KR102158066B1 - Method, Device, and Computer-Readable Medium for Optimizing Document Image - Google Patents

Abstract

The present invention derives drug-gene relationship data from biomedical literature data, and based on the derived drug-gene relationship data, a new drug for deriving a new drug predicted to have an effect or side effect to be identified is derived. It relates to a method, apparatus, and computer-readable medium.

Description

[Method, Device, and Computer-Readable Medium for Optimizing Document Image}]

The present invention relates to a method, an apparatus, and a computer-readable medium for deriving novel properties of drugs, and more particularly, to derive drug-gene relationship data from biomedical literature data, and derived drug-gene relationship data. It relates to a method, an apparatus, and a computer-readable medium for deriving novel properties of a drug for deriving a new drug that is predicted to have an effect or side effect to be determined based on the.

In order to develop a new drug, astronomical research costs and a long research period of 10 to 20 years are required. Currently, numerous researchers and pharmaceutical companies are working to find new efficacy of existing drugs as a way to reduce these costs and duration while developing new drugs. A typical example is Viagra. In the early stages of development, Viagra was studied to treat angina, but is now used as a treatment for erectile dysfunction.

As one of the methods to find new efficacy of drugs that have already been developed, research using clinical trial results is also conducted. However, it is true that there are numerous difficulties in obtaining a sufficient amount of data necessary for the study by obtaining consent from clinical trial subjects and proceeding with the study, and using drugs for a large number of subjects. Therefore, a new method is needed to solve these difficulties.

When the drug is used in the human body, it acts as a down-regulation or up-regulation effect on genes in the human body. In addition, when a disease occurs in the human body, it acts as an inhibitory or active action on genes in the human body. Therefore, a method to explore the relationship between medicine and disease was devised through genes as a medium.

Eventually, as the method through clinical trials has the above problems, the need for a method to explore such correlations using literature related to numerous existing biological studies, not just clinical trials, has grown.

In addition, drugs have certain side effects, and each side effect is derived through clinical trials, but there is enormous cost for this, and in the case of clinical trials to derive new uses or efficacy of the drug, it is also enormous cost. Is holding this.

An object of the present invention is to derive drug-gene relationship data from biomedical literature data, and based on the derived drug-gene relationship data, a novel characteristic of a drug to derive a new drug predicted to have an effect or side effect to be identified. It is to provide a method, apparatus, and computer-readable medium for derivation.

In order to solve the above problems, the present invention is a method for deriving a new characteristic of a drug, implemented by a computing device, in which DG related data including one or more related gene information for each of one or more drugs is stored in the computing device. Loading or extracting DG relevance data derived from biomedical literature data; And based on the matching rate of the gene information extracted from the DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information, the efficacy or side effect to be identified is unknown. It provides a method for deriving a new characteristic of a drug, including; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for the drug.

In an embodiment of the present invention, the DGP relevance derivation step includes: a text loading step of loading text from the biomedical document data; A sentence extraction step of extracting a text to be analyzed according to a preset rule from the loaded text; A sentence discrimination step of determining whether the analysis target text contains about or a gene; A valid word extraction step of extracting valid words related to the relationship between the drug and the gene from the analysis target text; And a first GRS calculation step of extracting DG related data by deriving the relationship between the drug and the gene from the effective word.

In one embodiment of the present invention, in the sentence extraction step, two or more words are extracted from the loaded text in units of phrases or clauses, and the text is analyzed as the text to be analyzed. It can be determined whether there is a description of drugs and genes in the text.

In one embodiment of the present invention, the effective word extraction step extracts a valid word group including valid words including verbs and nouns from the analysis target text, and the first GRS calculation step comprises: an element of the valid word group In the case where is corresponding to a biological term having a predetermined suppressive meaning, assigning a first parameter value to the effective word of the effective word group; Assigning a second parameter value to the effective word of the effective word group when the element of the effective word group does not correspond to a biological term of a predetermined suppressive meaning; And a first GRS derivation step of calculating a gene control score (GRS) for a drug and a gene based on one or more parameter values assigned to the effective word of the effective word group.

In an embodiment of the present invention, the first parameter value and the second parameter value correspond to numeric values having different signs, and in the step of deriving the first GRS, two or more parameter values assigned to valid words of the valid word group By multiplying, the gene control score can be calculated.

In an embodiment of the present invention, the step of deriving drug characteristics includes: a first step of calculating a matching rate for each DG topic extracted from the DG related data based on genetic information for the one or more drugs; A second step of deriving one or more representative DG topics based on a matching rate for the DG topic or a score derived from the matching rate of one or more drugs known to have an efficacy or side effect to be identified; A third step of deriving a new drug for which the target efficacy or side effect is predicted based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic; Can include.

In an embodiment of the present invention, the DG topic is a plurality of topics derived through a topic modeling method by considering each drug of the DG related data as a document, and considering the genetic information as a word, the first step May calculate a matching rate between gene information in each of the at least one DG related data and gene information included in each topic of the DG topic.

In an embodiment of the present invention, in the second step, for each of the one or more drugs known to have an efficacy or side effect to be identified, a matching rate for the DG topic or a score derived from the matching rate is equal to or greater than a preset criterion. DG topics can be derived as the representative DG topics.

In one embodiment of the present invention, the representative DG topic includes a plurality of topics, and the third step is a plurality of topics belonging to each of the representative DG topics that are not known to have the efficacy or side effects to be identified. Based on each matching rate for or a representative score derived from the score derived from the matching rate, a new drug for which the target efficacy or side effect is predicted among the one or more drugs that are not known to have the target efficacy or side effects can be derived. I can.

In an embodiment of the present invention, the representative score may be a matching rate or a sum of matching rates for a plurality of topics belonging to each of the representative DG topics, each of which is not known to have an efficacy or side effect to be determined.

In one embodiment of the present invention, the third step is a score derived from each matching rate or matching rate for a plurality of topics belonging to the representative DG topic of each of the at least one drug known to have an efficacy or side effect to be identified. On the basis of, by determining a score derived from each matching rate or matching rate for a plurality of topics belonging to the representative DG topic of each of the at least one drug that is not known to have an efficacy or side effect of the discrimination target, the discrimination target Among one or more drugs that are not known to have efficacy or side effects, new drugs with predicted efficacy or side effects can be derived.

In order to solve the above problems, as a device for deriving a new characteristic of a drug including one or more memories and one or more processors, DG-related data including one or more related gene information for each of one or more drugs is loaded from the memory. Or a DG-related data extraction unit derived from biomedical literature data; And based on the matching rate of the gene information extracted from the DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information, the efficacy or side effect to be identified is unknown. It provides an apparatus for deriving a new characteristic of a drug, including a drug characteristic derivation unit for determining whether there is an effect or side effect to be determined for the drug.

