WO2015194655A1 - METHOD, COMPUTER SYSTEM AND SOFTWARE FOR SELECTING Tag SNP, AND DNA MICROARRAY EQUIPPED WITH NUCLEIC ACID PROBE CORRESPONDING TO Tag SNP SELECTED BY SAID SELECTION METHOD - Google Patents

Abstract

The present invention addresses the problem of discovering a means for achieving the more proper selection of a Tag SNP to be contained in a nucleic acid probe, which is a probe contained in a DNA microarray or the like and used as a means for carrying out imputation, in the imputation of SNPs. Specifically, it is found that the problem can be solved by a method for selecting a Tag SNP that is used as a means for imputing information on SNPs of human genome using human genome information that includes information on a group of SNPs, in which genotypes of multiple individuals are specified, for the purpose of constituting a group of nucleic acid probes corresponding to the Tag SNP, wherein the sum total of mutual information amounts between Tag SNP candidates and Target SNPs for the candidates is employed as a measure for the selection of the Tag SNP. Thus, provided are: a computer system and a computer program which are developed on the basis of the above-mentioned principle; a DNA microarray which is equipped with a group of nucleic acid probes corresponding to a Tag SNP selected by the aforementioned means; and a method for producing the DNA microarray.

Description

Tag SNP selection method, selection computer system, selection software, and DNA microarray equipped with a nucleic acid probe corresponding to Tag SNP selected using the selection method

The present invention is an invention related to the field of genetic analysis based on nucleic acids, and more specifically, based on information about single nucleotide polymorphisms (SNPs) in the human genome, from less SNP information, with better accuracy, SNPs in individual human genomes It is an invention that provides means for deriving overall information.

It is known that the base sequence, which is the genetic code, differs from person to person in a considerable part so that our face, body shape, and personality are all different. The difference in cryptography is called polymorphism. Several types of polymorphisms are known, and among them, SNPs are currently attracting attention in relation to so-called custom-made medicine.

On the other hand, in the past medical care has been focused on investigating the cause of diseases and developing treatments. However, it is also known as reality that the same therapeutic effect is not manifested by every individual.

“Custom-made (tailor-made) medicine” means medical treatment that applies treatment methods suited to the individual patient's constitution in a custom-made manner, rather than capturing treatment methods uniformly. An essential element in knowing the constitution of each individual patient is the genetic information of the individual, and the relationship between various genetic information and constitutions and diseases is now being clarified through the decoding of the human genome. In such a situation, SNP is one of the human genetic elements that are currently focused on.

SNP is an abbreviation for single nucleotide polymorphism, which means a single base difference between individuals (single nucleotide polymorphism). Among the polymorphisms of genes, SNPs are the most numerous, and it is estimated that there are 30 million or more SNPs in the human genome. SNP is recognized as one of the most important factors when considering individual differences between humans. Currently, SNP has been analyzed for its relationship with illness and constitution, its relationship with the efficacy of drugs, and so on. Many results have been achieved.

Temporarily, personal genetic analysis is performed mainly on SNPs, and as a result, it is said that the genetic tendency of the individual, such as high blood pressure, diabetes, cancer, heart disease, stroke, and other lifestyle habits is large. If it is possible to identify the susceptibility to a disease, it is possible to take precautionary measures in advance by providing lifestyle guidance such as active meals and exercise in advance, making life more fruitful. In addition to helping, it is also expected to stop the increase in medical expenses. Even in the case of illness, it is possible to avoid unnecessary medication or dangerous medication in advance if the effectiveness of the drug and the risk of side effects are known in advance by SNP analysis.

On the other hand, it is becoming clear that there is not one kind of SNP that is directly involved in such a personal constitution, and in fact, a plurality of SNPs are related, and the analysis of SNPs is comprehensive. It has become increasingly clear that it is preferred.

Under such circumstances, attempts are now being made to apply DNA microarrays, which are used as a comprehensive analysis method of genes, to SNP analysis in the human genome.

When performing SNP analysis using a DNA microarray, the first problem is the number of SNP nucleic acid probes to be mounted on the DNA microarray. The SNP nucleic acid probe (hereinafter also referred to as “nucleic acid probe”) substantially includes a nucleotide sequence fragment on the human genome containing SNP bases or a complementary strand thereof. Currently known SNPs alone have more than 30 million, and it is technically and costly that the corresponding nucleic acid probes are comprehensively mounted on DNA microarrays and widely used for SNP detection. Have difficulty.

Therefore, the nucleic acid probes to be mounted on the DNA microarray are limited to those related to the human constitution or disease, and the nucleic acid probes to be mounted are narrowed down by performing a process called imputation. Attempts have been made.

This attempt focuses on the fact that SNPs in the genome are correlated with each other. SNPs with high correlation are concentrated in a limited region (haplotype block), and if the imputation selects an appropriate SNP (TagSNP) from the haplotype blocks, the SNP (TargetSNP) that strongly correlates with TagSNP Is a technique for narrowing down SNPs to be mounted on DNA microarrays based on the idea that genotypes can be estimated with high probability without performing typing by experiments.

The above-mentioned prior art document 1 discloses an attempt to select a TagSNP with an appropriately high probability chain using the relationship with the TargetSNP from TagSNP candidates.

However, at present, the number of nucleic acid probes mounted on a DNA microarray for SNP detection with high estimation accuracy exceeds 1 million, and the cost is high. On the other hand, if the number is smaller than that, the estimation accuracy is lowered, and there is a problem in providing predictability of diseases and the like based on accurate SNPs.

It is an object of the present invention to find a means for more appropriately selecting a Tag SNP included in a nucleic acid probe used as a means for performing imputation in a DNA microarray for SNP detection or the like when performing this imputation. It is an invention to do.

The present inventors have examined the use of “mutual information” used for prediction of RNA secondary structure and image alignment in medical image processing as an index for selecting appropriate TagSNPs. Surprisingly, the number of nucleic acid probes corresponding to Tag SNP used for DNA microarrays for SNP detection was greatly reduced, and imputation was performed based on the results obtained with the DNA microarrays. As a result, the present inventors have found that it is possible to maintain an accuracy equal to or higher than that of an existing commercial DNA microarray or the like, and completed the present invention. As described above, “SNP” in the present invention is an abbreviation for single nucleotide polymorphism (single nucleotide polymorphism) and means both singular and plural, like “nucleic acid probe”. The “group” in the “SNP group” or “nucleic acid probe group” conceptually means the presence of a large number of SNPs or nucleic acid probes, but strictly speaking, a plurality, that is, two or more SNPs or nucleic acids. It also means the presence of the probe. The “nucleic acid probe corresponding to Tag SNP” is a nucleic acid probe for specifying the SNP, and is specifically disclosed in the column of “array of the present invention” in item (3) of the embodiment of the invention. It is what is done.

The present invention provides the following inventions.

First, the present invention uses nucleic acid probe groups corresponding to Tag SNPs used as means for imputing human genome SNP information using human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified. A method for selecting the TagSNP to configure
a) Using the SNP group in the human genome information as a population, the SNPs present in the vicinity defined within a certain range from the locus of each SNP as a Tag SNP candidate are Target SNPs, and the Tag SNP candidates and these Target SNPs And calculate the sum of mutual information between
b) From all TagSNP candidates, select TagSNP candidates having a large sum of mutual information amounts as TagSNPs to be present in the nucleic acid probe used as the means for imputation, in order of increasing sum. ,
The present invention provides a method for selecting TagSNP (hereinafter also referred to as the selection method of the present invention).

Second, the present invention provides a DNA microarray (also referred to as an array of the present invention), characterized in that a nucleic acid probe corresponding to a TagSNP selected according to the selection method of the present invention is mounted. The array of the present invention can be produced by a DNA microarray production method (hereinafter also referred to as the array production method of the present invention), which comprises the following steps (1) and (2).
(1) a first step of selecting a TagSNP according to the selection method of the present invention;
(2) A second step of mounting a nucleic acid probe for detecting the genotype of the TagSNP in the human genome in the sample on the DNA microarray based on the TagSNP selected in the first step.

