KR20220043297A - Method, System, and Computer-Readable Medium for Predicting Side Effects of Drugs based on Similarity Measurement - Google Patents

Abstract

The present invention relates to a method, system, and computer-readable medium for predicting side effects of drugs based on similarity measurement, and more specifically, to a method, system, and computer-readable medium for predicting side effects of drugs based on similarity measurement, which derives similarity between a plurality of existing drugs known to have a predicted target side effect in order to predict the possibility of occurrence of the predicted side effect for a predicted target drug, and which also derives similarity between the predicted target side effect and the known existing side effect for the predicted target drug, so as to input information based on similarity to a learned inference model unit, thereby predicting the possibility of occurrence of the predicted side effect for the predicted drug.

Description

Method, System, and Computer-Readable Medium for Predicting Side Effects of Drugs based on Similarity Measurement

The present invention relates to a method, a system, and a computer-readable medium for predicting a side effect of a drug based on a similarity measurement, and more particularly, to predict the likelihood of occurrence of a predicted side effect for a predicted drug. Deriving the degree of similarity between a plurality of existing drugs known to be known It relates to a method, a system, and a computer-readable medium for predicting the likelihood of occurrence of a predicted adverse event for a drug, and predicting a side effect of a drug based on a similarity measurement.

Most of the new drug candidates developed in the pharmaceutical industry over the past few decades have not been approved by the US Food and Drug Administration (FDA) or other countries due to the side effects they can cause, so their R&D performance has been poor. In the drug development process, it is difficult to find activities that affect factors other than the target of a new drug candidate drug, which can cause side effects and further lead to drug development failure. As such, most fatal side effects are identified in pre-clinical or clinical trials, but some are discovered during the monitoring process after drug approval. Uncertainty about the potential side effects of these new drugs is an important issue not only for pharmaceutical companies, but also for patients who may be at risk to their health.

On the other hand, the conventional method used to predict the side effects of drugs is based on the assumption that similar drugs have similar properties in terms of chemical and biological properties such as drug structures or drug targets. Therefore, it calculates the similarity of the structure and target of a specific drug with other drugs, and predicts the side effects of a specific drug based on this, or predicts the side effects of a drug by further considering the similarity of the phenotypic information of the drug. methods have been presented.

However, the conventional methods for predicting the side effects of drugs by focusing on the chemical and biological properties of the drugs, when administered to the human body, because the drug is subject to complex effects such as metabolic transformation or pharmacokinetic changes, the side effects of the drug It is difficult to predict and the accuracy of the prediction is inevitably low to simply consider only the chemical and biological properties of drugs to predict the drug. Therefore, in order to predict the side effects caused by the drug, in addition to considering only the structure of the conventional drug or the target of the drug, the similarity of the drug is measured in various aspects to further facilitate the prediction of the possibility of side effects, and the prediction accuracy is also improved There is a need for research on new methods to do this.

The present invention relates to a method, a system, and a computer-readable medium for predicting a side effect of a drug based on a similarity measurement, and more particularly, to predict the likelihood of occurrence of a predicted side effect for a predicted drug. Deriving the degree of similarity between a plurality of existing drugs known to be known It relates to a method, a system, and a computer-readable medium for predicting the likelihood of occurrence of a predicted adverse event for a drug, and predicting a side effect of a drug based on a similarity measurement.

In order to solve the above problems, the present invention is a method for predicting drug side effects, implemented with a computing device, for each of two or more categories between a predicted drug and a plurality of existing drugs that are known to have a predicted target side effect deriving a first similarity value; extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect; deriving each second similarity value between the predicted side effect and a plurality of known side effects for the predicted drug; extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and inputting the first similarity representative information and the second similarity representative information to a learned inference model unit to derive information on the possibility of occurrence of the predicted target side effect for the predicted drug; It provides a method to predict the side effects of

In an embodiment of the present invention, the step of deriving the first similarity value includes, as a first category, a correlation between one or more other drugs that a drug affects or is affected by, as a first category, and the predicted drug affects The first similarity value for the first category between the correlation between one or more other drugs giving or affected can be derived

In one embodiment of the present invention, the step of deriving the first similarity value includes node information on genes and proteins included in an interaction path between a plurality of target genes targeted by the drug as a second category, and , it is possible to derive a first similarity value for the second category between the node information on the prediction target drug and the node information of each of the plurality of existing drugs.

In one embodiment of the present invention, the step of deriving the first similarity value comprises expressing gene information for one or more genes expressing SNP (Single Nucleotide Polymorphism) related to each of one or more diseases related to drugs into a third category. Including, it is possible to derive a first similarity value for the third category between the expressed gene information for the prediction target drug and the expressed gene information for each of the plurality of existing drugs.