In order to solve the above problems, as a computer-readable recording medium, the computer-readable recording medium stores instructions for causing a computing device to perform the following steps, the steps: DG relevance data extracting step of loading DG relevance data including one or more related gene information of the computer in the computing device or derived from biomedical literature data; And based on the matching rate of the gene information extracted from the DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information, the efficacy or side effect to be identified is unknown. It provides a computer-readable recording medium comprising; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for the drug.

According to an embodiment of the present invention, by extracting sentence information based on biomedical literature data and taking into account the down-regulation and up-regulation of the gene of the drug, the target efficacy or side effect For this unknown drug, it can exert the effect of discriminating whether there are any effects or side effects to be identified.

According to an embodiment of the present invention, it is possible to exert an effect of inducing a new discriminant effect or side effect of previously known drugs by indirectly using information on a drug-gene described in a vast amount of biomedical literature data.

According to an embodiment of the present invention, according to the update of new biomedical literature data, it is possible to exert an effect of deriving a drug having a new efficacy or side effect to be identified without a separate new work or experiment.

According to an embodiment of the present invention, it is possible to derive a drug having a new discriminant efficacy or side effect at low cost without a clinical trial.

According to an embodiment of the present invention, it is possible to derive a drug having a new discriminant effect or side effect by effectively utilizing the biomedical data.

1 schematically derives a process of deriving a new characteristic of a drug based on biomedical literature data according to an embodiment of the present invention.
FIG. 2 schematically shows a system including a computing device for deriving a new characteristic of a drug based on biomedical document data according to an embodiment of the present invention.
3 schematically shows the internal configuration of a computing device for deriving a new characteristic of a drug based on biomedical document data according to an embodiment of the present invention.
4 schematically shows the internal configuration of a DG-related data extracting unit according to an embodiment of the present invention.
FIG. 5 exemplarily shows a result of performing an operation of a valid word extractor according to an embodiment of the present invention.
6 exemplarily shows a biological term with a predetermined inhibitory meaning according to an embodiment of the present invention.
FIG. 7 exemplarily illustrates calculation of a genetic control score (GRS) using a first parameter value or a second parameter value assigned to a valid word according to an embodiment of the present invention.
8 schematically shows a process of a weak characteristic derivation step according to an embodiment of the present invention.
9 schematically illustrates a process of extracting topics from DG relevance data according to an embodiment of the present invention, and deriving a matching rate of topics for each drug.
10 schematically shows a process of deriving a score based on a matching rate for a topic for each drug according to an embodiment of the present invention.
FIG. 11 schematically illustrates a process of performing class marking for drugs having an efficacy or side effect to be identified among each drug according to an embodiment of the present invention.
12 is a diagram schematically showing a matching rate or a score derived from the matching rate of each drug for a representative DG topic according to an embodiment of the present invention.
13 is a diagram schematically showing a process of deriving a new drug for which an efficacy or side effect to be determined is predicted according to an embodiment of the present invention.
14 exemplarily illustrates an internal configuration of a computing device according to an embodiment of the present invention.

In the following, various embodiments and/or aspects are now disclosed with reference to the drawings. In the following description, for purposes of explanation, a number of specific details are disclosed to aid in an overall understanding of one or more aspects. However, it will also be appreciated by those of ordinary skill in the art that this aspect(s) may be practiced without these specific details. The following description and the annexed drawings set forth in detail certain illustrative aspects of one or more aspects. However, these aspects are exemplary and some of the various methods in the principles of the various aspects may be used, and the descriptions described are intended to include all such aspects and their equivalents.

Further, various aspects and features will be presented by a system that may include multiple devices, components and/or modules, and the like. It is also noted that various systems may include additional devices, components and/or modules, and/or may not include all of the devices, components, modules, etc. discussed in connection with the figures. It must be understood and recognized.

As used herein, “an embodiment,” “example,” “aspect,” “example,” and the like may not be construed as having any aspect or design described as being better or advantageous than other aspects or designs. . The terms'~part','component','module','system', and'interface' used below generally mean a computer-related entity, for example, hardware, hardware It can mean a combination of software and software, or software.

In addition, the terms "comprising" and/or "comprising" mean that the corresponding feature and/or element is present, but excludes the presence or addition of one or more other features, elements, and/or groups thereof. It should be understood as not.

In addition, terms including ordinal numbers such as first and second may be used to describe various elements, but the elements are not limited by the terms. These terms are used only for the purpose of distinguishing one component from another component. For example, without departing from the scope of the present invention, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element. The term and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

In addition, in the embodiments of the present invention, unless otherwise defined, all terms used herein including technical or scientific terms are commonly understood by those of ordinary skill in the art to which the present invention belongs. It has the same meaning as. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the embodiments of the present invention, an ideal or excessively formal meaning Is not interpreted as.

1 schematically shows a method of deriving a new property of a drug based on biomedical literature data according to an embodiment of the present invention.

According to the above embodiment, the DG relevance derivation step of deriving DG relevance data from biomedical document data is performed. By performing such DG relevance data, DG relevance data is derived from the biomedical document data.

The DG-related data is data indicating whether a specific gene is down-regulated or activated (up-regulated) by the use of a specific drug. These DG-related data become basic data for deriving new efficacy of existing drugs in the present invention.

Here, the “literature” of the biomedical literature data includes the same meaning as commonly understood, but is not limited thereto, and includes all expressions of a certain intention, idea, or information in writings or symbols such as papers, journals, and books. It is a concept. Preferably, it is preferable that the biomedical literature data is a summary or abstract information of the paper, journal, or book.

In addition, the method of deriving a new characteristic of a drug based on the biomedical literature data according to the present invention further includes the step of deriving a new drug having an efficacy or side effect to be determined from the DG-related data.

Hereinafter, an apparatus for deriving a new characteristic of a drug based on the biomedical literature data of the present invention will be described.

FIG. 2 schematically shows a system including a computing device for deriving a new characteristic of a drug based on biomedical document data according to an embodiment of the present invention.

The system includes a biomedical document data DB(A) connected from the outside with the computing device 1000, a synonym DB(B), a DB(C) storing DG-related data already derived about the relationship between a drug-gene, and a genetic symbol. It is composed of DB(D).

The computing device 1000 is a device that performs an operation function for deriving a new characteristic of a drug based on biomedical document data, and may include a processor, a bus, a network interface, and a memory. In another embodiment, a computing device for deriving a new characteristic of a drug based on biomedical literature data may include more components than these components.

The biomedical literature data DB(A) is a database storing a plurality of biomedical literature data that enables access to information necessary to perform computational functions of the computing device 1000 through an externally connected network. Includes the PubMed database, which provides an abstract of Alternatively, the biomedical document data may be accessed in a form stored in a memory of the computing device.

The drug synonym DB (B) is a database that stores drugs and synonym data that enable access to information necessary to perform computational functions of the computing device 1000 through an externally connected network, such as DrugBank or KEGG DRUG. May contain databases. Alternatively, the data of the weak synonym DB may be accessed in a form stored in the memory of the computing device.

