KR102482818B1 - Method for personalized prevention of adverse drug reaction of osteoporosis medication based on information of individual deleterious protein sequence variation - Google Patents

Abstract

The present invention relates to a method and system for selecting a therapeutic agent for osteoporosis based on individual protein damage information using personal genome sequencing. As a technology that can predict the side effects or risks of a specific drug, that is, osteoporosis treatment, for each individual through sequence analysis of the exon region, it is highly reliable and is also a general-purpose technology with a wide range of applications.

Description

Method for personalized prevention of adverse drug reaction of osteoporosis medication based on information of individual deleterious protein sequence variation}

The present invention relates to a method for selecting an osteoporosis drug based on individual protein damage information using personal genome sequencing to prevent side effects of the osteoporosis drug.

Due to the development of bioengineering technology, we have now reached the stage of predicting individual diseases by analyzing the whole genome sequence of humans and providing customized disease prevention and treatment.

Recently, as a result of comparing individual genome sequences, it has been found that different bases exist at the same location on the chromosome, and these differences in base sequences have been used to predict differences in individual responses to drugs. For example, since drug metabolism is slow or fast depending on specific genome sequence information possessed by an individual, effects and side effects of drugs may vary from person to person.

Therefore, there is an increasing social demand for personalized drug selection that can select drugs and doses suitable for patients using individual genome nucleotide sequence differences, and genomic information such as Single Nucleotide Polymorphism (SNP) is used as a marker. And pharmacogenomics or pharmacogenomics using the research results on the correlation between the marker and drug responsiveness/drug side effects are emerging.

In pharmacogenetics, the metabolism and response of drugs or chemicals in the general population or individuals is genetically analyzed and predicted. Some individuals may experience reactions other than the expected drug reactions to the drug. These drug side effects may be related to the severity of the disease being treated, drug interactions, patient's age, nutritional status, liver and kidney function, and environmental factors such as climate or food. Although it is caused by factors, genetic differences related to drug metabolism, for example, polymorphism of drug enzyme genes also have an effect, so related studies are being conducted.

On the other hand, an anti-osteoporosis drug, which is a kind of therapeutic agent for osteoporosis, is a drug that affects bone metabolism and maintains bone density by inhibiting osteoclasts. Bisphosphonate, one of the anti-bone resorbing agents, has been proven to inhibit bone density loss and prevent fractures, and is currently used as a first-line treatment for osteoporosis, and its use is widely increasing due to the increasing prevalence of osteoporosis with aging. According to previous studies, among the osteoporosis-related drugs used in Korea in 2008, bisphosphonates were the most used at 68%.

However, as side effects such as osteonecrosis of the jaw (ONJ) have recently been reported one after another due to long-term use of bisphosphonates, the safety of long-term administration of bisphosphonates has been raised. The case of jaw bone necrosis occurs clinically rarely (1/1000 to 1/100,000), but once it occurs, it is difficult to treat, so preventive measures are urgently needed. This is named Bisphosphonate-Related OsteoNecrosis of the Jaw (BRONJ), and various studies on its prevention and treatment have been conducted.

In addition, in 2014, the American Association of Oral and Maxillofacial Surgeon expanded the definition of bisphosphonate-related osteonecrosis to drug-related osteonecrosis of the jaw (MRONJ) through a written opinion, which This is because jaw necrosis in patients using existing bisphosphonates has been reported even when other anti-bone resorbing agents or anti-angiogenic agents such as denosumab are used as well as bisphosphonates. As such, jaw necrosis associated with drug use is continuously increasing, and a comprehensive approach to prevent and treat it is being attempted.

Accordingly, the above-mentioned pharmacogenomics can be used as a method for predicting side effects of a drug before drug administration. Previous studies have revealed that the risk of side effects such as bisphosphonate-related bone necrosis is related to genotype. As a result of whole genome analysis in previous studies, the ratio of rs17024608, a SNP present in the intron region of the RBMS3 gene, was statistically significantly high in the BRONJ patient group (p-value <7x10-8; odds ratio, 5.8; 95% confidence interval, 3.1 -11.1) (Paola N. et al., 2012).

It is currently known that genetic mutations in dozens to hundreds of drug metabolism-related genes increase or decrease drug metabolism by affecting the function of the corresponding protein in the metabolism of drugs in the body, and it is also necessary to identify new genetic mutations. Analysis of a small number of genes has limitations in identifying all of the various drug metabolism-related genes. Therefore, it goes beyond the method based on the results of population observation studies using markers such as minority genes or single nucleotide polymorphisms, and makes theoretical inferences about protein variations and their biological effects by directly utilizing individual genome sequence variation information. The necessity of introducing a methodology that provides useful and reliable information for individual osteoporosis treatment selection is strongly raised.

The present invention has been devised in view of the above, and analyzes individual genome sequence variation information, calculates individual protein damage scores from gene sequence variation information involved in the pharmacodynamics or pharmacokinetics of an osteoporosis treatment, and then calculates the individual protein damage score. By calculating individual drug scores in association with the correlation between protein and protein, we intend to provide a method for predicting the possibility of side effects of osteoporosis drugs and providing information for selecting osteoporosis drugs.

In one aspect, the present invention provides the steps of determining at least one gene sequence mutation information involved in the pharmaco-dynamics or pharmaco-kinetics of a therapeutic agent for osteoporosis from personal genome sequence information; Calculating individual protein damage scores using the gene sequence mutation information; and calculating an individual drug score by associating the individual protein damage score with a mutual relationship between a drug and a protein.

In another aspect, the present invention relates to a therapeutic agent for osteoporosis applicable to individuals, a database capable of searching or extracting information related to a gene or protein related to the therapeutic agent for osteoporosis; a communication unit capable of accessing the database; A first calculation module for calculating information on one or more gene sequence mutations involved in pharmacodynamics or pharmacokinetics of the osteoporosis treatment based on the information; a second calculation module for calculating individual protein damage scores using the gene sequence mutation information; a third calculation module for calculating an individual drug score by associating the individual protein damage score with a relationship between a drug and a protein; and a display unit for displaying the calculated value calculated by the calculation module, providing a system for selecting a treatment for osteoporosis using individual genome sequence mutation.

In another aspect, the present invention provides the steps of obtaining gene sequence mutation information involved in the pharmacodynamics or pharmacokinetics of a therapeutic agent for osteoporosis from personal genome sequence information; Calculating individual protein damage scores using the gene sequence mutation information; And calculating the individual drug score by associating the individual protein damage score with the correlation between the drug and the protein; providing a computer readable medium including an execution module that executes a processor that performs operations including; .

The method and system for selecting an individual osteoporosis treatment based on individual genome sequence variation information of the present invention is a specific drug for each individual, that is, through sequence analysis of the exon region of a gene encoding various proteins involved in the pharmacodynamics or pharmacokinetics of an osteoporosis treatment. It is highly reliable as a technology that can predict the response to osteoporosis treatment.

In addition, it is expected that the present invention based on pharmacogenomics will be widely used in the clinical field as it becomes easier to clinically use technologies (PCR, SNP chip, capillary sequencing, NGS, etc.) for finding out genotypes with the development of experimental methods. can In particular, the price of next-generation sequencing (NGS) is rapidly dropping, and it is possible to access it from the perspective of genome at a price ranging from tens of thousands of won to hundreds of thousands of won per patient.

While the method of using a small number of mutations or a small number of genes obtained through comparison between populations in existing pharmacogenomics as a marker has a large statistical error due to the selection of study subjects and differences between populations, the method of the present invention is Since molecular-level research and analysis results of proteins related to pharmacokinetics and pharmacodynamics related to drugs are directly applied to the selection of osteoporosis drugs, it can be applied without being significantly affected by differences between population groups, and it is more reliable than using a small number of markers. has the advantage of being high.

In particular, when using the method and system according to the present invention, the osteoporosis treatment, for example, by predicting the side effects or risks of bisphosphonate in advance, it can be usefully used to determine whether and how to use the osteoporosis treatment applied to an individual. It can lead to the prevention of side effects through proper dose administration in patients with osteoporosis.

Furthermore, if new knowledge about the drug-protein interaction of a therapeutic agent for osteoporosis, for example, a bisphosphonate, is discovered or provided, it can be easily added and applied to the method of the present invention, so that further improvement is made according to the accumulation of information as a result of future research. There is an advantage that a customized treatment method can be provided.

1 is a flowchart showing each step of a method for providing information for selecting an osteoporosis treatment using personal genome sequence variation according to an embodiment of the present invention.
2 is a schematic configuration diagram of an osteoporosis drug selection system using individual genome sequence variation according to an embodiment of the present invention (DB: database).
3A and 3B are flow charts showing ROC (Receiver Operating Curve) curves for verification of a method for selecting an osteoporosis treatment using genetic nucleotide sequence variation according to an embodiment of the present invention.

