JP2020178555A - Method for determining the risk of glaucoma - Google Patents

Abstract

To provide a method for determining the risk of glaucoma.SOLUTION: The present invention provides a method for determining the risk of glaucoma, on the basis of genotype information of a single nucleotide polymorphism set at least including rs12988934, rs801114, rs10074991, rs2857595, rs700651, rs228836, rs28357678, and rs364663.SELECTED DRAWING: Figure 1

Description

The present invention relates to a method for determining the risk of glaucoma.

Identification of the relationship between single nucleotide polymorphisms (hereinafter, also referred to as "SNP") and diseases is being promoted in order to use them for determining the risk of diseases. The NCBI SNP Database is a database that summarizes human SNPs, and manages SNPs with rs numbers. The rs number in the present specification also means the registration number in this NCBI SNP Database.

The relationship between the SNP specified by the rs number in the present specification and those disclosed in non-patent documents as diseases, pathological conditions, conditions, etc. related to the SNP is as follows.
rs12988934: SNP regarding monocyte count (Non-Patent Document 1)
rs801114: SNP for basal cell carcinoma (Non-Patent Document 2)
rs10074991: SNP related to gastric cancer (Non-Patent Document 3)
rs10074991: SNP for non-cardiac gastric cancer (Non-Patent Document 4)
rs2857595: SNP for laryngeal cancer (Non-Patent Document 5)
rs700651: SNP for cerebral aneurysm (Non-Patent Document 6)
rs228836: SNP regarding sitting height ratio (Non-Patent Document 7)
rs28357678: SNP on mitochondrial haplogroup (Non-Patent Document 8)
rs364663: SNP regarding the period of menarche (Non-Patent Document 9)

Okada Y, Hirota T, Kamatani Y, Takahashi A, Ohmiya H, Kumasaka N, et al. Identification of nine novel loci associated with white blood cell subtypes in a Japanese population. PLoS Genet. 2011; 7: e1002067. Stacey SN, Helgason H, Gudjonsson SA, Thorleifsson G, Zink F, Sigurdsson A, et al. New basal cell carcinoma susceptibility loci. Nat Commun 2015; 6: 6825. Hu N, Wang Z, Song X, Wei L, Kim BS, Freedman ND, et al. Genome-wide association study of gastric adenocarcinoma in Asia: a comparison of associations between cardia and non-cardia tumours. Gut 2016; 65: 1611 -8. Hu N, Wang Z, Song X, Wei L, Kim BS, Freedman ND, et al. Genome-wide association study of gastric adenocarcinoma in Asia: a comparison of associations between cardia and non-cardia tumours. Gut 2016; 65: 1611 -8. Wei Q, Yu D, Liu M, Wang M, Zhao M, Liu M, et al. Genome-wide association study identifies three susceptibility loci for laryngeal squamous cell carcinoma in the Chinese population. Nat. Genet. 2014; 46: 1110- Four. Bilguvar K, Yasuno K, Niemela M, Ruigrok YM, von Und Zu Fraunberg M, van Duijn CM, et al. Susceptibility loci for intracranial aneurysm in European and Japanese populations. Nat. Genet. 2008; 40: 1472-7. Chan Y, Salem RM, Hsu YH, McMahon G, Pers TH, Vedantam S, et al. Genome-wide Analysis of Body Proportion Classifies Height-Associated Variants by Mechanism of Action and Implicates Genes Important for Skeletal Development. Am. J. Hum . Genet. 2015; 96: 695-708. van Oven M, Kayser M. Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation. Hum Mutat. 2009; 30: E386-94. Tanikawa C, Okada Y, Takahashi A, Oda K, Kamatani N, Kubo M, et al. Genome wide association study of age at menarche in the Japanese population. PLoS ONE 2013; 8: e63821.

An object of the present invention is to provide a method for determining the risk of glaucoma (hereinafter, also referred to as "the disease").

The present inventors have diligently studied to solve the above problems. As a result, it was found that individual single nucleotide polymorphisms, which are seemingly not related to this disease at first glance, are related to this disease when viewed as one set. Then, by using the relationship, the present invention for determining the risk of the present disease has been completed.