The DG relevance data DB (C) is a DB in which weak-gene relevance data required to perform the computational function of the computing device 1000 through a network connected from outside is stored. In an embodiment of the present invention, the computing device may extract DG-related data from biomedical literature data, or, in a method according to an embodiment of the present invention, or in a different method, the relationship data for drugs and genes already accumulated in another DB. DG-related data can be collected from outside. Likewise, the DG-related data may be stored in a memory of a computing device and loaded, and a method of deriving a new characteristic of the drug of the present invention may be performed.

The genetic symbol DB(D) is a database storing genetic symbol data that enables access to information necessary for performing the computational function of the computing device 1000 through a network connected from the outside, and may include a database such as PharmGKB. have.

In the embodiment shown in FIG. 2, the biomedical literature data DB(A), about synonym DB(B), DG relation data DB(C), and genetic symbol DB(D) are of the computing device 1000. Although shown to operate independently from the outside, the present invention is not limited thereto, and the biomedical literature data DB(A), about synonym DB(B), DG relation data DB(C), and genetic symbol DB(D ) Corresponds to the inside of the computing device 1000, or information received from the biomedical literature data DB(A), about synonym DB(B), DG relevance data DB(C), and genetic symbol DB(D) May be stored in the computing device 1000 and implemented in a manner used in calculation of the computing device.

3 schematically shows the internal configuration of a computing device for deriving a new characteristic of a drug based on biomedical document data according to an embodiment of the present invention.

The computing device for deriving new characteristics of medicines based on biomedical literature data according to the above embodiment corresponds to a double arrow between the processor 100 and the bus (processor 100, memory 300, and network interface 200). ), a network interface 200 and a memory 300. The memory 300 includes an operating system, an execution code such as a DG-related data extraction unit 110, an execution code, a weak-characteristic extraction unit 120, an execution code, DG-related data, biomedical literature data, pharmacological synonym data, and gene symbol data. Can include. The processor 100 may include a DG-related data extraction unit 110 and a weak characteristic extraction unit 120. In other embodiments, a computing device for deriving a new characteristic of a drug based on biomedical literature data may include more components than those of FIG. 3.

The memory is a computer-readable recording medium and may include a permanent mass storage device such as a random access memory (RAM), a read only memory (ROM), and a disk drive. These software components may be loaded from a computer-readable recording medium separate from the memory using a drive mechanism (not shown). Such a separate computer-readable recording medium may include a computer-readable recording medium (not shown) such as a floppy drive, a disk, a tape, a DVD/CD-ROM drive, or a memory card. In another embodiment, software components may be loaded into a memory through the network interface 200 rather than a computer-readable recording medium.

The bus may enable communication and data transmission between components of a computing device that derive new properties of drugs based on biomedical literature data. The bus may be configured using a high-speed serial bus, a parallel bus, a storage area network (SAN) and/or other suitable communication technology.

The network interface 200 may be a computer hardware component for connecting a computing device for deriving new properties of a drug based on biomedical document data to a computer network. The network interface 200 may connect a computing device that derives new characteristics of a drug based on biomedical literature data to a computer network through a wireless or wired connection. Through such a network interface 200, a computing device for deriving a new characteristic of a drug based on biomedical literature data may be wirelessly or wiredly connected to the tactile interface device.

The processor may be configured to process instructions of a computer program by performing input/output operations of a computing device that derives new properties of drugs based on basic arithmetic, logic, and biomedical document data. Instructions may be provided to the processor by means of memory 300 or network interface 200 and via a bus. The processor may be configured to execute program codes for the DG-related data extraction unit 110 and the weak characteristic extraction unit 120. Such program codes may be stored in a recording device such as a memory.

The DG relevance data extraction unit 110 and the drug characteristics extraction unit 120 may be configured to perform a method of deriving new characteristics of a drug based on biomedical literature data, which will be described below. In the processor described above, some components may be omitted, additional components not shown may be further included, or two or more components may be combined according to a method of deriving new properties of a drug based on biomedical literature data.

Meanwhile, such a computing device preferably corresponds to a personal computer or a server, and in some cases, a smart phone, a tablet, a mobile phone, a video phone, and an e-book reader (e -book reader), desktop PC, laptop PC, netbook PC, personal digital assistant (PDA, hereinafter referred to as'PDA'), portable Multimedia player (portable multimedia player: PMP, hereinafter referred to as'PMP'), mp3 player, mobile medical device, camera, wearable device (for example, head-mounted Device (head-mounted device: HMD, for example, to be referred to as'HMD'), electronic clothing, electronic bracelet, electronic necklace, electronic appcessory, electronic tattoo, or smart watch ), etc.

The computing device receives the biomedical document data DB(A) from the outside through the network interface 200 or performs the process of the DG relevance data extraction unit 110 with respect to the biomedical document data previously stored in the memory 300. DG-related data can be used.

The DG-related data extracting unit 110 loads DG-related data including one or more related gene information for each of one or more drugs from the computing device or derives it from biomedical document data.

That is, in a preferred embodiment of the present invention, DG relevance data is continuously updated from biomedical document data, but in another embodiment, DG relevance data already constructed may be extracted or loaded and used.

The drug characteristic extraction unit 120 is based on the matching rate of the gene information extracted from the DG related data of each drug with respect to one or more DG topics extracted from the DG related data and constituted by one or more gene information, It is determined whether there is an efficacy or side effect to be identified for a drug whose efficacy or side effect to be identified is unknown.

4 schematically shows the internal configuration of the DG-related data extracting unit 110 according to an embodiment of the present invention.

The DG relevance data extraction unit 110 includes a text loading unit 111 for loading text from the biomedical document data; A sentence extraction unit 112 for extracting an analysis target text according to a preset rule from the loaded text; A sentence discrimination unit 113 that determines whether the analysis target text contains a drug and a gene; A valid word extracting unit 114 for extracting valid words related to a relationship between a drug and a gene from the analysis target text; A first GRS calculation unit 115 for deriving a relationship between a drug and an object of a gene from the effective word; A second GRS calculation unit 116 for deriving a relationship between a drug and a gene by synthesizing information from a plurality of documents or a plurality of texts to be analyzed using the result derived by the first GRS calculation unit 115; And a DG-related data loading unit 117 for loading DG-related data that has already been derived or constructed, rather than derives information on the weak-gene from the DG-related data extraction unit 110. The DG-related data is extracted from the biomedical document data or data already stored by performing the operation of the DG-related data extracting unit 110.

The text loading unit 111 loads text from the biomedical document data. The biomedical document data DB(A) is transmitted from the outside through the network interface 200 or the biomedical document data previously stored in the memory 300 is loaded into the DG relevance data extraction unit 110. Such biomedical literature data DB(A) includes the PubMed database, which provides an abstract of biomedical literature. For example, if the PubMed database is used through the network interface 200, about 1,450,000,000 abstracts can be loaded from about 7,000 journals, which means that the loaded text of the DG relevance data extraction unit 110 is do.

