KR101839572B1 - Apparatus Analyzing Disease-related Genes and Method thereof - Google Patents

Abstract

The present invention discloses a device for analyzing genetic relation, which can search candidate genes related to a specific disease by establishing a gene integrated network. The device for analyzing genetic relation comprises: a combination symbol detecting unit which searches text data including information on genes by using at least one gene symbol, and detects a gene combination symbol combined with the gene symbol from the searched text data; an integrated network establishing unit which establishes regional gene networks in which uses the gene symbol as a node and represents the genetic relation as an edge, based on the detected gene combination symbol, and establishes an integrated gene network by integrating the established regional gene networks; and an analyzing unit which determines the priority of gene symbols related to a disease to be analyzed by using the integrated gene network, and analyzes relations among the genes related to the disease according to the determined priority.

Description

[0001] Apparatus Analyzing Disease-related Genes and Methods [0002]

The present invention relates to an apparatus for analyzing a gene relationship. More particularly, the present invention relates to a device for searching genes associated with diseases and analyzing the relationship of the genes searched.

As much research on genes and diseases has progressed, the results have accumulated in a large amount of documentation. However, due to the nature of gene technology, research data are vast and necessary biologic documents are scattered. Therefore, there is a difficulty in genetic research by analyzing biological information at the right place.

Biological documents are generated as a result of a number of biological experiments, the results of which are stored in databases such as PubMed and OMIM. The stored database of these biological experiment results is useful for researchers to carry out new experiments. However, since the amount of the data is large and scattered, it is not easy for the researchers to access the data that is practically necessary.

Therefore, research has been carried out to analyze texts using text mining techniques and to understand the correlation between genes and proteins related to diseases. In the past research using text mining techniques, methods using synonyms and using verbs in sentences Methods already exist.

Prioritization is an important factor in text mining when considering the importance of genes in the identification of disease, and the development of techniques for inferring potential genes related to disease based on the phenotype of genes or the similarity of proteins Is in progress.

However, in the conventional technology, when searching for disease-related genes, information on useful genes may be missed by using a database related to a specific disease, and in collecting disease-related information for establishing a network, .

Therefore, there is a need to develop a technique for analyzing a large biological literature and analyzing the relationship between genes associated with diseases.

Korean Patent Laid-Open No. 10-2009-0007972 (published)

Disclosure of Invention Technical Problem [8] The present invention has been made to solve the above problems, and an apparatus and a method for analyzing gene relationships are disclosed. In particular, a gene network is constructed based on a gene already known to be related to a disease, and a device for analyzing a gene relationship related to a disease is disclosed.

According to an aspect of the present invention, there is provided a gene-relationship analyzing apparatus for searching text data including information on a gene using at least one gene symbol, A combined symbol detector for detecting a combined symbol combination of symbols; An integrated network construction unit for constructing an integrated gene network in which the gene symbol is a node and the relation of the gene is an edge based on the detected gene combination symbol; And an analysis unit for determining a gene symbol prioritization related to a disease to be analyzed using the constructed gene network and analyzing a relationship between genes related to the disease according to the determined priority; .

In the present invention, the gene relationship analyzing apparatus includes a seed gene symbol acquiring unit for acquiring a seed gene symbol already known as a gene related to a disease to be analyzed; And the combination symbol detector may search the text data using the obtained seed gene symbol and detect the gene combination symbol in the retrieved text data.

In the present invention, the combination symbol detector may include: a symbol extractor for extracting the genetic symbol from the sentence by classifying the searched text data on a sentence-by-sentence basis; analyzing the sentence; And when the at least two gene symbols are extracted in the sentence, the extracted gene symbols can be detected as a combination symbol.

 In the present invention, the integrated network construction unit may include a local network construction unit for constructing a local gene network using each of the nodes as a gene symbol included in the gene combination symbol detected using the seed gene symbol. And the integrated gene network can be constructed using the constructed local gene network.

In the present invention, the integrated network construction unit may include a network integrator for integrating the constructed local gene network according to a predetermined method, taking into consideration the type of gene symbol included in the gene combination symbol or the frequency at which the combined symbol is detected; And the integrated gene network can be constructed using the integrated local gene network.

According to the predetermined method of the present invention, when integrating the gene symbols included in the gene combination symbol, integrating each gene symbol of the combination symbol as a node increases the weight of the edge connecting the node As shown in FIG.

In the present invention, integration according to the predetermined method may be performed by combining the gene combination symbols with the matching gene symbol as a common node when one gene symbol included in the gene combination symbol coincides .

According to the method of the present invention, when the gene symbols included in the gene combination symbol do not coincide with each other, the non-matching gene symbol is integrated as a separate node and the gene combination symbols are not linked .

In the present invention, the analysis unit may include a priority determination unit for determining a priority of a gene symbol related to a disease to be analyzed based on the constructed integrated gene network; And a verification unit for arranging gene symbols related to the disease according to the determined priority and verifying the placed gene symbols with verification data related to the disease. , And the verified gene symbol can be used to analyze the relationship of the genes associated with the disease.

In the present invention, the priority determining unit may include: a first reference value calculating unit for calculating a first reference value related to the number of other nodes connected to one node based on the constructed integrated gene network; And a second reference value calculation unit for calculating a second reference value related to a weight of an edge connecting the one node with the other node; And the priority of the gene symbol related to the disease can be determined using the determined first reference value and the second reference value.

In the present invention, the priority determining unit may include: a third reference value calculating unit for calculating a third reference value related to a frequency at which the one node is exposed in a path connecting the arbitrary two nodes at the shortest distance on the integrated gene network; And the priority of the gene symbol related to the disease can be determined using the calculated third reference value.