Third, the present invention provides the following computer system (hereinafter also referred to as the computer system of the present invention). That is, the computer system of the present invention uses a nucleic acid probe corresponding to a Tag SNP used as means for imputing SNP information of a human genome using human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified. A computer system for selecting the TagSNP to form a group, comprising a recording unit and an arithmetic processing unit;
(A) In the recording unit, information on TagSNP candidates read from the human genome information, and information on SNPs present in the vicinity defined by a certain range from the gene loci of those TagSNP candidates are used as TargetSNP information.
(1) a locus on the human genome of each TagSNP candidate;
(2) Genotypes of TagSNP candidates in individual human genome information,
(3) the locus of TargetSNP on the human genome,
(4) Genotype of TargetSNP in individual human genome information,
Is recorded at least;
(B) The arithmetic processing unit calculates the sum of mutual information amounts with the Target SNP corresponding to each Tag SNP candidate based on the information of (1) to (4) of (A) from the recording unit. , Select the TagSNP candidate with the largest sum among them, and select it as the first TagSNP;
(C) TagSNP with the sum of the maximum mutual information by the step (B) again based on the TagSNP information and TargetSNP information from which the information of the TargetSNP group corresponding to the TagSNP selected so far has been extracted Select a candidate and select as a second Tag SNP;
(D) The steps (B) and (C) are repeated, and this repetition step is performed for selecting the Mth (M is a natural number) TagSNP, and the value of the natural number M is imputed by a predetermined imputation. Performing the remaining M-2 iterations until the expected number of nucleic acid probes corresponding to the selected TagSNP is used as a means for
This is a computer system for selecting a TagSNP.

This category of “computer system” is “thing” and can be replaced with “device”.

Fourth, the present invention provides the following computer program (hereinafter also referred to as the program of the present invention). That is, the program of the present invention uses nucleic acid probe groups corresponding to Tag SNPs used as means for imputing human genome SNP information using human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified. A computer program for selecting the TagSNP to configure the computer,
(A) Target SNP information, which is information on the Tag SNP candidates read from the human genome information, and information on SNPs present in the vicinity defined in a certain range from the locus of those Tag SNP candidates,
(1) a locus on the human genome of each TagSNP candidate;
(2) Genotypes of TagSNP candidates in individual human genome information,
(3) the locus of TargetSNP on the human genome,
(4) Genotype of TargetSNP in individual human genome information,
A first function for reading the information of (1) to (4) for processing in the arithmetic processing unit from the recording unit in which is recorded;
(B) Based on the information of (1) to (4) read out by the first function, the sum of mutual information amounts with the Target SNP corresponding to each Tag SNP candidate is calculated. A second function that selects the TagSNP candidate with the largest sum and selects it as the first TagSNP;
(C) Based on the Tag SNP information and Target SNP information from which the information of the Target SNP group corresponding to the Tag SNP selected so far has been extracted, the second function again causes the maximum mutual information amount to be added. A Tag SNP candidate is selected and selected as a second Tag SNP, and then the steps (B) and (C) are repeated, and this repetition step is repeated for the selection of the Mth Tag SNP (M is a natural number). A third function that is repeated until the value of the natural number M reaches a predetermined number of nucleic acid probes corresponding to the selected Tag SNP, which is used as a means for imputation that has been determined;
A computer program characterized by including an algorithm for realizing the above.

The present invention further provides a computer-readable recording medium (hereinafter also referred to as the recording medium of the present invention), in which the program of the present invention is recorded. A typical computer system of the present invention is characterized by executing the program of the present invention.

(I) In the selection method and computer system of the present invention, “Target SNP group used to calculate the sum of mutual information amounts for each Tag SNP candidate” is a Target SNP group that has been narrowed down in advance by an index other than the mutual information amount. It is suitable from the viewpoint of selection efficiency. From the same point of view, in the program of the present invention, in the previous stage of the algorithm for realizing the second function, the Target SNP group is selected by an index other than the mutual information amount and becomes the target that exhibits the second function. It is preferable that an algorithm that realizes preliminary narrowing of the TargetSNP group is provided.

Here, the “index other than mutual information amount” is a linkage disequilibrium value between a Tag SNP candidate and a Target SNP existing in the vicinity of the Tag SNP locus and a certain range, for example, an r ² linkage disequilibrium value. And d-chain disequilibrium values are typical. When selecting Tag SNPs, it is desirable to exclude SNPs whose linkage disequilibrium values are smaller than a specific threshold value, and use other SNPs as Target SNPs as targets for calculating mutual information for Tag SNP selection. Among the “indexes other than the mutual information amount” described above, it is preferable to use the “r ² linkage disequilibrium value”. When this “r ² linkage disequilibrium value” is used, the threshold value of the value is preferably in the range of 0.70 to 0.85. If this threshold value exceeds 0.85, the prior narrowing down becomes too strict, and there is a greater risk of excluding the originally preferred Tag SNP candidate from the selection target. If it is less than 0.70, the target for calculating the sum of mutual information There is a tendency for the selection process to become inefficient due to the excessive increase in the number of prior refinements.

(II) In the present invention (selection method, computer system, program), the “vicinity defined within a certain range” from the TagSNP candidate locus is preferably within 500 kbp upstream and downstream from the TagSNP locus. Is 100 to 500 kbp.

(III) In the present invention (selection method, computer system, program), the “number of Tag SNPs to be selected” is used as the number of Tag SNPs to be selected for a nucleic acid probe used as a means for imputation. It is required that the result of performing the imputation is not less than the number satisfying the predetermined performance. The definite index of the “predetermined performance” is not particularly limited, but is preferably an index that can more objectively reflect the imputation performance of the means using the TagSNP information. is there.

As an example of this suitable index, for SNPs with MAF (minor allele frequency) of 5% or more, the average of the squares of the correlation coefficient between the genotype typed by experiment and the locus estimated by imputation is 0. 0.94 or more, preferably 0.95 or more, more preferably 0.96 or more. If it is less than this number, it cannot be said that the correlation between the result of imputation based on the selected TagSNP typing result and the actual genotype is superior to that of the conventional product, and is expected in the present invention. It becomes difficult to fully demonstrate the usefulness of the conventional product. Further, the mean square of the correlation coefficient between the genotype and the actual genotype by SNP imputation at 3 to 5% of MAF is 0.82 or more, preferably 0.84 or more, more preferably 0.87 or more. The average of the squares of the correlation coefficient between the genotype and the actual genotype by the SNP imputation at 1 to 3% of MAF is 0.73 or more, preferably 0.75 or more, more preferably 0.79 It is also possible to use the above indicators.

Although the upper limit of the number is not particularly limited, it is 1 million or less at the time of completion of the present invention, and more than 700,000 is preferable from the viewpoints of both the economical efficiency due to the number used and the certainty of the predicted content for the SNP. is there. A specific lower limit of the number is about 300,000. As shown in the examples described later, it has been clarified that even if the number is 300,000, excellent imputation exceeding the basic level based on the MAF can be performed. Preferably, about 400,000 or more, more preferably about 500,000 or more, and very preferably about 600,000 or more are assumed. Depending on the expected performance of the array of the present invention, the above MAF may be used. An appropriate number can be selected by referring to the index or the like. In Japanese Patent Application No. 2014-223834, up to 675,000 TagSNPs in Japanese were actually identified and disclosed in the specification.

The “degree” indicating the number of SNPs such as “about 300,000 or about 400,000” is the same as “about”, but in particular, a specific number of tag SNPs such as “300,000” The imputation performance does not change substantially within a certain number of widths. Specifically, if the difference is within 1% of the specific TagSNP number, strictly within 0.5%, there is no difference in substantial imputation performance. This is a value that serves as a guide when several SNPs are removed from the Tag SNP group once selected. Further, when the SNP removed from the selected Tag SNP is not actually contributing to the imputation, the influence on the imputation performance is further reduced even if the SNP is removed.