The sentence extracting unit 112 extracts the text to be analyzed by extracting two or more words in units of phrases or clauses from the loaded text. This is because in deriving DG relevance data from the loaded text, the phrase or clause has the meaning of the smallest unit having necessary information.

The sentence determination unit 113 determines whether there are descriptions of drugs and genes in the text to be analyzed, and extracts a simultaneous use sentence. These simultaneous use sentences become a new text to be analyzed. Eventually, by performing the operation of the sentence determination unit 113, the description of the drug and the gene is derived as the DG relevance data showing the relationship between the drug and the gene.

In this process, the sentence determination unit 113 searches for the text to be analyzed with a seed including a drug synonym and a gene symbol to determine whether there are descriptions of such drugs and genes, and extracts a simultaneous use sentence. do. The about synonyms and data on the genetic symbols are transmitted from the outside through the weak synonym DB (B) and the genetic symbol DB (D) through the network interface 200 or previously stored in the memory 300 The about synonym data and the genetic symbol data are used.

The pharmacokinetics DB (B) and the drug synonym data 360 may include a database such as DrugBank or KEGG DRUG, and the gene symbol DB (D) and the gene symbol data 392 include a database such as PharmGKB. Can contain

For example, the genetic symbol and the disease synonym become seeds, search for an analysis target text, and from the analysis target text, “APOE” as one of the genetic symbols and “atorvastatin” as one of the disease synonyms Concurrent use sentences such as “APOE mRNA expression was reduced after atorvastatin treatment” can be extracted.

Preferably, the sentence discrimination unit 113 determines whether there are descriptions of drugs and genes in the analysis target text, extracts a simultaneous use sentence, and then, whether there is an expression related to a negative sentence in the analysis target text. Whether or not can be determined. The negative sentence determined in this way (not) is excluded from the analysis target text. This is, if there is a biological term among the words used in the text to be analyzed when performing the operation of the first GRS calculation unit 115 to be described later, whether the biological term is a biological term with a meaning of down-regulation or is active ( This is because it distinguishes whether or not it is a biological term of up-regulation meaning.If the biological term of the inhibitory meaning is written in a negative sentence and has an active meaning, the negative sentence (not) may also be included in the text to be analyzed, but it is not. . For example, the biological term “not activated” used in negative sentences does not necessarily have the meaning of “inhibited”, the biological term for inhibitory meaning.

The effective word extracting unit 114 extracts a valid word group including two or more valid words including verbs and nouns. The valid word group including such valid words is used when assigning a parameter value to the valid word in the operation of the first GRS calculator 115 to be described later. Here, the effective word means, first of all words used in the simultaneous use sentence, excluding the word corresponding to the seed containing the weak synonym and the genetic symbol, and the DG relevance data among the words used in the simultaneous use sentence. It refers to a word that has information indicating the relationship between two or more words that contain drugs or genes that are required to derive. Words having information indicating such relevance are preferably verbs or nouns other than parts of speech such as prepositions or adverbs among words used in the simultaneous use sentence. This is because the part of speech that has the main information of the sentence among the parts of speech classification based on the function of the word in the sentence composition becomes a verb and a noun.

FIG. 5 exemplarily shows a result of performing an operation of the effective word extracting unit 114 according to an embodiment of the present invention.

In the above embodiment, the effective words are about synonyms in “APOE mRNA expression was reduced after atorvastatin treatment” extracted with the simultaneous use sentence, and “APOE” and “APOE” corresponding to a seed containing a genetic symbol. Except for “atorvastatin”, it can be seen that it is an expression, reduced, and treatment corresponding to a verb or a noun among the words used in the simultaneous use sentence. That is, by performing the operation of the effective word extracting unit 114 in the simultaneous use sentence, the effective word group including the three valid words expression, reduced and treatment is generated.

The first GRS calculator 115 determines whether or not the effective word of the effective word group corresponds to a biological term having a predetermined suppressive meaning; If the element of the effective word group corresponds to a biological term of a preset inhibitory meaning, a first parameter value is assigned to the element of the effective word group, and the element of the effective word group is a biological term of a preset inhibitory meaning. If it does not correspond to, a second parameter value is assigned to the element of the valid word group; Based on the parameter values assigned to the effective words of the effective word group, the gene control score (GRS) for drugs and genes is calculated. Consequently, the DG-related data can be expressed as the gene control score (GRS) of the gene related to each drug.

The first GRS calculation unit 115 first determines whether the effective word of the effective word group corresponds to a biological term having a predetermined suppression meaning. According to an embodiment, the preset inhibitory meaning of biological terms (DRW, Down-regulation Relationship Words) is a biological term of inhibitory meaning and/or a drug-gene relationship ontology (PHARE ontology) and negative vocabulary group set by the user. It may include a biological term with a suppressive meaning consisting of words that are overlapping between them. The biological terms of the suppressive meaning are stored in the memory 300, and the first GRS calculation unit 115 is necessary in determining whether the valid words of the effective word group correspond to the biological terms of the preset suppression meaning. Data can be loaded from DRW data.

6 exemplarily shows a biological term with a predetermined inhibitory meaning according to an embodiment of the present invention.

In FIG. 6, words including Abolish, Antagonize, and Decrease correspond to the biological terms of the predetermined inhibitory meaning.

In this case, when the element of the effective word group corresponds to the preset biological term with a suppressive meaning, a first parameter value is assigned to the element of the effective word group. However, when the element of the effective word group does not correspond to the preset biological term of the suppressive meaning, a second parameter value is assigned to the element of the effective word group. In this way, a parameter value that is classified according to whether it corresponds to the biological term of the predetermined inhibitory meaning among the elements of the effective word group is assigned.

Finally, a gene control score (GRS) for drugs and genes or diseases and genes is calculated based on the parameter values assigned to the effective words of the effective word group. This gene control score (GRS) indicates whether the gene is down-regulated or activated (up-regulation) by the use of the drug according to the derived value, or the gene is suppressed by the onset of the disease (down-regulation). -regulation) or activation (up-regulation). Eventually, this gene control score is stored as the DG-related data 380 in the memory 300.

Preferably, the first parameter value and the second parameter value correspond to numeric values having different signs, and in the GRS derivation step, the genetic control score is multiplied by two or more parameter values assigned to the effective words of the effective word group. Calculate.

If the thus obtained gene control score (GRS) is the same code as the first parameter value, it can be seen that the gene is down-regulated due to the use of the drug or the onset of the disease, and the gene control score is the second parameter. If the code is the same as the value, it can be seen that the gene is up-regulated due to the use of the drug or the onset of the disease. That is, if the genetic control score is the same sign as the first parameter value, the effective word corresponding to the biological term of the preset suppression meaning is written odd times in the effective word group used in the simultaneous use sentence, and the simultaneous use sentence is suppressed. It indicates that it has a meaning, and when the gene control score is the same sign as the second parameter value, it indicates that it has an active meaning.

