KR102042823B1 - SNP marker set for predicting of prognosis of rheumatoid arthritis - Google Patents

Abstract

The present invention relates to a single nucleotide polymorphism (SNP) marker set for prognostic prediction of rheumatoid arthritis and a method for providing information of rheumatoid arthritis prognosis prediction using the marker set.

Description

SNP marker set for predicting of prognosis of rheumatoid arthritis}

The present invention relates to a single nucleotide polymorphism (SNP) marker set for prognostic prediction of rheumatoid arthritis and a method for providing information of rheumatoid arthritis prognosis prediction using the marker set.

Rheumatoid arthritis is a representative systemic chronic autoimmune disease mainly involving the joints. The incidence of rheumatoid arthritis is genetic-environmental, and genetic factors are known to contribute approximately 60%. Rheumatoid arthritis causes irreversible joint deformation and disorders over time if disease activity is not controlled, and joint destruction is affected by the degree of drug response to immunosuppressive agents and biological agents to control this.

In the treatment of rheumatoid arthritis, the paradigm is expected to change by selecting personalized medicine by predicting the prognosis of the disease and the response of the therapeutic drug in the future without reflecting individual differences. In particular, autoimmune diseases, including rheumatoid arthritis, account for 50% of patients who can receive personalized medical care, and the cost savings are estimated at US $ 6.8 billion (more than 7 trillion won). The single disease group is the one that can expect the greatest cost savings.

For this personalized medicine, it is very important to predict whether the patient will have a good prognosis or a poor prognosis in order to determine the treatment policy of a patient diagnosed with rheumatoid arthritis.

Recently, attempts have been made to discover genes for predicting the prognosis of rheumatoid arthritis, but genes for predicting the prognosis of rheumatoid arthritis are still unknown.

In addition, until now, research on genetic factors has been conducted mainly for foreign races, but the onset genetic factors and prognostic determinants have very different genetic backgrounds from race to race. Therefore, it is difficult to find risk factors and biomarkers through Korean genetic studies. It is essential.

Therefore, it is urgent to develop a marker for improving the accuracy of rheumatoid arthritis prognosis.

Accordingly, one object of the present invention is to provide a set of monobasic polymorphism (SNP) markers for prognostic prediction of rheumatoid arthritis, with a high accuracy of rheumatoid arthritis prognosis prediction.

An object of the present invention is to amplify a genetic sample of a subject, to obtain an amplified product of the SNP marker for predicting the prognosis of rheumatoid arthritis,

Analyzing the nucleotide sequence of the mononucleotide polymorphism site of the SNP marker, and

Providing information on the prognosis of rheumatoid arthritis prognosis comprising predicting the prognosis of rheumatoid arthritis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP marker and the clinical information of the subject to the rheumatoid arthritis prognosis model function. To provide a way.

Another object of the present invention is to contact a candidate compound with a sample of a subject,

Amplifying a genetic sample of a subject to obtain an amplification product of the SNP marker for prognostic prediction of the rheumatoid arthritis,

Analyzing the nucleotide sequence of the monobasic polymorphism site of the SNP marker,

Predicting the prognosis of rheumatoid arthritis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP marker and the clinical information of the subject to the rheumatoid arthritis prognosis model function; and

It provides a method for screening for a therapeutic agent for rheumatoid arthritis, comprising determining the candidate compound as a therapeutic agent for rheumatoid arthritis.

The present invention relates to a method for selecting a single nucleotide polymorphism (SNP) marker for predicting the prognosis of rheumatoid arthritis from a patient with rheumatoid arthritis, and providing information on the prognosis of rheumatoid arthritis prognosis using the marker.

One embodiment of the invention relates to a set of monobasic polymorphism (SNP) markers for prognostic prediction of rheumatoid arthritis, comprising one or more SNP gene markers selected from the group consisting of the following SNP gene markers:

rs1837 SNP gene marker, rs411337 SNP gene marker, rs9263800 SNP gene marker, rs2071543 SNP gene marker, rs9357155 SNP gene marker, rs2044102 SNP gene marker, rs12722588 SNP gene marker, rs3130100 SNP gene marker, rs17409828 SNP gene marker, rs32490 SNP gene marker rs10065841 SNP gene marker, rs10045084 SNP gene marker, rs154828 SNP gene marker, rs4699912 SNP gene marker, rs2929660 SNP gene marker, rs17030 SNP gene marker, rs2039223 SNP gene marker, rs2523537 SNP gene marker, rs12118313 SNP gene marker, rs12129787 SNP gene marker rs7323390 SNP gene marker, rs2281091 SNP gene marker, rs2235325 SNP gene marker, rs3132935 SNP gene marker, rs9263719 SNP gene marker, rs16822604 SNP gene marker, rs213204 SNP gene marker, rs3788368 SNP gene marker, rs10192014 SNP gene marker, rs315919 SNP gene marker rs3130014 SNP Gene Marker, rs804267 SN P gene marker, rs982764 SNP gene marker, rs4406737 SNP gene marker, rs4934433 SNP gene marker, rs213210 SNP gene marker, rs2072634 SNP gene marker, rs4421085 SNP gene marker, rs10858282 SNP gene marker, rs4766453 SNP gene marker, rs9889953 SNP gene marker, rs1256341 SNP Gene Marker, rs6583823 SNP Gene Marker, rs3824735 SNP Gene Marker, rs6602820 SNP Gene Marker, rs7255742 SNP Gene Marker, rs10754328 SNP Gene Marker, rs10932017 SNP Gene Marker, rs2834679 SNP Gene Marker, rs2280234 SNP Gene Marker, rs11894425 SNP Gene Marker, rs2020 SNP gene marker, rs3771300 SNP gene marker, rs12629751 SNP gene marker, rs7615916 SNP gene marker, rs5748470 SNP gene marker, rs16930094 SNP gene marker, rs12596236 SNP gene marker, rs478829 SNP gene marker, rs10828375 SNP gene marker, rs2229495 SNP gene marker, r1996 SNP gene marker, rs4475472 SNP Electronic marker, rs4739053 SNP gene marker, rs9358028 SNP gene marker, rs12677559 SNP gene marker, rs17722057 SNP gene marker, rs4920110 SNP gene marker, rs7595037 SNP gene marker, rs6723149 SNP gene marker, rs1558722 SNP gene marker, rs6818128 SNP gene marker, rs7671350 Gene marker, rs8109559 SNP gene marker, rs10169393 SNP gene marker, rs10185087 SNP gene marker, rs897477 SNP gene marker, rs375288 SNP gene marker, rs245478 SNP gene marker, rs2040623 SNP gene marker, rs9639288 SNP gene marker, rs6597642 SNP gene marker, rs7604115 Genetic marker, rs7901986 SNP genetic marker, and rs12777437.

