Matching base sequence s i The allele class association index between i ,s i ) is the matching base sequence to be identified w i Matching base sequence s i The gene matching excellence index between them, z is the base sequence S of the aligned gene in the homologous gene set S i The zth homologous allele of W(wi ,s z ) is the matching base sequence to be identified w i The gene matching excellence index between the matched base sequences corresponding to the z-th homologous allele, M is the base sequence S of the matched gene in the homologous gene set S i The number of similar alleles.

[0091] Furthermore, K1 comparison gene base sequences are randomly selected from the homologous gene set S, and the size of K1 is taken as an empirical value of 50, and the k1th (k1≤K1) comparison gene base sequence selected is recorded as S(k1). The set composed of all homologous alleles of each comparison gene base sequence in the homologous gene set S is taken as the homologous genome of each comparison gene base sequence, and the set composed of all non-homologous alleles of each comparison gene base sequence in the homologous gene set S is taken as the non-homologous genome of each comparison gene base sequence, and the homologous genome and non-homologous genome of the comparison gene base sequence S(k1) are recorded as G1 and G2 respectively.

[0092] For example, the alignment gene base sequence S(k1) and any alignment gene base sequence in the non-similar genome G2 are respectively used as the Needleman-Wunsch algorithm, and the gene matching excellence index between the alignment gene base sequence S(k1) and each alignment gene base sequence in the non-similar genome G2 is obtained according to the above steps, and the average of the gene matching excellence indexes between the alignment gene base sequence S(k1) and all the alignment gene base sequences in the non-similar genome G2 is used as the representative value of the non-similar genome G2; secondly, the average of the representative values ​​of the non-similar genomes corresponding to K1 alignment gene base sequences is used as the first decision threshold.

[0093] Secondly, the alignment gene base sequence S(k1) and any alignment gene base sequence in the same genome G1 are respectively used as the Needleman-Wunsch algorithm, and the gene matching excellence index between the alignment gene base sequence S(k1) and each alignment gene base sequence in the same genome G1 is obtained according to the above steps, and the average of the gene matching excellence indexes between the alignment gene base sequence S(k1) and all the alignment gene base sequences in the same genome G1 is used as the representative value of the same genome G1; secondly, the average of the representative values ​​of the same genome corresponding to K1 alignment gene base sequences is used as the second decision threshold;

[0094] The allele confidence coefficient of each compared gene base sequence is obtained according to the first decision threshold and the second decision threshold. Taking the first decision threshold as an example, when the matching base sequence to be identified is w i Matching base sequence s iThe gene matching excellence index W(w i ,s i ) is less than the first decision threshold, it is considered that the base sequence of the gene to be identified W and the base sequence of the compared gene S i With a higher probability of not being an allele, the gene base sequence S i The allele confidence coefficient I(S i ) is set to 1; when the matching base sequence to be identified is w i Matching base sequence s i The gene matching excellence index W(w i ,s i ) is greater than the first decision threshold and less than the second decision threshold, it is considered that the base sequence W of the gene to be identified and the base sequence S of the gene to be compared i With a smaller probability of being an allele, the gene base sequence S i The allele confidence coefficient I(S i ) is set to 2; when the matching base sequence to be identified is w i Matching base sequence s i The gene matching excellence index W(w i ,s i ) is greater than the second decision threshold, it is considered that the base sequence W of the gene to be identified and the base sequence S of the gene to be compared i With a greater probability of being an allele, the gene base sequence S i The allele confidence coefficient I(S i ) is set to 3.

[0095] Furthermore, the vector consisting of the gene matching excellence index, allele category association index, and allele confidence coefficient corresponding to each aligned matching base sequence is used as the allele feature vector of each aligned gene base sequence corresponding to each aligned matching base sequence, and is recorded as the aligned gene base sequence S i The allele characteristic vector is denoted as T i =(W(w i ,s i ),H(w i ,s i ),I(S i )).

[0096] At this point, the allele feature vector of each aligned gene base sequence is obtained, which is used to obtain the classification result of the aligned gene base sequence.

