KR20180090680A - Geneome analysis system - Google Patents

Abstract

The present invention relates to a genome analysis system. The dielectric analysis system according to an embodiment of the present invention includes a folksonomy construction system and an analysis part. The folksonomy construction system receives omix data for genes corresponding to the clinical variable of a specific disease. The folksonomy construction system extracts genome annotations from the omix data and classifies the genome annotations to generate a classification matrix. The folksonomy construction system generates mapping data corresponding to a folksonomy structure based on the classification matrix. The analysis part analyzes disease associations to genes based on the mapping data. The genome analysis system of the present invention can efficiently analyze a relationship between genome and disease.

Description

[0001] GENEOME ANALYSIS SYSTEM [0002]

The present invention relates to disease prediction and analysis, and more particularly to a genomic analysis system.

A genome is a collection of all the genetic information an organism has. Every cell that makes up an organism has the same number of chromosomes and genetic information. In the past, genomic analysis depended on qualitative methods, but recently it is possible to quantitatively analyze the genome using the next generation sequencing (NGS) technology. Unlike conventional sequence sequencing, the next-generation sequencing technology allows a large number of DNA fragments to be processed in parallel to analyze the base sequence at high speed. Thus, a variety of dielectric tin can be produced using the next generation sequencing technology.

BACKGROUND OF THE INVENTION With the development of electronic communication technology, electronic communication technology is frequently used in biotechnology and medicine. Thus, there is a need to analyze genomes using electronic communication technologies and to diagnose or predict human diseases. In addition, the development of biotechnology has led to the development of genetic information detection technology, which has led to the demand for understanding diseases based on various genome annotations.

Dielectric data should be effectively categorized to analyze correlations between genome annotations and disease, and to improve confidence in correlation. Therefore, development of various algorithms for effective classification of genomic data is required.

The present invention can provide a dielectric analysis system capable of effectively analyzing the relationship between a genome and a disease.

A dielectric analysis system according to an embodiment of the present invention includes a fountain-well construction system and an analysis unit.

The folksonomy building system receives the omix data for the genes corresponding to the clinical variables of a specific disease. The follyonomy construction system extracts the dielectric annotations from the omix data and classifies the dielectric annotations to generate a classification matrix. The phantom building system links gene and genome annotations based on the classification matrix and generates mapping data that links genomic annotations to clinical variables.

The analysis unit analyzes the disease association with the gene based on the mapping data. The analysis unit analyzes the association between genes and genetic annotations, the relationship between genetic annotations and clinical variables, or the relationship between genes and clinical variables.

The dielectric analysis system according to the embodiment of the present invention can ensure the reliability and efficiency of the correlation analysis between the genome and the disease by mapping the gene, the tin and the disease to the fuzzy nominal structure.

1 is a block diagram of a dielectric analysis system according to an embodiment of the present invention.
Figure 2 is a block diagram of the follyonomy building system of Figure 1;
3 is a diagram for explaining a structure linking genes, dielectric annotations, and clinical parameters.
4 is a flowchart showing a method of analyzing a dielectric using a dielectric analysis system.

Hereinafter, embodiments of the present invention will be described in detail and in detail so that those skilled in the art can easily carry out the present invention.

1 is a block diagram of a dielectric analysis system according to an embodiment of the present invention.

Referring to FIG. 1, a dielectric analysis system 1000 includes a clinical information database 100, a follyonomy construction system 200, an analysis unit 300, and a disease relevance database 400.

The clinical information database 100 stores a plurality of omix data od for a human body or a living body. The omix data (od) includes data on various biological information such as genomic data for organisms, transcript data, or protein body data. The omix data od stored in the clinical information database 100 may include information on an object of extraction of the omix data od. For example, when the omix data od is generated based on a genome or the like extracted from a patient for a specific disease, the omix data od may include clinical information of the patient. The clinical information database 100 may be an external database such as the Cancer Genome Atlas (TCGA), COSMIC, GEO, ArrayExpress, GWAS central, or CNVDB.

The follyonomy construction system 200 receives the omix data od from the clinical information database 100. The follyonomy construction system 200 can selectively receive the omix data od for specific clinical information from the clinical information database 100. [ That is, the folomonomy establishing system 200 can access the clinical information database 100 by specifying a patient group for a specific disease, and receive omix data (od) for a specified patient group based on clinical information.

The follyonomy construction system 200 establishes a classification system for genome data and clinical information based on the omix data (od). The follyonomy construction system 200 defines a connection relationship of genes, genome annotations, and clinical variables so as to correspond to the follyonomy structure using the connection relation of users, tags, and resources. The follyonomy construction system 200 extracts dielectric annotations corresponding to specific clinical variables from the genome data. The follyonomy construction system 200 generates the classification matrix by classifying the extracted dielectric annotations. The follyonomy construction system 200 establishes a connection relationship by linking genes and dielectric annotations based on the classification matrix and connecting dielectric annotations to clinical variables. The follyonomy construction system 200 generates the mapping data md based on this connection relationship. Details will be described later.