Another example of the present invention is to amplify a genetic sample of a subject to obtain an amplified product of the SNP marker for predicting the prognosis of the rheumatoid arthritis,

Analyzing the nucleotide sequence of the mononucleotide polymorphism site of the SNP marker, and

Providing information on the prognosis of rheumatoid arthritis prognosis, comprising predicting the prognosis of rheumatoid arthritis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP marker and the clinical information of the subject to the rheumatoid arthritis prognosis model function. It is about how to.

Another embodiment of the present invention comprises the steps of contacting the candidate compound to the sample of the subject,

Amplifying a genetic sample of the subject, obtaining an amplified product of the SNP marker for prognostic prediction of rheumatoid arthritis according to claim 1,

Analyzing the nucleotide sequence of the monobasic polymorphism site of the SNP marker,

Predicting the prognosis of rheumatoid arthritis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP marker and the clinical information of the subject to the rheumatoid arthritis prognosis model function; and

The present invention relates to a method for screening for a therapeutic agent for rheumatoid arthritis, comprising determining the candidate compound as a therapeutic agent for rheumatoid arthritis.

Hereinafter, the present invention will be described in more detail.

The present invention relates to a single nucleotide polymorphism (SNP) marker for predicting the prognosis of rheumatoid arthritis selected from patients with rheumatoid arthritis.

The term "single nucleotide polymorphism" (SNP) refers to the difference in single base-pair variation in DNA between an individual and an individual, and is the most common form of DNA polymorphism. Known. SNPs are known to occur once in a frequency of about 1,000 bp in humans (about 1 kb). If these SNPs affect a phenotype such as a disease, the polynucleotides comprising the SNPs can be used to diagnose the disease.

The term "Rheumatoid arthritis (RA)" is an autoimmune disease that causes chronic inflammation of the entire body, including joints, and is more common in women than men, and the cause of this has not been clarified yet, but only heredity. It is only assumed to be caused by environmental and environmental factors.

The term “prognosis” refers to the degree of joint deformity (injury of the joint) after the development of rheumatoid arthritis, or the rate of progression of joint destruction by rheumatoid arthritis, ie, the increase and decrease in the severity of rheumatoid arthritis.

The term "rs number" or "single nucleotide polymorphism (SNP) rs number" as used herein refers to a unique single nucleotide that is granted when a new SNP is found in the reference SNP cluster ID database. polymorphism) As a reference number, this unique number can be used to access the SNP Cluster ID database to identify the base sequence, SNP site, and mutation of the SNP. Therefore, the person skilled in the art can identify the chromosome number, locus, nucleotide sequence, SNP site, and mutant base in which the SNP exists by specifying the single nucleotide polymorphism (SNP) rs number. Single nucleotide polymorphism (SNP) markers are specified to predict the prognosis of arthritis.