The present invention is based on the discovery that highly safe drugs and dosage/usage can be selected in a personalized manner in drug treatment using osteoporosis drugs for each individual through analysis of individual genome sequence mutation information.

In one aspect, the present invention provides the steps of determining at least one gene sequence mutation information involved in the pharmaco-dynamics or pharmaco-kinetics of a therapeutic agent for osteoporosis from personal genome sequence information; Calculating individual protein damage scores using the gene sequence mutation information; and calculating an individual drug score by associating the individual protein damage score with a mutual relationship between a drug and a protein.

In the present invention, the osteoporosis treatment includes all substances exhibiting the same or similar pharmacological activity, such as drugs belonging to the drug class shown in Table 1 below, derivatives thereof, and pharmaceutically acceptable salts thereof, but are not limited thereto.

drug class drug bisphosphonates Alendronate, Clodronate, Ibandronate, Pamidronate, Risendronate, Zoledronate, Etidronate, Medronate, Tiludronate, Olpadronate, Cimadronate Anti-RANKL monoclonal antibody (anti-RANKL monoclonal Ab) Denosumab Cholecalciferol Ergocalciferol, Doxecalciferol, Alfacalcidol Vitamin D analogue Dihydrotachysterol Selective Estrogen Receptor Modulators (SERMs) Raloxifen, Bazedoxifene, Femarelle, lasofoxifene Hormone Replacement Therapy (HRT) Estrogen, Progesterone Calcitriol Calcipediol parathyroid hormone (PTH) Teriparatide Calcitonin Elcalcitonin, Salcatonin Vitamin K compound Menatetrenone, Menaquinone

Gene information involved in the pharmacodynamics or pharmacokinetics of the osteoporosis therapeutic agent in the present invention is DrugBank (http://www.drugbank.ca/) or KEGG Drug (http://www.genome.jp/kegg/drug/) or PharmGKB ( https://www.pharmgkb.org/) and the like, preferably FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1 ; , SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1, OBP2A, etc., but are not limited thereto. More specifically, the genes involved in the pharmacodynamics or pharmacokinetics of the osteoporosis therapeutic agent in the present invention are FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, It includes one selected from the group consisting of CHDH, AC018682.6, SLC12A4, ADARB2, CD3G, ZFYVE20, ASZ1, ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1 and ADCY3, preferably in the gene It may further include one or more selected from the group consisting of PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1 and OBP2A there is.

Genes/proteins in the present invention are indicated according to the nomenclature of the HGNC (HUGO Gene Nomenclature Committee) (Gray KA, Daugherty LC, Gordon SM, Seal RL, Wright MW, Bruford EA. genenames.org: the HGNC resources in 2013. Nucleic Acids Res.

Gene nucleotide sequence variation used as work information in the present invention refers to a variation or polymorphism of an individual's gene sequence. In the present invention, gene sequence mutation or polymorphism occurs in a gene region encoding a protein related to pharmacodynamics or pharmacokinetics of a therapeutic agent for osteoporosis, particularly in an exon region, but is not limited thereto.

The term “base sequence variation information” used in the present invention refers to information on substitution, addition, or deletion of bases constituting exons of a gene. Substitutions, additions, or deletions of these bases may occur for a variety of reasons, for example, due to structural differences including chromosomal mutations, truncations, deletions, duplications, inversions, and/or translocations.

On the other hand, nucleotide sequence polymorphism refers to individual differences in nucleotide sequences present in the genome, and the largest number of nucleotide sequence polymorphisms is Single Nucleotide Polymorphism (SNP), A, T, C, There is an individual difference in one base of the base sequence consisting of G. Sequence polymorphism is a form of polyalleic variation including Single Nucleotide Variation (SNV) including SNP, short tandem repeat polymorphism (STRP), or various number of tandem repeat (VNTR) and Copy number variation (CNV). can appear as

In the method of the present invention, nucleotide sequence variation or polymorphism information found in an individual's genome is collected in relation to proteins related to the pharmacodynamics or pharmacokinetics of an osteoporosis treatment. That is, the nucleotide sequence variation information used in the method of the present invention is one or more genes involved in the pharmacodynamics or pharmacokinetics of a therapeutic agent for osteoporosis among the individual genomic sequence information obtained, for example, a drug-related target protein, drug Mutation information found in the exon region of a gene encoding an enzyme protein involved in metabolism, a transporter protein, or a carrier protein, but is not limited thereto.

Individual genome sequence information used in the present invention can be determined using a known sequencing method, and commercialized services such as Complete Genomics, BGI (Beijing Genome Institute), Knome, Macrogen, DNALink, etc. may be used, but is not limited thereto.

In the present invention, gene sequence variation information present in an individual's genome sequence can be extracted using various methods, and a sequence comparison program with a reference group, for example, the genome sequence of HG19, for example, ANNOVAR ( Wang et al., Nucleic Acids Research, 2010; 38(16): e164), SVA (Sequence Variant Analyzer) (Ge et al., Bioinformatics. 2011; 27(14): 19982000), BreakDancer (Chen et al., Nat Methods. 2009 Sep;

The gene sequence mutation information may be received/obtained through a computer system, and in this respect, the method of the present invention may further include receiving genetic mutation information into a computer system. The computer system used in the present invention relates to genes involved in pharmacodynamics or pharmacokinetics of the osteoporosis treatment, for example, genes encoding drug-related target proteins, drug metabolism enzyme proteins, transporter proteins, or transporter proteins. It may contain or be accessible to one or more databases containing information. Such databases include, for example, DrugBank (http://www.drugbank.ca/), KEGG Drug (http://www.genome.jp/kegg/drug/), PharmGKB (http://www.pharmgkb.org /) may include open or closed databases or knowledge bases that provide information on gene/protein/drug-protein interactions, etc., but are not limited thereto.

In the present invention, the osteoporosis therapeutic agent may be information input by a user, information input from a prescription (previous), or information input from a database including information on an osteoporosis therapeutic agent. The prescription includes, but is not limited to, an electronic prescription.

The term “pharmaco-kinetics (pk) or pharmacokinetic parameters” used in the present invention refers to the characteristics of a drug related to absorption, movement, distribution, conversion, and excretion of a drug in the body for a certain period of time, and the distribution of the drug Volume (Vd), clearance rate (CL), bioavailability (F), absorption rate coefficient (ka), maximum plasma concentration (Cmax), time point of maximum plasma concentration (time point of maximum plasma concentration, Tmax), measurement of the area under the curve (AUC) of the change in blood drug concentration over a certain period of time, and the like.

The term "pharmacodynamics or pharmacodynamic parameters" used in the present invention refers to the physiological and biochemical action of a drug on the body and its action mechanism, that is, the characteristics related to the reaction or effect of the drug on the body.

As used in the present invention, the term "gene sequence variation score" means that when a genome sequence variation is found in the exon region of a gene encoding a protein, these individual variations are considered as amino acid sequence variation (substitution, addition) of the protein encoded by the gene. or deletion) or transcriptional regulatory mutation, which quantifies the degree of significant change or damage to the structure and/or function of the protein. The gene sequence mutation score is the evolutionary theory of amino acids on the genome sequence. It can be calculated by considering the degree of natural conservation and the degree of change in the structure or function of the protein according to the physical properties of the modified amino acid.

In the present invention, in order to be applied to calculating individual protein damage scores or individual drug scores, a method for calculating a gene sequence variation score may be performed using a method known in the art. For example, SIFT (Sorting Intolerant From Tolerant, Pauline C et al., Genome Res. 2001 May; 11(5): 863874; Pauline C et al., Genome Res. 2002 March; 12(3): 436446; Jing Hul et al., Genome Biol. 2012; 13(2): R9), PolyPhen, PolyPhen-2 (Polymorphism Phenotyping, Ramensky V et al., Nucleic Acids Res. 2002 September 1; 30(17): 38943900; Adzhubei IA et al., Nat Methods 7(4):248-249 (2010)), MAPP (Eric A. et al., Multivariate Analysis of Protein Polymorphism, Genome Research 2005;15:978986), Logre (Log R Pfam E- value, Clifford R.J et al., Bioinformatics 2004;20:1006-1014), Mutation Assessor (Reva B et al., Genome Biol. 2007;8:R232, http://mutationassessor.org/), Condel (Gonzalez- Perez A et al., The American Journal of Human Genetics 2011;88:440449, http://bg.upf.edu/fannsdb/), GERP (Cooper et al., Genomic Evolutionary Rate Profiling, Genome Res. 2005;15 :901-913, http://mendel.stanford.edu/SidowLab/downloads/gerp/), CADD (Combined Annotation-Dependent Depletion, http://cadd.gs.washington.edu/), Muta tionTaster, MutationTaster2 (Schwarz et al., MutationTaster2: mutation prediction for the deep-sequencing age. Nature Methods 2014;11:361362, http://www.mutationtaster.org/), PROVEAN (Choi et al., PLoS One. 2012;7(10):e46688), PMut (Ferrer-Costa et al., Proteins 2004;57(4):811-819, http://mmb.pcb.ub.es/PMut/), CEO (Combinatorial Entropy Optimization, Reva et al., Genome Biol 2007;8(11):R232), SNPeffect (Reumers et al., Bioinformatics. 2006;22(17):2183-2185, http://snpeffect.vib.be), fathmm (Shihab et al., Functional Analysis through Hidden Markov Models, Hum Mutat 2013;34 :57-65, http://fathmm.biocompute.org.uk/), and the like, can be used to calculate gene sequence variation scores from gene sequence variation information, but is not limited thereto.