That is, in the method of the present invention, a single nucleotide polymorphism set containing at least rs12988934, rs801114, rs10074991, rs2857595, rs700651, rs228836, rs28357678, and rs364663 found to be related to the present disease (hereinafter, "the present SNP set"". The risk of this disease is determined based on the genotype information of).

In the method of the present invention, the "single nucleotide polymorphism set" means one set of a plurality of single nucleotide polymorphisms, and the association with the present disease has been found by this one set. ..

Further, the "genotype information" in the method of the present invention is a single nucleotide polymorphism, which is classified into two homoconjugate types (AA, BB) and a heterozygous type (AB) in the single nucleotide polymorphism. It means genotype information, and "genotype information of this SNP set" means a set of genotype information of each single nucleotide polymorphism specified in this SNP set, in other words. For example, it is a set of information regarding the genotypic base of each SNP in the base sequence indicated by each rs number. The genotype information of this SNP set is as shown in FIG.

According to the present invention, the risk of this disease can be determined.

The genotype information of this SNP set is shown. An example of a conversion table showing the relationship between the genotype information of this SNP set and the values associated with the junction type for each SNP is shown. The ROC curve and AUC of the model using this SNP set are shown. Further, an SNP set containing N-1 SNPs obtained by arbitrarily removing one SNP from the present SNP set containing N SNPs is also referred to as a "comparison SNP set", and when representing each comparison SNP set, it is referred to as a "comparison SNP set". It is described as a comparative SNP set 1 and a comparative SNP set 2. The ROC curve and AUC of the model using the comparative SNP set 1 are shown. The ROC curve and AUC of the model using the comparative SNP set 2 are shown. The ROC curve and AUC of the model using the comparative SNP set 3 are shown. The ROC curve and AUC of the model using the comparative SNP set 4 are shown. The ROC curve and AUC of the model using the comparative SNP set 5 are shown. The ROC curve and AUC of the model using the comparative SNP set 6 are shown. The ROC curve and AUC of the model using the comparative SNP set 7 are shown. The ROC curve and AUC of the model using the comparative SNP set 8 are shown. The ROC curve and AUC of the model using the comparative SNP set 9 are shown. The ROC curve and AUC of the model using the comparative SNP set 10 are shown.

Embodiments of the present invention will be described. The following embodiments are examples for explaining the present invention, and the present invention is not intended to be limited to this embodiment. The present invention can be implemented in various forms as long as it does not deviate from the gist thereof.

In the present embodiment, glaucoma is a disease in which damage to the optic nerve progresses, leading to irreversible visual impairment.

Further, in the present embodiment, the present disease is generally a disease diagnosed in accordance with the guidelines published by the Medical Society regarding the present disease, and a disease described in the column of efficacy / effect in the attached document of the medical drug. Alternatively, it can be understood to mean at least one of the diseases understood as a general term in the pharmaceutical / medical industry.

In the method of the present embodiment, the risk of the disease is determined using a predetermined number of single nucleotide polymorphism sets that are seemingly unrelated to the disease.

The risk of this disease refers to the possibility of contracting this disease such as susceptibility to this disease and susceptibility to this disease. "Determining the risk" includes, for example, outputting the possibility of contracting the disease at present or in the future by dividing it into several levels, or outputting it numerically. Determining the risk of the disease includes assessing genetic factors or susceptibility to the disease, such as whether it is prone to disease or not.

In determining the risk of this disease, whether or not the subject who receives the determination of the risk of this disease is actually suffering (or developing) this disease at the time of determining the risk of this disease. Does not matter.

In the method of the present embodiment, the present SNP is a set of genotypes in which the genotypes of each SNP specified in the present SNP set are classified into two homozygous types (AA, BB) and heterozygous types (AB). Use the genotype information of the set. Then, the risk of the disease of the subject is determined based on the genotype information of the SNP set.

The SNP set used in the method of the present embodiment is a set including an SNP whose association with the disease has not been previously recognized. That is, usually, even if the SNPs included in this SNP set are analyzed individually, the risk of this disease cannot be determined. However, in the method of the present embodiment, the risk of this disease can be determined by analyzing the genotype information of each SNP included in this SNP set as a set. Further, when comparing the case where the present SNP set is analyzed and the case where the comparative SNP set is analyzed, a statistically significant result is obtained when the present SNP set is analyzed. That is, in the method of the present embodiment, by analyzing the SNP set to determine the risk of the disease, it is possible to provide a risk determination method having high accuracy or high predictive ability.