Among Tag SNP groups selected according to the selection method of the present invention, when they are actually converted into corresponding nucleic acid probes and mounted on a DNA microarray, they are not actually detected as SNPs in the population to which the present invention is applied. In addition, it is assumed that TagSNPs that do not exhibit appropriate imputation performance are slightly recognized. This is mainly clarified by ex-post verification, but it is also possible to remove such a non-functioning SNP from the Tag SNP group to be further used. Since the number of SNPs to be extracted is relatively small (about 0.1% at most), even if such extraction is performed, the above-mentioned “imputation performance is It will be well within the “substantially unchanged range”. In other words, when a specific number of Tag SNPs are selected in accordance with the selection method of the present invention, it is possible to anticipate the number of SNPs extracted corresponding to the ratio (%) as described above.

(IV) “Human genome information” used in the execution of the selection method and computer system of the present invention can be based on information in the human genome database, for example, a database for all human beings of the International 1000 Human Genome Project. However, the accuracy of SNP estimation based on TagSNP tends to be improved by using human genome information with a smaller category. Preferably, Asian Mongoloid, more specifically Japanese, Chinese, Malay, Polynesian, Micronesian, etc .; Caucasian, more specifically Italian, English, Iranian, Indian, Lap, etc .; Amerind, more specifically, Eskimo, Brazilian Indian, Alaska Indian, etc .; Negroid, more specifically, Nigerian, Bantu, Bushua, etc .; Australoid, more specifically, Australian natives, Papua New Guineans, etc. In addition, it is possible to make the category smaller, and it is also possible to accurately analyze and predict endemic disease by narrowing down to a specific region or a group of persons affected by a disease. However, in any case, the existence of specific human genome information is a prerequisite. In this example, the usefulness of the present invention was verified by performing verification based on a human genome database of 1070 Japanese people from “Tohoku University Tohoku Medical Megabank Organization (ToMMo)”.

(V) The genotype detected by the nucleic acid probe group corresponding to the TagSNP selected in the present invention (selection method, computer system, program) is used for imputing the SNP information of the human genome as described above. It is suitable as a thing. This “means for detecting the genotype detected by the nucleic acid probe group corresponding to Tag SNP” is not particularly limited as long as it can detect the genotype of SNP, and is currently provided. Or a nucleic acid detection means capable of detecting a SNP provided in the future. Specific examples include a DNA microarray, a next-generation sequencer NGS, a Sanger sequencer, and a mass array (registered trademark). Among these, one of the most suitable means at present is SNP detection using a DNA microarray provided by the above-described array of the present invention.

(VI) A specific method for producing an array of the present invention using a nucleic acid probe capable of detecting a base polymorphism in the TagSNP base can be performed according to a known method for producing a DNA microarray at the time of the present invention, It is also possible to apply a DNA microarray production method provided in the future.

(VII) Addition of other SNP Further, in the present invention, separately from Tag SNP selection, one or more other SNPs are selected and transferred in preference to the Tag SNP. It is possible to take

That is, in the selection method of the present invention, it is possible to select one or more other SNPs separately from the Tag SNP selection by the selection method of the present invention, and to carry over the Tag SNP with priority. It is also possible to mount nucleic acid probe groups corresponding to the other SNPs in the array of the present invention.

In addition, in the computer system of the present invention, one or more other SNPs should be selected separately from the Tag SNP selection by the selection method of the present invention, and these other SNPs should be selected. It is possible to carry in with priority as an SNP.

Further, in the program of the present invention, separately from the Tag SNP selection by the selection method of the present invention, one or more other SNPs are selected, and these other SNPs are selected as SNPs to be selected. It is possible to provide an algorithm that realizes the identification with priority. Hereinafter, when there is no notice in particular, "other SNP" shall mean "other 1 type (s) or 2 or more types SNP" mentioned above.

In the transfer of other SNPs as described above, it is preferable that the duplication between the other SNPs and the Tag SNP selected by the selection method of the present invention is removed. The method for removing one of the duplicated SNPs is not particularly limited. For example, the SNPs to be preferentially removed from the SNP population used when selecting the Tag SNPs may be removed in advance. Alternatively, take measures to be performed in advance, or remove SNPs that overlap with other SNPs among already selected Tag SNPs from other SNPs to be transferred later, or to remove them Such as taking or not.

As other SNPs, practically useful SNPs which are difficult to be selected by the selection method of the present invention are preferable. By preferentially using the nucleic acid probe for specifying these, an object such as further characterizing the DNA array can be achieved.

However, other SNPs are not introduced for the purpose of imputation based on them, but are used to directly detect their detection as an indicator of a specific disease or genetic substrate. . Therefore, when evaluating the imputation performance by the Tag SNP group selected by the selection method of the present invention, the extra SNP is excluded. Even if some of the other SNPs overlap with the Tag SNP, the number thereof is relatively small and can be ignored in the evaluation of the imputation performance. In Example 4-3 in Japanese Patent Application No. 2014-223834, the imputation performance is evaluated including other SNPs that have been deliberately introduced. However, this has a slight effect on the imputation performance when a considerable number (20,000 or more) of other SNPs out of about 650,000 SNPs, that is, SNPs other than Tag SNPs are included. It was done to confirm that. Specifically, 21,059 TagSNPs were extracted from the 675,000 TagSNP group, and the same number (21,059) of “other SNPs” was added instead. These "other SNP" dare including imputation performance was calculated an average value of ^{r 2} of the MAF1 ~ 3% of the SNP is 0.804, the MAF3 ~ 5% of the SNP 0.884, MAF5% Above, it was 0.959, indicating an excellent imputation performance over the existing commercial DNA array (OMNI2.5).

Practically useful SNPs that are candidates for use as “other SNPs” include (a) SNPs that have a weak linkage disequilibrium with Tag SNPs and are difficult to estimate genotypes with sufficient accuracy by imputation, (B) Y chromosome and mitochondrial SNPs, (c) SNPs that have been reported to be related to diseases from previous studies, (d) SNPs in the HLA region, (e) SNPs that have been reported to be related to drug metabolism, etc. Is mentioned. These will be described more specifically as follows.

(A) The degree of linkage disequilibrium with TagSNP is weak, and it is difficult to estimate genotype with sufficient accuracy by imputation:
Other SNPs in this classification correspond to SNPs having a low r ² linkage disequilibrium value (for example, r ² <0.2) with Tag SNPs of the present invention among Tag SNPs. From among these, it is practically preferable to select a SNP that affects the amino acid sequence of the protein.

(B) Y chromosome and mitochondrial SNPs:
For other SNP of this class, since the Y chromosome region does not occur genetic recombination, selection of TagSNP by r ² linkage disequilibrium value has no effect. Since these SNPs are small in number, it is relatively easy to select all the target SNPs regardless of the linkage disequilibrium value r ² .

(C) SNPs that have been reported to be associated with disease from previous studies:
Other SNPs of this classification are included in the database GWAS catalog (NHGRI GWAS Catalog) (http://www.genome.gov/gwastudies/: Welter, D. et al. The NHGRI GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids Res. 42, D1001-6 (2014).).

(D) HNP SNP:
For other SNP in this classification, HLA region is a region associated is reported many disease, be selected independently of the linkage disequilibrium values r ² from the TagSNP is practically suitable.

(E) SNPs reported to be associated with drug metabolism:
For other SNP in this ^{classification, Affymetrix (R) DMET TM plus} (Affymetrix, Inc) there is the following literature on the results was studied using the SNP described in these documents, as another SNP Can be used.