In an embodiment of the present invention, the step of deriving the first similarity value includes: target gene information for a plurality of target genes in which the fourth category and the drug target symptom information on one or more symptoms or diseases related to the drug a fifth category, and a first similarity value for the fourth category between the symptom information on the prediction target drug and the symptom information on each of the plurality of existing drugs and target gene information for the prediction target drug and A first similarity value for the fifth category between target gene information for each of the plurality of existing drugs may be derived.

In an embodiment of the present invention, the step of deriving the first similarity value comprises, for each category, an element of intersection between information on a corresponding category of the predicted drug and information on a corresponding category of each of the existing drugs A value obtained by dividing the number by the number of elements in the union may be calculated as the first similarity value.

In an embodiment of the present invention, the step of deriving the second similarity value includes, for each of a plurality of anatomically divided hierarchies, the type of hierarchical element in which the predicted target side effect is expressed, and the layer in which each of the existing side effects is expressed. The second similarity value may be derived based on the type of element.

In an embodiment of the present invention, the step of deriving the second similarity value comprises a set of hierarchical elements in which the predicted target side effect is expressed, and a layer in which each of the existing side effects is expressed, for each of a plurality of anatomically divided hierarchies. calculating a hierarchical similarity value obtained by dividing the number of elements in the intersection between sets of elements by the number of elements in the union; and calculating a value obtained by dividing the sum of the layer similarity values calculated for each of the plurality of layers by the number of the plurality of layers as the second similarity value.

In an embodiment of the present invention, the inference model unit, based on the first similarity representative information and the second similarity representative information, derives information on the first possibility of occurrence of the predicted side effect for the predicted drug. a plurality of first inference models; and a second inference model for deriving information on the second likelihood of the predicted side effect for the predicted drug based on the plurality of first likelihood information derived from each of the plurality of first inference models may include

In an embodiment of the present invention, each of the plurality of first inference models includes a different inference algorithm, and the inference algorithm includes a random forest, a logistic regression, and a naive Bayesian. and XGBoost.

In an embodiment of the present invention, the plurality of first inference models include a plurality of first samples and any one of a plurality of first samples paired with any one first drug and any one known side effect of the first drug. It is learned by the first similarity representative learning information and the second similarity representative learning information for each of a plurality of second samples that pair the second drug with any one side effect that is not known about the second drug, and the first similarity The representative learning information and the second similarity representative learning information may include: deriving a first similarity value for each of two or more categories between a prediction target drug and a plurality of existing drugs known to have a predicted target side effect; extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect; deriving each second similarity value between the predicted side effect and a plurality of known side effects for the predicted drug; It can be derived by; extracting the maximum value from the plurality of second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect.

In order to solve the above problems, the present invention is a system for predicting drug side effects, and a first similarity value is derived for each of two or more categories between a predicted drug and a plurality of existing drugs that are known to have a predicted target side effect. a first similarity derivation unit; a first similarity representative information derivation unit for extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect; a second similarity derivation unit for deriving each second similarity value between the predicted side effect and a plurality of existing side effects known to the predicted drug; a second similarity representative information derivation unit for extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and an expression probability prediction unit for inputting the first similarity representative information and the second similarity representative information to the learned inference model unit to derive information on the possibility of the occurrence of the predicted side effect for the prediction target drug; To provide a system for predicting side effects of drugs.

In order to solve the above problems, the present invention is a computer-readable medium for implementing a method of predicting drug side effects performed in a computing device having one or more processors and one or more memories, a predicted drug and a prediction deriving a first similarity value for each of two or more categories among a plurality of existing drugs known to have a target side effect; extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect; deriving each second similarity value between the predicted side effect and a plurality of known side effects for the predicted drug; extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and inputting the first similarity representative information and the second similarity representative information to a learned inference model unit, and deriving information on the possibility of occurrence of the predicted target side effect with respect to the predicted drug; - Provide readable media.

According to an embodiment of the present invention, since the first similarity value is derived by using the correlation between the drug and the drug as a category for each of the drug to be predicted and a plurality of existing drugs, the prediction target side effect of the drug to be predicted more accurately It is possible to exert an effect that can predict the possibility of manifestation.

According to an embodiment of the present invention, since the first similarity value is derived by using the node information between the plurality of target genes targeted by the drug as a category for each of the drug to be predicted and the plurality of existing drugs, the drug to be predicted more accurately It can exert the effect of predicting the possibility of the occurrence of the predicted target side effect.