[0097] Step S005, using a clustering algorithm to obtain classification results of all compared gene base sequences based on the allele feature vectors; and obtaining identification results of the alleles of the gene base sequence to be identified according to the classification results.

[0098] According to the above steps, the allele feature vectors of each aligned gene base sequence in the homologous gene set S are obtained respectively, and the allele feature vectors of all aligned gene base sequences are used as the input of the clustering algorithm. The k-shape clustering algorithm is used to divide all aligned gene base sequences in the homologous gene set S into K3 cluster clusters. The metric distance in the clustering process is the Euclidean distance. The k-shape clustering algorithm is a well-known technology, and the specific process is not repeated here. The allele identification result of the gene base sequence to be identified is obtained according to the obtained clustering result. The specific implementation process of the present invention is as follows Figure 3 shown.

[0099] Further, the alleles of the gene base sequence W to be identified are obtained based on K3 clusters. The specific process is as follows: Each allele feature vector contains three elements: gene matching excellence index, allele category association index, and allele confidence coefficient. i =(W(w i ,s i ),H(w i ,s i ),I(S i )) as an example, where the matching base sequence to be identified is w i Matching base sequence s i The higher the similarity between them, the higher the degree of base sequence matching. i ,s i ) value is larger; the base sequence W of the gene to be identified and the base sequence S of the gene to be compared i The greater the probability of being an allele, the greater the probability of H(w i ,s i ) value, the larger the value of I(S i ) value is bigger.The comparison gene corresponding to the cluster center corresponding allele characteristic vector of each cluster cluster is used as the representative gene of each cluster cluster, utilize PCR-RFLP (Polymerase Chain Reaction-Restriction Fragment Length Polymorphism) technology, gene to be identified, the DNA fragmentation of representative gene are copied respectively, then utilize specific enzyme that two DNA fragmentations that are copied are separated, secondly utilize electrophoresis technique to analyze and quantify the DNA fragmentation after the separation of gained, and the size and quantity of DNA fragmentation are compared, judge whether the representative gene is the allele of gene to be identified by comparing the result, and all comparison genes in the cluster cluster where the representative gene of the allele of gene to be identified is used as the allele of gene to be identified, obtain the gene information of gene to be identified based on all alleles of gene to be identified.Wherein PCR-RFLP technology and electrophoresis technique are known technologies to personnel in the art, and specific process is not repeated.