The purpose of the above-described algorithms is to determine how much each gene sequence variation affects protein function, and how much this effect causes damage to the protein, or whether there is no significant effect. They have something in common in that they basically determine the effect on the structure and / or function of the protein in consideration of the change that individual gene sequence mutations will cause in the amino acid sequence of the protein encoded by the gene.

In one embodiment according to the present invention, the SIFT (Sorting Intolerant From Tolerant) algorithm was used to calculate individual gene sequence variation scores. In the case of the SIFT algorithm, gene sequence variation information is input in a VCF (Variant Call Format) format file, and the degree to which each gene sequence variation damages a corresponding gene is scored. In the case of the SIFT algorithm, the closer the calculated score is to 0, the more severe the damage to the protein encoded by the gene is, so it is judged that the corresponding function is damaged. .

In the case of another algorithm, PolyPhen-2, the higher the calculated score, the greater the degree of functional damage to the protein encoded by the corresponding gene.

Recently, a study comparing and synthesizing SIFT, Polyphen2, MAPP, Logre, and Mutation Assessor and presenting the Condel algorithm (Gonzalez-Perez, A. & Lopez-Bigas, N. Improving the assessment of the outcome of nonsynonymous SNVs with a consensus deleteriousness score, Condel. The American Journal of Human Genetics, 2011;88(4):440-449.) was published, and in this study, gene sequence mutations that damage proteins and gene sequence mutations with little effect Comparing the above five algorithms using HumVar and HumDiv (Adzhubei, IA et al., A method and server for predicting damaging missense mutations. Nature methods, 2010;7(4):248-249), a set of known data did As a result, 97.9% of HumVar's gene sequence mutations causing protein damage and 97.3% of gene sequence mutations with little effect were detected in at least three algorithms out of the five algorithms, and 99.7% of HumDiv's protein damage was detected. Gene sequence mutations that cause sequencing and gene sequence mutations with a small effect of 98.8% were equally detected in at least three of the five algorithms. In addition, as a result of drawing ROC (Receiver Operating Curve) curves showing the accuracy of the results calculated by integrating the five algorithms and each algorithm for HumDiv and HumVar, AUC ( Area Under the Receiver Operating Curve) was confirmed. That is, although the above-described various algorithms have different calculation methods, the calculated gene sequence mutation scores are significantly correlated with each other. Therefore, applying the gene sequence variation score calculated by applying the above-described algorithms or a method applying the algorithms to the individual protein damage score and individual drug score calculation step according to the present invention is mutually exclusive in calculating each gene sequence variation score. Regardless of the type of other algorithm, it is within the scope of the present invention.

When a gene sequence mutation occurs in an exon region of a gene encoding a protein, it can directly affect the structure and/or function of the protein. Therefore, the gene sequence mutation information can be related to the degree of protein function damage. In this aspect, in the next step, the method of the present invention calculates individual protein damage scores based on the gene sequence variation scores described above.

As used herein, the term "protein damage score" means that two or more significant nucleotide sequence mutations are found in a gene region encoding one protein, so that one protein has two or more gene sequence mutation scores, It refers to a score calculated by integrating the gene sequence variation scores, and if there is only one significant sequence variation in a gene region encoding a protein, the gene sequence variation score and the protein damage score are the same. At this time, when there are two or more gene sequence mutations encoding a protein, the protein damage score is calculated as the average value of the gene sequence mutation scores calculated for each mutation, and this average value is, for example, geometric mean, arithmetic mean, harmonic mean , arithmetic geometric mean, arithmetic harmonic mean, geometric harmonic mean, Pythagorean mean, quartile mean, quadratic mean, cut mean, Winsorized mean, weighted mean, weighted geometric mean, weighted arithmetic mean, weighted harmonic mean, function mean, power mean , generalized f-means, percentiles, maximums, minimums, modes, medians, median ranges, measures of central tendency, simple products or weighted products, or functional operations of the above calculations. , but not limited thereto.

In one embodiment according to the present invention, the protein damage score was calculated by Equation 1 below, and Equation 1 below can be modified in various ways and is not limited thereto.

In Equation 1, S _g is the protein damage score of the protein encoded by gene g , n is the number of nucleotide sequence variations to be analyzed among the nucleotide sequence variations of gene g , v _i is the gene sequence of the ith gene sequence variation is the variance score, and p is a non-zero real number. In Equation 1, when the value of p is 1, it becomes an arithmetic average, when the value of p is -1, it becomes a harmonic average, and when the value of p approaches 0, it becomes a geometric average.

In another embodiment according to the present invention, the protein damage score was calculated by Equation 2 below, and Equation 2 below can be modified in various ways and is not limited thereto.

In Equation 2, S _g is the protein damage score of the protein encoded by gene g , n is the number of nucleotide sequence variations to be analyzed among the nucleotide sequence variations of gene g , v _i is the gene sequence of the i th gene sequence variation A variance score, w _i , is a weight given to v _i . When all weights w _i have the same value, the protein damage score S _g becomes the geometric mean value of the gene sequence variation _scores vi . The weight may be given in consideration of the type of the corresponding protein, the pharmacokinetic or pharmacodynamic classification of the corresponding protein, the pharmacokinetic parameter of the corresponding drug enzyme protein, and the population or racial distribution.

The term “pharmacokinetic parameters of drug enzyme proteins” used in the present invention includes Vmax, Km, Kcat/Km and the like. Vmax is the maximum enzyme reaction rate at very high substrate concentrations, and Km is the substrate concentration that allows the reaction to reach 1/2 Vmax. Km can be viewed as an affinity between the enzyme and the substrate, and the smaller the Km, the stronger the bond between the enzyme and the substrate. Kcat, also called the metabolic turnover rate of an enzyme, means the number of substrate molecules metabolized in 1 second per each enzyme active site when the enzyme is active at its maximum rate, and it means how fast the corresponding enzyme reaction actually occurs.

In the next step, in the method of the present invention, individual drug scores are calculated by associating the above-described protein damage scores with the mutual relationship between drugs and proteins.

As used herein, the term “drug score” refers to a target protein involved in the pharmacodynamics or pharmacokinetics of a given drug, for example, a therapeutic agent for osteoporosis, an enzyme protein involved in drug metabolism, a transporter protein, or a transporter. After finding the proteins, calculating the protein damage score of the proteins, and putting them together again, it refers to the value calculated for one drug.

In the present invention, the drug score is calculated as the average value of the protein damage scores when there are two or more protein damages involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment, and this average value is, for example, geometric mean, arithmetic mean, harmonic mean, arithmetic mean Geometric mean, arithmetic harmonic mean, geometric harmonic mean, Pythagorean mean, quartile mean, quadratic mean, cut mean, Winsorized mean, weighted mean, weighted geometric mean, weighted arithmetic mean, weighted harmonic mean, function mean, power mean, generalization It can be calculated by the calculated f-mean, percentile, maximum value, minimum value, mode value, median value, median range, measures of central tendency, simple product or weighted product, or function operation of the above calculated values, but accordingly Not limited.

The drug score may be calculated by adjusting the weights of target proteins involved in pharmacodynamics or pharmacokinetics of the drug, that is, osteoporosis treatment, enzyme protein involved in drug metabolism, transporter protein, and transporter protein, in consideration of pharmacological properties, The weight may be assigned considering pharmacokinetic parameters of the drug enzyme protein, distribution by population or race, and the like. In addition, the protein damage scores of proteins that do not directly interact with the drug, but interact with the precursor or metabolites of the drug, for example, proteins participating in the pharmacological pathway, are also considered. A drug score can be calculated. In addition, a comprehensive drug score may be calculated by considering protein damage scores of proteins involved in pharmacodynamics or pharmacokinetics of the drug and proteins significantly interacting with each other. Information on proteins participating in the pharmacological pathway of the corresponding drug, interacting significantly with the corresponding proteins, or participating in the signaling pathway is found in PharmGKB (Whirl-Carrillo et al., Clinical Pharmacology & Therapeutics 2012;92( 4):414-4171), The MIPS Mammalian Protein-Protein Interaction Database (Pagel etl al., Bioinformatics 2005;21(6):832-834), BIND (Bader et al., Biomolecular Interaction Network Database, Nucleic Acids Res 2003 Jan 1;31(1):248-50), Reactome (Joshi-Tope et al., Nucleic Acids Res. 2005 Jan 1;33(Database issue):D428-32). can do.

