JP2004152035A - Research gene product data classification system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a research gene product data classification system for efficiently carrying out research target retrieval by classifying respective data such as gene or compound/medicine and disease according to working, effects, and symptoms or the like, and for removing noise data, a data group whose number of times of document excerpt is large, and data which do not belong to any classification in an early stage. <P>SOLUTION: Gene information, compound/medicine information, disease information and document information associated with the respective information are stored so as to be associated with each other in a data base, and when any of the gene, compound/medicine and disease from the data base is used as a keyword, the other information is scored in the order of high relevance, and ranking is prepared. Then, second ranking data are prepared from the first ranking data, and the second ranking information is arranged for each of the first ranking information. The similarity of the first ranking data is calculated by using the second ranking data, and the first ranking data are grouped according to the similarity. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a system for efficiently detecting genes and compounds that can be useful candidates for studying genes and their products by examining the relationship between genes, compounds and diseases using a computer.
[0002]
[Prior art]
In conducting research on genes and their products, it is effective to find genes and their products that seem to be the most important. To do so, we must select gene candidates that we think are important. Genes considered to be important are genes targeted for drug discovery for a specific disease. Therefore, candidate low molecular compounds are identified while considering the relationship between genes and diseases, and finally screening tests are performed. In this work, when performing the work of identifying the association between a disease and a gene and the work of creating a compound candidate for a selected gene group, since there is no database that directly links the gene and the compound, These genes and compound candidates are created with reference to the work and literature. In fact, he had used his experience to find out its importance, referring to published papers, public gene databases and product databases.
[0003]
In the past, it was possible to search one-dimensionally for information on genes and information on compounds using literature and gene databases, but the relationship between them, that is, the relationship between compounds involved in a specific gene and The only way to find out was to rely on intuition and experience.
[0004]
For example, when researching information on genes and their products while referring to the research results of others, manually search the product, gene database, and literature database for gene names, compound names, and search the relevant gene database, He printed the summary and looked at it to find compounds that were empirically relevant and repeated the experiment.
[0005]
In this case, it was impossible to objectively study, it was impossible to read a large amount of literature, and judgment was often abandoned, and unnecessary experiments were performed, which wasted research expenses. . Furthermore, gene databases and literature databases are frequently updated and cannot be implemented manually.
In order to know which genes and compounds are often studied for a specific disease, it is necessary to examine the strength of the association between these three factors. I couldn't go and see.
[0006]
For that reason, it is impossible to find genes and compounds that are likely to become research subjects, important compounds and genes, those that have already completed the research stage, and genes and compounds that are useless even after research. there were.
In addition, since the amount of research data on genes has increased rapidly in recent years, the amount of data that can be retrieved is extremely large, and it requires experience and a lot of time to empirically know these relationships. Was.
[0007]
In order to solve these problems, the database of genes and literature, which is updated daily, finds the relevance of diseases and compounds related to genes using a computer, and examines which genes and compounds in relation to specific diseases in the future. In order to provide researchers with information to help guide what genes and compounds are useless even if they do research, the combination of compounds / drugs and genes, or genes and diseases, Using the number of documents discussed in a specific combination, the score of the number of documents is used to rank the data according to the score (number of documents) that is strongly related to the specific data. The method of measuring the scale of is being taken.
[0008]
[Patent Document 1]
Japanese Patent Application No. 2001-235345
[0009]
[Problems to be solved by the invention]
However, even if the above-described improvement method is adopted, it is possible to extract data relating to specific data, but it is not possible to find relevance between the extracted data because attention is not paid to the tendency of the extracted data.
Therefore, in order to determine a research target from simply ranked data, after ranking, the characteristics of each piece of data are investigated, and what kind of experimental process or data to be used in the research field is determined. Knowing surveys need to be done and time may be spent on target determination.
[0010]
The present invention has been conceived in order to solve the above-described problems, and classifies each data of a gene, a compound, a drug, and a disease by action, efficacy, symptom, and the like, thereby improving the efficiency of research target search and reducing noise. An object of the present invention is to provide a research gene product data classification system capable of removing data, a data group with a large number of references, and data not belonging to any classification at an early stage.
[0011]
[Means for Solving the Invention]
In order to achieve the above object, the research gene product data classification system of the present invention comprises a means for storing gene information, compound / drug information, disease information and literature information relating to each of these information in association with each other, / A system that assigns a score to other information in the order of strong relevance using either the drug or the disease as a keyword, and creates a ranking. The first ranking information arranged and arranged, and the information having a large number of documents shared with the document information relating to each information of the first ranking information are ranked and arranged for each of the first ranking information. It is characterized by creating second ranking information.