Hereinafter, in relation to each SNP included in this SNP set, the rs number and the chromosome number in which each SNP exists (in the case of a sex chromosome, it is indicated by X or Y, and in the case of mitochondrial origin, it is indicated by MT). , The position on the chromosome of each SNP and the base sequence corresponding to the rs number are listed. In the base sequence indicated by each rs number, the SNP is indicated by enclosing it in []. For example, when it is expressed as [A / G], it indicates that there is a single nucleotide polymorphism of A or G at the position of the base sequence. In addition, information such as a base sequence and a disease related to each SNP can be obtained, for example, by searching the NCBI SNP Database based on the rs number. Such information can be referred to by the relevant Database and is incorporated herein by reference. The positions on the chromosome described below correspond to the version GRCh37 of the assembly genome.

rs12988934
Chromosome number 2
Position on the chromosome 1823236665
Nucleotide sequence TTTTTGCAGTCATTCTTCTCAATGTCT [T / C] AGATTTGTAGCAAGTTGAAAAAATGT (SEQ ID NO: 1)

rs801114
Chromosome number 1
Position on the chromosome 2289977835
Nucleotide sequence ACTGCATTCTGCCCTTTTCTGACCCA [T / G] GTCCCATATTTAAAGGCTTCACATA (SEQ ID NO: 2)

rs20069911
Chromosome number 5
Position on the chromosome 407900551
Nucleotide sequence TAGGGCTTTTTGCTCCCAGCCTATA [A / G] TGTTCCAGCTCGTGTGGTTTTCCGC (SEQ ID NO: 3)

rs28557595
Chromosome number 6
Position on the chromosome 315846969
Nucleotide sequence GATAAGACTTGCCAGAGAAGTCAC [T / C] CAATCTCACCCTCTGCCAGCACGTAG (SEQ ID NO: 4)

rs700651
Chromosome number 2
Position on the chromosome 198631714
Nucleotide sequence GGAAACAGTTAATGTTTATAATATAGT [A / G] TTTAGAAACTGAATGAATAGTCGTAT (SEQ ID NO: 5)

rs228836
Chromosome number 20
Position on the chromosome 50065648
Nucleotide sequence CTCGGGGCTAAGATTAGCCCAGGAG [A / G] CTGTCACCTTTCCTTAATGTGAGAA (SEQ ID NO: 6)

rs283757678
Chromosome number MT
Position on the chromosome 14668
Nucleotide sequence CCACACTCAACAGAAACAAAGCATA [T / C] ATCATTATTTCCGCACGGACTACAA (SEQ ID NO: 7)

rs364663
Chromosome number 6
Position on the chromosome 10543189
Nucleotide sequence CACTCGTATGCTTGCCCTCCCTTCTC [A / T] CCCTCTACAAACCAATAGTCTACCAA (SEQ ID NO: 8)

In the method of the present embodiment, each SNP constituting the SNP set can be specified by referring to the base sequence specified by the rs number, but the rs number described in the present specification is merged with another rs number. When a new rs number is assigned, the corresponding rs number in the present specification also means the rs number after the merger and another rs number to be merged. Further, when the rs number described in the present specification is a number assigned by merging a plurality of rs numbers, the corresponding rs number in the present specification also means another original rs number.

Further, the above-mentioned base sequence indicated by each rs number related to SNP is shown as a specific base sequence, but even if the base sequence in the portion other than the corresponding SNP in the base sequence is changed due to differences in race or the like. Good.

The method of the present embodiment can be used for subjects of any race, but can be particularly preferably used for Asians. Among Asians, it can be preferably used by East Asian subjects such as Japanese. In addition, the method of the present embodiment may be used for a subject of any gender.

Hereinafter, one aspect of a method for determining the risk of this disease by analyzing the genotype information of this SNP set will be described. However, the determination method is not limited to the following.