Single nucleotide polymorphism markers for prognostic prediction of rheumatoid arthritis in one embodiment of the present invention, rs1837 SNP gene marker, rs411337 SNP gene marker, rs9263800 SNP gene marker, rs2071543 SNP gene marker, rs9357155 SNP gene marker, rs2044102 SNP gene marker, rs12722588 SNP gene marker, rs3130100 SNP gene marker, rs17409828 SNP gene marker, rs32490 SNP gene marker, rs10065841 SNP gene marker, rs10045084 SNP gene marker, rs154828 SNP gene marker, rs4699912 SNP gene marker rs2929660 SNP gene marker, rs17030 SNP gene marker, rs2039223 SNP gene marker, rs2523537 SNP gene marker, rs12118313 SNP gene marker, rs12129787 SNP gene marker, rs7323390 SNP gene marker, rs2281091 SNP gene marker, rs2235325 SNP gene marker, rs3132935 SNP gene marker rs9263719 SNP Gene Marker, rs16822604 SNP Gene Marker , rs213204 SNP gene marker, rs3788368 SNP gene marker, rs10192014 SNP gene marker, rs315919 SNP gene marker, rs3130014 SNP gene marker, rs804267 SNP gene marker, rs982764 SNP gene marker, rs4406737 SNP gene marker, rs4934433 SNP gene marker, rs213210 SNP gene marker , rs2072634 SNP gene marker, rs4421085 SNP gene marker, rs10858282 SNP gene marker, rs4766453 SNP gene marker, rs9889953 SNP gene marker, rs1256341 SNP gene marker, rs6583823 SNP gene marker, rs3824735 SNP gene marker, rs6602820 SNP gene marker, rs7255742 , rs10754328 SNP gene marker, rs10932017 SNP gene marker, rs2834679 SNP gene marker, rs2280234 SNP gene marker, rs11894425 SNP gene marker, rs2066804 SNP gene marker, rs3771300 SNP gene marker, rs12629751 SNP gene marker, rs7615916 SNP gene marker, rs5748470 , rs16930094 SNP gene marker, rs12 596236 SNP Gene Marker, rs478829 SNP Gene Marker, rs10828375 SNP Gene Marker, rs2229495 SNP Gene Marker, rs1996913 SNP Gene Marker, rs4475472 SNP Gene Marker, rs4739053 SNP Gene Marker, rs9358028 SNP Gene Marker, rs12677559 SNP Gene Marker, rs17722057 SNP Gene Marker, rs17722057 rs4920110 SNP gene marker, rs7595037 SNP gene marker, rs6723149 SNP gene marker, rs1558722 SNP gene marker, rs6818128 SNP gene marker, rs7671350 SNP gene marker, rs8109559 SNP gene marker, rs10169393 SNP gene marker, rs10185087 SNP gene marker, rs897477 SNP gene marker, rs897477 at least one SNP gene selected from the group consisting of rs375288 SNP gene marker, rs245478 SNP gene marker, rs2040623 SNP gene marker, rs9639288 SNP gene marker, rs6597642 SNP gene marker, rs7604115 SNP gene marker, rs7901986 SNP gene marker, and rs12777437 SNP gene marker Markers (Tables 1a-i ).

In Tables 1a to i, the chromosome number indicates the chromosome number where SNP is present, the nucleotide sequence position indicates the position of the nucleotide sequence where SNP exists, and the changed nucleotide sequence indicates a substituted base of SNP for each SNP rs number. The first nucleotide sequence of is a major allel, except for the first nucleotide sequence is called the minor allel (Minor allel), the nearest gene is the closest in sequence to the position of the SNP The human gene is located, and the functional gene is a gene that is affected by expression, tertiary structure, etc. when the SNP is located.

TABLE 1a

TABLE 1b

TABLE 1c

TABLE 1D

TABLE 1e

TABLE 1f

Table 1g

*

Table 1h

TABLE 1i

The single nucleotide polymorphism marker for prognostic prediction of rheumatoid arthritis in one embodiment of the present invention is PLINK (multivariate conditional logistic models) from the genetic information of 250 Korean cohorts within 2 years of illness collected from Hanyang University Rheumatoid Arthritis Hospital. )) Was used to screen 35,926 SNPs from 2,551,841 SNP genome-wide association studies (GWAS), and to collect disease gene information previously associated with rheumatoid arthritis. Rheumatoid arthritis prognosis predictive target genes were collected based on the collected disease-disease association information, and 85 SNPs belonging to the collected rheumatoid arthritis related gene region were selected as a final RNP marker set for predicting prognosis of rheumatoid arthritis.

Of the 250 cohorts, the number of patients with good rheumatoid arthritis prognosis is 125, and the number of patients with poor rheumatoid arthritis prognosis is 125.

In the present invention, `` a good prognosis of rheumatoid arthritis '' means that the amount of change per year (ΔSHS / year) of the Sharp-van der Heijde (SHS) score indicating the degree of deformation (joint damage) of the joint after onset of rheumatoid arthritis is 33 % Of patients, with a poor prognosis of rheumatoid arthritis, means patients with a ΔSHS / year value of the top 33%.

Using the selected SNP marker set, the nucleotide sequence information of the SNP marker and the clinical information of the subject may be applied to the rheumatoid arthritis prognosis model function to predict the prognosis of rheumatoid arthritis according to the P-value obtained.

The present invention is to amplify a genetic sample of a subject, to obtain an amplified product of the SNP marker for predicting the prognosis of the rheumatoid arthritis,

Analyzing the nucleotide sequence of the mononucleotide polymorphism site of the SNP marker, and

A method for providing information on the prognosis of rheumatoid arthritis prognosis comprising predicting the rheumatoid arthritis prognosis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP marker and the clinical information of the subject to the rheumatoid arthritis prognosis model function. It is about.

The term "diagnosis" as used herein may include any type of analysis used to predict or predict disease prognosis and determine disease severity.

The SNP marker for prognostic prediction of rheumatoid arthritis may refer to the selected 85 SNP combinations.

The term "P value" used in the present invention refers to a probability value having a poor prognosis when applied to a prognostic prediction model function constructed by a researcher. If the calculated P value is more than 0.5, the patient predicts a poor prognosis, and if the calculated P value is less than 0.5, the prognosis is good.

In one embodiment of the present invention, in order to predict the prognosis of rheumatoid arthritis using SNP markers and clinical information, (1) to build a rheumatoid arthritis prognosis model using SNP markers, and (2) rheumatoid arthritis prognosis prediction model using clinical information (3) Rheumatoid arthritis prognosis prediction model using the constructed SNP markers and rheumatoid arthritis prognosis prediction model using clinical information were integrated.