According to an embodiment of the present invention, the first similarity value is derived by using the expression gene information in which the SNP is expressed for each of one or more diseases related to the drug for each of the prediction target drug and the plurality of existing drugs as a category, so that the first similarity value is more accurately It can exert the effect of predicting the possibility of the occurrence of the predicted side effect for the predicted drug.

According to an embodiment of the present invention, since the second similarity value is derived between the predicted side effect and a plurality of existing side effects known for the predicted drug, it is possible to more accurately predict the likelihood of the predicted side effect for the predicted drug can have an effect.

According to an embodiment of the present invention, the learned inference model unit for deriving information on the likelihood of occurrence includes a second inference model implemented by ensemble learning in which the result values derived from the plurality of first inference models are input. , it can exert the effect of improving the performance of the inference model unit.

1 schematically illustrates a process for predicting the likelihood of occurrence of a predicted side effect for a predicted drug based on a plurality of databases according to an embodiment of the present invention.
2 schematically illustrates a computing device implementing a method for predicting a side effect of a drug according to an embodiment of the present invention.
3 schematically shows detailed steps of a method for predicting a side effect of a drug according to an embodiment of the present invention.
4 schematically illustrates the interaction between a drug and a drug according to an embodiment of the present invention.
5 schematically shows an interaction pathway between a plurality of target genes targeted by a drug according to an embodiment of the present invention.
6 schematically shows expression gene information for a drug according to an embodiment of the present invention.
7 schematically shows symptom information related to a drug according to an embodiment of the present invention.
8 schematically illustrates hierarchical elements in which side effects are expressed according to a plurality of anatomically divided hierarchies according to an embodiment of the present invention.
9 schematically illustrates a process of deriving first similarity representative information and second similarity representative information from a plurality of first similarity values and a plurality of second similarity values for each category according to an embodiment of the present invention.
10 schematically shows an internal configuration of an inference model unit according to an embodiment of the present invention.
11 schematically illustrates a process in which the first inference model learns from training data according to an embodiment of the present invention.
12 schematically illustrates a process of deriving similarity representative learning information according to an embodiment of the present invention.
13 schematically shows Area Under the ROC Curve (AUC) values for the configuration of various inference model units according to an embodiment of the present invention.
14 schematically illustrates an internal configuration of a computing device according to an embodiment of the present invention.

Hereinafter, various embodiments and/or aspects are disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of one or more aspects. However, it will also be recognized by one of ordinary skill in the art that such aspect(s) may be practiced without these specific details. The following description and accompanying drawings set forth in detail certain illustrative aspects of one or more aspects. These aspects are illustrative, however, and some of various methods may be employed in the principles of the various aspects, and the descriptions set forth are intended to include all such aspects and their equivalents.

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

As used herein, “embodiment”, “example”, “aspect”, “exemplary”, etc. may not be construed as an advantage or advantage in any aspect or design being described over other aspects or designs. . The terms '~part', 'component', 'module', 'system', 'interface', etc. used below generally mean a computer-related entity, for example, hardware, hardware A combination of and software may mean software.

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

Also, terms including an ordinal number, such as first, second, etc., may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. 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 generally understood by those of ordinary skill in the art to which the present invention belongs. have the same meaning as Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in an embodiment of the present invention, an ideal or excessively formal meaning is not interpreted as

1 schematically illustrates a process for predicting the likelihood of occurrence of a predicted side effect for a predicted drug based on a plurality of databases according to an embodiment of the present invention.

As shown in FIG. 1, a plurality of databases for drugs or side effects are used to predict the likelihood of occurrence of a predicted target side effect with respect to a predicted target drug, and the plurality of databases are considered as a method for predicting a conventional side effect. In addition to the database of information on the structure of the drug and the target gene of the drug, the database on the relationship between the drug and the drug, the protein-protein interaction (PPI) network between the multiple target genes of the drug By additionally using a database, a database for genes expressing disease-related Single Nucleotide Polymorphism (SNP), a database for drug indications, and a database for anatomical characteristics of side effects The likelihood of occurrence can be predicted.

The databases shown in FIG. 1 may correspond to a separate external database including the above information, such as DrugBank, Search Tool for the Retrieval of Interacting Genes/Proteins (STRING), PubChem, TTD, CTD, and repoDB. The computing device 1000 implementing the method of predicting drug side effects may receive information directly from the external databases, or may receive information on the external database from a user or an administrator of the computing device 1000 .