First, the genotype of each SNP contained in the present SNP set in the sample is identified by using the sample of the subject. The sample used for detecting SNP is not particularly limited as long as it is a sample containing chromosomal DNA. Examples of such a sample include body fluid samples such as saliva, blood, and urine; cell samples such as oral mucosa; and body hair such as hair. For the detection of SNP, the chromosomal DNA isolated by a conventional method from these samples may be directly used, or the isolated chromosomal DNA may be amplified and the amplified chromosomal DNA may be used.

Detection of SNPs can be performed by conventional gene polymorphism analysis methods. For example, DNA chip method (DNA microarray), sequence analysis using a conventional sequencer using a Sanger method, a next-generation sequencer (NGS; Next Generation Sequencer), PCR (Polymerase Chain Reaction), hybridization, invader method, etc. However, it is not limited to these.

In the DNA chip method, a DNA chip in which a large number of DNA fragments (probes) including an SNP site are arranged on a substrate is used, chromosomal DNA is hybridized with a probe on the chip, and the binding site is detected by fluorescence or current. The sequence of chromosomal DNA is analyzed by. Examples of the DNA chip used for SNP analysis include a chip on which an oligonucleotide probe capable of detecting a base sequence containing an SNP site is arranged.

In addition, sequence analysis can be performed by a normal Sanger method. For example, a sequence reaction is performed using a primer set at a position of several tens of bases on the 5'side of a base showing a polymorphism, and the analysis result is used to determine what kind of base the corresponding position is. Can be done. Prior to the sequence reaction, it is preferable to amplify the fragment containing the SNP site in advance by PCR or the like. From the viewpoint of efficiency, NGS technology may be used.

Further, the detection of SNP can be performed, for example, by examining the presence or absence of amplification by conventional PCR. For example, a primer having a sequence corresponding to a region containing a base showing a polymorphism and having a 3'end corresponding to each polymorphism is prepared. PCR can be performed using each primer to determine which type of polymorphism is present depending on the presence or absence of amplification products. In addition, the LAMP method (Loop-Mediated Isothermal Amplification; Patent No. 33133358), the NASBA method (Nucleic Acid Sequence-Based Amplification; Patent No. 2843586; Patent No. 2843586), and the ICAN method (ISAN). It is also possible to check the presence or absence of amplification by referring to Japanese Patent No. 343929). In addition, a single chain amplification method or an analysis method using NGS may be used.

It is also possible to amplify a DNA fragment containing an SNP site and determine which type of polymorphism it is based on the difference in mobility of the amplification product in electrophoresis. Examples of such a method include the PCR-SCSP (single-strand conformation polymorphism) method (Genomics. 1992 Jan 1; 12 (1): 139-146.). Specifically, first, the DNA containing the target SNP is amplified, and the amplified DNA is dissociated into a single-stranded DNA. The dissociated single-stranded DNA can then be separated on a non-denatured gel and the type of polymorphism can be determined by the difference in mobility of the separated single-stranded DNA on the gel.

Furthermore, when a base showing a polymorphism is included in the restriction enzyme recognition sequence, it can be analyzed by the presence or absence of cleavage by the restriction enzyme (RFLP (Restriction Fragment Length Polymorphism) method). In this case, first, the DNA sample is cleaved with a restriction enzyme. The DNA fragment can then be separated and the size of the detected DNA fragment can be used to determine what type of polymorphism it is.

It is also possible to analyze the types of polymorphisms by examining the presence or absence of hybridization. That is, it is also possible to check which base the SNP is by preparing a probe corresponding to each base and checking which probe it hybridizes to.

In this way, the genotype data of the subject can be determined for each SNP of this SNP set. Here, the "genotype data of the subject" means the genotype information possessed by the subject.

The risk of the disease is then determined based on the genotype information of the SNP set. Any model can be used to determine the risk. The model is not particularly limited, but for example, a logistic regression model that uses the genotype information of this SNP set, inputs a feature amount calculated from the genotype data of the subject, and outputs the risk of this disease. Can be used. In the logistic regression model, parameters are machine-learned in advance using genotype data of humans suffering from this disease and genotype data of humans not suffering from this disease as learning data.