The clinical information includes anti-CCP antibody values, ESR (Erythrocyte sedimentation rate), BMI (Body mass index), HAQ (Health Assessment Questionnaire), Baseline Sharp-van der Heijde Score, and disease duration (year) At least one selected from the group consisting of, but is not limited thereto.

In one embodiment of the present invention (1) to build a predictive model for rheumatoid arthritis prognosis using the SNP markers, the selected 85 SNPs of the 250 Korean cohort genome within 2 years of the disease collected from Hanyang University rheumatoid arthritis hospital In the nucleotide sequence of the same position, when the corresponding SNP is isomorphic and is a major allel (AA) 0, if it is heterozygous (Aa) 1, isomorphic and is a minor allel (aa ) Was converted to a value of 2 and applied to the rheumatoid arthritis prognosis prediction model function to construct a rheumatoid arthritis prognosis prediction model using an SNP marker. Preferably, the rheumatoid arthritis prognosis prediction model function is specifically defined using an SVM function. Enter the value obtained by the above conversion into the SVM function provided in the e1071 package in the R program to obtain a rheumatoid arthritis prognosis prediction model. In this case, the "kernel = radial" option was used to use the radial basis function kernel which is generally known to have the best performance among various available kernel functions.

The cross-validation analysis was performed to verify the accuracy of the rheumatoid arthritis prognosis prediction model using the constructed SNP marker. As a result, the rheumatoid arthritis prognosis prediction model using the constructed SNP marker was It was confirmed that the accuracy of the prediction of rheumatoid arthritis prognosis of 0.6015 was shown.

The rheumatoid arthritis prognosis model function may be a support vector machine (SVM), a neural network, a random forest, or a discriminant analysis, and preferably a SVM function, but is not limited thereto.

In the nucleotide sequence of the same position as each of the 85 selected SNPs in the rheumatoid arthritis prognosis prediction model using the constructed SNP marker, when the corresponding SNP is isotype and is the major allel (AA) 0, heterozygous In the case of (Aa) 1, isomorphic, and in the case of a minor allel, the value that is obtained by converting the value to (aa) 2 and inputting the value is called P _SNP .

The SVM is an algorithm for estimating a function that satisfies a given condition based on the Lagrangian optimization theory. Among them, a SVM is a support vector for applying a classification analysis method using a maximum margin classifier. This is referred to as Support Vector Classification (SVC).

In one embodiment of the present invention in order to verify the rheumatoid arthritis prognosis prediction model using the constructed SNP markers divided 250 Korean cohorts within 2 years of the disease period from Hanyang University Rheumatoid Arthritis Hospital into 10 groups of 25 people each, After selecting one test group from ten groups and training the other nine groups, the SNP model information of the training set of the nine groups is added to the SVM model as described above. Subsequent SVM model analysis was performed to build a model for predicting rheumatoid arthritis prognosis using SNP markers, and to predict the presence or absence of SNP presence information of 25 patients selected for the test set. The value was calculated by substituting the model function, and among the above 10 groups for cross-validation. SVM was randomly selected, and the remaining 9 groups were used to construct a rheumatoid arthritis prognostic prediction model function, and the selected one group was performed 10 times to verify the accuracy of the constructed rheumatoid arthritis prognostic prediction model function. The average accuracy of the prognostic prediction of the rheumatoid arthritis patient of the model is represented numerically, and as a result, the rheumatoid arthritis prognosis prediction model using the constructed SNP marker according to an embodiment of the present invention shows that the accuracy of the rheumatoid arthritis prognosis prediction is 0.6015. there was.

In one embodiment of the present invention (2) to build a rheumatoid arthritis prognosis prediction model using the clinical information, the clinical information obtained from Hanyang University rheumatoid arthritis hospital and input to the SVM (Support Vector Machine) using the clinical information rheumatoid arthritis A prognostic prediction model was constructed.

In the step of constructing the rheumatoid arthritis prognosis prediction model using the clinical information, the rheumatoid arthritis prognosis prediction model is input by inputting the numerical value of the clinical information to the SVM function in the same manner as the method of constructing the rheumatoid arthritis prognosis model using the SNP marker. Built.

The clinical information includes anti-CCP antibody values, ESR (Erythrocyte sedimentation rate), BMI (Body mass index), HAQ (Health Assessment Questionnaire), Baseline Sharp-van der Heijde Score, and disease duration (year) At least one selected from the group consisting of: ESR (Erythrocyte sedimentation rate), BMI (Body mass index), HAQ (Health Assessment Questionnaire), Baseline Sharp-van der Heijde Score (SHS), and disease duration (year) Can be obtained using a known method, and the blood extracted from the cohort 250 is applied to an anti-CCP kit (CCPlus®IMMUNOSCAN, Uro Diagnostica) for the quantification of the anti-CCP antibody, Enzyme immunoassay (Triturus system, Grifols) can be obtained by assigning a value of 1 if the result is 25 or more and a value of 0 if it is less than 25.

P _clinic is a value that is generated when a numerical value of clinical information of a subject is input into a rheumatoid arthritis prognosis prediction model using the constructed clinical information.