[0012]
Further, the present invention provides a means for storing gene information, compound / drug information, and disease information in association with literature information relating to each of these information, and a method for associating a gene, compound / drug, or disease with a keyword. Means for assigning a score to other information in descending order to create a ranking, wherein the first ranking information and the first ranking information are arranged by ranking the document information related to the keyword and the information having the largest number of documents to be shared. And ranking the document information related to each piece of information of the first ranking information and the information having a large number of documents to be shared to create second ranking information arranged for each of the first ranking information, The second ranking information is compared with information of a predetermined rank or higher in the ranking information, and a similarity measure is measured based on the number of matching information.
[0013]
Further, according to the present invention, a means for associating and storing gene information, compound / drug information, and disease information with literature information relating to each of these information, and a related means using any one of the gene, compound / drug, and disease as a keyword Means for assigning a score to other information in descending order to create a ranking, wherein the first ranking information and the first ranking information are arranged by ranking the document information related to the keyword and the information having the largest number of documents to be shared. And ranking the document information related to each piece of information of the first ranking information and the information having a large number of documents to be shared to create second ranking information arranged for each of the first ranking information, By calculating the distance between the pieces of ranking information using the score of the second ranking information, a measure of the degree of similarity between the pieces of information is measured.
[0014]
With the system equipped with each of the above means, it becomes possible to classify each data of genes, compounds / drugs, diseases by action, efficacy, symptoms, etc., and to make research target search more efficient, noise data, Many data groups and data that does not belong to any classification can be removed at an early stage.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 shows an outline of an embodiment of the present invention. The outline of the present invention will be described below with reference to FIG. 1, and the detailed configuration will be described with reference to FIG.
First, as a premise, the database used in the detection system according to the present invention is constructed by, for example, the following method.
[0016]
(1) Construction of a basic database
(A) Gene database structure
The gene database stores genetic information for all organisms having genes including humans. At the time of storage, a gene ID is assigned to each gene. In addition to the gene ID, data items to be stored include a gene name and a gene symbol, a symbol alias (synonym), and a synonym using Mesh or the like as essential items, and may include a LOCATION item.
(B) Composition of compound / drug database
The compound / drug database stores compound and drug information. At the time of storage, a compound / drug ID is assigned to each compound / drug. The data items to be stored in addition to the compound / drug ID are mandatory items such as compound names or drug names, molecular weights, molecular weights, CAS numbers, and meshes.
(C) Structure of disease database
The disease database stores disease information. At the time of storage, a disease ID is assigned to each disease. The data items to be stored other than the disease ID are required to be synonyms using the disease name and Mesh.
(D) Document (article) database structure
The literature (article) information database extracted from the literature is a database in which document information is stored for each data item. The data items to be stored are required to be a document (article) ID, an article name, an author name, an article journal name, a publication date, an abstract, a chemical list, a CAS number, and a Mesh assigned to each article.
[0017]
(2) Construction of a database that associates genes / compounds / drugs / disease with literature
(A) Structure of Gene / Document Correlation Database DB1
Determine whether keywords related to the genes included in the gene database are included in the article information in the document information database, and if the relationship between the gene and the article is determined, store the data that associates the gene database with the document information database. I do. Data such as the gene ID, the article ID, the number of appearances of the gene name in the abstract, the presence or absence of the gene name in the article title, whether or not the synonym such as Mesh matches the gene name, and the publication date of the article are stored.
(B) Configuration of compound / drug / document association database DB2
It is determined whether a keyword related to a compound or drug included in the compound / drug database is included in the article information in the literature information database, and if the relationship between the compound / drug and the article is determined, the compound / drug database and the document information are determined. Stores data associated with a database.
Data such as the compound ID, the article ID, the number of occurrences of the compound in the abstract, the presence or absence of the compound name in the article title, whether or not the chemical list matches the compound, and the publication date of the article are stored.
(C) Structure of disease / document association database DB3
If the keyword related to the disease included in the disease database is included in the article information of the document information database and the relationship between the disease and the article is determined, the data in which the disease database and the document information database are associated is stored. .
Data such as the disease ID, the article ID, the number of appearances of the disease name in the abstract, the presence or absence of the disease name in the article title, whether or not the synonym such as Mesh matches the disease name, and the date of publication of the article are stored.
[0018]
In some cases, the database contains unfinished article data. In the case of unfinished data, information indicating that the article is unfinished is entered so that the document (article) data can be updated at the next update. Keep it. The article ID from which information could not be obtained, such as the article data being deleted, is deleted. The article ID for which information could not be obtained is deleted from each of the above databases DB1, DB2, and DB3.