In addition, as a model for determining the risk of a disease, instead of the logistic regression model, an arbitrary kernel function such as a multi-layer perceptron, a neural network such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network), or a Gaussian kernel is used. Various other models such as support vector machines, random forests modeled as recurrent trees, multiple regression analysis, models using hidden Markov models, statistical models and probabilistic models can also be adopted. It is also possible to adopt a model that makes a comprehensive judgment by combining various models.

Next, an example of risk determination of this disease using a model will be described. First, the genotype data of the subject who determines the risk of this disease is converted into a feature amount that can be input to the model. The feature amount in the method of the present embodiment is, for example, whether the genotype data of the subject is homozygous (AA), homozygous (BB), or heterozygous (AB) for each SNP of the SNP set. It is a parameter indicating whether or not. The genotype is represented by nucleotides such as "GG" indicating that both SNPs of homologous chromosomes are G (guanine) and "AG" indicating that one is G (guanine) and the other is A (adenine). Therefore, the genotype data of the subject is converted into parameters that can be input to the model using the genotype information of this SNP set. However, if the model does not need to be converted to such parameters, then the above conversion is not required.

The conversion of the genotype data of the subject into a feature amount can be performed, for example, by adding a value to the genotype data of the subject with respect to each SNP included in this SNP set. For example, for each SNP, the value (eg, for example) of the SNP depends on whether the subject's genotype data corresponds to homozygous (AA), homozygous (BB), or heterozygotes (AB). , 0 or 1). This makes it possible to convert the genotype data of the subject into a feature amount. In the following, the case where the value corresponding to each SNP is 0 or 1 will be described as an example, but the value corresponding to the SNP is not limited to two values of 0 or 1.

The value associated with the junction type can be determined for each SNP. For example, one SNP associates a value of 1 when the subject's genotype data is homozygous (AA) and associates a value of 0 when it is homozygous (BB) and heterozygous (AB). Thus, the other SNPs associate a value of 1 when the subject's genotype data is heterozygous (AB) and a value of 0 when they are homozygous (AA) and homozygous (BB). May be associated with each other. In addition, if the genotype data of the subject is heterozygous (AB) and homozygous (BB), the value 1 is associated, and if the genotype data is homozygous (AA), the value 0 is associated. Good.

As described above, the genotype data of the subject can be converted into a feature amount. The value used for the association in the conversion to the feature amount can be arbitrarily determined. For example, based on the above non-patent document, a value 1 is originally associated with a genotype that is highly related to a disease associated with each SNP, and a genotype that is not related to a disease associated with each SNP is assigned. The value 0 can be associated.

The relationship between the junction type for each SNP and the value associated with the junction type can be expressed as, for example, a conversion table as shown in FIG. 2 based on the genotype information of the present SNP set as shown in FIG. it can. In the conversion table of FIG. 2, if it matches the shaded genotype, the value associated with the SNP is set to 1, and if it does not match, the value associated with it is set to 0. In the notation of specific genotypes in FIGS. 1 and 2, A indicates adenine, G indicates guanine, C indicates cytosine, and T indicates thymine. However, the format of the feature amount conversion table is not limited to FIG.

Finally, the risk of the disease in the subject is determined based on the genotype information of the SNP set. More specifically, using the conversion table based on the genotype information of this SNP set, the genotype data of the subject is calculated as a feature amount converted so that it can be input to the model, and the feature amount is calculated as a predetermined determination model. You can enter in to determine the subject's risk of this disease.

In the judgment model, the feature amount is weighted to indicate that there is a positive correlation with the risk of this disease or a weight indicating that there is a negative correlation with the risk of this disease for each SNP of this SNP set. can do. For example, the values (features) associated with rs2857595 are weighted to indicate that they have a positive correlation with the risk of this disease, and the values associated with rs700651, rs364663, rs10074991, rs28357678, rs228836, and rs801114 ( The feature amount) is weighted to indicate that it has a negative correlation with the risk of this disease, and the value (feature amount) associated with rs12988934 has a positive or negative correlation with the risk of this disease. Can be weighted to represent.