According to another aspect of the present invention, there is provided a method for analyzing gene relationships, comprising the steps of: searching for text data including information on a gene using at least one gene symbol; Detecting a symbol; Constructing an integrated gene network based on the detected gene combination symbol, the gene symbol being a node and the relation of the gene as an edge; And determining the gene symbol priorities related to the disease to be analyzed using the constructed gene network and analyzing the relationship of the genes related to the disease according to the determined priority; .

In the present invention, the gene relational analysis method comprises: obtaining a seed gene symbol already known as a gene related to a disease to be analyzed; And searching the text data using the obtained seed gene symbol and detecting the gene combination symbol in the retrieved text data.

According to an embodiment of the present invention, the detecting step may include: dividing the searched text data into sentences, analyzing the separated sentences, and extracting gene symbols from the sentences; And when the at least two gene symbols are extracted in the sentence, the extracted gene symbols can be detected as a combination symbol.

In the present invention, the step of constructing the integrated gene network includes constructing a local gene network having gene nodes included in the gene combination symbols detected using the seed gene symbols as respective nodes. And the integrated gene network can be constructed using the constructed local gene network.

In the present invention, the step of constructing the integrated gene network may include integrating the constructed local gene network according to a predetermined method in consideration of the type of the gene symbol included in the gene combination symbol or the frequency with which the combined gene symbol is detected, ; And the integrated gene network can be constructed using the integrated local gene network.

According to the present invention, when integrating according to the predetermined method, when all of the gene symbols included in the gene combination symbol coincide with each other, each of the gene symbols of the combination symbol is used as a node to increase the weight of the edge connecting the node And combining the gene combination symbols with the matching gene symbol as a common node when one gene symbol included in the gene combination symbol coincides with each other.

According to the present invention, in the case where the gene symbols included in the gene combination symbol do not coincide with each other, the unmatched gene symbols are combined as a separate node and the gene combination symbols are not linked .

In the present invention, the analyzing step may include analyzing the number of other nodes connected to one node based on the constructed integrated gene network, the weight of the edge connecting the one node to the other node, Determining a priority of a gene related to a disease to be analyzed in consideration of the frequency of exposure of the one node in the path connecting the node at the shortest distance; And placing the gene symbol related to the disease according to the determined priority and verifying the placed gene symbol with verification data related to the disease; , And the verified gene symbol can be used to analyze the relationship of the genes associated with the disease.

The present invention also discloses a computer program stored in a computer readable recording medium for executing the above-described gene relationship analysis method in a computer.

According to the present invention, there is an advantage that the relationship of the gene related to the disease can be analyzed.

In particular, it has the advantage of integrating gene networks and deriving proven candidate genes.

1 is a block diagram of a gene-relationship analyzing apparatus according to an embodiment of the present invention.
2 is an enlarged block diagram of the combination symbol detection unit in the embodiment of FIG.
3 is an enlarged block diagram of the integrated network construction unit in the embodiment of FIG.
4 is an exemplary diagram illustrating a process of establishing an integrated network in the embodiment of FIG.
5 is an exemplary diagram illustrating an integrated gene network constructed by the integrated network building unit in the embodiment of FIG.
Figure 6 is an exemplary diagram illustrating a scaled integrated gene network.
7 is an enlarged block diagram of the analysis unit in the embodiment of FIG.
8 is an enlarged block diagram of the priority determining unit in the embodiment of FIG.
FIG. 9 is an illustration showing disease-related genes listed according to gene symbol priorities. FIG.
FIG. 10 is an exemplary diagram showing a text detected by a verification unit in the embodiment of FIG. 7 to verify a candidate gene symbol. FIG.
11 is a flowchart of a method of analyzing a gene relationship according to an embodiment of the present invention.
12 is an enlarged flow chart of the step of detecting in the embodiment of Fig.
13 is an enlarged flow chart of a step of constructing an integrated gene network in the embodiment of FIG.
14 is an enlarged flow chart of the step of analyzing in the embodiment of Fig.
15 is a reference diagram showing a flow of a method of analyzing a gene relationship according to another embodiment of the present invention.
16 is a reference diagram showing the gene analysis performance of the present invention as compared with other gene analysis methods.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the following description with reference to the accompanying drawings, the same or corresponding components are denoted by the same reference numerals, and a duplicate description thereof will be omitted.

In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope. The singular expressions include plural expressions unless the context clearly dictates otherwise. Each of the steps described below may be implemented by one or a plurality of software modules, or hardware that is responsible for each function, or a combination of software and hardware.

Specific meanings and examples of the terms will be described below in accordance with the order of each drawing.

Hereinafter, a configuration of a gene relationship analyzing apparatus according to an embodiment of the present invention will be described in detail with reference to the related drawings.

1 is a block diagram of a gene-relationship analyzing apparatus 10 according to an embodiment of the present invention.

The gene relational analysis apparatus 10 includes a seed gene symbol acquisition unit 100, a combination symbol detection unit 200, an integrated network construction unit 300, and an analysis unit 400. For example, the gene relational analysis apparatus 10 can analyze a gene relationship related to a disease to be analyzed using a seed gene and a gene network, which are genes known to be related to a specific disease. The gene-relationship analyzing apparatus 10 analyzes the gene-related genes related to the disease by inputting the seed gene, derives candidate genes expected to be related to the disease, and verifies the candidate genes.