Hereinafter, a process in which the operation of the first GRS calculator 115 is performed will be described according to the exemplary embodiment of FIGS. 5 and 6.

When it is determined whether the elements of the effective word group including expression, reduced, and treatment in FIG. 5 correspond to the biological terms of the preset inhibitory meaning illustrated in FIG. 6, reduced corresponds to the biological term of the preset inhibitory meaning. . On the contrary, other words, expression and treatment, do not correspond to the pre-set biological terms of inhibitory meaning. Therefore, a first parameter value is given to reduced, and a second parameter value distinguished from the first parameter value is given to expression and treatment.

FIG. 7 exemplarily illustrates calculation of a genetic control score (GRS) using a first parameter value or a second parameter value assigned to a valid word according to an embodiment of the present invention.

Referring to FIG. 7,'-1' corresponding to a negative real number may be assigned as the first parameter value, and a'+1' value corresponding to a positive real number may be assigned as the second parameter value. That is, among the elements of the effective word group of FIG. 5, a first parameter value of -1 is assigned to a reduced corresponding to the biological term of the preset suppression meaning, and a second parameter is applied to expressions and treatments corresponding to other words. A value of +1 is given, and a given value of -1 by multiplying all parameter values assigned to the effective word becomes the gene control score (GRS). It can be seen that the gene control score has the same sign as the first parameter value. That is, as described above, it can be seen that APOE, a specific gene, is down-regulated using atorvastatin, a specific drug shown in FIG. 7.

Preferably, when two or more gene control scores are derived for the same drug and gene in the first GRS calculation unit, the sum of gene control scores (GRS _sum ) from the two or more gene control scores can be updated with the gene control score. have. If a number of trials are repeated and different results are derived due to errors in each trial, the larger the sample of these trials, the more accurate results can be derived. That is, the gene control score sum (GRS _sum ) of the two or more gene control scores has the advantage of showing more accurate results than the gene control score.

For example, if the gene control score for a specific gene APOE of a specific drug A is +1, +1, and -1 from documents 1 to 3, which are biomedical literature data, all of these gene control scores are +1 (+1 +1-1) is the sum of the gene control points. The derived gene control score +1 can be updated with a new gene score.

Preferably, in the second GRS calculation unit, only the sign of the sum of gene control scores in a plurality of documents is taken as data. For example, when the gene control score for a specific gene B for a specific drug A is +1, +1, +1, +1, +1 in five documents, the second GRS calculation unit It is preferable that the updated gene control score or the gene control score sum is updated to +1 instead of +5.

8 schematically shows a process of a weak characteristic derivation step according to an embodiment of the present invention.

The drug characteristic derivation step comprises: a first step (S10, S20) of calculating a matching rate for each DG topic extracted from the DG relevance data based on genetic information for the at least one drug; A second step (S30, S40) of deriving one or more representative DG topics based on a score derived from a matching rate or a matching rate for the DG topic of one or more drugs known to have an efficacy or side effect to be identified; A third step of deriving a new drug for which the target efficacy or side effect is predicted based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic (S50 ); includes.

In step S10, a DG topic is extracted. The DG topic is a plurality of topics derived through a topic modeling method by considering each drug of the DG-related data as a document and considering the gene information as a word.

In recent years, as the number of documents has increased rapidly, techniques have been studied for researchers to deal with only necessary documents. As such a technique, a technique called topic modeling is currently researched and developed. In topic modeling, the main keywords of the document are extracted based on the content of the document, and it is based on an algorithm for grouping documents having corresponding keywords.

Topic modeling is a statistical-based modeling algorithm that extracts a main topic or main theme from a plurality of documents, and a document is assumed to be a complex of a plurality of topics. In topic modeling, a set of words frequently mentioned at the same time in several documents are classified into one topic. A topic forms a new category, and the meaning of words constituting the topic gives meaning to the category.

In a general data clustering method, one document corresponds to one cluster, but in topic modeling, one document may correspond to a plurality of topics, and the matching for each topic may be designated as a probability value other than O/X. have.

The overall process of LDA (Latent Dirichlet Allocation), which is an exemplary algorithm of topic modeling, may be accomplished by the following equation.

이고,

는 단어 집단(vocabulary)에서의 분포이다. d번째 문서에 대한 토픽 비율(proportion)은

이고,

는 d번째 문서의 k 토픽의 토픽 비율을 나타낸다. d번째 문서에 대한 토픽 할당(topic assignment)는

이고,

는 문서 d의 n 번째 단어의 토픽 할당에 해당한다. 문서 d의 관측된 단어는

이고,

는 문서 d의 n번째 단어를 의미하고, 이는 고정된 단어 집단의 요소에 해당한다.The topic here is

ego,

Is the distribution in the vocabulary. The topic proportion for the d-th document is

ego,

Represents the topic ratio of the k topic of the d-th document. The topic assignment for the d-th document is

ego,

Corresponds to the topic assignment of the nth word of document d. The observed word in document d is

ego,

Denotes the nth word of document d, which corresponds to an element of a fixed group of words.

In step S10, topic modeling in the LDA method as described above may be performed or topic modeling may be performed in another method.

On the other hand, in another embodiment of the present invention, the DG topic may not be extracted by performing topic modeling from the DG related data as in step S10, but the already extracted DG topic may be loaded.

The DG topics as described above are composed of a plurality, and each topic includes a plurality of gene information.

For example, topic A may be configured as follows.

Gene A: +

Gene B: +

Gene C:-

Gene D: +

Gene E:-

Gene F: +

Gene H: +

… Gene Z: +

In one embodiment of the present invention, there are about 2200 drugs currently approved by the FDA, and in the present invention, topics were extracted through topic modeling for 684 drugs having genetic literature information. The appropriate number of topics was derived through the LOG-LIKELIHOOD technique, and the appropriate number of topics may be set in different ways or arbitrarily.

In S20, a matching rate of gene information in each of the at least one DG related data and gene information included in each topic of the DG topic is calculated. It is preferable that the matching rate is calculated as a probability value, and more preferably, the sum of each drug for each DG topic is normalized to a specific number (for example, 1).

In S30, a drug group with known efficacy or side effects to be identified is derived. For example, if the target efficacy or side effect is breast cancer, among the drugs included in the DG-related data, a group of drugs with efficacy or side effects for breast cancer is derived.

For example, if the number of drug-genes (usually having multiple genetic information for one drug) group or drug included in the DG-related data is 600, it is said that there are already efficacy against breast cancer. There may be as many as six known drugs. In this case, the six drugs correspond to a group of drugs with known efficacy or side effects.

In S40, one or more DG topics in which the score derived from the matching rate or the matching rate for the DG topic for each of the one or more drugs known to have an efficacy or side effect to be determined is greater than or equal to a preset criterion are derived as the representative DG topic.