On the other hand, the publication date data of the dissertation is acquired from each of the databases DB1, DB2, and DB3 from the dissertation information and updated. By utilizing this, a chart of the number of publications of documents (papers) is displayed at the time of screen output.
Now, each database file of the gene / document association database DB1, the compound / drug / document association database DB2, and the disease / document association database DB3 is schematically shown in FIG. 2 (A), FIG. 2 (B), and FIG. It is configured as in C).
[0019]
In DB1, if a gene ID or a symbol, an alias, or a synonym is described in a document (paper) for the gene ID assigned to the gene name, the ID assigned to the document is It is stored in a state enumerated for the gene ID. FIG. 2A shows that there are T14, T25, T27, T90, and other documents in the documents in which information on the gene ID G1 is described. Documents describing information on G2 include T2, T8, T25, T30, and the like, and it can be seen that the document of T25 describes both information on G1 and G2. The additional information column stores data such as the number of appearances of gene names in the abstract of the document, the presence or absence of the gene name in the title of the article, whether or not the gene name matches a synonym such as Mesh, and the date of publication of the article. Is done.
[0020]
Similarly, FIG. 2B shows that, for example, documents describing information on C1 of the compound / drug ID include T1, T4, T14, T100, and other documents. A comparison between FIG. 2B and FIG. 2A shows that the document of T14 describes the information of G1 and C1, and the document of T90 describes the information of G1 and C2. In the additional information column, data such as the number of occurrences of the compound / drug in the abstract of the document, the presence / absence of the compound / drug in the title of the paper, whether the chemical list matches the compound / drug, the date of publication of the paper, etc. Is stored.
[0021]
FIG. 2C shows that, for example, T35, T46, T58, T77, and the like exist in the literature in which information on the disease ID D1 is described. Comparing FIG. 2 (C), FIG. 2 (A), and FIG. 2 (B), it can be seen that the document of T14 describes information of G1, C1, and D2. The additional information column stores data such as the number of occurrences of the disease name in the abstract of the document, the presence or absence of the disease name in the title of the article, whether or not the synonym such as Mesh matches the name of the disease, and the date of publication of the article. Is done.
Next, ranking (ranking) will be described as the first step of measuring the similarity.
In this ranking method, ranking is performed in descending order of shared article IDs. For example, referring to FIG. 2, when ranking a disease using G1 of the gene ID as a keyword, the literature information on G1 and the disease IDs D1, D2,. ) Are sequentially compared, and the number of common documents is accumulated, so that the diseases are arranged in descending order of the total number of cases.
[0022]
When comparing the document information of G1 and D1, there is no common document from the part displayed in FIG. 2, but when comparing the document information of G1 and D2, the part displayed in FIG. T14 and T90 are obtained as common document information. In this way, the data stored in the database is successively referred to, and the disease IDs having the largest number of common papers are arranged in order and stored.
The case of ranking diseases using specific genes as keywords has been described.However, when ranking compounds and drugs using specific genes as keywords, ranking genes or diseases using specific compounds and drugs as keywords Is performed, the same processing is performed when ranking genes or compounds / drugs using a specific disease as a keyword.
Next, the data classification system will be described with reference to FIG. 1. In S1, data serving as a keyword is selected from any of the categories of gene, compound / drug, and disease, and a category to be ranked is selected.
[0023]
Using the data of the gene / document correspondence database DB1, the compound / drug / document correspondence database DB2, and the disease / document correspondence database DB3, the ranking was selected in S2 for the specific keyword by the method described above, and the selected keyword was selected. The first ranking data bn (n = 1, 2, 3,... N) for the category is obtained (S3).
In S4, specific data bi is further selected from the ranking data bn as a keyword, and a category to be ranked is selected. Using each data of the gene / document correspondence database DB1, the compound / drug / document correspondence database DB2, and the disease / document correspondence database DB3, the ranking is executed for the selected keyword data bi by the method described above, and the selected keyword data bi is selected. The second ranking data cim (m = 1, 2, 3,... M) is obtained for the category (S6). Then, the above ranking process is executed until ranking data for each data from b1 to bn is obtained in S7, and when the ranking process is completed, the similarity between each data of bn is calculated in S8, and the similarity is calculated by the similarity. Classify the bn data. A registration is made for each classified data, and p groups (categories) are created. Thus, the data classification process ends.