The genetic relationship analyzer 10 analyzes a variety of biological documents produced as a result of a biological experiment through a biological experiment. For this purpose, a database containing biological information such as PubMed, OMIM, GHR, KEGG, LuGend and IGDB.NSCLC Can be used. For example, the gene relational analysis apparatus 10 may acquire gene symbols known to be associated with lung cancer in the OMIM database and use them as seed gene symbols to search for seed gene symbols related to a specific disease, Using seed gene symbols, we can extract text data including information about genes in PubMed. We can construct a gene network by detecting the gene combination symbol in the extracted text data, and extract candidates derived from the constructed gene network The genes can be verified using the GHR, KEGG, LuGend and IGDB.NSCLC databases.

The apparatus 10 for analyzing a gene relationship includes a database for acquiring input values, a database for acquiring desired gene symbols using input values, and a database for verifying acquired gene symbols, In the present invention, a gene symbol includes an identification means capable of identifying a gene, such as a genetic code representing a gene and a gene name.

The seed gene symbol acquisition unit 100 acquires a seed gene symbol already known as a gene symbol related to the disease to be analyzed. For example, the seed user's symbol acquisition unit 100 may acquire a seed gene symbol from a database in which biological information including gene information already known to be related to disease is stored in advance. A database in which the seed gene symbol acquisition unit 100 acquires seed gene symbols and a database used by the combination symbol detection unit 200 for searching text data including information on genes using seed gene symbols But may be provided in other databases.

Table 1 shows 32 seeded gene symbols, which are 32 gene symbols already known to be associated with lung cancer acquired by the seed gene symbol acquisition unit 100. In Table 1, Literature means the number of text data retrieved using each seed gene symbol. For example, the seed gene symbol acquiring unit 100 acquires a known seed gene symbol related to a specific disease from a first database (OMIM database), and the combination symbol detecting unit 200 uses seed gene symbols And the text data including the information about the gene can be retrieved from the second database (PubMed database).

The combined symbol detecting unit 200 includes a preprocessing unit 220 and a symbol extracting unit 240. For example, the combination symbol detector 200 searches for text data including information on a gene using at least one gene symbol, and detects a gene combination symbol combined with the gene symbol in the retrieved text data. The combination symbol detection unit 200 can search text data including information about a gene using at least one seed gene symbol as a search keyword and detect a combination gene symbol in the retrieved text data. In the present invention, a combined gene symbol means at least two or more gene symbols present in a sentence in the retrieved text data.

For example, the combination symbol detector 200 detects ALK-BCL2 as a gene combination symbol if two gene symbols ALK and BCL2 exist in the retrieved text data. This is one of the text mining techniques and corresponds to the co-occurrence analysis method. It is based on a methodology that assumes that mutually appearing gene symbols are related to each other in a sentence.

The preprocessing unit 220 preprocesses the retrieved text data and extracts at least a part of the retrieved text data. For example, the preprocessing unit 220 may extract only the abstract text using the keyword MeSH Term in the retrieved text data. The gene relational analysis apparatus 10 can reduce the amount of processing data through a preprocessing process performed in the preprocessing unit 220. [

The symbol extracting unit 240 extracts the genetic symbols in the sentence by analyzing the separated sentences by classifying the searched text data on a sentence-by-sentence basis. For example, when the preprocessing unit 220 performs a preprocessing process, the symbol extracting unit 240 extracts the preprocessed text data on a sentence-by-sentence basis, and extracts the gene symbols in the sentence by analyzing the separated sentences. The combined symbol detection unit 200 detects the extracted gene symbol as a combined gene symbol when the extracted symbol is at least two or more within one sentence. Will be described with reference to FIG.

The integrated network construction unit 300 includes a local network construction unit 320 and a network integration unit 340. For example, the integrated network construction unit 300 constructs a local gene network that uses the gene symbol as a node and the relation of the gene as an edge based on the detected gene combination symbol, Integrate and build an integrated gene network. In the present invention, the integrated gene network is a network containing gene symbols related to a specific disease, and can be constructed by connecting each gene symbol as a node and connecting the nodes as an edge.

The local network construction unit 320 constructs a local gene network generated by the seed gene symbol by using the gene symbol included in the gene combination symbol detected using the seed gene symbol as each node. For example, the local network construction unit 320 constructs a network based on the gene combination symbol detected in the text data retrieved using the obtained seed gene symbol, Network. For example, the local network constructed using the seed gene symbol ALK includes only the network in which the gene symbols detected from the text data retrieved using the seed gene symbol ALK are each node. That is, the local network construction unit 320 can construct 32 local gene networks using 32 seed genes.

The network integration unit 340 integrates the constructed local gene network according to a predetermined method in consideration of the type of the gene symbol included in the gene combination symbol or the frequency at which the gene combination symbol is detected. For example, the network integration unit 340 integrates a local gene network constructed based on one seed gene symbol to generate an integrated gene network. Will be described with reference to FIG.

The network integrating unit 340 may integrate the genes according to a predetermined method. First, if the gene symbols included in the gene combination symbol are identical (326) It is possible to integrate the network by using each gene symbol of the combination symbol as a node and increasing the weight of the edge connecting the node. As shown in FIG. 4, when the types of the combination gene symbols included in the network 1 322 and the network 2 324 are all the same, the integrated network 342 is used as a method of increasing the weights of the gene combination symbols Can be generated.

For example, when one gene symbol included in the gene combination symbol is identical, the network integration unit 340 may integrate the corresponding gene combination symbol as a common node . As shown in FIG. 4, when only one gene combination symbol (328, gene symbol b) in the respective gene combination symbols of network 1 322 and network 2 324 coincides (328, gene symbol b) And connecting the nodes corresponding to the gene symbols b and c on both sides to integrate the network.