Specifically, if you want to derive a new drug for breast cancer, determine the value of each DG topic for the six drugs, and among them, the high matching rate or the score derived from the matching rate (normalized from the matching rate or a specific function). DG topics with high scores) are collected. As a result, the collected DG topics may correspond to information on genetic topics related to breast cancer, and this is referred to as a representative DG topic.

For example, a DG topic having a matching rate of more than a preset value or a score derived from the matching rate from six drugs may be extracted as a representative DG topic.

For example, it is assumed that there are drugs from about 1 to about 600, topics are from topic A to topic Z, and breast cancer corresponds to the efficacy to be identified, and that about 1 to about 6 are known to have the target efficacy for breast cancer. do.

For example, in about 1, topic A has the highest matching rate or score derived from the matching rate, in about 2, topic B has the highest matching rate or score derived from the matching rate, and in about 3, topic C In about 4, Topic D has the highest matching rate or score derived from the matching rate, and In about 5, Topic E has the highest matching rate or matching rate. With a score, if topic F has the highest matching rate or a score derived from the matching rate in about 6, the representative DG topics can be topics A, B, C, D, E, and F. This suggests that topics A, B, C, D, E, and F are topics related to breast cancer.

On the other hand, representative DG topics are not only one that can be derived from one drug, but two or more DG topics that exceed the standard value are extracted, or two or more DG topics that have a score derived from the matching rate or matching rate of the same or similar range. May be extracted.

In step S50, a new drug for which the target efficacy or side effect is predicted is derived based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic.

Specifically, in step S50, the matching rate or the score derived from the matching rate for the DG topic belonging to the representative DG topic group of each drug whose discrimination target efficacy or side effect is unknown are considered.

For example, if topics A, B, C, D, E, and F are representative DG topics as in the above example, the topics A, B, C, D, E, and F for each of about 7 to 600 New drugs are derived based on the matching rate or score based on the matching rate.

Preferably, on the basis of a representative score derived from a score derived from each matching rate or a score derived from a matching rate for a plurality of topics belonging to the representative DG topic of each of the at least one drug that is not known to have an effect or side effect to be identified, Among the at least one drug that is not known to have the effect or side effect to be identified, a new drug for which the effect or side effect to be identified is predicted is derived.

Preferably, the representative score corresponds to the sum of the scores derived from each matching rate or matching rate for a plurality of topics belonging to each of the representative DG topics that are not known to have the efficacy or side effects to be identified. I can.

For example, if the sum of the scores derived from the matching rates for topics A, B, C, D, E, and F of about 1 to 6 correspond to 0.6, 0.7, 0.6, 0.6, 0.7, and 0.9, respectively, , 0.6, 0.7, 0.6, 0.6, 0.7, 0.9, a representative value (for example, mean value, median value, etc.) obtained by performing statistical analysis may be a standard for drugs having similar genetic topics for breast cancer.

That is, in deriving a new drug, in the step S50, a score derived from each matching rate or matching rate for a plurality of topics belonging to the representative DG topic of each of the one or more drugs known to have an efficacy or side effect to be identified is A reference value can be set as a reference. In an embodiment of the present invention, the reference value may be set through various statistical techniques. For example, Z-score may be used.

Thereafter, in step S50, a score derived from each matching rate or matching rate for a plurality of topics belonging to the representative DG topic of each of the at least one drug that is not known to have an efficacy or side effect to be determined based on the reference value is calculated. Discriminate.

For example, for about 7 to 600, the matching rate for the topics A, B, C, D, E, F, or the sum or sum of scores derived from the matching rate was converted using a statistical technique (Z -score) can be applied to derive new drugs.

For example, for about 10, 11, 12, 13, the matching rate for the topics A, B, C, D, E, F, or the sum of scores derived from the matching rate is the topic of about 1 to about 6 If it is close to the matching rate for A, B, C, D, E, or F or the sum of the scores derived from the matching rate, about 10, 11, 12, and 13 have similar genetic information to about 1 to 6 and from this 10, 11, 12, 13 can also be predicted to have efficacy against breast cancer.

In a preferred embodiment of the present invention, statistical transformation (eg, normal distribution) is performed on the first conversion value (eg, sum) of the matching rate for the representative DG topic of each drug or the value derived from the matching rate. And a drug for which the discriminant efficacy or side effect is unknown, having a first transformed value corresponding to the first conversion value of the drug for which the discriminant efficacy or side effect is known, or in a preset range, can be derived. .

9 schematically illustrates a process of extracting topics from DG relevance data according to an embodiment of the present invention, and deriving a matching rate of topics for each drug.

Drugs 1, 2, 3, etc. on the left side of FIG. 9 are DG-related data, and indicate related gene information for each drug. In the present invention, each drug is viewed as a document in such DG-related data, and the genetic information of each drug is determined as a word, and topic modeling is performed.

The red, yellow, green, and blue squares of the Topic Assignment of FIG. 9 are Topics derived from DG-related data of drugs, and such topics are as described above, Gene A:+, Gene B:+, Gene C: -It can have the shape of a back.

Rather than determining whether the genetic information of each drug 100% matches each topic, in the present invention, it is determined that the genetic information of each drug is common from them, and the matching rate for each extracted topic is extracted. . For example, in FIG. 9, assuming that 4 topics have been derived, the bar graph shows a matching rate or probability value for 4 topics for a specific drug (eg, Drug 2).

The table on the right side of FIG. 9 corresponds to a table that summarizes probability values or matching rates for each derived DG topic of each drug. As described above, in the present invention, when there are N drugs in the DG related data, the matching rate for the DG topics of the number of effective ranges derived for each of the N drugs is derived.

In FIG. 9, P _I,X denotes a matching rate for the X-th topic of the I-th drug or a probability value for the matching rate.

10 schematically shows a process of deriving a score based on a matching rate for a topic for each drug according to an embodiment of the present invention.

In an embodiment of the present invention, the matching rate may be used as it is, or the matching rate may be converted into a specific score as shown in FIG. 10. Preferably, for each drug, it is possible to derive a ranking for the matching rate for each topic and convert it into a score for each ranking.

For example, assuming that there are 6 topics, each topic if the matching rate or probability value of the topics A, B, C, D, E, F of Drug 2 is 0.1, 0.2, 0.3, 0.4, 0.5, 0.6. For the rankings 6, 5, 4, 3, 2, and 1 are assigned, 1 point for the 1st, 0.8 points for the 2nd, 0.6 points for the 3rd, 0.4 points for the 4th, and 5th 0.2 points for the 6th rank and 0 points for the 6th rank may be given.

Here, R _I,X corresponds to a score derived from the matching rate for the X-th topic of the I-th drug or the probability value for the matching rate.

FIG. 11 schematically illustrates a process of performing class marking for drugs having an efficacy or side effect to be identified among each drug according to an embodiment of the present invention. For example, when the efficacy of the drug to be discriminated is breast cancer and Drugs 1, 3, and 4 are known to be effective against breast cancer, as in FIG. 11, the Class item for Drugs 1, 3, 4 is indicated as T. Has been. This is because drugs having a matching rate for gene topics similar to Drugs 1, 3, and 4 can be expected to have the efficacy of breast cancer.