Meanwhile, the computing device 1000 derives similarity representative information between the predicted drug and the predicted side effect by using the information of the plurality of databases described above. Specifically, the similarity representative information includes first similarity representative information and second similarity representative information, and the first similarity representative information corresponds to the degree of similarity between the predicted drug and a plurality of existing drugs known to have the predictable side effect. and the second similarity representative information corresponds to a degree of similarity between the predicted side effect and a plurality of existing side effects known for the predicted drug.

As such, the present invention does not simply consider the similarity between the drug and the drug, but also considers the similarity between the side effect to be predicted and the side effect known for the predicted drug, so the expression of the predicted target side effect for the predicted drug It can have the effect of more accurately predicting the possibility.

The predicted drug described in the present invention means a drug that is a target for predicting which side effects may occur, such as a new drug candidate drug, and the predicted target side effect means a side effect to be predicted with respect to the predicted target drug. In addition, the existing drug means a drug that is already known to cause the expected side effect, and the existing side effect means a side effect that is already known to be caused by the predicted drug.

The computing device 1000 derives information on the possibility of occurrence of the predicted side effect with respect to the prediction target drug based on the similarity representative information between the predicted target drug and the predicted target side effect. Specifically, the computing device 1000 inputs the similarity representative information to the machine-learned inference model unit 1700 to derive information on the likelihood of occurrence, and the information on the likelihood of occurrence is , or may include various forms such as quantified information about the possibility of expression.

Hereinafter, a method of predicting a drug side effect implemented in the computing device 1000 will be described in detail.

2 schematically illustrates a computing device 1000 implementing a method for predicting a side effect of a drug according to an embodiment of the present invention.

As shown in FIG. 2 , the computing device 1000 implementing the method for predicting the side effects of other drugs according to an embodiment of the present invention may communicate with an external reference database 2000 .

The reference database 2000 may include a plurality of databases, and the plurality of databases may include a first reference database 2100, a second reference database 2200, and an N-th reference database 2300, As described above, each of the reference databases 2100 to 2300 includes information related to drugs or side effects.

In this way, the reference database 2000 may communicate with the computing device 1000 implementing a method for predicting side effects of drugs, and the computing device 1000 is included in the reference database 2000. information can be accessed.

Meanwhile, the computing device 1000 includes a database processing unit 1100 , a first similarity deriving unit 1200 , a first similarity representative information deriving unit 1300 , a second similarity deriving unit 1400 , and a second similarity representative. It may include an information derivation unit 1500 , a likelihood prediction unit 1600 , an inference model unit 1700 , and a DB 1800 .

Specifically, the database processing unit 1100 accesses the reference database 2000 to receive stored information, or a first similarity value between the predicted drug and the existing drug and the predicted target side effect and the existing information in the computing device 1000 . The received information may be processed to easily derive a second similarity value between side effects. Meanwhile, the database processing unit 1100 may receive information about the reference database 2000 directly from the user and process the corresponding information, or may directly receive processed information from the user.

The first similarity deriving unit 1200 derives a first similarity value corresponding to the degree of similarity between the predicted drug and a plurality of existing drugs known to have the predicted target side effect. Specifically, the first similarity deriving unit 1200 derives a first similarity value for each of the prediction target drug and the plurality of existing drugs according to a plurality of preset categories, and derives a first similarity value according to the category A specific method will be described later.

The first similarity representative information deriving unit 1300 is based on a plurality of first similarity values between the prediction target drug and the plurality of existing drugs derived from the first similarity deriving unit 1200, first similarity representative information to derive Specifically, the first similarity representative information derivation unit 1300 derives the first similarity representative information by extracting a maximum value from among a plurality of first similarity values derived for each category of the prediction target drug and the plurality of existing drugs. Therefore, the first similarity representative information may include first similarity values corresponding to maximum values for each category.

The second similarity derivation unit 1400 derives a second similarity value corresponding to the degree of similarity between the predicted target side effect and a plurality of existing side effects known for the predicted target drug. Specifically, the second similarity derivation unit 1400 derives a second similarity value for each of the predicted target side effect and the plurality of existing side effects according to a preset category. A detailed method of deriving the second similarity value will be described later. to do it

The second similarity representative information derivation unit 1500 is configured to obtain second similarity representative information based on a plurality of second similarity values between the predicted target side effect and the plurality of existing side effects derived from the second similarity derivation unit 1400 . to derive Specifically, the second similarity representative information derivation unit 1500 derives the second similarity representative information by extracting a maximum value among a plurality of second similarity values derived for each of the predicted target side effect and the plurality of existing side effects. Therefore, the second similarity representative information may include a second similarity value corresponding to the maximum value of the preset category.