In another embodiment, when the gene symbols included in the gene combination symbol do not match (330), the network integration unit 340 does not connect the gene combination symbol with the inconsistent gene symbol as a separate node Network can be integrated. For example, as shown in FIG. 4, if at least one of the gene symbols included in the network 1 322 and the network 2 324 does not match (330) And the network combination symbol of the network 2 can be integrated unchanged. Will be described with reference to Figs. 5 and 6. Fig.

The integrated gene network constructed by the integrated network construction unit 300 has a structure in which a plurality of gene symbols are connected to each node and each of the nodes is connected to an edge, and a node corresponding to a gene node symbol having a high frequency is connected to another gene symbol node . For example, when the number of gene symbols included in the gene integrated network constructed in the integrated network construction unit 300 is large, each gene symbol may be difficult to observe. In this case, It is possible to re-scale easily. Will be described with reference to FIG.

The analysis unit 400 includes a priority determination unit 420 and a verification unit 440. For example, the analysis unit 400 determines a gene symbol prioritization related to a disease to be analyzed using the constructed integrated gene network, and analyzes the relation of genes related to the disease according to the determined priority. The analysis unit 400 determines the priority of the gene symbols corresponding to the nodes appearing on the constructed integrated gene network in consideration of specific reference values and analyzes the relation of the genes related to the disease according to the determined priority. Also, the analysis unit 400 can derive candidate genes deduced to be related to diseases using the constructed integrated gene network. Will be described with reference to FIG.

The priority determining unit 420 includes a first reference value calculating unit 422, a second reference value calculating unit 424, and a third reference value calculating unit 426. E.g. The priority determining unit 420 determines the priority of the gene symbol related to the disease to be analyzed based on the constructed integrated gene network. The manner in which the priority determining unit 420 determines the priority order may be changed depending on what kind of reference value is used and prioritized.

The first reference value calculation unit 422 calculates a first reference value related to the number of other nodes connected to one node based on the constructed integrated gene network. For example, the first reference value calculated by the first reference value calculation unit 422 may be set high when the number of other nodes connected to one node determined on the integrated gene network is large. The second reference value calculation unit 422 calculates a second reference value related to the weight of the edge connecting the one node with the other node. In the case where all the gene symbols included in the gene combination symbol in the network integration unit 340 described above match (326), the weight of the edge connecting the corresponding gene symbols can be increased as described above. The second reference value calculation unit 422 calculates a second reference value representing a weight of an edge connecting one node and the other nodes on the integrated gene network.

The third reference value calculation unit 423 calculates a third reference value related to the frequency at which the one node is exposed in a path connecting the two nodes at the shortest distance on the integrated gene network. For example, any two nodes on an integrated gene network have one shortest path connecting each node, measuring the shortest path between any two nodes that can be combined on an integrated gene network, Can be generated. The second reference value calculation unit 423 may calculate the third reference value by measuring the frequency at which each node is exposed in the generated integrated gene path map. Will be described with reference to FIG.

The priority determining unit 420 determines the priority of the gene related to the disease on the integrated gene network using the calculated first reference value, second reference value, and third reference value, and the analyzing unit 400 determines the priority Depending on the ranking, the top 20 gene symbols can be placed on the integrated gene network. The priorities determined by the priority decision unit 420 include Degree Centrality, Weighted Degree Centrality, Betweness and EigenVector.

For example, the degree of connection, which is one of the priorities used by the priority determining unit 420, reflects characteristics related to the number of other nodes connected to one node on the integrated gene network. . The analysis unit 400 can list the gene symbols existing on the integrated gene network according to the degree of connection, and it is found that the most nodes are connected to the EGFR gene symbol on the integrated gene network related to lung cancer have. As shown in FIG. 9, when the analyzer 400 lists the gene symbols existing on the integrated gene network using the degree of connection, the second largest number of nodes are connected to the BRAF gene symbol 516 . For example, Weighted Degree Centrality, which is one of the priorities used by the priority determining unit 420, is a function of the number of other nodes connected to one node on the integrated gene network and the weights of edges connected to one node And can be determined using the first reference value and the second reference value.

In addition, the mediation centency (Betweens), which is one of the priorities used by the prioritization unit 420, can be determined using the third reference value as a priority order considering the frequency with which each node is exposed in the shortest path map. The eigenvector as one of the priorities includes a number of other nodes connected to one node on an integrated gene network, a weight of an edge connecting the one node and another node, The priority can be determined using the first reference value and the second reference value. The Eigenvector can be expressed by the following equation.

는 상수이며, x_i는 i노드의 고유벡터로서 주변 노드 들의 고유벡터의 평균을 의미하고, A_ij는 i노드와 j노드를 연결하는 엣지의 가중치, x_j는 i노드에 연결된 주변 노드들의 고유벡터를 의미한다. 우선 순위 결정부(420)는 고유벡터 중심성(EigenVector)을 이용하여 통합 유전자 네트워크상에서 하나의 노드에 연결된 다른 노드의 영향력까지 고려하여 우선순위를 결정할 수 있다.Wherein I and j are integers from 1 to n,

Is constant, and, x _i is unique from i close to mean the average of the eigenvectors of the neighboring nodes as a unique vector of the node and, A _ij is the weight, x _j of the edge connecting the i node and j nodes connected to the i-node Vector. The priority determining unit 420 may determine the priority by taking into consideration the influence of other nodes connected to one node on the integrated gene network using the eigenvector.