The analysis unit 300 analyzes the association between the genome and disease information based on the mapping data md. The analysis unit 300 may receive the mapping data md mapped by the follyonomy construction system 200. Alternatively, the analysis unit 300 may receive the mapping data md stored in the disease relevance database 400. [ An association analysis between disease information may include predicting genomic annotations in a disease group, analyzing clinical trends, or predicting a set of related genes.

The analysis unit 300 can predict a gene having a high correlation with a clinical parameter. For example, the analysis unit 300 quantitatively analyzes a link between a gene and a clinical variable. The analysis unit 300 extracts the common points of the genes linked to the specific clinical variables, that is, consensus markers. In addition, the analysis unit 300 extracts a difference between the genes and the genes linked to specific clinical variables, i.e., specific markers. The analysis unit 300 can predict a gene having a high correlation with a clinical parameter based on the intersection of specific markers and consensus markers.

The analyzer 300 can predict a gene showing a pattern commonly using specific dielectric tin. For example, the analysis unit 300 quantitatively analyzes a link between a gene and a dielectric annotation. The analysis unit 300 extracts the common points, i.e., consensus markers, of the genes linked to specific dielectric tins. In addition, the analysis unit 300 extracts a difference between the genes linked to specific genomes and other genes, that is, specific markers. The analyzer 300 can predict a gene showing a pattern commonly using specific dielectric tins based on the intersection of specific markers and consensus markers.

The analysis unit 300 can predict a gene having a common genome annotation for the same clinical variables. For example, the analysis unit 300 quantitatively analyzes the linkage between the genome annotation and the clinical variables, and extracts the same clinical variables and the common points of genes linked to the same genome annotation, and differences with other genes. Based on these common points and differences, the analysis unit 300 can predict a gene having a common gene annotation for the same clinical variable.

The analysis unit 300 may analyze the association between the gene and the disease using the specific marker or the consensus marker described above and perform a permutation test or an enrichment test for the quantitative analysis . The analysis unit 300 generates analysis data including analysis result information of the association between the gene and the disease based on the mapping data md.

The disease relevance database 400 stores mapping data md and analysis data. The disease association database 400 includes a mapping database for storing mapping data (md) relating to the linkage relationship of genes, genome annotations, and clinical variables in a database, and stores analysis data on the association between genes and diseases And an analysis database for storing the converted data.

The disease relevance database 400 may receive the mapping data md from the phonemic building system 200 and receive the analysis data from the analysis unit 300. The disease relevance database 400 periodically receives the mapping data md from the phonemic building system 200 to update the linkage information of the gene, the genome annotation, and the clinical variables, The analysis result information can be updated by receiving the analysis data. The user can access the disease association database 400 to refer to the association between the gene and the disease.

The user can use the genome analysis system 1000 to predict a disease possibility or perform a medical diagnosis. For example, a user such as a doctor can detect a gene of a subject such as a patient or a visiting customer, and analyze the gene using the genome analysis system 1000. Dielectric tin and clinical variables for samples with genes similar to the gene of interest can be referred to, and genetic tin of the gene can be extracted for disease relevance. Based on the results of the disease association analysis, the user can suggest the possibility and prevention of the disease of the subject.

Figure 2 is a block diagram of the follyonomy building system of Figure 1;

2, the follyonomy construction system 200 includes a communication interface 210, a system interface 220, a dielectric annotation extraction unit 230, a classification matrix generation unit 240, and a mapping unit 250 .

The communication interface 210 receives the omix data od from the clinical information database 100 at the request of the user. The communication interface 210 provides a request signal for specific clinical information to the clinical information database 100 and receives the omix data od corresponding to the specific clinical information from the clinical information database 100. The communication interface 210 receives a request signal for specific clinical information from the system interface 220. The communication interface 210 provides the omix data od to the dielectric annotation extractor 230.

The system interface 220 controls the communication interface 210 to receive the omix data od from the clinical information database 100 according to a user's request. The system interface 220 receives an external command USER from the user and controls the communication interface 210 based on an external command USER. The external command (USER) may be an error data collection command for clinical information on a particular disease. The external command USER may include a plurality of clinical information and an omnix data collection command for the multiple analysis of the mapping data md. In this case, the communication interface 210 can receive the omix data od corresponding to a plurality of pieces of clinical information.