In one embodiment according to the present invention, the drug score was calculated by Equation 3 below, and since Equation 3 below can be modified in various ways, it is not limited thereto.

In Equation 3, S _d is the drug score of drug d , n is a protein directly involved in the pharmacodynamics or pharmacokinetics of the drug d or interacting with the precursor or metabolites of the drug, for example, pharmacological The number of proteins encoded by one or more genes selected from the gene group participating in the pathway, g _i is a protein directly involved in the pharmacodynamics or pharmacokinetics of the drug d or interacting with the precursor or metabolites of the drug , For example, a protein damage score of a protein encoded by one or more genes selected from a gene group participating in a pharmacological pathway, and p is a non-zero real number. In Equation 3, when the value of p is 1, it becomes an arithmetic average, when the value of p is -1, it becomes a harmonic average, and when the value of p approaches 0, it becomes a geometric average.

In another embodiment according to the present invention, the drug score was calculated by Equation 4 below, and since Equation 4 below can be modified in various ways, it is not limited thereto.

In Equation 4, S _d is the drug score of drug d , n is a protein directly involved in the pharmacodynamics or pharmacokinetics of the drug d or interacting with the precursor or metabolites of the drug, for example, pharmacological The number of proteins encoded by one or more genes selected from the gene group participating in the pathway, g _i is a protein directly involved in the pharmacodynamics or pharmacokinetics of the drug d or interacting with the precursor or metabolites of the drug , For example, a protein damage score of a protein encoded by one or more genes selected from a gene group participating in a pharmacological pathway, and w _i is a weight given to the g _i . When all weights w _i have the same value, the drug score S _d becomes the geometric mean value of the protein damage score g _i . The weight may be given in consideration of the type of the corresponding protein, the pharmacokinetic or pharmacodynamic classification of the corresponding protein, the pharmacokinetic parameter of the corresponding drug enzyme protein, and the population or racial distribution.

In the case of the geometric mean calculation method used in the method of one embodiment according to the present invention, all weights were given the same regardless of the characteristics of the relationship between the drug and the protein, but as in another embodiment according to the present invention, the relationship between the drug and the protein It is possible to calculate the drug score by assigning a weight considering each characteristic. For example, different scores may be assigned to a drug's target protein and a drug-related transporter protein. In addition, for example, it is possible to calculate a drug score by assigning weights to the pharmacokinetic parameters Km, Vmax, and Kcat/Km of the corresponding drug enzyme protein. In addition, for example, since the target protein is judged to be more important in pharmacological action than the transporter protein, a high weight can be assigned, and the transporter protein or carrier protein can be given a high weight for a drug that is sensitive to concentration, , but not limited thereto. Weights can be carefully adjusted according to the relationship between the drug and the drug-related protein, and the nature of the drug-protein interaction. Sophisticated algorithms can be used that assign weights to the properties of drug-protein interactions, such as assigning 2 points to a target protein and 1 point to a transporter protein.

In the above description, only proteins directly interacting with drugs were taken as examples, but as in one embodiment according to the present invention, proteins interacting with precursors of the drug or metabolites of the drug, involved in the pharmacodynamics or pharmacokinetics of the drug. It is possible to improve the predictive ability of the formula by utilizing information on proteins that significantly interact with proteins that interact with each other and related proteins participating in the signal transduction pathway. That is, information on various proteins involved in protein-protein interaction networks or pharmacological pathway information can be used. In other words, no significant mutation was found in the protein that directly interacts with the drug, so there is no calculated value for the protein damage score or no damage (for example, 1.0 points when the SIFT algorithm is applied), but it does not interact with or interact with the protein. The average value (eg, geometric average) of protein damage scores of related proteins participating in the same signal transduction pathway may be used instead as the protein damage score of the corresponding protein to calculate the drug score.

The individual drug score may be calculated for all drugs for which information on one or more associated proteins can be obtained or for selected drugs among them. In addition, these individual drug scores can be converted into a rank score.

The method of the present invention may further include determining whether to use a drug applied to the individual, that is, a therapeutic agent for osteoporosis, using the above-described individual drug score.

The individual drug score of the present invention can be individually applied to all drugs, but it is more useful if it is applied to each category of disease, clinical characteristics, or mode of action, or among drugs subject to medical comparison. The drug classification system usable in the present invention, for example, ATC (Anatomical Therapeutic Chemical Classification System) code, list of 15 frequently prescribed drugs in the United States (top 15 frequently prescribed drug classes during 2005 ~ 2008 in the United States (Health, United States, 2011, Centers for Disease Control and Prevention)), a list of drugs with known pharmacogenomic markers that may affect drug action information listed on drug labels, or drugs withdrawn from the market due to side effects, etc. can include

The method of the present invention may further include calculating a prescription score.

The term "prescription score" as used in the present invention refers to the drug score determined for each drug when two or more drugs are administered simultaneously or with a short time interval sufficient to significantly affect each other's pharmacological actions. It is a score that is calculated as a sum of points. In the present invention, the prescription score may be calculated by summarizing the drug scores determined for each drug when two or more drugs determined by the order of priority among the drugs are required and simultaneous administration is required. The calculation of the prescription score may be calculated by simply averaging, summing, or multiplying drug scores of the plurality of drugs, for example, when there is no protein that interacts with the plurality of drugs in common. If there is a protein that interacts with multiple drugs in common, each drug score is calculated by assigning, for example, a 2-fold weight to the protein damage score of the common interacting protein, and then the corresponding drug scores are calculated. By summing, the prescription score can be calculated.

The prescription score is intended to go beyond the effects of individual drugs and determine the appropriateness or risk of multiple drug prescriptions included in the prescription (previous) applied to the individual. In this respect, the method of the present invention may further include a step of determining the appropriateness or risk of the prescription (previous) applied to the individual.

The method of the present invention includes, but is not limited to, performed for the purpose of preventing side effects of an osteoporosis treatment agent.

1 is a flowchart showing each step of a method for providing information for selecting an osteoporosis treatment using personal genome sequence variation according to an embodiment of the present invention. In one embodiment according to the present invention, (1) input or receive genome sequence information of an individual user (S100), (2) input or receive osteoporosis treatment-related information (S110), (3) gene sequence mutation information of an individual user Determination (S120), (4) Calculation of individual protein damage score for osteoporosis treatment (S130), (5) Calculation of individual drug score for osteoporosis treatment (S140), (6) Display of drug score, sorting or prioritizing drug score A method for providing information for selecting an osteoporosis treatment proceeds in order of determining the ranking (S150) and (7) selecting an osteoporosis treatment considering drug scores and priorities and calculating a prescription score (S160).

The method according to the present invention provides the pharmacogenomic calculation process and basis for calculating the drug score as information such as pictures, diagrams, and explanations to help the prescribing doctor's judgment when selecting the drug scores sorted by rank. can be further included. That is, the method according to the present invention provides one or more of gene sequence variation information, gene sequence variation score, protein damage score, drug score, and information used for the calculation, which are the basis for drug ranking of the present invention. An additional step may be included.

In another aspect, the present invention relates to a therapeutic agent for osteoporosis that can be applied to an individual, a database capable of searching or extracting information related to a gene or protein related to the therapeutic agent for osteoporosis; a communication unit capable of accessing the database; A first calculation module for calculating information on one or more gene sequence mutations involved in pharmacodynamics or pharmacokinetics of the osteoporosis treatment based on the information; a second calculation module for calculating individual protein damage scores using the gene sequence mutation information; a third calculation module for calculating an individual drug score by associating the individual protein damage score with a relationship between a drug and a protein; and a display unit for displaying the calculated value calculated by the calculation module.

In the present invention, a module may mean a functional and structural combination of hardware for implementing the technical idea according to the present invention and software for driving the hardware. For example, the module may mean a logical unit of a predetermined code and a hardware resource for executing the predetermined code, and does not necessarily mean a physically connected code or one type of hardware. is apparent to those skilled in the art.