The verification unit 440 arranges the gene symbols related to the disease according to the determined priority, and verifies the placed gene symbols with the verification data related to the disease. For example, the verification unit 440 may include a database used for searching for seed gene symbols, and data including genetic information other than the database in which the obtained seed gene symbols are extracted from text data including information on genes using input keywords The base can be used to verify at least some of the gene symbols on the integrated gene network.

Table 2 shows a list of genes detected as being associated with lung cancer in the verification database used by the verifying unit 440. GHR, KEGG, LuGend, and IGDB.NSCLC are the database for verification, Gene is the number of gene symbols derived from the gene associated with lung cancer in each verification database, and Total is the number of gene symbols derived from each verification database This is the sum of the lists.

For example, the verification unit 440 may include an OMIM database for searching for seed gene symbols that are already known to be associated with a specific disease, and text data including information about genes using the obtained seed gene symbols At least some of the gene symbols on the integrated gene network can be verified using data including information about genes in the GHR, KEGG, LuGend, and IGDB.NSCLC databases, which are databases other than the PubMed database for extraction.

EGFR-BRAF-EGF-KRAS-PIK3CA-ERBB2-PTEN-EGF-KRAS-EGFR-KRAS- (SEQ ID NO: 3), which is located at the upper right upper left of the gene symbols arranged as TP53-CASP8-NARS-PARK2-CDKN2A-APC-AKT1-IRF1-CCND1-STAT3-FAS- 502 denotes a seed gene symbol used as an input keyword, and a gene symbol 510 located in a cell shaded to the lower left and upper right of the upper left corner indicates a gene symbol verified by the verification data. Among the placed gene symbols, a gene symbol located in a non-hatched cell means a candidate gene symbol, which means a gene symbol that does not correspond to a seed gene symbol nor exists in a verification database.

For example, the verification target to be verified by the verification unit 440 may be provided as a candidate gene among the gene symbols arranged according to the priority determined by the priority decision unit 420. [ The verifying unit 440 verifies whether or not the candidate gene is a genetic symbol that does not correspond to the seed gene symbol but does not exist in the verification database, And verifying the disease-relatedness of the candidate gene based on the verified verification texts.

2 is an enlarged block diagram of the combination symbol detector 200 in the embodiment of FIG.

The combined symbol detecting unit 200 includes a preprocessing unit 220 and a symbol extracting unit 240. For example, the combination symbol detector 200 searches for text data including information on a gene using at least one gene symbol, and detects a gene combination symbol combined with the gene symbol in the retrieved text data. The combination symbol detection unit 200 detects at least two or more gene symbols as a combination symbol in one sentence. The specific method of detecting the combination symbol detection unit 200 by the combination symbol detection unit 200 is the same as described above, and thus is omitted.

3 is an enlarged block diagram of the integrated network construction unit 300 in the embodiment of FIG.

The integrated network construction unit 300 includes a local network construction unit 320 and a network integration unit 340. For example, the integrated network construction unit 300 constructs an integrated gene network that links the gene symbols to the nodes and connects the nodes to the edges based on the detected gene combination symbols. The concrete method of constructing the integrated gene network is as described above, so it is omitted.

4 is an exemplary diagram illustrating a process of establishing an integrated network in the embodiment of FIG.

The method for constructing the local gene network constructed using the gene combination symbols detected from the text data retrieved for each seed gene can be roughly divided into three methods. The network integration method used by the network integration unit 340 to construct the integrated gene network by integrating the local gene network is a network integration method that uses the gene combination symbols detected from the text data retrieved for each seed gene, It can also be used to build gene networks. The method of constructing the unified network 342 by combining the network 1 322 and the network 2 324 is the same as described above, and thus will not be described here.

5 is an exemplary diagram illustrating an integrated gene network constructed by the integrated network building unit in the embodiment of FIG.

The integrated gene network constructed by the integrated network construction unit 300 includes the nodes that make the gene symbols as nodes and connect the nodes with specific weights. The gene relational analysis apparatus 10 can acquire gene symbols of the highest rank according to the priority order using the constructed integrated gene network and analyze the relation of gene symbols related to the disease to be analyzed. In one embodiment, the integrated gene network constructed in the present invention includes 1339 nodes and 4092 edges, and in ranking according to the degree of connection among the priorities, EGFR, BRAF, and KRAS are ranked, May be represented by nodes larger than the symbols.

Figure 6 is an exemplary diagram illustrating a scaled integrated gene network.

The integrated network construction unit 300 scales the weights of edges on the integrated gene network because the integrated gene network is too complex and difficult to acquire candidate genes related to diseases when the integrated gene network is constructed with too many gene symbols Re-scale) to rebuild the integrated gene network. Referring to FIG. 6, the scaled integrated gene network includes 60 nodes and 100 edges.

7 is an enlarged block diagram of the analysis unit in the embodiment of FIG.

The analysis unit 400 includes a priority determination unit 420 and a verification unit 440. For example, the analysis unit 400 determines a gene symbol prioritization related to a disease to be analyzed using the constructed integrated gene network, and analyzes the relation of genes related to the disease according to the determined priority. The analyzing unit 400 analyzes the user-related relationship related to the disease using the integrated gene network as described above, so that it is omitted.

8 is an enlarged block diagram of the priority determining unit in the embodiment of FIG.