[Technology reviews]
・ Burmester J. K., et al. DMET microarray technology for pharmacogenomics-based personalized medicine.Methods in Molecular Biology 632: 99-124 (2010).
・ Sissung T. M., et al. Clinical pharmacology and pharmacogenetics in a genomicsera: the DMET platform.Pharmacogenomics 11 (1): 89-103 (2010).
・ Deeken J. F. The Affymetrix DMET platform and pharmacogenetics in drug development.Current Opinion in Molecular Therapeutics 11 (3): 260-268 (2009).

[Identification of new drug-related biomarkers]
Caldwell M. D., et al. CYP4F2 genetic variant alters required warfarin dose.Blood 111 (8): 4106-12 (2008).
McDonald M. G., et al. CYP4F2 Is a vitamin K1 hydroxylase: A molecular explanation for altered warfarin dose in carriers of the functionally defective V433M variant.15th North American Regional ISSX meeting Abstract 67 (2008).

[Drug development and safety research]
Mega J. L., et al. Cytochrome p-450 polymorphisms and response to clopidogrel.New England Journal of Medicine 360 (4): 354-62 (2009).
・ US Food and Drug Administration.Early communication about an ongoing safety review of clopidogrel bisulfate (marketed as Plavix).
・ Dumaual C., et al. Comprehensive assessment of metabolic enzyme and transporter genes using the Affymetrix Targeted Genotyping System.Pharmacogenomics 8 (3): 293-305 (2007).
・ Daly T. M., et al. Multiplex assay for comprehensive genotyping of genes involved in drug metabolism, excretion, and transport.Clinical Chemistry 53 (7): 1222-30 (2007).

[Genotype / phenotype databasing]
・ Man M., et al. Genetic variation in metabolizing enzyme and transporter genes: Comprehensive assessment in 3 major East Asian subpopulations with comparison to Caucasians and Africans.Journal of Clinical Pharmacology doi: 10.1177 / 0091270009355161 (2010).
・ UNC's McCleod discusses 'practical' approach to bringing pharmacogenetics to all countries.GenomeWeb Pharmacogenomics Reporter (2010).

According to the present invention, the number of Tag SNPs used in a means for imputation, such as a DNA microarray for SNP detection, can be greatly saved, and the imputation performance based on the results obtained in the means can be improved with existing commercial DNA. Provided are means capable of maintaining an accuracy equal to or higher than that of a microarray, a DNA microarray produced by the means, and a production method thereof. More specifically, the present invention makes it possible to select a nucleic acid probe for SNP detection at a low cost based on the above significant savings in the number of Tag SNPs and excellent imputation performance. Can be offered at low cost. In addition, it is possible to downsize the array detection unit necessary for excellent imputation performance by greatly reducing the number of nucleic acid probes, which will improve the performance of future gene analysis technology. It is thought that it will greatly contribute. In addition, although the examples described below disclose the results of using Japanese as a population, the present invention is inherently applicable to a population based on any race and is further different. It can also be applied to race imputation.

It is the flowchart which outlined the content of the program of this invention. It is the flowchart which expressed FIG. 1 more concretely.

One of the objects of the present invention is that, as described above, the number of Tag SNPs corresponding to the nucleic acid probes mounted on the array used in the means for performing the imputation using the DNA microarray for SNP detection is greatly increased. Select a Tag SNP group that can save and maintain an accuracy of imputation performance based on the result obtained by the means equivalent to or better than that of existing commercial DNA microarrays, etc., and select corresponding nucleic acid probes. The purpose is to prepare an on-board DNA microarray. This object can be achieved according to the selection method of the present invention described above. The selection method of the present invention can be performed by executing the program of the present invention in the computer system of the present invention.

(1) Selection method of the present invention In “human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified” in the selection method of the present invention, the means for specifying the SNP group is a next-generation sequencer (NGS) or the like. It can carry out using a well-known statistical process from the base sequence of several human genomes using.

In addition, in order to obtain linkage disequilibrium values such as “mutual information” and “r ² linkage disequilibrium value” which are indices of the selection method of the present invention, It is necessary that the frequency of the genotypes of TagSNP and TargetSNP is calculated from “genotype”. The frequency can be obtained by a conventional method. When the haplotype of the SNP group is specified, it is possible to calculate the linkage disequilibrium value and mutual information of the SNP group more precisely, which is preferable. In this case, the frequency of the aforementioned genotype is replaced with the frequency of alleles constituting the genotype, and the frequency of the combination of genotypes between the two SNPs may be replaced with the frequency of the specified haplotype. Furthermore, “fading processing”, which is a haplotype specifying means, is known.

¡Fading processing methods are roughly divided into the following two methods.

(A) Method utilizing linkage disequilibrium between segregated loci (polymorphic loci) (SHAPEIT2: Delaneau et al., Improved whole chromosome phasing for disease and population genetic studies, Nature Methods, 2013; MaCH: Li et al., MaCH: using sequence and genotype data to estimate haplotypes and unobserved genotypes, Genetic Epidemiology, 2010)
This method is a method of statistically fading, usually using genotype data of a population of 1000 or more, and is highly accurate at loci with mutations having high allele frequency (5% or more), but low allele frequency. As for the locus, the accuracy tends to be low due to the lack of the number of data, and in order to obtain high accuracy, the genotype of a huge sample population is required.

(B) Method using sequencer read information (GATK Read Backed Phasing (Developer: Broad Institute); HapCompass: Aguiar D., and Istrail S., Hapcompass: a fast cycle basis algorithm for accurate haplotype assembly of sequence data, Journal of Computational Biology, 2012)
In this method, when sequencer reads are obtained across the heterozygous loci, fading is performed by examining the bases in the leads, and fading is possible even at loci with low allele frequencies. The length of the sequencer lead is usually limited to a few hundred bp at the longest, so the fading range tends to be limited. However, with the advancement of next-generation sequencer technology, the lead length is being extended.

In the selection method of the present invention,
a) A SNP group in the human genome database as a population, and SNPs existing in the vicinity within a certain range from the locus of each SNP that is a Tag SNP candidate are set as Target SNPs, and the Tag SNP candidates and these Target SNPs To calculate the sum of mutual information.

Mutual information is defined by the following equation when two random variables x and y follow probability distributions P (x) and P (y), and the joint probability of x and y follows P (x, y). Amount.

In the present invention, x and y are two different SNP genotypes, and p (x) and p (y) correspond to their frequencies. p (x, y) is the frequency with which these genotypes are observed simultaneously in two SNPs. According to this definition, the “mutual information amount between Tag SNP candidate and Target SNP” can be calculated. In other words, as a premise for calculating the mutual information amount, in addition to the frequency of the genotype of each TagSNP candidate, the genotype of each TargetSNP existing within the vicinity defined within a certain range from the locus of each TagSNP candidate It is also necessary to calculate the frequency at which the two are simultaneously observed. However, when the haplotype of the SNP group is specified, the genotype frequency is replaced with the allele frequency constituting the genotype, and the frequency at which genotypes are simultaneously observed in the two SNPs is replaced with the haplotype frequency. That's fine.

It is possible to obtain the essential elements of the index in the selection method of the present invention by calculating the sum of each “mutual information amount between Tag SNP candidates and these Target SNPs” calculated in this way.

And b) Tag SNP candidates having a large value of the sum of mutual information among all Tag SNP candidates, as Target SNPs to be present in the nucleic acid probe used as the means for imputation, in the descending order of the sum. By selecting, the selection method of the present invention can be performed.

As described above, in the selection method of the present invention, it is preferable that the TargetSNP group is configured by being narrowed down in advance by an index other than the mutual information amount, from the viewpoint of improving the efficiency of TagSNP selection. The “r ² linkage disequilibrium value (R square value or R ^ 2)” that is particularly suitable among them is Pearson's correlation coefficient regarding the frequency of genotype between two SNPs, and a value of 0 to 1 Is an index indicating that there is a stronger linkage disequilibrium as the value approaches 1. However, when the haplotype of the SNP group is specified, the genotype frequency is replaced with the allele frequency constituting the genotype, and the frequency at which genotypes are simultaneously observed in the two SNPs is replaced with the haplotype frequency. That's fine.