12 is a diagram schematically showing a matching rate or a score derived from the matching rate of each drug for a representative DG topic according to an embodiment of the present invention, and FIG. 13 is a discrimination object according to an embodiment of the present invention It is a diagram schematically showing the process of deriving a new drug with predicted efficacy or side effects.

12 is a topic that satisfies the score criteria derived from a preset matching rate or matching rate for Drugs 1, 3, 4, for example, when it is said that Drugs 1, 3, and 4 are known as drugs with target efficacy (representative DG topic) is Topic a, b, c, d, e.

As in the Sum item in Fig. 12, in this step, for each drug, a score derived from the matching rate or matching rate for Topic a, b, c, d, e corresponding to the representative DG topic (Fig. 12 is Corresponds to this). Here, Sum can be a sum value, but it can be a matching rate for topics a, b, c, d, e, or a numerical value derived from scores derived from the matching rate.

In the embodiments shown in FIGS. 12 and 13, the score derived from the matching rate of each representative DG topic is derived as Sum. In this case, Drugs 1, 3, and 4 are expected to have higher Sum values than other drugs, and drugs with similar Sum values can be expected to have similar efficacy to Drugs 1, 3 and 4.

In one embodiment of the present invention, Sum for the scores of the representative DG topics of each drug is derived by the summation method, the normal distribution for the derived Sum is derived, and the Z-score for the Sum of each drug is derived. After that, drugs having a Z-score having a range corresponding to the Z-score of Drugs 1, 3, and 4 can be predicted as drugs having an efficacy or side effect to be identified.

In the present invention, it is not limited to Sum, but in other ways, drugs having a similar pattern in which the matching rate, probability value, or score derived from the matching rate and probability value of drugs known to have a discriminant effect or side effect on a representative DG topic It can be predicted that there are effects or side effects to be identified.

14 exemplarily illustrates an internal configuration of a computing device according to an embodiment of the present invention.

14, a computing device 11000 includes at least one processor 11100, a memory 11200, a peripheral interface 11300, and an input/output subsystem ( I/Osubsystem) 11400, a power circuit 11500, and a communication circuit 11600 may be included at least. In this case, the computing device 11000 may correspond to the user terminal A connected to the tactile interface device or the aforementioned computing device B.

The memory 11200 may include, for example, high-speed random access memory, magnetic disk, SRAM, DRAM, ROM, flash memory, or nonvolatile memory. have. The memory 11200 may include a software module, an instruction set, or other various data necessary for the operation of the computing device 11000.

In this case, accessing the memory 11200 from another component such as the processor 11100 or the peripheral device interface 11300 may be controlled by the processor 11100.

The peripheral device interface 11300 may couple input and/or output peripheral devices of the computing device 11000 to the processor 11100 and the memory 11200. The processor 11100 may execute various functions for the computing device 11000 and process data by executing a software module or instruction set stored in the memory 11200.

The input/output subsystem 11400 may couple various input/output peripherals to the peripherals interface 11300. For example, the input/output subsystem 11400 may include a monitor, a keyboard, a mouse, a printer, or a controller for coupling a peripheral device such as a touch screen or a sensor to the peripheral device interface 11300 as needed. According to another aspect, the input/output peripheral devices may be coupled to the peripheral device interface 11300 without going through the input/output subsystem 11400.

The power circuit 11500 may supply power to all or part of the components of the terminal. For example, the power circuit 11500 may include a power management system, one or more power sources such as batteries or alternating current (AC), a charging system, a power failure detection circuit, a power converter or inverter, a power status indicator or power. It may contain any other components for creation, management, and distribution.

The communication circuit 11600 may enable communication with another computing device using at least one external port.

Alternatively, as described above, the communication circuit 11600 may enable communication with other computing devices by transmitting and receiving an RF signal, also known as an electromagnetic signal, including an RF circuit, if necessary.

The embodiment of FIG. 14 is only an example of the computing device 11000, and the computing device 11000 omits some of the components shown in FIG. 14, further includes additional components not shown in FIG. 14, or 2 It can have a configuration or arrangement that combines two or more components. For example, a computing device for a communication terminal in a mobile environment may further include a touch screen or a sensor, in addition to the components shown in FIG. 14, and various communication methods (WiFi, 3G, LTE) in the communication circuit 1160 , Bluetooth, NFC, Zigbee, etc.) may include a circuit for RF communication. Components that may be included in the computing device 11000 may be implemented in hardware, software, or a combination of hardware and software including one or more signal processing or application-specific integrated circuits.

Methods according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computing devices and recorded in a computer-readable medium. In particular, the program according to the present embodiment may be composed of a PC-based program or an application dedicated to a mobile terminal. An application to which the present invention is applied may be installed on a user terminal through a file provided by the file distribution system. For example, the file distribution system may include a file transmission unit (not shown) that transmits the file according to the request of the user terminal.

The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments are, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA). , A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, such as one or more general purpose computers or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of these, configuring the processing unit to operate as desired or processed independently or collectively. You can command the device. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodyed in a transmitted signal wave. The software may be distributed over networked computing devices and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

Although the embodiments have been described by the limited embodiments and drawings as described above, various modifications and variations are possible from the above description to those of ordinary skill in the art. For example, the described techniques are performed in a different order from the described method, and/or components such as a system, structure, device, circuit, etc. described are combined or combined in a form different from the described method, or other components Alternatively, even if substituted or substituted by an equivalent, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and claims and equivalents fall within the scope of the claims to be described later.

Claims (13)