Cross-validation analysis performed to verify the accuracy of rheumatoid arthritis prognosis prediction model using the SNP markers to verify the accuracy of the rheumatoid arthritis prognosis prediction model using the clinical information. The experiment was performed in the same way, and as a result, it was confirmed that the rheumatoid arthritis prognosis model using the constructed clinical information indicates the accuracy of the prediction of rheumatoid arthritis prognosis of 0.7.

In one embodiment of the present invention (3) in order to integrate the rheumatoid arthritis prognosis prediction model using the constructed SNP markers and the rheumatoid arthritis prognosis prediction model using clinical information, a weighting constant used to add each SVM algorithm value ( α) is the same sequence as each of the 85 selected SNPs of 250 genomes of Korean cohorts within 2 years of disease collection from Hanyang University Rheumatoid Arthritis Hospital. ) Is (AA) 0, heteroa (Aa) 1, isomorphic, and in case of minor allel, (aa) 2 is converted to the value of the above (1) rheumatoid using the SNP marker The PSNP value is obtained by substituting the arthritis prognosis model, and the numerical value of the clinical information of the subject is input into the rheumatoid arthritis prognosis prediction model using the constructed (2) clinical information. In this case, after obtaining P _Clinic , a P _sum value is obtained by applying a weight sum method of adding two weights to the two probability values as shown in Equation 1 below, and the P _sum value is 0.5. When the value of 1 to 1, the prognosis of the rheumatoid arthritis prognosis of the patient is determined to be bad, and when the value of P _sum is less than 0.5, a model for determining the prognosis of the rheumatoid arthritis of the patient is established, and then the constructed model is established. With respect to the weight α of the following Equation 1, the experiment was carried out in the same manner as in the above-described cross analysis while increasing the weight α of 0.3 from 0.6 to 0.01 to obtain the weight value α of the most accurate case.

The α value, which is the weight when the accuracy of the most prognostic prediction of rheumatoid arthritis determined in the process of obtaining the α value, which is the weight, is the highest, is 0.40.

[Equation 1]

P _sum = αP _SNP + (1-α) P _Clinic

In Equation 1,

P _SNP is a P-value obtained by applying the nucleotide sequence information of the SNP marker to the rheumatoid arthritis prognosis model function,

P _clinic is the P-value obtained by applying the clinical information of the subject to the predictive model function of rheumatoid arthritis.

P _sum is a P-value obtained by weighted _sum of the P _SNP value and P _clinic value,

α is a constant used when the weighted sum of the P _SNP value and the P _clinic value.

In an embodiment of the present invention, cross-validation is performed in the same manner as described above to verify the rheumatoid arthritis prognosis prediction model using the constructed SNP markers and clinical information. The prediction of rheumatoid arthritis prognosis using the rheumatoid arthritis prognosis model was improved by 131% compared to the prediction of rheumatoid arthritis prognosis using the SNP marker. Using the rheumatoid arthritis prognosis prediction model, it was confirmed that the accuracy of the rheumatoid arthritis prognosis prediction was improved by 113% compared to the prediction of the rheumatoid arthritis prognosis.

In the base sequence at the same position as each of the selected 85 SNPs of the subject, the SNP is homozygous for the rheumatoid arthritis prognostic model using the constructed SNP marker. ) 0, heteromorphic (Aa) 1, homomorphic, minor allel, (aa) 2, the value obtained by substituting the corresponding value (P _SNP ), and the constructed clinical information by the prognosis prediction model using RA substitutes the clinical information value of the subject can be obtained by using the obtained _sum P value (P _clinic).

In the rheumatoid arthritis prognosis prediction model using the constructed SNP marker and clinical information, when the value of P _sum value of Equation 1 is 0.5 to 1, the prognosis of rheumatoid arthritis in the subject may be judged to be bad, and the constructed SNP marker In the rheumatoid arthritis prognosis prediction model using clinical information, when the value of P _sum value of Equation 1 is less than 0.5, the prognosis of rheumatoid arthritis in a subject may be judged to be good.

The present invention comprises the steps of contacting a candidate compound to a sample of a subject, amplifying an SNP marker for prognostic prediction of the rheumatoid arthritis from the sample, analyzing the nucleotide sequence of the monobasic polymorphism site of the marker, the assay A method for screening a rheumatoid arthritis therapeutic agent comprising applying nucleotide sequence information and clinical information of the subject to a rheumatoid arthritis prognosis model function, and determining the candidate compound as a rheumatoid arthritis therapeutic agent.

The SNP markers for predicting the prognosis of rheumatoid arthritis selected from the patients with rheumatoid arthritis may be a combination of a single nucleotide polymorphism marker for prognostic prediction of the 85 rheumatoid arthritis screened in one embodiment of the present invention.

In the present invention, after treating an unknown candidate compound for rheumatoid arthritis patients, rheumatoid using SNP markers and clinical information constructed using a single nucleotide polymorphism (SNP) marker for predicting the prognosis of rheumatoid arthritis selected from the patients with rheumatoid arthritis The candidate compound may be determined as a therapeutic agent for rheumatoid arthritis if the rheumatoid arthritis prognosis of the patient who has contacted the candidate compound is better than the rheumatoid arthritis prognosis of the patient who has not contacted the candidate compound using an arthritis prognosis prediction model.