The term “calculation module” used in the present invention refers to the gene sequence mutation score, protein damage score, drug score and information that is the basis for calculating the gene sequence mutation score, protein damage score, and drug score for the osteoporosis therapeutic agent and gene to be analyzed according to the method of the present invention. It may mean a predetermined code that calculates each score based on the information and a logical unit of hardware resources for executing the predetermined code, and necessarily means a code that is physically connected, or a type of hardware. does not mean

The system according to the present invention may further include a fourth calculation module for determining whether to use the osteoporosis treatment applied to the individual by using the individual drug score calculated in the third calculation module.

The system according to the present invention also adds a fifth calculation module for calculating the prescription score by integrating the drug scores determined for each drug when two or more drugs determined by the priority between drugs are required and simultaneous administration is required can include

The system according to the present invention is a user interface in which the user inputs a list of treatments for osteoporosis, or accesses a database including information on an osteoporosis treatment, extracts related information, and calculates and provides a drug score of the drug accordingly. may additionally include.

The system according to the present invention may further include a display unit for additionally displaying a calculation process in which the value calculated by each calculation module or the priority between drugs is determined and information based on the calculation or calculation.

In the system according to the present invention, the server including the database or its access information, the calculated information, and the user interface device connected thereto may be used in conjunction with each other.

The system according to the present invention is immediately updated when new pharmacological/biochemical information on the drug-protein interaction is generated, so that it can be used to select a more improved osteoporosis treatment. In one embodiment according to the present invention, the gene sequence mutation information, gene sequence mutation score, protein damage score, drug score, and information based on the calculation are stored in each of the calculation modules according to the update of the database or knowledge base. is updated

2 is a schematic configuration diagram of an osteoporosis drug selection system using individual genome sequence mutations according to an embodiment of the present invention. The system 10 of the present invention includes a database (DB) 100 capable of searching or extracting information related to a gene or protein related to a therapeutic agent for osteoporosis, a communication unit 200, a user interface or terminal 300, a calculation unit 400, and It may be configured to include the display unit 500 .

In the system according to the present invention, the user interface or terminal 300 may request osteoporosis treatment selection process using personal genome sequence mutation from the server, receive and/or store the result, and may be a smart phone, a PC (Personal Computer), or a tablet PC. , a personal digital assistant (PDA), a terminal equipped with a memory means such as a web pad and equipped with a microprocessor and equipped with a mobile communication function equipped with an arithmetic capability.

In the system according to the present invention, the server is a means for providing access to the database 100 for osteoporosis drugs, gene mutations, or drug-protein interactions, and is connected to the user interface or terminal 300 through the communication unit 200. It is configured to exchange various types of information. Here, the communication unit 200 is not only communication in the same hardware, but also a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), the Internet, 2G, It may include 3G, 4G mobile communication networks, Wi-Fi, Wibro, etc., and the communication method does not cover wired or wireless, and any communication method may be used. The database 100 may also be directly installed on the server and may be connected to various life science databases accessible through the Internet or the like depending on the purpose.

In the system according to the present invention, the calculation unit 400 is a first calculation module 410 for calculating one or more gene mutation information involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment using the collected/input information as described above, It may include a second calculation module 420 that calculates a protein damage score and a third calculation module 430 that calculates an individual drug score.

The method according to the present invention may be implemented in hardware, firmware, or software or a combination thereof. When implemented as software, the storage medium includes any medium that stores or transmits data in a form readable by a device such as a computer. For example, a computer readable medium may include read only memory (ROM); random access memory (RAM); magnetic disk storage media; optical storage media; It includes flash memory devices and other electrical, optical or acoustic signal transmission media.

In this aspect, the present invention comprises the steps of obtaining gene sequence mutation information involved in the pharmacodynamics or pharmacokinetics of a therapeutic agent for osteoporosis from personal genome sequence information; Calculating individual protein damage scores using the gene sequence mutation information; and an execution module that executes a processor that performs an operation including calculating the individual drug score by associating the individual protein damage score with a mutual relationship between a drug and a protein.

The processor may further include determining whether to use the osteoporosis treatment applied to the individual by using the individual drug score.

In another aspect, the present invention provides a biomarker composition for predicting side effects of a therapeutic agent for osteoporosis.

Genes that may be included in the biomarker composition according to the present invention include FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682. 6, SLC12A4, ADARB2, CD3G, ZFYVE20, ASZ1, ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, ADCY3, PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1, OBP2A, and the like, but are not limited thereto. By analyzing mutations of the gene or its protein, it is possible to predict the occurrence of side effects of the osteoporosis treatment, and it can be used as a marker using an agent capable of detecting it.

Hereinafter, the present invention will be described in more detail through the following examples. These examples are intended to explain the present invention in detail, and the scope of the present invention is not limited by these examples.

Example 1. Analysis and application of individual genomic sequence mutation information of osteoporosis patients who showed warning signs of serious side effects in treatment with bisphosphonates, an osteoporosis treatment drug

Osteoporosis refers to a condition in which overall bone can be easily destroyed due to a decrease in bone density and a change in microstructure in bone tissue. It is common for patients to be diagnosed with osteoporosis along with a fracture, and in the case of osteoporosis patients, vertebral compression fractures or hip fractures are common. Osteoporosis in postmenopausal women, which is most common, occurs when the function of osteoclasts is enhanced and the function of osteoblasts is reduced due to a decrease in estrogen. In addition, osteoblast function is reduced due to aging and osteoporosis may occur due to lack of calcium in the body, and other secondary causes such as renal hypercalciuria, Cushing's syndrome, androgen insensitivity syndrome, inflammatory bowel disease, and bone metastasis of malignant tumors may occur. may also appear

On the other hand, bisphosphonate, a kind of osteoporosis treatment, is a drug that affects bone metabolism and maintains bone density by inhibiting osteoclasts. Bisphosphonates have been proven to inhibit bone density loss and prevent fractures, and are currently used as the primary treatment for osteoporosis, and their use is widely increasing due to the increasing prevalence of osteoporosis with aging. According to previous studies, among the osteoporosis-related drugs used in Korea in 2008, bisphosphonates were the most used at 68%.

However, with the recent long-term use of bisphosphonates, side effects such as osteonecrosis of the jaw (ONJ) have been reported one after another, raising questions about the safety of long-term use of bisphosphonates. The case of jaw bone necrosis occurs clinically rarely (1/1000 to 1/100,000), but once it occurs, it is difficult to treat, so preventive measures are urgently needed.

Accordingly, the above-mentioned pharmacogenomics can be used as a method for predicting side effects of a drug before drug administration. Previous studies have shown that the risk of side effects such as Bisphosphonate-Relate OsteoNecrosis of the Jaw (BRONJ) is related to genotype. As a result of whole genome analysis in previous studies, the ratio of rs17024608, a SNP present in the intron region of the RBMS3 gene, was statistically significantly high in the BRONJ patient group (p-value <7x10-8; odds ratio, 5.8; 95% confidence interval, 3.1 -11.1) (Paola N. et al., 2012).

In order to confirm whether a group at risk of responding to bisphosphonate treatment, which is a type of osteoporosis treatment, can be distinguished using the method for providing selection and provision of an osteoporosis treatment using individual genome sequence mutations of the present invention, the following experiment was performed.

Among osteoporosis patients, a genetic sequence comparison analysis was performed by dividing a group (n = 38) with side effects (MRONJ, Medication-Related OsteoNecrosis of Jaw) and a group without side effects (n = 30) during bisphosphonate treatment.

In order to obtain information on individual genome sequence mutations of the 68 cases, 80-fold whole exome sequencing was performed using Life Technology's Ion AmpliSeq Exome Kit. At this time, in addition to the above methods, Whole Genome Sequencing, which obtains information on the entire genome of an individual in addition to the exome, or Targeted Exome Sequencing for 500-1000 osteoporosis-related major genes, is an alternative can be done with

The analyzed nucleotide sequence fragments are SAM (Sequence Alignment Map) and BAM (Binary Alignment Map) files aligned to the human reference sequence (eg, HG19) through the process of data cleaning and quality check. output in the form The clean alignment result is detected by software tools such as SAMTools:pileup, SAMTools:mpileup, GATK:recalibration, and GATK:realignment to detect mutations such as SNVs (Single Nucleotide Variants) and InDels. It was output as a file in VCF (Variant Calling Format) format.

After receiving the VCF file containing the gene sequence mutation information, the gene sequence mutation score vi value described above was calculated for each mutation using the SIFT algorithm, and then the individual protein damage score Sg was calculated using Equation 2. Subsequently, the individual protein damage scores were compared for each gene in 38 cases with side effects and 30 cases in the healthy control group. Fifty genes showing a statistically significant difference between the side effect group and the control group were selected as follows, and 32 genes with higher statistical significance based on p-value were selected as the first group. The following genes correspond to genes highly related to the pharmacodynamics or pharmacokinetics of bisphosphonates and their metabolites.