[0100] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An information-based method for identifying alleles, characterized in that: The method comprises the following steps: Respectively obtaining the base sequence of the gene to be identified and the base sequence of the comparison gene; The Needleman-Wunsch algorithm is used to obtain the gene similarity score coefficient between the matching base sequence to be identified and each aligned matching base sequence, which is used to characterize the matching degree of each sequential base in the matching base sequence to be identified and each aligned matching base sequence; The matching base sequence to be identified and each matching base sequence are divided into a secondary sequence to be identified and a secondary sequence of the matched base respectively of the same length by even segmentation; the sequence matching score coefficient between the secondary sequence to be identified and the secondary sequence of the matched base is obtained according to the matching characteristics of the bases between the secondary sequence to be identified and the secondary sequence of the matched base; A threshold segmentation algorithm is used to obtain a segmentation threshold of all the sequence matching score coefficients between the identification matching base sequence and each comparison matching base sequence; the comparison base secondary sequences whose sequence matching score coefficients are greater than the segmentation threshold are marked as similar sequences, and the comparison base secondary sequences whose sequence matching score coefficients are less than the segmentation threshold are marked as deviation sequences; The set of each adjacent deviation sequence in all aligned base secondary sequences divided by each aligned matching base sequence is regarded as a gene deviation group; The absolute value of the difference between the order value of the last deviation sequence in the current gene deviation group and the order value of the first deviation sequence in the next gene deviation group adjacent to the current gene deviation group is taken as the gene deviation distance of the current gene deviation group; Among them, for the last gene deviation group in each aligned matching base sequence, the average of the gene deviation distances from the first gene deviation group to the penultimate gene deviation group in each aligned matching base sequence is taken as the gene deviation distance of the last gene deviation group; Obtain the gene difference feedback coefficient between the to-be-identified matching base sequence and each matching base sequence according to the gene deviation distance of all gene deviation groups in each matching base sequence According to the gene difference feedback coefficient between the matching base sequence to be identified and all the matched base sequences, the gene matching excellence index between the matching base sequence to be identified and each matched base sequence is obtained; according to the gene matching excellence index and the allele group of each matched base sequence, the allele category association index between the matching base sequence to be identified and each matched base sequence is obtained; according to the value of the gene matching excellence index corresponding to each matched base sequence, the allele feature vector of each matched gene base sequence is obtained; each allele feature vector Gene matching excellence index , allele class association index , allele confidence coefficient Three elements; ,in, is the matching base sequence to be identified, is the ith matched base sequence, is the ith aligned gene base sequence; the matching base sequence to be identified Matching base sequences with alignment The higher the similarity between them, the higher the degree of base sequence matching. The larger the value of; the base sequence of the gene to be identified W, the base sequence of the gene to be compared The greater the probability of the allele, The larger the value of The larger the value of ; The homologous genome and non-homogeneous genome of each aligned gene base sequence are obtained from all aligned gene base sequences by random sampling; Obtaining the average of the gene matching excellence indexes between each compared gene base sequence and all compared gene base sequences in its corresponding non-similar genome as the representative value of the non-similar genome, and taking the average of the representative values ​​of a preset number of the non-similar genomes as the first decision threshold; Obtaining the average of the gene matching excellence indexes between each compared gene base sequence and all compared gene base sequences in its corresponding similar genome as the representative value of the similar genome, and taking the average of the representative values ​​of a preset number of the similar genomes as the second decision threshold; Obtaining the allele confidence coefficient of each aligned matching base sequence according to the comparison results of the gene matching excellence index corresponding to each aligned matching base sequence and the first decision threshold and the second decision threshold; A clustering algorithm is used to obtain classification results of all compared gene base sequences based on the allele feature vectors; and an identification result of the allele of the gene base sequence to be identified is obtained according to the classification results.

2. The information-based identification method for alleles according to claim 1, characterized in that: The method for obtaining the gene similarity score coefficient between the to-be-identified matching base sequence and each aligned matching base sequence using the Needleman-Wunsch algorithm is as follows: The base sequence of the gene to be identified and each of the compared gene base sequences are used as inputs of the Needleman-Wunsch algorithm, and the optimal matching base sequence of the base sequence of the gene to be identified output by the Needleman-Wunsch algorithm is recorded as a matching base sequence to be identified; the optimal matching base sequence of each compared gene base sequence output by the Needleman-Wunsch algorithm is recorded as an compared matching base sequence; The gene similarity score coefficient between the to-be-identified matching base sequence and each of the compared matching base sequences is obtained according to the base matching results between the to-be-identified matching base sequence and each of the compared matching base sequences.

3. The information-based identification method for alleles according to claim 2, characterized in that: The method for obtaining the gene similarity score coefficient between the to-be-identified matching base sequence and each of the compared matching base sequences according to the base matching results between the to-be-identified matching base sequence and each of the compared matching base sequences is: The matching scores of the bases in each order between the to-be-identified matching base sequence and each comparison matching base sequence that meet the preset conditions and do not meet the preset conditions are set to 1 and -1 respectively; Determine the matching score weight of the bases in each order according to the number of blank bases in each order of the matching base sequence to be identified and each comparison matching base sequence; The product of the matching score of the base in each order and the matching score weight is taken as the first cumulative factor, and the accumulation of the first cumulative factor in all orders of the matching base sequence to be identified and each compared matching base sequence is taken as the gene similarity score coefficient between the matching base sequence to be identified and each compared matching base sequence.