Meaning that the prognosis of rheumatoid arthritis in patients who contacted the candidate compound is better than that of rheumatoid arthritis in patients not contacting the candidate compound means that SNP markers and clinical information are used after treating an unknown candidate compound for rheumatoid arthritis patients. It may mean that the value of P _sum of the value of Equation 1 of the rheumatoid arthritis prognosis prediction model is less than 0.5.

The present invention relates to a single nucleotide polymorphism (SNP) marker set for prognostic prediction of rheumatoid arthritis and a method for providing information of prognostic prediction of rheumatoid arthritis using the marker set, wherein the single base for prognostic prediction of rheumatoid arthritis of the present invention A set of polymorphic markers can be used to predict the prognosis of rheumatoid arthritis.

1 is a schematic diagram of a final rheumatoid arthritis prediction model constructed by integrating 85 prognostic predictive SNP markers of 85 rheumatoid arthritis and existing clinical information according to an embodiment of the present invention.
2 is a diagram showing the verification of the prognostic model for rheumatoid arthritis using the SNP marker according to an embodiment of the present invention.
3 is a diagram showing confirming the optimum weight α in accordance with an embodiment of the present invention.
4 is a diagram showing the verification of the prognostic model for rheumatoid arthritis using SNP markers and clinical information according to an embodiment of the present invention.

Hereinafter, the present invention will be described in more detail with reference to the following examples. However, these examples are only for illustrating the present invention, and the scope of the present invention is not limited by these examples.

Example  One. Korean Wave Arthritis  Prognostic SNPs Marker  Selection

SNP markers for use in predicting the prognosis of rheumatoid arthritis were selected using the following method.

Specifically, 250 Korean cohorts were collected from Hanyang University Rheumatoid Arthritis Hospital within 2 years of the disease period, and genomic DNA was extracted from the blood and the genomic DNA was extracted using multivariate conditional logistic models. We applied the program to select statistically significant 35,926 SNPs in patients with good prognosis and patients with poor prognosis of 2,551,841 SNP genome-wide association study (GWAS), and the disease genes previously associated with rheumatoid arthritis. To collect the information, the target genes for predicting rheumatoid arthritis prognosis were collected based on the gene-disease association information collected by DisGeNet and the disease-disease association information collected by MimMiner, and 85 SNPs belonging to the collected rheumatoid arthritis related gene sites were collected. End Korean Rheumatism SNP markers for predicting arthritis prognosis were selected.

85 SNP markers for predicting rheumatoid arthritis prognosis selected by the above method are described in Tables 1a to i. In Table 1a to i, the chromosome number indicates the chromosome number where the SNP is present, the nucleotide sequence position indicates the position of the nucleotide sequence where the SNP exists, and the changed nucleotide sequence indicates the position and substituted base of the SNP for each SNP rs number. The close gene means a human gene located closest in the sequence from the position of the SNP, and the functional gene means a gene affected by expression, tertiary structure, etc. when the SNP is located.

Of the 250 cohorts, the number of patients with good rheumatoid arthritis prognosis is 125, and the number of patients with poor rheumatoid arthritis prognosis is 125. `` Rheumatoid arthritis prognosis '' means 33% of patients with a small change in value per year (ΔSHS / year) of the Sharp-van der Heijde (SHS) score, which indicates the degree of joint deformity (injury of the joint) after rheumatoid arthritis. Poor prognosis for rheumatoid arthritis' means patients with a ΔSHS / year value of the top 33%.

The characteristics of the cohort are shown in Table 2 below.

TABLE 2

Example  2. SNP Marker  Used Rheumatoid arthritis  Prognosis prediction method

2-1. SNP Marker  Used Rheumatoid arthritis  Prognostic Prediction Model

In order to construct a rheumatoid arthritis prognosis prediction model using the selected SNP markers, an SVM (Support Vector Machine) model analysis using 85 SNP sets selected in Example 1 was performed.

Specifically, in the base sequence at the same position as each of the 85 SNPs selected in Example 1 in the 250 cohort genome used in Example 1, when the corresponding SNP is homologous and is a major allel (AA) 0 , If it is heteromorphic (Aa) 1, isomorphic, and if it is a minor allel, convert it to the value of (aa) 2 and input the value into the SVM function provided by the e1071 package in the R program to predict rheumatoid arthritis prognosis model In this case, the "kernel = radial" option was used to use the radial basis function kernel, which is generally known to have the best performance among various available kernel functions.

2-2. SNP Marker  Used Rheumatoid arthritis  Validation of Prognostic Prediction Model

In order to verify the accuracy of the rheumatoid arthritis prognostic model function using the SNP marker constructed in Example 2-1, the following experiment was performed.

Specifically, 250 cohorts used in Example 1 were divided into 10 groups of 25 persons, one group of the 10 groups was randomly selected (Test set), and the remaining 9 groups were combined (Training set). SNP model analysis was performed by assigning the SNP presence information of the training set of 9 groups into the SVM model in the same manner as in Example 2-1, constructing a rheumatoid arthritis prognosis predictive model function using the SNP marker, and selecting the randomly. (Test set) The SNP presence information of 25 patients was substituted into the rheumatoid arthritis prognosis prediction model function using the constructed SNP markers, and the value was calculated. Among the 10 groups for cross-validation, One group is randomly selected, and the remaining nine groups are used to build a rheumatoid arthritis prognostic model function. The procedure of verifying the accuracy of the rheumatoid arthritis prognosis prediction model function was performed ten times, and the average accuracy of the prognosis prediction of the rheumatoid arthritis patients of the SVM model constructed above was numerically represented.