(first group)

FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682.6, SLC12A4, ADARB2, CD3G, ZFYVE20, ASZ1, ARHGEF FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, ADCY3

(2nd group)

PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1, OBP2A

For the gene group selected as described above, individual drug scores were calculated using the method of the present invention. More specifically, after calculating the gene sequence variation score using the SIFT algorithm from the individual gene sequence variation information, individual protein damage scores for the 50 genes were calculated using Equation 2, and Equation 4 After calculating individual drug scores for bisphosphonates using , statistical analysis was performed for each group. The results are shown in Table 2.

Osteoporosis drug side effects group (n=38) normal control
(n=30) p-value Mean drug score (group 1 drug) 0 0.344 <0.001 Mean drug score (group 1 and 2 drugs) 0.031 0.472 <0.001 ENSEMBL HUGO Protein damage score (group 1) ENSG00000160867 FGFR4 0.32 0.52 <0.001 ENSG00000132740 IGHMBP2 0.48 0.78 <0.001 ENSG00000114790 ARHGEF26 0.46 0.65 <0.001 ENSG00000134248 HBXIP 0.37 0.80 <0.001 ENSG00000158477 CD1A 0.65 0.92 <0.001 ENSG00000171970 ZNF57 0.29 0.51 <0.001 ENSG00000112769 LAMA4 0.27 0.57 <0.001 ENSG00000213445 SIPA1 0.24 0.67 <0.001 ENSG00000076716 GPC4 0.36 0.82 <0.001 ENSG00000110801 PSMD9 0.64 0.93 <0.001 ENSG00000106948 AKNA 0.50 0.71 0.001 ENSG00000151458 ANKRD50 0.69 0.89 0.001 ENSG00000079393 DUSP13 0.50 0.65 0.001 ENSG00000145675 PIK3R1 0.79 0.95 0.001 ENSG00000084092 C4orf14 0.66 0.89 0.001 ENSG00000147138 GPR174 0.45 0.75 0.001 ENSG00000016391 CHDH 0.23 0.44 0.001 ENSG00000187600 AC018682.6 0.39 0.72 0.001 ENSG00000124067 SLC12A4 0.45 0.82 0.001 ENSG00000185736 ADARB2 0.29 0.55 0.002 ENSG00000160654 CD3G 0.44 0.73 0.002 ENSG00000131381 ZFYVE20 0.84 0.98 0.002 ENSG00000154438 ASZ1 0.36 0.68 0.002 ENSG00000136002 ARHGEF4 0.87 1.00 0.002 ENSG00000196159 FAT4 0.46 0.60 0.002 ENSG00000162727 OR2M5 0.63 0.93 0.002 ENSG00000107521 HPS1 0.68 0.92 0.003 ENSG00000159917 ZNF235 0.48 0.79 0.003 ENSG00000185624 P4HB 0.84 0.97 0.003 ENSG00000101974 ATP11C 0.80 0.98 0.003 ENSG00000131018 SYNE1 0.44 0.56 0.004 ENSG00000138031 ADCY3 0.80 0.90 0.004 Protein damage score (group 2) ENSG00000159527 PGLYRP3 0.76 0.97 0.005 ENSG00000213401 MAGEA12 0.32 0.59 0.005 ENSG00000177854 TMEM187 0.53 0.69 0.005 ENSG00000176555 OR4S1 0.30 0.62 0.005 ENSG00000241598 KRTAP5-4 0.42 0.72 0.005 ENSG00000114204 SERPINI2 0.43 0.68 0.005 ENSG00000204290 BTNL2 0.76 0.89 0.005 ENSG00000221870 CXorf1 0.45 0.77 0.006 ENSG00000116690 PRG4 0.72 0.94 0.006 ENSG00000118640 VAMP8 0.67 0.85 0.006 ENSG00000198930 CSAG1 0.47 0.72 0.006 ENSG00000157181 C1orf27 0.79 0.97 0.006 ENSG00000221953 C1orf229 0.74 0.97 0.006 ENSG00000198739 LRRTM3 0.68 0.93 0.007 ENSG00000124664 SPDEF 0.61 0.88 0.007 ENSG00000145649 GZMA 0.43 0.58 0.008 ENSG00000049323 LTBP1 0.64 0.86 0.009 ENSG00000122136 OBP2A 0.55 0.68 0.009

As shown in Table 2, as a result of calculating the protein damage score and individual drug score from the gene sequence mutation information of the selected 32 group 1 genes, there was a statistically significant difference between the normal control group and the bisphosphonate side effect group. confirmed.

In addition, as a result of calculating individual drug scores from a total of 50 gene sequence mutation information including gene sequence mutation information of group 2 genes, there was a statistically significant difference in individual drug scores between the two groups. (p-value < 0.05).

Through the above results, it is possible to significantly distinguish the group in which drug side effects occur during bisphosphonate treatment and the group in which it does not, by using the individual drug score calculation through the analysis of individual genome sequence variation information according to the present invention, thereby preventing unwanted side effects. It was confirmed that it can be prevented in advance.

Therefore, by using the method of the present invention, it is possible to predict a group with a high possibility of side effects when bisphosphonate is administered among patients with osteoporosis in the future, and for a high-risk group, the concentration of the drug can be adjusted or alternative treatments or interventions can be induced to be used. will be.

Example 2. Verification of validity of personalized drug selection method based on individual genome sequence variation information

Until now, reliable research results on individual differences in individual genome sequence variation information and pharmacological responses are very limited. Studies to date have followed the paradigm of case-control observational studies in which individual differences in responsiveness are studied by comparing groups positive or negative for specific mutations by drug. In this research paradigm, costly case-control studies must be performed for all combinations consisting of numerous nucleotide sequence variations and numerous drug pairs, but it is practically impossible. On the other hand, the personalized drug selection method according to the present invention not only targets all gene sequence variations, but also does not require expensive case-control design observational studies, and individual protein damage by pure calculation of genome sequence variations. Since it proposes a method of calculating and applying scores and individual drug scores, it has a great advantage that it is possible to make inferences for personalized drug selection for combinations between all genome sequence variations and all drugs.

In order to evaluate the validity of the calculation result of personalized drug selection according to the method of the present invention, 497 frequently prescribed drugs were selected based on the following criteria; (1) Among the drugs included in the ATC code of the top 15 frequently prescribed drug classes during 2005~2008 in the United State (Health, United States, 2011, Centers for Disease Control and Prevention) Drugs for which at least one gene related to pharmacodynamics or pharmacokinetics is known, (2) Drugs for which the action of an established pharmacogenomic genome sequence mutation marker has been applied to the drug labeling of the US Food and Drug Administration, (3) Withdrawal from the market due to drug side effects, etc. Drugs known to the DrugBank database to be

As the validity evaluation criterion data, 650 (65.9%) of the 987 gene sequence mutation-drug interaction pairs that have at least one connection with the 497 drugs were extracted from the established knowledge of 987 gene sequence mutation-drug interaction pairs provided by PharmGKB. Considering that the present invention targets nucleotide sequence variation in the exon region, overlapping parts between the data to be verified and the evaluation standard data were removed for fair evaluation. More specifically, a fairer evaluation was performed by removing all 36 nucleotide sequence mutations located in the exon region among the 650 pairs and selecting only the nucleotide sequence mutations in the non-coding region. In conclusion, 614 pairs were selected as the final gold standard for evaluation.

Next, by analyzing the genome sequences of 1092 people provided by The 1000 Genomes Project and applying the method according to the present invention to each of the 1092 people, pharmacogenomic risk of each individual and pharmacogenomics by gene sequence mutation registered in PharmGKB Each risk was calculated.

Sensitivity, specificity, and area under the receiver operating curve were used for validity evaluation. After ranking 497 drugs based on individual drug scores and setting thresholds by rank at 496 split positions between each rank, (1) if the drug score rank of the drug is higher than the threshold and the PharmGKB variant is found in the individual genome true positive when present in the variant, (2) true negative when the drug's drug score ranks below the threshold and the PharmGKB variant is not in the individual genomic variant, (3) the drug's drug score ranks above the threshold but PharmGKB False positive when the mutation was not in the individual genome variant, (4) false negative when the drug score rank of the drug was lower than the threshold but the PharmGKB variant was in the individual genome variant, true positive for each rank threshold L in each individual, The number of true negatives, false positives, and false negatives was calculated, and sensitivity and specificity were calculated as in the following formula.

D is a set of 497 total drugs, GS is a set of personalized PharmGKB drugs that are used as gold standards for each individual because individual genetic sequence variations match the risk alleles of PharmGKB, D _L is a set of drugs with high ranking thresholds , and vertical bar brackets indicate the number of elements in the set.