The selection method of the present invention can be efficiently performed by pre-selecting a Target SNP group having a linkage disequilibrium value greater than a certain value in the linkage disequilibrium value such as the r ² linkage disequilibrium value. The threshold for selection of r ² linkage disequilibrium values has been described above. Furthermore, the “neighboring defined within a certain range” and the “number of Tag SNPs to be selected” are also described above. The “addition of other SNPs” is also described above.

(2) Computer system and computer program of the present invention The computer system of the present invention is a system that serves as means for performing the above-described selection method of the present invention, and the program of the present invention selects the present invention in the computer system of the present invention. A computer program with an algorithm for performing the method. “Algorithm” means a formalized representation of a procedure for solving a problem, as in the general concept of the computer field.

The computer system of the present invention can include hardware related to a normal computer system. That is, in addition to a “recording unit” corresponding to a normal hard disk drive and an “arithmetic processing unit” corresponding to a CPU, for example, a “temporary storage unit” corresponding to a RAM, an “operation unit” corresponding to a keyboard, a mouse, a touch panel, etc. A display unit corresponding to a display, an input / output interface (IF) unit corresponding to a serial or parallel interface corresponding to an operation unit, a video memory and a D / A converter, A “communication interface (IF) unit” that outputs a corresponding analog signal is provided. The communication IF unit can exchange data with external information, particularly human genome information such as a human genome database.

Hereinafter, unless otherwise specified, the description will be made as processing performed by the “arithmetic processing unit” of the computer system of the present invention. The “arithmetic processing unit” is operated by operating the “operation unit” to acquire the data of the human genome database in particular through the “communication IF unit” and record the data in the “recording unit”. After reading into the “temporary storage unit” and performing a predetermined process, the result is recorded in the “recording unit” again. The “arithmetic processing unit” creates screen data for prompting operation of the “operation unit” and screen data for displaying the processing results, and displays these images on the “display unit” via the video RAM of the input IF unit. To do. The program of the present invention is used or recorded in advance in the “recording unit”, or recorded in an external hardware resource, and in the “arithmetic processing unit” according to the algorithm described in the “arithmetic processing unit” as necessary. Is done.

FIG. 1 is a flowchart outlining the contents of the program of the present invention, and FIG. 2 is a flowchart more specifically expressing FIG. Step S1 is common to FIG. 1 and FIG. 2, “Target SNP, Tag SNP candidates, and genes of those loci from an input file containing information on the site (chromosome, position) for each SNP and the genotype of each individual. This is a step of “reading a type”. In the examples described below, this input file is a mutation found in a reference panel, that is, a data file of 1070 Japanese full-length genomes determined by NGS (Next Generation Sequencer) at Tohoku Medical Megabank Organization (ToMMo). A file composed of information on chromosome sites was used as an example of human genome information.

This step S1 describes the first function of the program of the present invention. That is, this step S1 is performed in human genome information including a plurality of individual genotypes.
(A) a locus on the human genome of each TagSNP candidate;
(B) Genotype of TagSNP candidate in individual human genome information,
(C) a locus on the human genome of TargetSNP,
(D) Target SNP genotype in individual human genome information,
Describes a “first function” for reading the information (a) to (d) for processing in the arithmetic processing unit from the recording unit in which is recorded.

As described above, it is possible to provide a step for giving priority to “other SNPs” as a step before step S1. In this case, it is preferable to provide a step of extracting from the other SNP candidates from the above Tag SNP candidates. This pre-feeding step is preferably provided as an alternative to the post-feeding step described later.

Step S1 'shown in FIG. 2 shows an initial setting state for the Tag SNP and Target SNP that are selected thereafter. In step S <b> 1 ′, “s” indicates the number of Tag SNPs to be selected. At this time, “s = 0”, that is, no Tag SNP is selected. In this context, “S = [0,..., 0]” is the number of Tag SNP candidates (the number of SNPs in which the number of 0s in the row [] is to be scrutinized. 1 indicates that the SNP indicated by 1 is selected as a Tag SNP candidate). “T = [0,..., 0]” indicates the same content by replacing the “TagSNP candidate” with “TargetSNP”.

Step S2 in FIG. 1 is a step of “calculating scores for all unselected Tag SNP candidates” from the human genome information read from the recording unit in step S1. In this step S2, the first half of the second function of the program of the present invention is described. Steps S2-1 (1), S2-2, S2-3 (1), S2-4, S2-5, S2-3 (2), and S2-1 (2) in FIG. Corresponds to step S2. These are collectively described as “Step S2”. Steps S2-1 (1) / (2) and Steps S2-3 (1) / (2) are each a set of loop ends.

In step S2, based on the information of (1) to (4) read out by the first function, the sum of mutual information amounts with the Target SNP corresponding to each Tag SNP candidate is calculated. A function that uses the sum as a score is described. The mutual information amount is an information concept calculated by numerical calculation of the above-described contents. As a premise for calculation, in addition to the frequency of the genotype of each Tag SNP candidate, the mutual information amount is within a certain range from the locus of each Tag SNP candidate. It is also necessary to calculate the frequency of the combination of the TagSNP candidate and the target SNP candidate genotype in each Target SNP existing in the determined neighborhood, and these frequency calculations are preferably performed in this step S2. It is.

In the present example shows a preferred embodiment for performing the threshold that defines the lower limit of r ² linkage disequilibrium values TargetSNP narrowing of calculating the mutual information for each TagSNP (R ^ 2). The method for calculating the r ² linkage disequilibrium value is as described above, and the preferable range of the threshold is also as described above, but in the examples described later, “r ² > 0.8” was used as the threshold.

Step S2-1 (1) shown in FIG. 2 is the first loop end for selecting M TagSNP candidates “i” one by one. “Score: I (i) = 0” in step S2-2 indicates initialization of the TagSNP candidate “i” selected in step S2-1 (1) at this time. Step S2-3 (1) is the first loop end for selecting N TargetSNPs “j” one by one.

Step S2-4 indicates that this is a step for determining whether or not to calculate a score. In a combination of a TagSNP candidate “i” and a TargetSNP “j” to be examined in combination with it, “L [i, j] <= L0” means that the TagSNP candidate “i” and the TargetSNP “j” are on the genome. The distance (bp) “L0” is equal to or less than a specific value. That is, “L0” indicates a distance within a predetermined range from the locus of the TagSNP candidate. The distance is as described above. Further, “R [i, j]> = R0” indicates that the r ² linkage disequilibrium value between the TagSNP candidate “i” and the TargetSNP “j” is a threshold value “R0 or more”. The threshold value is also as described above. T [j] indicates 1 if the already examined TargetSNP “j” is already covered by one or more TagSNP candidates, and indicates 0 if it is not covered. That is, if T [j] = 0, it indicates that the selected TargetSNP “j” is not covered by the TagSNP candidate “i” to be paired. In step 2-4, if the condition in the determination box is “Yes”, the process proceeds to the next step S2-5. If “No”, the process returns to step S2-3 (1) again. It is described as a step.

Step S2-5 is a step of calculating a score when “Yes” is determined in Step S2-4 and adding the value to the TagSNP candidate “i”. As described above, the “score” is the mutual information amount between the TagSNP candidate “i” and the TargetSNP “j” covered as a pair.

Step S2-3 (2) is the loop termination of step S2-3 (1) for selecting the above-described Target SNP, and step S2-1 (2) is step S2-1 (1) for selecting the above-described Tag SNP candidate. ) Loop end. These loops update the TagSNP candidate and TargetSNP pairs to be scrutinized.

Step S3 shown in FIG. 1 is a step of “selecting a Tag SNP candidate having the maximum score calculated in step S2”. This step S2 describes the second half of the second function of the program of the present invention. Steps S3-1, S3-2 (1), S3-3, and S3-2 (2) shown in FIG. ) Step S3-2 (1) / (2) is a set of loop ends.