The results are shown in FIG.

As shown in FIG. 2, the cross-validation of the randomly selected group showed the accuracy of the rheumatoid arthritis prognosis prediction of 0.6015. From the above results, it can be seen that the model constructed in Example 2-1 can predict the patient's prognosis with a high probability of a certain level or more even when a new sample is introduced.

Example  3. Using clinical information Rheumatoid arthritis  Prognosis prediction method

3-1. Using clinical information Rheumatoid arthritis  Prognostic Prediction Model

In order to establish a prognostic model for rheumatoid arthritis prognosis using clinical information, SVM model analysis was performed by obtaining clinical information of the following six factors from Hanyang University rheumatoid arthritis hospital.

Specifically, ESR (Erythrocyte sedimentation rate), BMI (Body mass index), HAQ, which were quantified using a known method, for 250 Korean cohorts within 2 years of illness collected from Hanyang University Rheumatoid Arthritis Hospital used in Example 1 (Health Assessment Questionnaire), Baseline Sharp-van der Heijde Score (SHS), and clinical information on disease duration (years) were obtained from Hanyang University Rheumatoid Arthritis Hospital, and blood extracted from the cohort 250 for quantification of anti-CCP antibodies. Is applied to anti-CCP kit (CCPlus®IMMUNOSCAN, Uro Diagnostica), and the result is obtained using Enzyme immunoassay (Triturus system, Grifols). In this case, a value of 0 was assigned, and the clinical information numerical values of the six factors, which were quantified, were inputted to the SVM function in the same manner as in Example 2-1 to predict the prognosis of rheumatoid arthritis. It was built.

3-2. Using clinical information Rheumatoid arthritis  Validation of Prognostic Prediction Model

In order to verify the accuracy of the rheumatoid arthritis prognosis prediction model function using the clinical information constructed in Example 3-1, the following experiment was performed.

Specifically, cross-validation of the prognostic prediction model of rheumatoid arthritis using the clinical information constructed by the method of Example 3-1 was performed using the same method as in Example 2-2.

The result is shown at the left end of FIG. 4 (Clinical Only).

As shown in the left side of Figure 3, it was confirmed that the accuracy of the prediction of rheumatoid arthritis of 0.7 appears when using a model constructed using six clinical information.

Example  4. SNP Marker  And clinical information Rheumatoid arthritis  Prognosis prediction

4-1. SNP Marker  And clinical information Rheumatoid arthritis  Prognostic Prediction Model

Rheumatoid arthritis prognosis prediction model using the SNP marker constructed in Example 2-1 and rheumatoid arthritis prognosis prediction model using the clinical information constructed in Example 3-1, and the rheumatoid arthritis prognosis prediction model using the SNP marker and clinical information In order to construct the following method was used.

Specifically, in the base sequence at the same position as each of the 85 SNPs selected in Example 1 of the 250 cohort genomes used in Example 1 to obtain the weight constant α used to add the respective SVM algorithm values, If the SNP is isomorphic and is a major allel (AA) 0, if it is heteromorphic (Aa) 1, if it is isomorphic and is a minor allel, convert it to a value of (aa) 2 The P _SNP value was obtained by substituting the predictive model for rheumatoid arthritis prognosis using the SNP marker constructed in Example 2-1, and the numerical value of clinical information of 250 Korean cohorts used in Example 3-1 was constructed in Example 3-1. a clinical information with the obtained after the P values are substituted in rheumatoid arthritis _Clinic prognosis prediction model, to the two weighted adding a couple of different weight to the probability value for the sum (weight sum) method is applied as shown in equation (1) W P showed the _sum value, the P _sum value of 0.5 to 1 If the value of the judgment that the rheumatoid arthritis prognosis of the patients bad, and when the P _sum value out a value of less than 0.5 is rheumatoid arthritis prognosis of the patient A model which judges to be good was constructed, and then, cross-analysis was carried out in the same manner as in Example 2-2, with the weight of the following Equation 1 increasing by 0.01 from 0.3 to 0.6 for the constructed model. The α value, which is the weight of the case, was obtained.

The α value, which is the weight when the accuracy of the most prognostic prediction of rheumatoid arthritis determined in the process of obtaining the α value that is the weight, is the highest, is 0.40.

The results are shown in FIG.

 [Equation 1]

P _sum = αP _SNP + (1-α) P _Clinic

4-2. SNP Marker  And clinical information Rheumatoid arthritis  Validation of Prognostic Prediction Model

In order to verify the accuracy of the rheumatoid arthritis prognostic model function using the SNP markers and clinical information constructed in Example 4-1, the following experiment was performed.

Specifically, cross-validation of the rheumatoid arthritis prognosis prediction model using the SNP markers and the clinical information constructed by the method of Example 4-1 was performed using the same method as in Example 2-2. The results are shown in FIG.

As shown in Figure 4, compared with the case of using the SNP markers to predict the prognosis of rheumatoid arthritis using the rheumatoid arthritis prognosis model, the accuracy of the prediction of rheumatoid arthritis prognosis was improved by 131%, using the rheumatoid arthritis prognosis prediction model using clinical information As a result, it was confirmed that the accuracy of rheumatoid arthritis prognosis was improved by 113% compared to the prediction of rheumatoid arthritis prognosis.