As a result of the calculation, 18 patients did not have any mutation that matched that of PharmGKB, so it was not possible to define a set of personalized PharmGKB drugs used as the gold standard, so they were excluded from this feasibility analysis. A ROC curve was drawn by calculating , and specificity, and AUC was calculated. More specifically, first, gene sequence variation scores were calculated using the SIFT algorithm for the entire population of 1092 people, and then Equations 2 and 4 were applied to calculate protein damage scores and drug scores, respectively. In addition, in order to determine the usefulness of weight application according to racial distribution, racial-specific sensitivity, specificity, and AUC value calculation based on this were calculated for the four races specified in The 1000 Genomes Project (African (AFR, n=246), American (AMR, n = 181), Asian (ASN, n = 286), and European (EUR, n = 379) were performed in the same manner, and then race-specific sensitivity, specificity, and AUC were obtained. The results are shown in Table 3, Table 4 and Figure 3.

Distribution by protein class and calculation of average protein damage score protein family number of proteins number of drugs involved Number of protein-drug pairs Average protein damage score target protein 440 486 2357 0.798 carrier protein 10 50 65 0.728 metabolic enzyme protein 74 330 1347 0.733 transporter protein 54 176 457 0.733 total 545 497 4201 0.783

Calculation of drug scoring validity (AUC) by protein group and race using The 1000 Genomes Project data Total AFR AMR ASN EUR Validity of drug scoring ( AUC ) target protein 0.617 0.634 0.608 0.614 0.614 carrier protein 0.554 0.511 0.599 0.485 0.594 metabolic enzyme protein 0.587 0.642 0.580 0.558 0.579 transporter protein 0.497 0.492 0.488 0.489 0.512 Validity of drug score calculation with or without weighting for each protein group ( AUC ) simple geometric mean 0.666 0.744 0.650 0.634 0.653 weighted geometric mean 0.667 0.742 0.652 0.633 0.654

Table 3 shows the distribution of each protein group for the 497 drugs used in this example, and the number of protein-drug pairs and average protein damage score for each group are shown together.

Table 4 shows the individual drug score calculation validity (AUC) calculated when the drug score was calculated using Equation 4, when weights for each protein group were not applied (simple geometric mean) and when applied (weighted geometric mean). are shown separately for each protein group and each race.

More specifically, taking the entire population as an example, weight wi is not assigned in Equation 4 to calculate the drug score (weight wi = 1), and each protein group, such as target protein, carrier protein, metabolic enzyme protein, transporter protein, etc. The calculated AUC values were 0.617, 0.554, 0.587, and _0.497 , respectively, and the individual weighted geometric mean drug score calculation validity (AUC=0.667 ) was obtained (see FIG. 3b). As a result, the validity of calculating the weighted geometric mean drug score for each individual to which the weight for each protein group was applied was applied to the simple geometric mean calculation formula without weighting (weight wi = 1), and the validity of calculating the simple geometric mean drug score for each individual (AUC = 0.666) was confirmed to be improved by 0.001 points (see FIG. 3a).

In addition, as shown in FIG. 3A, as another example of weight application, as a result of performing an individual drug score calculation validity (AUC) analysis by applying weights according to the number of people for each race, when racial specificity was considered (thick line), the total The AUC value of the population group (Total) was 0.666 (African 0.744, American 0.650, Asian 0.631, European 0.653), when racial specificity was not considered (dotted line) , the AUC value of the entire population was 0.633 (Africans 0.623, Americans 0.629, Asians 0.64, Europeans 0.636), confirming that the validity of drug score calculation considering racial specificity was further improved compared to the case where it was not.

In addition, as shown in FIG. 3B, when only weights for each protein group are applied without considering racial specificity (dotted line), the validity AUC of calculating individual drug scores of the present invention is 0.634, and when weights for each protein group are applied together with racial specificity ( thick line), the AUC of the validity of the individual drug score calculation of the present invention is 0.667, indicating the usefulness of different weights.

Although the exemplary embodiments of the present application have been described in detail above, the scope of the present application is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concepts of the present application as defined in the following claims are also within the scope of the present application. that belongs to

All technical terms used in the present invention, unless defined otherwise, are used with the same meaning as commonly understood by one of ordinary skill in the art related to the present invention. The contents of all publications incorporated herein by reference are incorporated herein by reference.

Claims (32)

Determining at least one gene sequence mutation information involved in pharmaco-dynamics or pharmaco-kinetics of a therapeutic agent for osteoporosis from personal genome sequence information;
Calculating individual protein damage scores using the gene sequence mutation information; and
Calculating an individual drug score by associating the individual protein damage score with a correlation between an osteoporosis treatment and a protein,
The above individual protein damage The score is calculated by Equation 2 below,
[Equation 2]

In Equation 2, S _g is the protein damage score of the protein encoded by gene g , n is the number of nucleotide sequence variants to be analyzed among the nucleotide sequence variants of the gene g , and v _i is the gene of the ith nucleotide sequence variant to be analyzed. A base sequence variation score, w _i is a weight given to the gene sequence variation score v _i of the ith base sequence variation to be analyzed, i is an integer from 1 to n ,
The individual drug score is calculated by Equation 4 below,
[Equation 4]

In Equation 4, S _d is the calculated drug score of drug d , n is the number of proteins encoded by one or more genes involved in the pharmacodynamics or pharmacokinetics of drug d , g _i is involved in the pharmacodynamics or pharmacokinetics of drug d is the protein damage score of a protein encoded by one or more genes, w _i is a weight given to the protein damage score g _i of a protein encoded by the ith gene, and i is an integer from 1 to n , a personal genome sequence Method for providing information for selection of osteoporosis treatment using mutation.
According to claim 1,
The osteoporosis treatment agent consists of bisphosphonate, anti-RANKL monoclonal antibody, golecalciferol, vitamin D analog, selective estrogen receptor modulator, calcitriol, calcipediol, calcitonin, vitamin K compound, estrogen, progesterone and parathyroid hormone A method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, which is a drug belonging to at least one class selected from the group.
According to claim 1,
Genes involved in the pharmacodynamics or pharmacokinetics are FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682.6, SLC12A4, ADARB2 , CD3G, ZFYVE20, ASZ1, ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, and ADCY3, a method for providing information for selecting an osteoporosis treatment using genetic sequence mutation of one or more individuals selected from the group consisting of .
According to claim 3,
The gene further comprises at least one selected from the group consisting of PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1 and OBP2A A method for providing information for selecting an osteoporosis treatment using personal genome sequence variation, comprising:
According to claim 1,
Genes involved in the pharmacodynamics or pharmacokinetics are FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682.6, SLC12A4, ADARB2 , CD3G, ZFYVE20, ASZ1, ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, ADCY3, PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27 , C1orf229, LRRTM3, SPDEF, GZMA, at least one selected from the group consisting of LTBP1 and OBP2A, a method for providing information for the selection of osteoporosis treatment using individual genome sequence mutation.
According to claim 1,
The gene sequence mutation information is a method for providing information for selecting an osteoporosis treatment using individual genome sequence mutation, which is a substitution, addition or deletion of a base constituting an exon of a gene.
According to claim 1,
Method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, wherein the genetic sequence variation information is obtained through comparative analysis with the genome sequence of the reference group
According to claim 1,
The protein damage score or the drug score is SIFT (Sorting Intolerant From Tolerant), PolyPhen (Polymorphism Phenotyping), PolyPhen-2, MAPP (Multivariate Analysis of Protein Polymorphism), Logre (Log R Pfam E-value), MutationAssessor, MutationTaster, One selected from the group consisting of MutationTaster2, PROVEAN (Protein Variation Effect Analyzer), PMut, Condel, GERP (Genomic Evolutionary Rate Profiling), GERP++, CEO (Combinatorial Entropy Optimization), SNPeffect, fathmm, and CADD (Combined Annotation-Dependent Depletion) A method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, which is calculated from one or more gene sequence variation scores calculated using the above algorithm.
According to claim 1,
The protein damage score or drug score is calculated from the genetic sequence variation score, a method for providing information for selecting an osteoporosis treatment using individual genome sequence variation.
According to claim 9,
The gene sequence variation score is SIFT (Sorting Intolerant From Tolerant), PolyPhen (Polymorphism Phenotyping), PolyPhen-2, MAPP (Multivariate Analysis of Protein Polymorphism), Logre (Log R Pfam E-value), MutationAssessor, MutationTaster, MutationTaster2, One or more algorithms selected from the group consisting of PROVEAN (Protein Variation Effect Analyzer), PMut, Condel, GERP (Genomic Evolutionary Rate Profiling), GERP++, CEO (Combinatorial Entropy Optimization), SNPeffect, fathmm, and CADD (Combined Annotation-Dependent Depletion) A method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, which is calculated by applying to gene sequence variation.
According to claim 1,
The protein damage score, when there are two or more target sequence mutations found in the gene encoding the protein, is calculated as the average value of the gene sequence mutation scores. How to provide information for.
According to claim 11,
The average value is geometric mean, arithmetic mean, harmonic mean, arithmetic geometric mean, arithmetic harmonic mean, geometric harmonic mean, Pythagorean mean, Heron mean, inverse harmonic mean, root mean square deviation, centroid mean, quartile mean, quadratic mean, cut mean , Winsorized mean, weighted mean, weighted geometric mean, weighted arithmetic mean, weighted harmonic mean, mean of a function, power mean, generalized f-mean, percentile, maximum, minimum, mode, median, median range, central tendency A method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, which is calculated by at least one selected from the group consisting of measures of central tendency, simple products and weighted products.
delete delete According to claim 1,
The drug score, when there are two or more protein damages involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment, is calculated as the average value of the protein damage scores, providing information for selecting an osteoporosis treatment using individual genome sequence mutation method.
According to claim 15,
The average value is geometric mean, arithmetic mean, harmonic mean, arithmetic geometric mean, arithmetic harmonic mean, geometric harmonic mean, Pythagorean mean, Heron mean, inverse harmonic mean, root mean square deviation, centroid mean, quartile mean, quadratic mean, cut mean , Winsorized mean, weighted mean, weighted geometric mean, weighted arithmetic mean, weighted harmonic mean, mean of a function, power mean, generalized f-mean, percentile, maximum, minimum, mode, median, median range, central tendency A method for providing information for selecting an osteoporosis treatment using individual genome sequence variation, which is calculated by at least one selected from the group consisting of measures of central tendency, simple products and weighted products.
delete delete According to claim 1,
The protein damage score or drug score is calculated by weighting a value determined by considering the type of the protein, pharmacodynamics or pharmacokinetic classification of the protein, pharmacokinetic parameters of the drug-metabolizing enzyme, or distribution by population or race Method for providing information for selection of osteoporosis treatment using human and individual genome sequence mutation.
According to claim 1,
The method further includes a subsequent step of determining whether to use or how to use the osteoporosis treatment applied to the individual using the individual drug score, providing information for selecting an osteoporosis treatment using individual genome sequence mutation method.
According to claim 1,
The method further includes a preceding step of receiving nucleotide sequence mutation information of a gene involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment into a computer system,
The computer system includes a database containing genetic information involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment, or provides information for selecting an osteoporosis treatment using individual genome sequence variation that is accessible to the database.
The method of any one of claims 1 to 12, 15, 16 and 19 to 21,
The method for providing information for selecting an osteoporosis therapeutic agent using individual genome sequence mutation, characterized in that performed for the purpose of preventing side effects of the osteoporosis therapeutic agent.
Regarding the osteoporosis treatment applicable to individuals, a database capable of searching or extracting information related to genes or proteins related to the osteoporosis treatment;
a communication unit capable of accessing the database;
A first calculation module for calculating one or more gene sequence mutation information involved in the pharmacodynamics or pharmacokinetics of the osteoporosis treatment based on the information;
a second calculation module for calculating individual protein damage scores using the gene sequence mutation information;
a third calculation module for calculating an individual drug score by associating the individual protein damage score with a relationship between a drug and a protein; and
And a display unit for displaying the calculated value calculated by the calculation module,
The above individual protein damage The score is calculated by Equation 2 below,
[Equation 2]