The priority determining unit 420 includes a first reference value calculating unit 422, a second reference value calculating unit 424, and a third reference value calculating unit 426. E.g. The priority determining unit 420 determines the priority of the gene symbol related to the disease to be analyzed based on the constructed integrated gene network. A specific method of determining the priority order using the calculated first, second, and third reference values by the priority determining unit 420 is the same as described above, and thus will not be described.

FIG. 9 is an illustration showing disease-related genes listed according to gene symbol priorities. FIG. Will be described with reference to FIG.

The verification unit 440 arranges the gene symbols related to the disease according to the determined priority, and verifies the placed gene symbols with the verification data related to the disease.

EGFR-BRAF-EGF-KRAS-PIK3CA-ERBB2-PTEN-EGFR-EGFR-EGFR-EGFR is constructed when the gene symbols on the integrated gene network associated with lung cancer constructed in the integrated network construction unit 300 are arranged according to the degree of connection. TP53-CASP8-NARS-PARK2-CDKN2A-APC-AKT1-IRF1-CCND1-STAT3-FAS-MET-MGMT. As shown in FIG. 9, the seed gene symbol 502 is a gene symbol used as an input keyword to construct an integrated gene network related to lung cancer, and is a gene symbol located in a hatched cell of an upper left hemisphere, Symbol 504 corresponds to a gene associated with lung cancer present in the verification database. This was not present in a database (e.g., PubMed) for constructing a gene symbol list related to lung cancer shown in Fig. 9, but was associated with lung cancer present in the verification database (GHR, KEGG, LuGend and IGDB.NSCLC) Because it is a gene symbol, it is a reliable gene symbol in terms of disease relatedness. However, the gene symbol located in the non-hatched cell is a candidate gene symbol 506, which means a gene symbol that does not correspond to a seed gene symbol and does not exist in a verification data base.

The chart shown in FIG. 9 is analyzed according to the types of priorities determined by the priority determining unit 420 as follows.

In Table 3, the number of seed genes is the number of seed genes used as input keywords to construct an integrated gene network related to lung cancer, the number of inferred genes is not the verified gene, the number of candidate genes not corresponding to the seed gene, The percentage of inferred genes is the ratio of candidate genes out of the 20 ranked genes, the number of confirmed inferred genes is the number of verified gene symbols, the percentage of confirmed genes and the number of candidate genes and the percentage of confirmed genes Means the extent of the seed gene and the proven gene symbol among the 20 ranked genes.

FIG. 10 is an exemplary diagram showing a text detected by a verification unit in the embodiment of FIG. 7 to verify a candidate gene symbol. FIG.

The verifying unit 440 may include a verification text detecting unit for detecting a sentence including a content related to a disease related to the candidate gene, an experimental result, and the like, for verifying the disease relevance of the candidate gene symbol. Based on the verified verification texts The disease-relatedness of candidate genes can be verified. For example, in order to construct an integrated gene network related to lung cancer, the verification text detection unit can detect a sentence including a disease and an experimental result in a PubMed database that retrieves text data using a seed gene as an input keyword .

For example, the Verification Text Detector is part of a study that suggests that the APC gene symbol may be used as a potential diagnostic marker for lung cancer, "Hypermethylation of the APC gene promoter in a potential diagnostic marker for lung cancer diagnosis" And can be used as a basis for disease-relatedness of candidate genes.

11 is a flowchart of a method of analyzing a gene relationship according to an embodiment of the present invention.

In S100, the seed gene symbol acquisition unit 100 acquires a seed gene symbol already known as a gene related to the disease to be analyzed. The specific method of acquiring the gene symbol already known to be related to the disease by the seed gene symbol acquiring unit 100 is as described above and is omitted.

In S200, the combination symbol detection unit 200 searches text data including information on a gene using at least one gene symbol, and detects a gene combination symbol in which the gene symbol is combined in the retrieved text data. For example, the combining symbol detector 200 may search for text data including information on a gene using a seed gene symbol, and may detect a gene combination symbol in the retrieved text data. The combination symbol detection unit 200 can detect the gene combination symbol using different databases, as described above, and thus will be omitted.

In step S300, the integrated network construction unit 300 constructs a local gene network that uses the gene symbol as a node and the relation of the gene as an edge based on the detected gene combination symbol, and integrates the constructed local gene network To build an integrated gene network. The concrete method of integrating the network by the integrated network building unit 300 is the same as described above, and thus will not be described.

In step S400, the analysis unit 400 determines a gene symbol prioritization related to a disease to be analyzed using the constructed gene network, and analyzes the relationship of genes related to the disease according to the determined priority. The method for determining the priority in the analysis unit 400 and the specific method for detecting at least one gene symbol arranged according to the priority determined using the verification database are the same as described above and will not be described here.

12 is an enlarged flow chart of the step of detecting in the embodiment of Fig.

In step S220, the preprocessing unit 220 preprocesses the retrieved text data and extracts at least a part of the retrieved text data. For example, the preprocessing unit 220 can extract only the abstract text using the keyword MeSH Term in the retrieved text data as described above.

In step S240, the symbol extracting unit 240 extracts the genetic symbols in the sentence by classifying the searched text data on a sentence-by-sentence basis. The symbol extracting unit extracts the gene symbols in the separated sentences, and extracts the gene symbols using the preset gene symbol list.

13 is an enlarged flow chart of a step of constructing an integrated gene network in the embodiment of FIG.

In step S320, the local network construction unit 320 constructs a local gene network generated on the basis of the seed gene symbol by using the gene symbol included in the gene combination symbol detected using the seed gene symbol as each node. The local network construction unit 320 can construct a separate local gene network for each seed gene symbol, and a specific method for constructing the local network is as described above.