[0024]
FIG. 3 shows second ranking data cnm (n = 1, 2, 3,...) Created for each of the above-described first ranking data bn (n = 1, 2, 3,... N). .. N, m = 1, 2, 3,... M) are arranged in descending order of ranking. Here, the method of measuring the similarity between each data of bn is as follows.
For example, consider only the top 100 data cnm (m ≦ 100). The threshold of how many data items to select is selected as appropriate.
By comparing the ranking data cnm (m ≦ 100) of each of the bn data (a group of these data is referred to as a b group), the similarity is higher as the data of the b group having a larger number of data matches is higher. For example, when measuring the similarity between b1 and b2, the ranking data for b1 from c11 to c1m (m ≦ 100) and the ranking data for b2 from c21 to c2m (m ≦ 100) are compared and the same. Calculate how many data you have. For example, if c11 and c23 are the same data and c14 and c21 are the same data, two cases are counted at this time. In the same way, check whether there is any other identical data. Finally, how many pieces of matching data are found is calculated. Similarly, the similarity between b1 and b3 is calculated, and the same operation is repeated until bn. Next, the similarity is calculated by comparing b2 with b3 to bn. The process ends when the similarity between bn-1 and bn is calculated, that is, when the similarity is calculated for all combinations of data belonging to the group b.
[0025]
Next, a method of measuring the second similarity will be described.
Assuming that data of group b from b1 to bn has m-dimensional coordinates, b1 has coordinates (c11... c1m), and bn has coordinates (cn1. Cnm)). Therefore, the distance between each point (each data) of the b group is calculated by the following formula.
distance between bi and bj = Σ (cim−cjm)²= (Ci1-cj1)²+ (Ci2-cj2)²... + (cim-cjm)²
Or
(Σ (cim-cjm)²)^1/2= ((Ci1-cj1)²+ · · + (Cim-cjm)²)^1/2Use any of
As described above, the distance between the data is calculated, and the closer (shorter) the distance, the higher the similarity. However, since cnm indicates an ID number of a gene, a compound / drug, or a disease, it cannot be used as a coordinate for calculating a distance as it is. Assign.
[0026]
The score corresponds to Freq or [%] shown in FIG. The example of FIG. 4 shows a ranking result in a case where a compound (chemical) called HALOPERIDOL is selected as a keyword, and gene (gene) is selected in the item of select next display. Here, Count indicates the number of documents in which the gene indicated in the column of gene name is described, and Freq indicates the number of documents in which the gene and the compound HALOPERIDOL are described. [%] Means a numerical value represented by the following equation.
[%] = 100 × (Freq) / (Count)
In terms of the relationship between the first ranking data and the second ranking data, Freq describes information bi (i = 1, 2,... N) and cim (m = 1, 2, 3). ,... M) indicates the number of documents in which the information is also described. "Count" indicates the number of documents in which information "cim" is described. [%] Has the same meaning as in the above formula.
As described above, when the similarity is calculated by the first method or the second method, those having similar similarities are classified into several groups as one group.
[0027]
An example of classification into groups by similarity will be described with reference to FIGS.
When used by the user, enter a gene (symbol name is also possible), a compound / drug (CAS number is also acceptable), or any part of a disease or a part thereof on the keyword input screen as shown in FIG. Category, gene (gene), compound / drug (chemical), or disease (disease). In the item of select nextdisplay, a selection is made as to which of the categories, gene, chemical, and disease, a ranking list is displayed.
[0028]
After making the above selections, when the GO button is clicked, as described above, the database of the category of the keyword input above (any of DB1, DB2, DB3) and the database of the category for which the ranking list is to be displayed are used. Next, the number of common documents is detected and displayed in descending order of the number of common documents.
The example of FIG. 4 shows a ranking result in a case where a compound (chemical) called HALOPERIDOL is selected as a keyword, and gene (gene) is selected in the item of select next display. The displayed genes correspond to the first ranking data of the present invention. The display is from the first place to the 15th place, but in fact, all the genes stored in the database DB1 are ranked. Although numbers are arranged beside the gene name, as described above, Count indicates the number of documents in which the gene is described, Freq describes the gene, and also describes a compound called HALOPERIDOL Indicates the number of documents.
[%] Means a numerical value represented by the following equation.
[%] = 100 × (Freq) / (Count)
Further, the item of Time chart indicates a temporal change in the number of publications of documents in which information of a compound named HALOPERIDOL and information of the gene are both described. The publication date of the document is obtained, and the total number of publications in the publication year is displayed in a graph.