For example, when weighting features, if the genotype of rs2857595 is TC and the genotype of rs12988934 is TC, the weighting indicates that there is a positive correlation with the risk of this disease. The genotype of rs700651 is AG, the genotype of rs364663 is TT or AT, the genotype of rs10074991 is AG, the genotype of rs28357678 is TT, the genotype of rs228836 is AG, the genotype of rs801114 is TG, and the genotype of rs12988934. When the genotype is CC, weighting can be made to indicate that there is a negative correlation with the risk of the disease. In addition, in the case of the genotype of each SNP associated with the value 0 as the feature amount, it can be evaluated that there is no correlation with the risk of this disease or it is low enough to be ignored.

Such weighting representing the correlation with the risk of this disease is a parameter using genotype data of humans suffering from this disease and genotype data of humans not suffering from this disease as learning data. Is identified by machine learning. At this time, if a certain model is weighted to indicate that a certain SNP has a positive correlation with the risk of this disease, the SNP has a positive correlation with the risk of this disease in other models as well. It is usual that a weight is given to indicate that there is. That is, it is difficult to imagine a situation in which the correlation with the risk of this disease is reversed in a certain SNP depending on the type of model or the like. The specific value of the weighting differs depending on the model and is not particularly limited.

Here, a group of SNPs that are weighted to indicate that there is a positive correlation with the risk of this disease in this SNP set is called a "positive correlation SNP set" and has a negative correlation with the risk of this disease. A group of SNPs that are weighted to represent is called a "negatively correlated SNP set". This SNP set includes a positively correlated SNP set and a negatively correlated SNP set, and based on such genotype information of this SNP set, the risk of this disease of the subject is increased. It is possible to comprehensively judge both the factors that reduce the risk.

The judgment result obtained as described above is also used as an aid when a specialist of this disease diagnoses this disease. In addition, the determination result of the risk of the present disease may be corrected based on the risk of the present disease determined as described above and the result of the questionnaire from the subject. In addition, advice on improving life may be output to the subject based on the risk of this disease and the results of a questionnaire from the subject.

The present invention can also provide test reagents such as primers and probes. Examples of such a probe include a probe that includes the SNP site and can determine the type of base of the SNP site depending on the presence or absence of hybridization. In addition, examples of the primer include a primer that can be used for PCR for amplifying the SNP site, and a primer that can be used for sequence analysis of the SNP site. In addition to these primers and probes, the test reagent of the present embodiment may include a polymerase or buffer for PCR, a reagent for hybridization, and the like.

Hereinafter, this embodiment will be described in more detail with reference to Examples. However, this embodiment is not limited to these examples.

The relationship between this SNP set and this disease was verified as follows.

With the consent of more than 73,000 users of the gene analysis service, saliva samples and information on the morbidity of various diseases were collected. The morbidity information is, for example, a numerical value that becomes 1 when suffering from this disease and 0 when not suffering from this disease. The genotype data for each user was identified from the saliva sample, and a database was constructed in which the genotype data of the user was associated with various morbidity information. From this database, a case-control set of 229 subjects suffering from this disease and 229 controls not suffering from the disease was constructed.

Next, the genotypes of each SNP in this SNP set of subjects and controls were classified into two homozygous types (AA, BB) and heterozygous types (AB). Then, when the genotype matches the genotype of the shaded conversion table shown in FIG. 2, the value of x _i is set to 1, and when the genotype does not match, it is set to 0, and x _{1 to} x _N are set to the following mathematical formulas. It was used as an explanatory variable of the logistic regression model represented by (1). For example, in the case of rs801114, the value of x ₁ was set to 1 when the genotype was "TG", and the value of x ₁ was set to 0 when the genotype was "TT" or "GG". In this embodiment, N = 10. In addition, the objective variable of the logistic regression model represented by the following mathematical formula was a value p (morbidity information) between 0 and 1 representing the probability of suffering from this disease.
α = 0.1

1. 1. Model verification by AUC The accuracy of the judgment method using this SNP set will be described. From the above database, a data set in which the genotype information of the user and the morbidity information were associated with each other was created for the test. The genotypes of each SNP in this SNP set of each user in the data set are classified into homozygous types (AA, BB) and heterozygotes (AB), and the classified genotypes are shaded as shown in FIG. When the genotype was matched, the value of x _i was evaluated as 1, and when they did not match, the value was evaluated as 0, and x ₁ to x _N were calculated as feature quantities.