The dielectric annotation extractor 230 receives the omix data od from the communication interface 210. The dielectric annotation extractor 230 extracts the dielectric annotation from the omix data od. The dielectric annotation extractor 230 can extract various dielectric annotations by filtering various parameters from the omix data od. For example, the dielectric annotation extractor 230 may extract the SNVs using copy number variation (CNV) filtering, single nucleotide variation (SNV), methyl filtering, miRNA filtering, or mRNA (protein) The dielectric annotation can be extracted from the omix data od. However, the present invention is not limited to this, and the dielectric tin extraction unit 230 may have various types of dielectric tin extraction methods. Dielectric tin may include SNV, CNV, methyl, miRNA, insertion-deletion (INDEL) of the nucleotide sequence, protein, or differential expression gene. The dielectric annotation extractor 230 generates dielectric annotation data based on the extracted dielectric annotations.

The classification matrix generator 240 receives the dielectric annotation data from the dielectric annotation extractor 230. The classification matrix generator 240 may classify the dielectric annotation data to generate a classification matrix. For example, the classification matrix generator 240 classifies the genomic annotation data corresponding to clinical variables based on specific clinical information into CNV, SNV, methyl, miRNA, and the like, and determines the relationship between the gene corresponding to the clinical variable and the genome annotation Can be represented by a matrix. That is, one matrix can be generated that represents the relationship of gene and dielectric annotations to one clinical variable.

The classification matrix generation unit 240 may generate a matrix of the relationship between the gene and the clinical variable for the dielectric annotation or a matrix of the relationship between the genome annotation for the gene and the clinical variable without generating the matrix based on the clinical variable have. In this case, the follyonomy construction system 200 may provide a request signal for a plurality of clinical information to the clinical information database 100 to secure data for a plurality of clinical variables, It is possible to receive the data od.

The classification matrix generation unit 240 may generate a classification matrix using various statistical analysis methods. For example, the classification matrix generator 240 may generate a classification matrix using t-test (t-test), survival, or machine learning.

The mapping unit 250 receives the classification matrix from the classification matrix generation unit 240. The mapping unit 250 defines the linkage between the gene, the dielectric annotation, and the clinical variable based on the classification matrix. The linkage relationship between genes, genome tin, and clinical variables corresponds to the fungus structure. The follyonomy structure can correspond to a structure in which a user tags a tag to access a resource. The mapping unit 250 can map the classification matrix to a structure in which genes tag the genome annotations to access the clinical variables. For example, the mapping unit 250 may construct a network structure that links the dielectric annotation with the clinical variables based on the dielectric tin components included in the classification matrix. The mapping unit 250 can construct a network structure linking the gene and the dielectric annotation based on the intersection value of the dielectric tin component and the gene component included in the classification matrix. The mapping unit 250 generates mapping data (md) for a relation linking these genes, dielectric annotations, and clinical variables. The mapping data md is provided to the analysis unit 300 or the disease relevance database 400.

3 is a diagram for explaining a structure linking genes, dielectric annotations, and clinical parameters.

Referring to FIG. 3, the follyonomy construction system 200 of FIG. 2 may have received three pieces of omics data od. In other words, the follyonomy construction system 200 stores the omix data od for the first gene (G1), the second gene (G2), and the third gene (G3) 100).

The follyonomy construction system 200 extracts dielectric tins from the first to third genes G1 to G3. Dielectric tin extracted from the first to third genes (G1 to G3) includes SNV, CNV, methyl (Methyl), and miRNA. Dielectric tin corresponding to the first clinical variable R1 includes a first SNV (T11), a first CNV (T12), a first Methyl (T13), and a first miRNA (T14). The dielectric tin corresponding to the second clinical variable R2 includes a second SNV (T21), a second CNV (T22), a second Methyl (T23), and a second miRNA (T24). Dielectric tin corresponding to the third clinical variable R3 includes the third SNV (T31), the third CNV (T32), the third Methyl (T33), and the third miRNA (T34). The dielectric tin extracted from the first gene (G1) includes a first SNV (T11), a first CNV (T12), a first miRNA (T14), a second CNV (T22), and a third Methyl (T33). The dielectric tin extracted from the second gene G2 includes a first CNV (T12), a second SNV (T21), a second miRNA (T24), and a third CNV (T32). Dielectric tin extracted from the third gene (G3) includes the first Methyl (T13), the second Methyl (T23), the third SNV (T31), the third Methyl (T33), and the third miRNA (T34).

The dielectric tin and the clinical variables may be network structures constructed by the phantom building system 200. In this case, FIG. 3 may be for analyzing the disease association from the first to third genes (G1 to G3) extracted from the three subjects. The analysis unit 300 may analyze the disease association from the genetic annotations of the first to third genes G1 to G3. For example, the first SNV (T11), the first CNV (T12), and the first miRNA (T14) of the dielectric tin corresponding to the first clinical variable are detected from the subject having the first gene (G1) The possibility of disease expressed by the first clinical variable is higher than that of the other subjects. The third SNV (T31), the third Methyl (T33), and the third miRNA (T34) among the dielectric tin corresponding to the third clinical variable were detected in the subject having the third gene (G3) The possibility that the disease is expressed by the patients is higher than that of the other subjects.