[0029]
Among the genes ranked in FIG. 4, the second [DRD1]: dopamine receptor D1 is set as a keyword, gene is selected as a keyword category, and disease (disease) is selected in the item of select next display, and an example of ranking results Is displayed in FIG. The disease ranking corresponds to the second ranking data of the present invention. Here also, the meaning of the numbers is the same as described above, Count indicates the number of documents in which the disease is described, and Freq indicates the number of documents in which the gene DRD1 is also described in addition to the disease. [%] Is as follows as described above.
[%] = 100 × (Freq) / (Count)
In this way, the diseases are ranked for all of the genes ranked in FIG. Then, the similarity is calculated by the above-described method, and grouping (category) is performed according to the similarity.
[0030]
FIG. 6 shows an example of the result of classification into groups. Circles, crosses, squares, and triangles are attached to the numbers representing the rankings, and those with the same mark are genes belonging to the same group. In the example of FIG. 6, classification is performed by adding a circle or the like to the number representing the ranking, but actually, the same color or pattern is assigned to the column or character of the gene belonging to the same group. be able to.
This is plotted on a two-dimensional plane and visualized in FIG.
FIG. 8 shows the size of the range of the plot data of FIG. For example, if the distribution state in FIG. 8 is an example of a gene, the interferon gene (a factor indicating an antiviral effect) is located near the median value on the horizontal axis if the data classified into the predetermined group has characteristics. It can be seen that there are many pseudo genes in the lower part of the vertical axis.
The gene data group affected by HALOPERIDOL of a specific drug is classified according to the onset disease.
Also, if the classified data is a drug, it can be seen that the toxicity is higher as going to the right side of the horizontal axis, and the antihypertensive effect is higher as going up the vertical axis. Also, if the classified data is a disease, it can be seen that the frequency of concurrent occurrence with cancer increases toward the right side of the horizontal axis, and that the tendency tends to occur in women toward the upper side of the vertical axis.
In addition, they are classified according to various effects, efficacy, symptoms, and the like.
[0031]
【The invention's effect】
According to the research gene product data classification system of the present invention, specific data can be classified by indication and function, and it is possible to classify each data of genes, compounds, drugs, and diseases by action, efficacy, symptoms, and the like. Become. In addition to improving the efficiency of research target search, noise data, data groups with a high number of citations, and data that does not belong to any classification can be removed at an early stage.
[Brief description of the drawings]
FIG. 1 is a flowchart showing a process of a data classification process of the present invention.
FIG. 2 is a diagram showing a file configuration of a database according to the present invention.
FIG. 3 is a diagram illustrating a relationship between first ranking data and second ranking data according to the present invention.
FIG. 4 is a diagram showing an example of first ranking data of the present invention.
FIG. 5 is a diagram showing an example of second ranking data of the present invention.
FIG. 6 is a diagram showing an example of grouping first ranking data according to the present invention.
FIG. 7 is a diagram plotted on a two-dimensional plane for visualizing data grouped in the present invention.
FIG. 8 is a diagram showing a group range of data plotted on a two-dimensional plane in the present invention.

Claims (3)

A means for storing gene information, compound / drug information, and disease information in association with literature information relating to each of these information, and using any of the gene, compound / drug, or disease as a keyword to sort other information in ascending order of relation Means for assigning a score to create a ranking, wherein first ranking information arranged by ranking the document information related to the keyword and information having a large number of documents to be shared, and the first ranking Research gene product data classification characterized by ranking the document information related to each piece of information and the information having a large number of documents to be shared, and creating second ranking information arranged for each of the first ranking information. system. A means for storing gene information, compound / drug information, and disease information in association with literature information relating to each of these information, and using any of the gene, compound / drug, or disease as a keyword to sort other information in ascending order of relation Means for assigning a score to create a ranking, wherein first ranking information arranged by ranking the document information related to the keyword and information having a large number of documents to be shared, and the first ranking The ranking is determined from the document information related to each piece of information and the information having a large number of documents to be shared, and second ranking information arranged for each of the first ranking information is created, and a predetermined ranking of the second ranking information is determined. Research gene product data characterized by comparing the above information between the second ranking information and measuring the similarity scale based on the number of matching information. Classification system. A means for storing gene information, compound / drug information, and disease information in association with literature information relating to each of these information, and using any of the gene, compound / drug, or disease as a keyword to sort other information in ascending order of relation Means for assigning a score to create a ranking, wherein first ranking information arranged by ranking the document information related to the keyword and information having a large number of documents to be shared, and the first ranking The ranking is performed from the document information related to each piece of information and the information having a large number of documents to be shared, the second ranking information arranged for each of the first ranking information is created, and the distance between the first ranking information is calculated. By using the score of the second ranking information to measure a similarity measure between the information. System.

Images