In step S340, the network integrating unit 340 integrates the constructed local gene network according to a predetermined method in consideration of the type of the gene symbol included in the gene combination symbol or the frequency with which the combined symbol is detected. For example, when the network integration unit 340 integrates according to a predetermined method, when all of the gene symbols included in the gene combination symbol coincide with each other, each of the gene symbols of the combination symbol is used as a node, And if the one gene symbol included in the gene combination symbol coincides with the corresponding one of the gene combination symbols, the combining of the gene combination symbols using the corresponding gene symbol as a common node may be performed. In another embodiment, the network integrating unit 340 integrates according to a predetermined method in the case where the gene symbols included in the gene combination symbol do not coincide with each other, using the inconsistent gene symbol as a separate node, In a non-connected manner.

14 is an enlarged flow chart of the step of analyzing in the embodiment of Fig.

In step S420, the priority determining unit 420 determines the number of other nodes connected to one node based on the constructed integrated gene network, the weight of the edge connecting the one node to the other node, The priorities of the genes related to the disease to be analyzed are determined in consideration of the frequency of exposure of the one node in the path connecting the two nodes at the shortest distance. For example, the priority determining unit 420 may set the number of other nodes connected to one node as a first reference value, the weight of an edge connecting one node to the other node as a second reference value, The degree of exposure of the one node is calculated as a third reference value in a path connecting the two nodes of the first node to the shortest distance, and the degree of connection, the degree of connection, and the like are calculated using the calculated first, second, Weighted Degree Centrality, Betweness, and EigenVector can be determined.

In S440, the verifying unit 440 arranges the gene symbol related to the disease according to the determined priority, and verifies the placed gene symbol with the verification data related to the disease. The verifying unit 440 verifies the placed gene symbol with the verification data related to the disease as described above, and thus will be omitted.

15 is a reference diagram showing a flow of a method of analyzing a gene relationship according to another embodiment of the present invention.

In step S100, the seed gene symbol acquisition unit 100 may acquire a seed gene symbol, which is already known as a gene related to the disease to be analyzed, in the OMIM database 2000. In S220, the combination symbol detector 200 searches for text data including genetic information using the obtained seed gene symbol. In S240, the combination symbol detection unit detects a pair of gene symbols as a gene combination symbol if there are two gene symbols in one sentence in the retrieved text data.

In step S320, the local network construction unit 320 constructs a local gene network using each of the genes included in the gene combination symbol detected using the seed gene symbol. In step S340, the network integrating unit 340 integrates the constructed local gene network according to a predetermined method in consideration of the type of the gene symbol included in the gene combination symbol or the frequency with which the combined symbol is detected.

16 is a reference diagram showing the gene analysis performance of the present invention as compared with other gene analysis methods.

As a result of analyzing the gene symbols related to lung cancer using the gene analysis apparatus 10 according to the present invention, it was found that more lung cancer-related genes (seed gene symbols + proven genes) than the other methods including SSL Method and RWRHN Gene symbol) can be derived. The gene analysis apparatus (10) of the present invention analyzed 9 genes associated with lung cancer, but the SSL method can analyze 8 lung cancer genes and the RWRHN can analyze 4 lung cancer related genes. The gene-relationship analyzing apparatus 10 of the present invention can be used not only for analyzing the relationship between genes related to a specific disease, but also for searching for various biological data or drugs related to a specific disease, and for analyzing the relationship between drugs and genes.

All or part of the method of an embodiment of the present invention described above can be implemented in the form of a computer-executable recording medium (or a computer program product) such as a program module executed by a computer. Here, the computer-readable medium may include computer storage media (e.g., memory, hard disk, magnetic / optical media or solid-state drives). Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media.

Also, all or part of the method according to an embodiment of the present invention may include instructions executable by a computer, the computer program comprising programmable machine instructions to be processed by a processor, Language, an object-oriented programming language, an assembly language, or a machine language.

Means or module in the present specification may mean hardware capable of performing the functions and operations according to the respective names described herein and may be implemented by computer program code , Or may refer to an electronic recording medium, e.g., a processor or a microprocessor, having computer program code embodied thereon to perform particular functions and operations. In other words, a means or module may mean a functional and / or structural combination of hardware for carrying out the technical idea of the present invention and / or software for driving the hardware.

Thus, a method according to an embodiment of the present invention may be implemented by a computer program as described above being executed by a computing device. The computing device may include a processor, a memory, a storage device, a high-speed interface connected to the memory and a high-speed expansion port, and a low-speed interface connected to the low-speed bus and the storage device. Each of these components is connected to each other using a variety of buses and can be mounted on a common motherboard or mounted in any other suitable manner.

It will be apparent to those skilled in the art that various modifications, changes, and substitutions are possible, without departing from the essential characteristics and spirit of the invention as disclosed in the accompanying claims. will be. Therefore, the embodiments disclosed in the present invention and the accompanying drawings are intended to illustrate and not to limit the technical spirit of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments and the accompanying drawings . The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

Claims (20)