In step S3-1, the number of the TagSNP candidate having the maximum score calculated in step S2 is set to “k”, and one of the above S value rows is set to “1” as the TagSNP to be selected. . Step S3-2 (1) starts recording that all TargetSNPs (j = 1,..., N) corresponding to the TagSNP “k” having the maximum score are covered by the TagSNP “k”. Step S3-3 is a step of determining whether or not to perform update description to T [j] = 1 in the next step S3-4. That is, the score the maximum TagSNP "k" at the moment, when r ² linkage disequilibrium values between one TargetSNP in TargetSNP group "j" corresponding to this is the threshold value "R0 or more" A determination of “yes” is made, and the process proceeds to the next step S3-4, where it is determined that the TargetSNP “j” is already covered as a TargetSNP in TagSNP “k” and updated as T [j] = 1. The Next, in step S3-2 (2), which is the loop end of step 3-2 (1), the process returns to step S3-2 (1) again, and the next Target SNP is confirmed, and all of the Target SNP groups are checked. When these confirmations for the target SNP are completed, the loop is finished and the process can proceed to the next step S4. In contrast, if TargetSNP the ^{r 2} linkage disequilibrium values "less than R0" threshold for "j", the judgment of "no" is made in step S3-3, the flow returns to step S3-2 (1) The target SNP “j” is not covered and the same confirmation is performed for the next Target SNP.

Step S4 is common to FIGS. 1 and 2 and is a step of “determining whether the total number of selected Tag SNP candidates has reached the predetermined number”. In FIG. 2, it is described as a determination that the number of mounted devices is “S0”. This step S4 describes the third function of the program of the present invention. That is, based on the Tag SNP information and Target SNP information from which the information of the Target SNP group selected in Steps S2 and S3 performing the second function has been extracted, the sum of the maximum mutual information amount is again obtained in Steps S2 and S3. (As mentioned above, in this example, the Tag SNP with the threshold of the r ² linkage disequilibrium value is selected again and selected as the second Tag SNP. Thereafter, Steps S2 and S3 are repeated, A third function is described in which this iterative process is performed until reaching the “predetermined number of means for imputing a DNA microarray for SNP detection”.

As described above, as a step after step S4, a step for giving priority to “other SNP” can be provided. In this case, it is preferable to provide a step of extracting the already selected Tag SNP from the other SNP. This post-feeding step is preferably provided alternatively to the above-described pre-feeding step.

The program of the present invention can be described in, for example, C language, Java (registered trademark), Perl, Python, etc., and can be a multi-platform.

Furthermore, the program of the present invention can be stored in a computer-readable recording medium or a recording medium that can be connected to a computer, and these recording media are also provided as the storage medium of the present invention. Examples of these recording media include magnetic media such as flexible disks, flash memories, and hard disks, optical media such as CD, DVD, and BD, magneto-optical media such as MO and MD, and the like. is not.

(3) Array of the present invention The array of the present invention is equipped with a nucleic acid probe corresponding to the TagSNP information (first step) selected using the selection method or computer system of the present invention described above (second step). ), That is, (a) a first step of selecting a TagSNP according to the selection method of the present invention; (b) detecting the genotype of the TagSNP in the human genome in the sample based on the TagSNP selected by the first step For producing the nucleic acid probe for the DNA microarray. As the second step, known methods can be widely used, and new means for producing a DNA microarray to be provided in the future can be used as long as the effects of the present invention are not impaired.

Nucleic acid probes can be prepared using, for example, gene amplification methods such as PCR and RNA PCR, DNA chemical synthesis, etc., using appropriate amplification primers for the base sequence of the human genome containing the target SNP base. By carrying out the procedure, a DNA fragment serving as a probe group can be obtained. The base length of the DNA fragment is not particularly limited, but is 10 to 100 bases, more preferably 10 to 40 bases. If the base length of the DNA fragment is long, the ability of the probe to capture the target nucleic acid containing the SNP base increases, but it tends to be unsuitable for high-density DNA microarrays. On the other hand, if the base length is short, the target nucleic acid capturing ability tends to be poor. Considering these advantages and disadvantages, the base length of the nucleic acid probe mounted on the DNA microarray can be designed and manufactured. For use as a nucleic acid probe, the DNA fragment may be modified, and a known modification method can be used. What is used for modification may be appropriately used those used in this field, such as various fluorescent dyes and coloring dyes, and is not limited thereto.

As described above, a nucleic acid probe capable of generating a capture signal on a DNA microarray by capturing a tag SNP selected based on the present invention as a target and contacting with a sample-derived DNA sample is prepared. The

It is possible to produce a DNA microarray on which a desired nucleic acid probe is mounted by attaching and immobilizing a nucleic acid probe prepared in advance on a carrier. Examples of the carrier include a solid phase carrier made of a material such as glass, plastic (eg, polypropylene, nylon, etc.), polyacrylamide, nitrocellulose, gel, other porous material or non-porous material.

Examples of the method for attaching the nucleic acid probe to the surface of the carrier include a printing method on a plate. Furthermore, as a technique for producing a high-density array, a technique for generating an array containing thousands of oligonucleotides complementary to a defined sequence at a defined position on the surface in situ using photolithography synthesis technology, Examples thereof include a method of rapidly synthesizing a predesigned DNA strand and directly attaching it to a carrier, and it is also possible to produce a DNA microarray using a masking technique. It can also be produced by an inkjet printing apparatus for oligonucleotide synthesis, and a DNA microarray using fluorescent beads or magnetic beads can be produced.

By making full use of these techniques, it is possible to produce a DNA microarray capable of detecting the TagSNP selected according to the present invention. In addition to in-house production, for example, it is possible to obtain a “commercial product” by requesting a company that has contracted production of microarrays.

The array of the present invention produced as described above detects the presence of base substitution in the Tag SNP selected by the present invention in the DNA specimen by contacting with the DNA specimen as a signal of each spot, so that the SNP can be detected. It can be confirmed whether it is a homotype or a heterotype. By integrating and organizing the obtained results and performing imputation, it is possible to estimate TargetSNP information that is not mounted on the DNA microarray, that is, other than TagSNP. Is possible. The DNA sample to be used is not particularly limited as long as it is a target from which human genomic DNA can be obtained even in a small amount. For example, blood, saliva, urine, feces, sweat, nails, hair, skin, oral tissue, semen, marrow Fluid, lymph and the like. A DNA sample can be obtained by purifying the genomic DNA in these original samples.

Examples of the present invention will be disclosed below.

[Example 1] Selection of Tag SNP As described above, a chromosomal site in which mutations were found in a data file of the entire genome of 1070 Japanese people determined using NGS (Next Generation Sequencer) at Tohoku Medical Megabank Organization (ToMMo) The computer program having the contents shown in FIG. 1 was executed on the file composed of the above information to select TagSNPs to be included in the nucleic acid probe mounted on the DNA microarray.

Here, the threshold value of “r ² linkage disequilibrium value” used for narrowing down prior Tag SNP candidates is “r ² > 0.8”, and “neighboring defined within a certain range” The selection method of the present invention was carried out at ± 500 kbp from the locus. The number of TagSNPs used for nucleic acid probes to be mounted on the DNA microarray is 675,000. The current Tag SNP candidate and Target SNP were selected from about 9.4 million SNP groups that have been analyzed in advance in Affymetrix DNA microarrays, but such prior selection is not necessarily performed. For example, the selection method of the present invention can be performed by randomly imagining a Tag SNP group and a Target SNP group from an arbitrary SNP group. It is also an efficient means to exclude a SNP having a low MAF from Tag SNP candidates in advance. Furthermore, the selection method of the present invention can also be performed based on an existing list of TagSNPs.