Claims (9)

A composition for predicting prognosis of rheumatoid arthritis, comprising an agent capable of detecting a set of monobasic polymorphism (SNP) gene markers for prognostic prediction of rheumatoid arthritis,
The SNP gene marker set is rs1837 SNP gene marker, rs411337 SNP gene marker, rs9263800 SNP gene marker, rs2071543 SNP gene marker, rs9357155 SNP gene marker, rs2044102 SNP gene marker, rs12722588 SNP gene marker, rs3130100 SNP gene marker, rs17409828 SNP gene Comprising a marker, and rs32490 SNP gene marker, a composition for predicting prognosis of rheumatoid arthritis. The method of claim 1, wherein the SNP gene marker set is used to predict the rheumatoid arthritis prognosis according to the P-value obtained by applying the nucleotide sequence information of the SNP gene marker set and the clinical information of the subject to the rheumatoid arthritis prognosis model function. Will, composition. 3. The P _sum of claim 2 obtained by applying the P-value to the following equation. Is 0 to less than 0.5 A composition wherein the prognosis of rheumatoid arthritis is determined to be good:
[Equation 1]
P _sum = αP _SNP + (1-α) P _Clinic
In Equation 1,
P _SNP is a P-value obtained by applying the nucleotide sequence information of the SNP gene marker set to the rheumatoid arthritis prognosis model function,
P _clinic is the P-value obtained by applying the clinical information of the subject to the predictive model function of rheumatoid arthritis.
P _sum is a P-value obtained by weighted _sum of the P _SNP value and P _clinic value,
α is a constant used when the weighted sum of the P _SNP value and the P _clinic value. Amplifying the genetic sample of the subject to obtain an amplified product of a set of SNP gene markers for prognostic prediction of rheumatoid arthritis,
Analyzing the nucleotide sequence of the monobasic polymorphism site of the SNP gene marker set, and
Predicting rheumatoid arthritis prognosis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP gene marker set and the clinical information of the subject to a rheumatoid arthritis prognosis model function. As a method of providing
The SNP gene marker set for prognostic prediction of rheumatoid arthritis is rs1837 SNP gene marker, rs411337 SNP gene marker, rs9263800 SNP gene marker, rs2071543 SNP gene marker, rs9357155 SNP gene marker, rs2044102 SNP gene marker, rs12722588 SNP gene marker, rs3130100 SNP A method of providing information of rheumatoid arthritis prognosis prediction, comprising a genetic marker, an rs17409828 SNP gene marker, and an rs32490 SNP gene marker. The method of claim 4, wherein the clinical information includes a numerical value of anti-CCP antibody, ESR (Erythrocyte sedimentation rate), Body Mass Index (BMI), Health Assessment Questionnaire (HAQ), Baseline Sharp-van der Heijde Score (SHS), And at least one selected from the group consisting of disease periods. According to claim 4, P _sum obtained by applying the nucleotide sequence information of the SNP gene marker set and the clinical information of the subject to the rheumatoid arthritis prognosis model function applied to Equation 1 below Is 0 to less than 0.5 Methods of providing information of rheumatoid arthritis prognosis prediction, in which a rheumatoid arthritis prognosis is determined to be good:
[Equation 1]
P _sum = αP _SNP + (1-α) P _Clinic
In Equation 1,
P _SNP is a P-value obtained by applying the nucleotide sequence information of the SNP gene marker set to the rheumatoid arthritis prognosis model function,
P _clinic is the P-value obtained by applying the clinical information of the subject to the predictive model function of rheumatoid arthritis.
P _sum is a P-value obtained by weighted _sum of the P _SNP value and P _clinic value,
α is a constant used when the weighted sum of the P _SNP value and the P _clinic value. The method according to claim 6, wherein the P _SNP is the same as each of the 10 SNP markers included in the set of SNP gene markers of the subject, the corresponding SNP is homologous, and if the allele (Major allel) (AA) 0, heteromorphic (Aa) 1, homozygous, minor allel (aa) 2, and converted to the value of (aa) 2 obtained by assigning the corresponding value to the rheumatoid arthritis prognostic model using the constructed SNP marker set Value, the method. The method of claim 4, wherein the rheumatoid arthritis prognosis model function is a support vector machine (SVM), Neural Network, Random Forest, or Discriminant Analysis. Contacting the candidate compound with a sample of a subject,
Amplifying the genetic sample of the subject to obtain an amplified product of a set of SNP gene markers for prognostic prediction of rheumatoid arthritis,
Analyzing the nucleotide sequence of the mononucleotide polymorphism site of the SNP gene marker set,
Predicting the prognosis of rheumatoid arthritis according to the P-value obtained by applying the nucleotide sequence information of the analyzed SNP gene marker set and the clinical information of the subject to the rheumatoid arthritis prognosis model function, and
A method of screening for a therapeutic agent for rheumatoid arthritis, comprising determining the candidate compound as a therapeutic agent for rheumatoid arthritis,
The SNP gene marker set for prognostic prediction of rheumatoid arthritis is rs1837 SNP gene marker, rs411337 SNP gene marker, rs9263800 SNP gene marker, rs2071543 SNP gene marker, rs9357155 SNP gene marker, rs2044102 SNP gene marker, rs12722588 SNP gene marker, rs3130100 A method for screening a rheumatoid arthritis therapeutic agent comprising an SNP gene marker, an rs17409828 SNP gene marker, and an rs32490 SNP gene marker.

Images