A seed gene symbol acquisition unit acquiring a seed gene symbol already known as a gene symbol related to a disease to be analyzed;
A combination symbol detector for searching text data including information about the seed gene using the obtained seed gene symbol and for detecting a gene combination symbol combined with the gene symbol in the retrieved text data;
An integrated network construction unit for constructing an integrated gene network by integrating the constructed local gene network based on the detected gene combination symbol as a node and representing the relationship of the gene as an edge, ; And
An analysis unit for determining gene symbol priorities related to a disease to be analyzed using the integrated gene network and analyzing a relationship between genes related to the disease according to the determined priority; Lt; / RTI >
Wherein the analyzer comprises: a priority determining unit for determining a priority of a gene symbol related to a disease to be analyzed based on the constructed integrated gene network; Further comprising:
Wherein the priority determination unit comprises: a first reference value calculation unit for calculating a first reference value related to the number of other nodes connected to one node based on the constructed integrated gene network; A second reference value calculation unit for calculating a second reference value related to a weight of an edge connecting the one node with the other node; And a third reference value calculation unit for calculating a third reference value related to a frequency at which the one node is exposed in a path connecting the arbitrary two nodes at the shortest distance on the integrated gene network. Further comprising:
Wherein the gene symbol priority is determined using at least one of the first reference value, the second reference value, and the third reference value. delete The apparatus of claim 1, wherein the combination symbol detector
A symbol extracting unit for dividing the searched text data by a sentence unit, analyzing the separated sentences, and extracting a gene symbol from the sentence; Further comprising:
Wherein the extracted gene symbols are detected as combination symbols when at least two or more gene symbols are extracted in the sentence. [2] The apparatus of claim 1,
A local network construction unit for constructing a local gene network generated by the seed gene symbol by using the gene symbol included in the gene combination symbol detected using the seed gene symbol as each node; Further comprising:
And constructing the integrated gene network using the constructed local gene network. 5. The system according to claim 4, wherein the integrated network building unit
A network integrator for integrating the constructed local gene network according to a predetermined method in consideration of the type of gene symbol included in the gene combination symbol or the frequency at which the combined gene symbol is detected; Lt; / RTI >
And the integrated gene network is constructed using the integrated local gene network. 6. The method of claim 5, wherein incorporating according to the predetermined method
And integrating the gene symbols of the combination symbol as a node and increasing a weight of an edge connecting the node when the gene symbols included in the gene combination symbol coincide with each other. 6. The method of claim 5, wherein incorporating according to the predetermined method
And combining the gene combination symbols by using the matching gene symbol as a common node when one gene symbol included in the gene combination symbol coincides with each other. 6. The method of claim 5, wherein incorporating according to the predetermined method
Wherein when the gene symbols included in the gene combination symbol do not coincide with each other, the non-matching gene symbol is integrated as a separate node and the gene combination symbols are not connected. The apparatus of claim 1, wherein the analyzing unit
A verifying unit for arranging a gene symbol related to the disease according to the determined gene symbol priority and verifying the placed gene symbol with verification data related to the disease to check the relation with the disease; Further comprising:
And analyzing a relationship between genes related to the disease using the verified gene symbol. delete delete Acquiring a seed gene symbol already known as a gene symbol associated with the disease to be analyzed;
Searching for text data including information on the seed gene using the obtained seed gene symbol, and detecting a gene combination symbol in which the gene symbol is combined in the retrieved text data;
Constructing a local gene network that represents the gene symbol as a node and the relation of the gene as an edge based on the detected gene combination symbol, and constructing an integrated gene network by integrating the constructed local gene network; And
Determining a gene symbol prioritization related to a disease to be analyzed using the constructed integrated gene network and analyzing a relationship between genes related to the disease according to the determined priority; Lt; / RTI >
Wherein the analyzing comprises: determining a gene symbol prioritization related to a disease to be analyzed based on the constructed integrated gene network; Further comprising:
Wherein the determining comprises determining a first reference value for the number of other nodes connected to one node based on the constructed integrated gene network, a second reference value for a weight of an edge connecting the one node to the other node, Wherein the priority of the gene symbol is determined by using at least one of a third reference value related to the frequency with which one node is exposed in a path connecting an arbitrary two nodes at the shortest distance on the integrated gene network Way. delete 13. The method of claim 12, wherein detecting
Dividing the searched text data into sentences, analyzing the separated sentences, and extracting gene symbols from the sentences; Further comprising:
Wherein when the at least two gene symbols are extracted from the sentence, the extracted gene symbols are detected as a combination symbol. 13. The method of claim 12, wherein the step of constructing the integrated gene network comprises
Constructing a local gene network generated by the seed gene symbol by using the gene symbol included in the gene combination symbol detected using the seed gene symbol as each node; Further comprising:
And constructing the integrated gene network using the constructed local gene network. 16. The method of claim 15, wherein constructing the integrated gene network comprises:
Integrating the constructed local gene network according to a predetermined method in consideration of the type of the gene symbol included in the gene combination symbol or the frequency at which the combined gene symbol is detected; Further comprising:
Wherein the integrated gene network is constructed using the integrated local gene network. 17. The method of claim 16, wherein incorporating according to the predetermined method
And when each of the gene symbols included in the gene combination symbol coincides with each other, each of the gene symbols of the combination symbol is used as a node, the weight of the edge connecting the node is increased,
And combining the gene combination symbols by using the matching gene symbol as a common node when one gene symbol included in the gene combination symbol coincides with each other. 17. The method of claim 16, wherein incorporating according to the predetermined method
If the gene symbols included in the gene combination symbol do not match, integrating the non-matching gene symbols as separate nodes and not linking the gene combination symbols. 13. The method of claim 12, wherein analyzing comprises:
Disposing a gene symbol related to the disease according to the determined gene symbol priority, and verifying the placed gene symbol with verification data related to the disease; Further comprising:
And analyzing a relationship between genes related to the disease using the verified gene symbol. A program stored in a computer-readable recording medium for realizing the method for analyzing a gene relationship according to any one of claims 12 to 14 through being executed by a processor.

Images