In this example, 675,000 Tag SNP groups selected as described above (hereinafter abbreviated as 675,000 in principle) are used for 131 SNP genotypes different from the above 1070. The performance was evaluated by imputation. First, NGS was used to identify the SNP loci and the genotypes of 131 individuals, and the genotype information corresponding to the 675,000 Tag SNP groups selected in this example was selected from the SNP loci. . Here, specifying the genotype corresponding to the TagSNP group based on the analysis result of NGS corresponds to specifying the genotype using a DNA microarray. Next, 131 SNP genotypes were estimated for 131 human genotypes corresponding to this Tag SNP group by referring to the aforementioned human genome information for 1070 people (imputation). In order to evaluate this estimation result, the square (r ² ) of the correlation coefficient between 131 genotypes estimated by imputation and the genotype specified by NGS was calculated. If the result of the estimated results are identified experimentally (NGS, etc.) is completely consistent with 131 people all, r ² becomes 1.0, will be fully estimate the true genotype, reverse the smaller the value of about r ² more different estimation has been made subject to the true genotype. The average of r ² calculated in this way to evaluate the selection result of Tag SNP was calculated as the average value for each MAF of the SNP to be estimated. As a result, the average value of ^{r 2} of the MAF1 ~ 3% of the SNP is 0.81, MAF3 ~ 5% in the SNP 0.88, the results show a very good imputation performance of 0.96 in MAF5% or more was gotten.

The above 675,000 Tag SNP groups are disclosed in Example 4 (Examples 4-1 and 4-2) of Japanese Patent Application No. 2014-223834.

[Example 2] Comparison with existing commercial DNA microarray (1)
As a comparison with the above examples, using the SNPs mounted on the existing commercial DNA microarray, the genotypes of 131 SNPs as in this example were estimated by imputation. As a result, Illumina's Human Omni 2.5-8 (hereinafter, simply referred to as OMNI2.5) average of MAF1 ~ 3 percent of SNP ^{r 2} is imputation using SNP information of 0.80, MAF3 ~ It was 0.87 for 5% SNP and 0.96 for MAF over 5%. This result is an imputation performance almost equivalent to that of the above example, but the number of SNPs mounted on the commercial DNA microarray is about 2.3 million (exactly 2,338,671). Greatly exceeded the 675,000 units. That is, if the imputation using the Tag SNP group selected by the method of the above embodiment is performed, the genotype of the SNP is much higher efficiency than when the SNP mounted on the existing commercial DNA microarray is used. It was shown that there is a great advantage in that it can be estimated.

[Example 3] Comparison with existing commercial DNA microarray (2)
Next, verification of how much the imputation performance is recognized with the number of mountings less than the above 675,000, in addition to the number of TagSNPs of 675,000, 300,000 (hereinafter, 30 Abbreviated as 10,000), 400,000 (hereinafter abbreviated as 400,000), 500,000 (hereinafter abbreviated as 500,000), and 600,000 (hereinafter abbreviated as 600,000). In the case of MAF 1 to 3%, 3 to 5%, and 5% or more. The results are shown in Table 1. In Example 4-1 of Japanese Patent Application No. 2014-223834, “300,000”, Example 4-2-1 “400,000”, and Example 4-2-2 “500,000”. In Example 4-2-3, “600,000” and Example 4-2-4 “675,000” were specifically disclosed TagSNPs used here. Yes.

From the results in Table 1, the following can be understood.
1. In view of the relative values in Table 1 above, if the number of probes mounted according to the present invention is 500,000 or more, an imputation performance equivalent to or higher than that of OMNI 2.5 can be obtained.
2. Even when the number of probes mounted according to the present invention is further reduced to 400,000, the same performance as OMNI 2.5 can be obtained.
3. Even when the number of probes obtained according to the present invention is further reduced to 300,000, although the performance is slightly inferior to that of OMNI 2.5, an equivalent performance can be obtained. The basic performance of is maintained.

From the above, by designing the DNA microarray with the probe obtained according to the present invention, the probe is mounted to about 1/10 compared to the number of 2.3 million OMNI2.5 probes mounted. It was revealed that a DNA microarray having almost the same performance as OMNI2.5 can be designed even if the number is reduced.

S1... Step S1 ′ describing the first function of the program of the present invention S1 ′... Step S2 describing the initial setting state of Tag SNP and Target SNP selected after S1. Step S2-1 (1) describing the first half of the function of Step S2-1 describing the function as the first first loop end in S2 Step S2 describing initialization of the TagSNP candidate -3 (1) ... step S2-4 describing the function as the second first loop end in S2 ... step S2-5 describing the determination as to whether or not to calculate the score Step S2-3 (2) describing addition of the score of the TagSNP for which the score has been calculated Step S2-1 (2) describing that it is the end of the loop of S2-3 (1) Step S3 describing that it is the end of the loop of S2-1 (1) above. Step S3-1 describing the selection of one TagSNP candidate with the maximum score calculated in S2. Step S3-2 (1) describing the maximum Tag SNP candidate number Step S3-3 describing the function as the first loop end in S3 Judgment whether or not to perform update description in the next step Step S3-4 for describing the step S3-2 (2) for describing the function for performing the update description Step S4 for describing the end of the loop of the above S3-2 (1) Step of describing determination of whether or not the number of selected TagSNP candidates has reached the predetermined number

Claims (38)