4. The information-based identification method for alleles according to claim 1, characterized in that: The method for obtaining the sequence matching score coefficient between the secondary sequence to be identified and the secondary sequence of the aligned base according to the matching characteristics of the bases between the secondary sequence to be identified and the secondary sequence of the aligned base is: The sum of the matching scores of the secondary sequence to be identified and the bases in all orders in each aligned base secondary sequence is taken as the sequence matching score coefficient between the secondary sequence to be identified and each aligned base secondary sequence.

5. The information-based identification method for alleles according to claim 4, characterized in that: The method for obtaining the gene difference feedback coefficient between the to-be-identified matching base sequence and each matching base sequence according to the gene deviation distance of all gene deviation groups in each matching base sequence is as follows: The difference between the mean of the sequence matching score coefficients corresponding to all similar sequences in each aligned matching base sequence and the mean of the sequence matching score coefficients corresponding to all deviation sequences is taken as the first difference; the calculation result with the natural constant as the base and the first difference as the exponent is taken as the first multiplication factor; The difference between the segmentation threshold and the sequence matching score coefficient corresponding to each deviation sequence in each gene deviation group is accumulated on each gene deviation group as the numerator, the gene deviation distance of each gene deviation group is used as the denominator, and the ratio of the numerator to the denominator is used as the group deviation coefficient of each gene deviation group; the sum of the group deviation coefficients of all gene deviation groups in each aligned matching base sequence is used as the second multiplication factor; The gene difference feedback coefficient between the to-be-identified matching base sequence and each comparison matching base sequence consists of a first multiplication factor and a second multiplication factor, wherein the gene difference feedback coefficient is in direct proportion to the first multiplication factor and the second multiplication factor.

6. The information-based identification method for alleles according to claim 1, characterized in that: The method for obtaining the gene matching excellence index between the to-be-identified matching base sequence and each of the compared matching base sequences according to the gene difference feedback coefficient between the to-be-identified matching base sequence and all the compared matching base sequences is: Obtaining the minimum value between the base sequence length of the gene to be identified and the base sequence length of each compared gene, and using the natural constant as the base and the absolute value of the difference between the minimum value and the length of the matching base sequence to be identified as the exponent as the second difference; The sum of the opposite number of the second difference and the preset parameter is used as the numerator, the calculation result with the natural constant as the base and the absolute value of the length difference between each aligned matching base sequence as the exponent is used as the denominator, and the ratio of the numerator to the denominator is used as the gene sequence length matching coefficient between the matching base sequence to be identified and each aligned matching base sequence; The difference between the gene similarity score coefficient corresponding to each aligned matching base sequence and the minimum value of the gene similarity score coefficient between the matching base sequence to be identified and all aligned matching base sequences is taken as the numerator, and the ratio of the numerator to the gene difference feedback coefficient corresponding to each aligned matching base sequence is taken as the first component factor; The gene matching excellence index between the to-be-identified matching base sequence and each comparison matching base sequence is composed of two parts: the gene sequence length matching coefficient and the first component factor, wherein the gene matching excellence index is proportional to the gene sequence length matching coefficient and the first component factor.

7. The information-based identification method for alleles according to claim 1, characterized in that: The method for obtaining the allele category association index between the to-be-identified matching base sequence and each aligned matching base sequence according to the gene matching excellence index and the allele group of each aligned matching base sequence is: Obtaining the same category allele group of each aligned gene base sequence from the gene database; taking the ratio of the sum of the gene matching excellence indexes corresponding to the aligned matching base sequences corresponding to each aligned gene base sequence in the same category allele group to the number of all aligned gene base sequences in the same category allele group as the third component factor; The sum of the gene matching excellence index corresponding to each aligned matching base sequence and the third component factor is used as the allele category association index between the matching base sequence to be identified and each aligned matching base sequence.

8. The information-based identification method for alleles according to claim 1, characterized in that: The method of using a clustering algorithm to obtain the classification results of all aligned gene base sequences based on the allele feature vector is: The allele feature vectors of all aligned gene base sequences are used as input of the clustering algorithm, and the clustering algorithm is used to obtain the classification results of the allele feature vectors of all aligned gene base sequences.

Images