A method of deriving a new property of a drug, implemented as a computing device,
DG relevance data extracting step of loading DG relevance data including one or more related gene information for each of one or more drugs in the computing device or derived from biomedical literature data; And
Drugs with unknown efficacy or side effects, based on the matching rate of gene information extracted from DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information Including; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for
The DG-related data extraction step,
A text loading step of loading text from the biomedical document data;
A sentence extraction step of extracting a text to be analyzed according to a preset rule from the loaded text;
A sentence discrimination step of determining whether the analysis target text contains about or a gene;
A valid word extraction step of extracting valid words related to the relationship between the drug and the gene from the analysis target text; And
Including; a first GRS calculation step of extracting DG related data by deriving the relationship between the drug and the gene from the effective word; and
In the effective word extraction step, a valid word group including valid words including verbs and nouns is extracted from the analyzed text,
The first GRS calculation step,
If the element of the effective word group corresponds to a biological term having a predetermined suppressive meaning, assigning a first parameter value to the effective word of the effective word group;
Assigning a second parameter value to the effective word of the effective word group when the element of the effective word group does not correspond to a biological term of a predetermined suppressive meaning; And
A method for deriving a new characteristic of a drug, comprising the step of deriving a GRS for calculating a gene control score (GRS) for a drug and a gene, based on one or more parameter values assigned to the effective word of the effective word group. delete A method of deriving a new property of a drug, implemented as a computing device,
DG relevance data extracting step of loading DG relevance data including one or more related gene information for each of one or more drugs in the computing device or derived from biomedical literature data; And
Drugs with unknown efficacy or side effects, based on the matching rate of gene information extracted from DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information Including; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for
The weak characteristic extraction step,
A first step of calculating a matching rate between gene information in each of the at least one DG related data and gene information included in each topic of a DG topic;
A second step of deriving one or more representative DG topics based on a matching rate for the DG topic or a score derived from the matching rate of one or more drugs known to have an efficacy or side effect to be identified;
A third step of deriving a new drug for which the target efficacy or side effect is predicted based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic; Including,
The DG topic is a method of deriving a new characteristic of a drug, which is a plurality of topics derived through a topic modeling method by considering each drug in the DG relevance data as a document, and regards genetic information as a word.
delete The method of claim 3,
The second step,
For each of the one or more drugs known to have an efficacy or side effect to be identified, a new drug for deriving at least one DG topic with a score derived from the matching rate or matching rate for the DG topic is equal to or higher than a preset criterion as the representative DG topic. How to derive characteristics.
The method of claim 3,
The representative DG topic includes a plurality of topics,
The third step,
Based on a representative score derived from a score derived from each matching rate or a score derived from the matching rate for a plurality of topics belonging to the representative DG topic of each of the at least one drug that is not known to have the target efficacy or side effect, the determination target A method of deriving new properties of a drug by deriving a new drug with predicted efficacy or side effect among one or more drugs that are not known to have efficacy or side effects.
The method of claim 6,
The representative score is the sum of the scores derived from the respective matching rates or matching rates for a plurality of topics belonging to the representative DG topic of each of the at least one drug that is not known to have an efficacy or side effect to be identified, derives new characteristics of a drug How to.
An apparatus for deriving new properties of a drug comprising at least one memory and at least one processor,
A DG-related data extracting unit for loading DG-related data including one or more related gene information for each of one or more drugs from a memory or derived from biomedical literature data; And
Drugs with unknown efficacy or side effects, based on the matching rate of gene information extracted from DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information It includes a drug characteristic extraction unit that determines whether there is an efficacy or side effect to be determined for,
The DG related data extracting unit,
A text loading unit for loading text from the biomedical document data;
A sentence extracting unit for extracting an analysis target text according to a preset rule from the loaded text;
A sentence discrimination unit for determining whether the analysis target text contains a drug or a gene;
A valid word extracting unit for extracting valid words related to a relationship between a drug and a gene from the analysis target text; And
Including; a first GRS calculation unit for extracting DG related data by deriving the relationship between the drug and the gene from the effective word; and
The effective word extracting unit extracts a valid word group including valid words including verbs and nouns from the analyzed text,
The first GRS calculation unit,
If the element of the effective word group corresponds to a biological term having a predetermined suppressive meaning, assigning a first parameter value to the effective word of the effective word group;
Assigning a second parameter value to the effective word of the effective word group when the element of the effective word group does not correspond to a biological term of a predetermined suppressive meaning; And
A device for deriving a new characteristic of a drug by performing a GRS derivation step of calculating a gene control score (GRS) for a drug and a gene based on one or more parameter values assigned to the effective word of the effective word group.
An apparatus for deriving new properties of a drug comprising at least one memory and at least one processor,
A DG-related data extracting unit for loading DG-related data including one or more related gene information for each of one or more drugs from a memory or derived from biomedical literature data; And
Drugs with unknown efficacy or side effects, based on the matching rate of gene information extracted from DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information It includes a drug characteristic extraction unit that determines whether there is an efficacy or side effect to be determined for,
The weak characteristic extraction unit,
A first step of calculating a matching rate between gene information in each of the at least one DG related data and gene information included in each topic of a DG topic;
A second step of deriving one or more representative DG topics based on a matching rate for the DG topic or a score derived from the matching rate of one or more drugs known to have an efficacy or side effect to be identified;
A third step of deriving a new drug for which the target efficacy or side effect is predicted based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic; Perform,
The DG topic is a device for deriving a new characteristic of a drug, which is a plurality of topics derived through a topic modeling method by considering each drug of the DG relevance data as a document and regards genetic information as a word.
As a computer-readable recording medium,
The computer-readable recording medium stores instructions that cause a computing device to perform the following steps, the steps being:
DG relevance data extracting step of loading DG relevance data including one or more related gene information for each of one or more drugs in the computing device or derived from biomedical literature data; And
Drugs with unknown efficacy or side effects, based on the matching rate of gene information extracted from DG related data of each drug, to one or more DG topics extracted from the DG-related data and constituted by one or more gene information Including; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for
The DG-related data extraction step,
A text loading step of loading text from the biomedical document data;
A sentence extraction step of extracting a text to be analyzed according to a preset rule from the loaded text;
A sentence discrimination step of determining whether the analysis target text contains about or a gene;
A valid word extraction step of extracting valid words related to the relationship between the drug and the gene from the analysis target text; And
Including; a first GRS calculation step of extracting DG related data by deriving the relationship between the drug and the gene from the effective word; and
In the effective word extraction step, a valid word group including valid words including verbs and nouns is extracted from the analyzed text,
The first GRS calculation step,
If the element of the effective word group corresponds to a biological term having a predetermined suppressive meaning, assigning a first parameter value to the effective word of the effective word group;
Assigning a second parameter value to the effective word of the effective word group when the element of the effective word group does not correspond to a biological term of a predetermined suppressive meaning; And
A computer-readable recording medium comprising a GRS derivation step of calculating a gene control score (GRS) for a drug and a gene based on one or more parameter values assigned to the effective word of the effective word group. As a computer-readable recording medium,
The computer-readable recording medium stores instructions that cause a computing device to perform the following steps, the steps being:
DG relevance data extracting step of loading DG relevance data including one or more related gene information for each of one or more drugs in the computing device or derived from biomedical literature data; And
Drugs whose discrimination target efficacy or side effect is unknown based on the matching rate of gene information extracted from DG related data of each drug to one or more DG topics extracted from the DG-related data and constituted by one or more gene information. Including; a drug characteristic derivation step of determining whether there is an effect or side effect to be determined for
The weak characteristic extraction step,
A first step of calculating a matching rate between gene information in each of the at least one DG related data and gene information included in each topic of a DG topic;
A second step of deriving one or more representative DG topics based on a matching rate for the DG topic or a score derived from the matching rate of one or more drugs known to have an efficacy or side effect to be identified;
A third step of deriving a new drug for which the target efficacy or side effect is predicted based on the score derived from the matching rate or the matching rate of one or more drugs that are not known to have the target efficacy or side effects for the representative DG topic; Including,
The DG topic is a plurality of topics derived through a topic modeling method by considering each drug of the DG relevance data as a document, and considering genetic information as a word, a computer-readable recording medium. delete delete

Images