The feature amount related to this SNP set for each user is input to the above logistic regression model (hereinafter, also referred to as "judgment model"), it is predicted whether or not each user has this disease, and the false positive rate is predicted. The rate and the true positive rate were calculated, and the ROC (Receiver Operating Characteristic) curve and the AUC (Area Under the Curve) were obtained, respectively. More specifically, the judgment model is subjected to 5-fold cross-validation, five ROC curves (ROC fold 1 to ROC fold 5) are obtained, and the mean (Mean ROC) and standard deviation (± 1st d. Dev.) Are obtained. I asked. The broken line (Luck) in FIG. 3 is a case where whether or not the patient is suffering from this disease is randomly output, and corresponds to the ROC curve of the model having no predictive ability.

Further, in the same manner, a logistic regression model (hereinafter, also referred to as “comparison determination model”) was created for each comparative SNP set excluding one SNP from the present SNP set in the same manner as described above. Then, the feature amount related to each comparative SNP is input to each comparative judgment model, it is predicted whether or not each user has this disease, the false positive rate and the true positive rate are calculated, and the ROC curve and AUC are obtained. I asked for each. The results are shown in FIGS. 4 and later.

When this disease was judged using this SNP set, the AUC was 0.75 ± 0.04, which was significantly higher than that in the case of random output (AUC = 0.5), and the judgment using this SNP set. It can be confirmed that the predictive ability of the model is high.

On the other hand, in the case of the comparison determination model using each comparison SNP set, the AUC is lower than that in the case of using the present SNP set. Therefore, although the AUC of the comparison judgment model using each comparison SNP set is higher than that in the case of random output (AUC = 0.5), the AUC of the judgment model using this SNP set (0.75 ± 0.04). It can be confirmed that it is generally lower than.

Therefore, by making a judgment using all the SNPs included in this SNP set, it is possible to determine whether or not the patient is suffering from this disease with higher accuracy than when using each comparative SNP set excluding one SNP from this SNP set. It turned out that it can be predicted with.

2. Wilcoxon rank sum test verification Wilcoxon rank, which is a type of nonparametric test, to confirm that the judgment model using this SNP set is significantly superior to the comparative judgment model using each comparative SNP set. The sum test was performed. Specifically, the null hypothesis that there is no difference between the AUC of the judgment model using this SNP set and the AUC of the comparison judgment model using each comparative SNP set is set, and the Wilcoxon rank is set with the significance level as 0.01. The sum test was performed.

As a result, the p-values were all 3.96 × 10 ^-18 , and it was confirmed that the null hypothesis was rejected. That is, there is a statistically significant difference between the AUC of the judgment model using this SNP set and the AUC of the comparison judgment model using each comparison SNP set, and the judgment model using this SNP set uses each comparison SNP set. It can be said that the model is superior to the comparison judgment model used.

As described above, the method of the present embodiment has the effect that the accuracy of predicting whether or not the patient has the disease is significantly higher than the accuracy of the random prediction. In addition, the method of the present embodiment has a significant difference between the result of determination of this disease based on the genotype information of this SNP set and the result of determination of this disease based on the genotype information of the comparative SNP set. Has the effect of being. The effect is believed to be based on a potential correlation previously unseen between the genotype information of the SNP set and the disease. The logistic regression model and other models exemplified above are based on the genotype information of this SNP set, and learn data and disease information on the genotypes of humans suffering from this disease and those not suffering from this disease. It is obtained by machine learning the parameters by using as. That is, each model is only one phenotype representing the above-mentioned potential correlation, and the type of model used in the implementation of the method of the present embodiment is not particularly limited.

The method of the present invention determines the risk of the disease and contributes to its prevention and / or treatment in the fields related to medical treatment and healthcare.

Claims (1)

A method for determining the risk of glaucoma based on the genotype information of a single nucleotide polymorphism set containing at least rs12988934, rs801114, rs10074991, rs2857595, rs700651, rs228836, rs28357678, and rs364663.