The first to third genes G1 to G3 may be genes corresponding to at least one of the first to third clinical variables R1 to R3. In this case, FIG. 3 may be for additionally databaseing the connection between the clinical variables and the genome annotation from three subjects having the disease expressed by at least one of the first to third clinical variables R1 to R3 . The analysis unit 300 can analyze the connection between the dielectric annotation and the clinical variables from the dielectric annotations of the first to third genes G1 to G3. For example, when genetic tin extracted from a subject having a disease represented by the first clinical variable (R1) and having a first gene (G1) was analyzed, the first SNV (T11), the first CNV T12) and when the first miRNA (T14) is extracted, it can be analyzed that the disease potential expressed by the first clinical variable is high. The correlation between the first Methyl (T13) and the first clinical parameter can be analyzed to have a lower weight than the first SNV (T11), the first CNV (T12), and the first miRNA (T14).

Thus, when analyzing the disease association using the structure as in FIG. 3, the association between the genome annotation and the clinical variable may vary with data accumulation. The number of accesses of genome annotations linked to specific clinical variables can be used to measure the association between genome annotations and clinical variables. Based on the accumulated data, the genome tin detected from the gene of the subject can be analyzed and the possibility of disease can be predicted.

4 is a flowchart showing a method of analyzing a dielectric using a dielectric analysis system.

Referring to FIG. 4, a dielectric analysis method (S1000) includes a step of extracting dielectric tins (S100), a step of creating a classification matrix (S200), a step of constructing a follyonomy structure (S300), and a step of analyzing a disease association (S400). The dielectric analysis method (S1000) can be performed by the dielectric analysis system 1000 of FIG.

The step of extracting the dielectric tin (SlOO) may be performed by the follooming system 200 of FIG. 1 or the dielectric annotation extractor 230 of FIG. In step S100 of extracting the dielectric tin, the follyonomy construction system 200 extracts the dielectric tin from the omix data od. The dielectric annotation may be extracted from the omix data (od) corresponding to a particular clinical variable.

The step S200 of generating the classification matrix may be performed by the fumnomy building system 200 of FIG. 1 or the classification matrix generator 240 of FIG. In step S200 of generating a classification matrix, the falseonomy construction system 200 generates a classification matrix by classifying the dielectric annotations. For example, a classification matrix may represent the relationship between a gene and a dielectric annotation corresponding to a particular clinical variable.

Step S300 of constructing the follyonomy structure may be performed by the follooming system 200 of FIG. 1 or the mapping unit 250 of FIG. In step S300 of constructing the follyonomy structure, the follyonomy construction system 200 may generate mapping data md defining relationships linking genes, genome annotations, and clinical variables based on the classification matrix. In other words, a gene can be linked to a genome annotation, and a genomic annotation linking the genome annotation with a clinical variable can be constructed.

The step S400 of analyzing the disease association can be performed by the analysis unit 300 of FIG. In step S400 of analyzing the disease association, the analysis unit 300 may analyze the association between the genome and the disease based on the mapping data md. For example, associations between disease groups and genetic annotations, disease groups and clinical trends, or trends between disease groups and genes can be analyzed. Specifically, in step S400 of analyzing the disease association, it is possible to predict a gene having a high correlation with a clinical variable, predict a gene using a specific genome annotation in common, and use a common genome annotation Can be predicted.

The dielectric analysis system 1000 and the dielectric analysis method S1000 of the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer readable medium. The computer-readable medium includes a recording medium such as a hard disk, a magnetic medium, and an optical recording medium, and may include a storage device such as ROM, RAM, flash memory and the like.

The above description is a concrete example for carrying out the present invention. The present invention includes not only the above-described embodiments, but also embodiments that can be simply modified or easily changed. In addition, the present invention includes techniques that can be easily modified by using the above-described embodiments.

1000: Dielectric analysis system 100: Clinical information database
200: foolishness building system 300: analysis unit
400: Disease association database

Claims (1)

The method comprising the steps of: receiving omix data for a gene corresponding to a clinical condition of a specific disease; extracting genomic annotations from the omix data; generating a classification matrix by classifying the genomic annotations; A follyonomy construction system for connecting the dielectric tins and generating mapping data for connecting the dielectric tins and the clinical variables; And
And an analysis unit for analyzing the association between the gene and the dielectric annotations based on the mapping data, the association between the dielectric annotations and the clinical variable, or the association between the gene and the clinical variable.

Images