1. In order to construct a nucleic acid probe group corresponding to a Tag SNP used as a means for imputing human genome SNP information using human genome information including information on SNP groups in which genotypes of a plurality of individuals are specified, A method of selecting a Tag SNP,
   a) Using the SNP group in the human genome information as a population, the SNPs present in the vicinity defined within a certain range from the locus of each SNP as a Tag SNP candidate are Target SNPs, and the Tag SNP candidates and these Target SNPs And calculate the sum of mutual information between
   b) From all TagSNP candidates, select TagSNP candidates having a large sum of mutual information amounts as TagSNPs to be present in the nucleic acid probe used as the means for imputation, in order of increasing sum. ,
   A method for selecting a TagSNP, characterized in that
2. 2. The method of selecting a Tag SNP according to claim 1, wherein the human genome information is human genome database information including information on SNP groups in which genotypes of a plurality of individuals are specified.
3. 3. The TagSNP selection method according to claim 1 or 2, wherein a TargetSNP group used for calculating a sum of mutual information amounts for each of the TagSNP candidates is narrowed down in advance using an index other than the mutual information amount. .
4. 4. The method of selecting a Tag SNP according to claim 3, wherein the index other than the mutual information amount is a linkage disequilibrium value with a Target SNP group existing in the vicinity defined within a certain range from the Tag SNP candidates.
5. Linkage disequilibrium values, characterized in that it is a r ² linkage disequilibrium values, tagSNP selection method of claim 4.
6. 6. The TagSNP selection method according to any one of claims 1 to 5, characterized in that the vicinity defined in a certain range is within 500 kbp upstream and downstream from the TagSNP base.
7. The number of TagSNPs selected for the nucleic acid probe used as the means for imputation described above is equal to or greater than the number that satisfies the predetermined performance as a result of performing the imputation by the means. Item 7. The method for selecting a Tag SNP according to any one of Items 1 to 6.
8. The predetermined performance is characterized in that an average of squares of correlation coefficients between MAF 5% SNP genotype and actual genotype estimated by imputation is 0.94 or more. The method for selecting TagSNP described in 1.
9. 9. The TagSNP selection method according to claim 1, wherein the human genome information is derived from a specific race or a human group belonging to a smaller category.
10. In addition to the selection of a Tag SNP by the selection method, one or more other SNPs are selected, and these other SNPs are transferred in preference to the Tag SNP. The method for selecting TagSNP according to any one of 1 to 9.
11. The TagSNP selection method according to any one of claims 1 to 10, wherein the nucleic acid probe group is a nucleic acid probe group to be mounted on a DNA microarray.
12. A DNA microarray on which a nucleic acid probe corresponding to a TagSNP selected according to the TagSNP selection method according to claim 11 is mounted.
13. A method for producing a DNA microarray, comprising the following steps (1) and (2):
    (1) A first step of selecting a TagSNP according to the selection method according to claim 11;
    (2) A second step of mounting a nucleic acid probe for detecting the genotype of the TagSNP in the human genome in the sample on the DNA microarray based on the TagSNP selected in the first step.
14. In order to construct a nucleic acid probe group corresponding to TagSNP used as means for imputing SNP information of human genome using human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified, A computer system comprising: a recording unit and an arithmetic processing unit;
    (A) In the recording unit, information on TagSNP candidates read from the human genome information, and information on SNPs present in the vicinity defined by a certain range from the gene loci of those TagSNP candidates are used as TargetSNP information.
    (1) a locus on the human genome of each TagSNP candidate;
    (2) Genotypes of TagSNP candidates in individual human genome information,
    (3) the locus of TargetSNP on the human genome,
    (4) Genotype of TargetSNP in individual human genome information,
    Is recorded at least;
    (B) The arithmetic processing unit calculates the sum of mutual information amounts with the Target SNP corresponding to each Tag SNP candidate based on the information of (1) to (4) of (A) from the recording unit. , Among these, select the TargetSNP candidate with the largest sum and select it as the first Tag SNP;
    (C) TagSNP with the sum of the maximum mutual information by the step (B) again based on the TagSNP information and TargetSNP information from which the information of the TargetSNP group corresponding to the TagSNP selected so far has been extracted Select a candidate and select as a second Tag SNP;
    (D) The steps (B) and (C) are repeated, and this repetition step is performed for selecting the Mth (M is a natural number) TagSNP, and the value of the natural number M is imputed by a predetermined imputation. Performing the remaining M-2 iterations until the expected number of nucleic acid probes to be used as a means for achieving is reached;
    The computer system which selects TagSNP characterized by the above-mentioned.
15. The computer system for selecting a Tag SNP according to claim 14, wherein the human genome information is human genome database information including information on SNP groups in which genotypes of a plurality of individuals are specified.
16. In calculating the mutual information amount in the arithmetic processing unit, the genotype of the target SNP group is determined, and (1) the frequency of the genotype of each Tag SNP candidate, and (2) the gene locus of each Tag SNP candidate The frequency of genotypes of each TargetSNP existing in the vicinity defined within a certain range, and (3) the frequency of the combination of genotypes of the TagSNP candidate and TargetSNP candidate are calculated. The computer system which selects TagSNP of 14 or 15.
17. The TagSNP according to any one of claims 14 to 16, characterized in that, for each of the TagSNP candidates, a TargetSNP group used for calculating a sum of mutual information amounts is narrowed down in advance by an index other than the mutual information amount. Choose computer system.
18. 18. The computer system for selecting a Tag SNP according to claim 17, wherein the index other than the mutual information is a linkage disequilibrium value with a Target SNP group existing in the vicinity defined within a certain range from the Tag SNP candidates. .
19. Linkage disequilibrium values, characterized in that it is a r ² linkage disequilibrium values, the computer system for selecting a TagSNP of claim 18.
20. 20. The computer system for selecting a TagSNP according to any one of claims 14 to 19, characterized in that the vicinity defined in a certain range is within 500 kbp upstream and downstream from the TagSNP base.
21. The number of Tag SNPs to be selected for a nucleic acid probe used as a means for imputation is equal to or greater than a number that satisfies a predetermined performance as a result of imputation by the means. A computer system for selecting the TagSNP according to any one of 14 to 20.
22. The predetermined performance is characterized in that the mean square of the correlation coefficient between the SNP genotype of MAF 5% estimated by imputation and the actual genotype is 0.94 or more. The computer system which selects TagSNP of description.
23. Separately from the selection of Tag SNP in the computer system, one or more other SNPs are selected, and the other SNPs are preferentially introduced as SNPs to characterize the nucleic acid probe. A computer system for selecting the TagSNP according to any one of claims 14 to 22.
24. The computer system for selecting TagSNP according to any one of claims 14 to 23, wherein the nucleic acid probe group is a nucleic acid probe group to be mounted on a DNA microarray.
25. In order to construct a nucleic acid probe group corresponding to TagSNP used as means for imputing SNP information of human genome using human genome information including information of SNP groups in which genotypes of a plurality of individuals are specified, A computer program for selecting
    (A) Target SNP information, which is information on the Tag SNP candidates read from the human genome information, and information on SNPs present in the vicinity defined in a certain range from the locus of those Tag SNP candidates,
    (1) a locus on the human genome of each TagSNP candidate;
    (2) Genotypes of TagSNP candidates in individual human genome information,
    (3) the locus of TargetSNP on the human genome,
    (4) Genotype of TargetSNP in individual human genome information,
    A first function for reading the information of (1) to (4) for processing in the arithmetic processing unit from the recording unit in which is recorded;
    (B) Based on the information of (1) to (4) read out by the first function, the sum of mutual information amounts with the Target SNP corresponding to each Tag SNP candidate is calculated. A second function for selecting the Target SNP candidate with the largest sum and selecting it as the first Tag SNP;
    (C) Based on the Tag SNP information and Target SNP information from which the information of the Target SNP group corresponding to the Tag SNP selected so far has been extracted, the second function again causes the maximum mutual information amount to be added. A Tag SNP candidate is selected and selected as a second Tag SNP, and then the steps (B) and (C) are repeated, and this repetition step is repeated for the selection of the Mth Tag SNP (M is a natural number). A third function that is performed until the natural number M reaches a predetermined number of nucleic acid probes to be used as a means for imputation.
    The computer program characterized by including the algorithm which implement | achieves.
26. 26. The computer program according to claim 25, wherein the human genome information is human genome database information including information on SNP groups in which genotypes of a plurality of individuals are specified.
27. In the second function, (1) the frequency of the genotype of each Tag SNP candidate, (2) the frequency of the genotype of each Target SNP existing within a certain range from the locus of each Tag SNP candidate, 27. The computer program according to claim 25 or 26, further comprising: (3) an algorithm for calculating a frequency of the combination of the Tag SNP candidate and the Target SNP candidate genotype.
28. Algorithm for realizing preliminary narrowing down of TagSNP candidate groups to be subjected to the second function by selecting TagSNP candidates by an index other than the mutual information amount in a previous stage of the algorithm for realizing the second function. The computer program according to any one of claims 25 to 27, wherein:
29. 29. The computer program according to claim 28, wherein the index other than the mutual information amount is a linkage disequilibrium value with a TargetSNP group existing in the vicinity defined within a certain range from the TagSNP candidate.
30. Linkage disequilibrium values, characterized in that it is a r ² linkage disequilibrium values, the computer program of claim 29.
31. The computer program according to any one of claims 25 to 30, wherein the vicinity defined in a certain range is within 500 kbp upstream and downstream from the TagSNP base.
32. The number of Tag SNPs to be selected for a nucleic acid probe used as a means for imputation is equal to or greater than a number that satisfies a predetermined performance as a result of imputation by the means. The computer program according to any one of 25 to 31.
33. The predetermined performance is characterized in that an average of squares of correlation coefficients between MAF 5% SNP genotype and actual genotype estimated by imputation is 0.94 or more. The computer program described.
34. Independent of the selection of Tag SNPs, an algorithm is provided that enables one or more other SNPs to be selected and these other SNPs to be specified preferentially as SNPs to be selected. The computer program according to any one of claims 25 to 33, characterized by comprising:
35. The computer program according to any one of claims 25 to 34, wherein the nucleic acid probe group is a nucleic acid probe group to be mounted on a DNA microarray.
36. A computer-readable recording medium on which the computer program according to any one of claims 25 to 35 is recorded.
37. A computer system for selecting a TagSNP according to any one of claims 14 to 24, wherein the computer program according to any one of claims 25 to 35 is executed.
38. The TagSNP according to any one of claims 14 to 24 and 37, wherein the human genome information is derived from a specific race or a human group belonging to a smaller category. Computer system to do.

Images