In Equation 2, S _g is the protein damage score of the protein encoded by gene g , n is the number of nucleotide sequence variants to be analyzed among the nucleotide sequence variants of the gene g , and v _i is the gene of the ith nucleotide sequence variant to be analyzed. A base sequence variation score, w _i is a weight given to the gene sequence variation score v _i of the ith base sequence variation to be analyzed, i is an integer from 1 to n ,
The individual drug score is calculated by Equation 4 below,
[Equation 4]

In Equation 4, S _d is the calculated drug score of drug d , n is the number of proteins encoded by one or more genes involved in the pharmacodynamics or pharmacokinetics of drug d , g _i is involved in the pharmacodynamics or pharmacokinetics of drug d is the protein damage score of a protein encoded by one or more genes, w _i is a weight given to the protein damage score g _i of a protein encoded by the ith gene, and i is an integer from 1 to n , a personal genome sequence Osteoporosis treatment selection system using mutation.
24. The method of claim 23,
The system further comprises a fourth calculation module for determining whether to use the osteoporosis treatment applied to the individual using the individual drug score calculated in the third calculation module, osteoporosis treatment using individual genome sequence mutation. selection system.
24. The method of claim 23,
The system further comprises a user interface for calculating and providing individual drug scores for the osteoporosis treatment according to the input of the osteoporosis treatment by the user.
24. The method of claim 23,
The display unit further comprises a display unit for additionally displaying the value calculated in each of the calculation modules, the calculation process, or information based on the calculation, the osteoporosis treatment selection system using the individual genome sequence mutation.
27. The method of any one of claims 23 to 26,
Each of the calculation modules stores the gene sequence mutation information, protein damage score, drug score, and information that is the basis for the calculation,
An osteoporosis treatment selection system using individual genome sequence mutation, in which information of each calculation module is updated according to the update of the database.
A computer-readable storage medium comprising an execution module that executes the functions of the processor, wherein the functions are:
Obtaining at least one gene sequence mutation information involved in the pharmacodynamics or pharmacokinetics of an osteoporosis treatment from personal genome sequence information;
Calculating individual protein damage scores using the gene sequence mutation information; and
Calculating an individual drug score by associating the individual protein damage score with a mutual relationship between a drug and a protein;
The above individual protein damage The score is calculated by Equation 2 below,
[Equation 2]

In Equation 2, S _g is the protein damage score of the protein encoded by gene g , n is the number of nucleotide sequence variants to be analyzed among the nucleotide sequence variants of the gene g , and v _i is the gene of the ith nucleotide sequence variant to be analyzed. A base sequence variation score, w _i is a weight given to the gene sequence variation score v _i of the ith base sequence variation to be analyzed, i is an integer from 1 to n ,
The individual drug score is calculated by Equation 4 below,
[Equation 4]

In Equation 4, S _d is the calculated drug score of drug d , n is the number of proteins encoded by one or more genes involved in the pharmacodynamics or pharmacokinetics of drug d , g _i is involved in the pharmacodynamics or pharmacokinetics of drug d is the protein damage score of a protein encoded by one or more genes, w _i is a weight given to the protein damage score g _i of a protein encoded by the ith gene, and i is an integer from 1 to n , computer readable storage media.
29. The method of claim 28,
Wherein the processor determines whether or not to use the osteoporosis treatment applied to the individual using the individual drug score; performing a function that further includes, a computer readable storage medium.
Determining at least one gene sequence mutation information involved in pharmaco-dynamics or pharmaco-kinetics of a therapeutic agent for osteoporosis from personal genome sequence information;
Calculating individual protein damage scores using the gene sequence mutation information; and
Calculating an individual drug score by associating the individual protein damage score with a correlation between an osteoporosis treatment and a protein,
Genes involved in the pharmacodynamics or pharmacokinetics are FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682.6, SLC12A4, ADARB2 , CD3G, ZFYVE20, ASZ1, ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, ADCY3, PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27 , C1orf229, LRRTM3, SPDEF, GZMA, at least one selected from the group consisting of LTBP1 and OBP2A, a method for providing information for the selection of osteoporosis treatment using individual genome sequence mutation.
31. The method of claim 30,
The gene further comprises at least one selected from the group consisting of PGLYRP3, MAGEA12, TMEM187, OR4S1, KRTAP5-4, SERPINI2, BTNL2, CXorf1, PRG4, VAMP8, CSAG1, C1orf27, C1orf229, LRRTM3, SPDEF, GZMA, LTBP1 and OBP2A A method for providing information for selecting an osteoporosis treatment using personal genome sequence variation, comprising:
31. The method of claim 30,
The gene is FGFR4, IGHMBP2, ARHGEF26, HBXIP, CD1A, ZNF57, LAMA4, SIPA1, GPC4, PSMD9, AKNA, ANKRD50, DUSP13, PIK3R1, C4orf14, GPR174, CHDH, AC018682.6, SLC12A4, ADARB2, CD3G, ZFYVE20, ASZ1 , ARHGEF4, FAT4, OR2M5, HPS1, ZNF235, P4HB, ATP11C, SYNE1, and ADCY3 further comprising one or more selected from the group consisting of, a method for providing information for selecting a therapeutic agent for osteoporosis using individual genome sequence mutations.

Images