On the other hand, the above-described first similarity deriving unit 1200, the first similarity representative information deriving unit 1300, the second similarity deriving unit 1400 and the second similarity representative information deriving unit 1500 are predicted In addition to deriving the information input to the inference model unit 1700 in order to derive the possibility of occurrence of the predicted target side effect for the target drug, the first similarity representative corresponding to the learning data for the inference model unit 1700 to learn Learning information and second similarity representative learning information may be derived.

The likelihood prediction unit 1600 is based on the first similarity representative information and the second similarity representative information derived with respect to the predicted drug and the predicted target side effect, the expression of the predicted target side effect with respect to the predicted drug Derive information about possibilities. Specifically, the likelihood prediction unit 1600 may derive information on the likelihood by inputting the first similarity representative information and the second similarity representative information to an inference model unit 1700 to be described later.

The inference model unit 1700 receives the first similarity representative information and the second similarity representative information for the predicted drug and the predicted side effect from the likelihood prediction unit 1600, and derives information on the likelihood of occurrence. do. Specifically, the inference model unit 1700 may derive information on the possibility of occurrence based on a preset rule, but preferably, the inference model unit 1700 is a machine learning algorithm that performs learning by a predetermined method. Based on the above, information on the possibility of expression is derived.

Meanwhile, in FIG. 2 , the inference model unit 1700 is illustrated as being included in the computing device 1000 , but the present invention is not limited thereto. In another embodiment of the present invention, the inference model unit 1700 is the computing device It may be included in another server or other computing device separate from 1000, and the likelihood prediction unit 1600 calls the inference model unit 1700 included in the other server or other computing device to express the expression. information about the possibilities can be derived.

The DB 1800 may store information necessary for predicting the likelihood of occurrence of a predicted side effect for the predicted drug, and specifically, a plurality of first similarity values and a plurality of second similarity values for the predicted drug and the predicted target side effect. It is possible to store similarity representative information including a similarity value, first similarity representative information and second similarity representative information, and also learning data for the inference model unit 1700 to learn, specifically, first similarity representative learning information and It is possible to store the learning data including the second similarity representative learning information. Additionally, the DB 1800 may store the prediction result derived from the likelihood prediction unit 1600 , that is, information on the likelihood.

3 schematically shows detailed steps for a method for predicting a side effect of a drug according to an embodiment of the present invention.

As shown in FIG. 3 , as a method for predicting drug side effects, implemented by the computing device 1000 , a first for each of two or more categories between a predicted drug and a plurality of existing drugs that are known to have a predicted target side effect 1 deriving a similarity value (S100); extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect (S110); deriving each second similarity value between the predicted side effect and a plurality of existing side effects known to the predicted drug (S120); extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect (S130); and inputting the first similarity representative information and the second similarity representative information into the learned inference model unit 1700, and deriving information on the possibility of occurrence of the predicted target side effect for the predicted drug (S140) ; may be included.

Specifically, the computing device 1000 implementing a method for predicting drug side effects includes a specific drug (prediction target drug) for predicting side effects from a user and a specific side effect (prediction target side effect) to predict for the predicted drug information can be entered.

Thereafter, the first similarity derivation unit 1200 of the computing device 1000 derives a first similarity value for each of a plurality of existing drugs known to be capable of expressing the predicted target drug and the predicted target side effect ( S100) is performed.

Specifically, in the step of deriving the first similarity value ( S100 ), the first similarity value for each category is derived according to two or more categories preset for each of the prediction target drug and the plurality of existing drugs. Accordingly, the types of first similarity values for the prediction target drug and the specific existing drug may be derived for each of two or more preset categories, and in each category, a plurality of first similarity values corresponding to the number of a plurality of existing drugs are derived. can be

That is, in the step of deriving the first similarity value (S100), the similarity between the predicted drug and other drugs may be measured in order to derive information on the possibility of occurrence of the predicted target side effect for the predicted drug.

On the other hand, the information on the existing drug known to have the predicted side effect may be input in advance into the computing device 1000 by the user, or the information may be accessed or received by an external database such as the reference database 2000 . .

The first similarity representative information deriving unit 1300 performs a step of deriving the first similarity representative information based on the plurality of first similarity values derived from the first similarity deriving unit 1200 ( S110 ).

Specifically, in the step of deriving the first similarity representative information ( S110 ), a first similarity value having a maximum value among a plurality of first similarity values derived for each category is derived as the first similarity representative information. Accordingly, the first similarity representative information may include first similarity values having a maximum value for each category.

In this way, the first similarity representative information derived through the first similarity representative information deriving unit 1300 may be included in the input information to be input to the inference model unit 1700, and specifically, prediction among the input information. It may correspond to an input factor for the degree of similarity between the target drug and a plurality of existing drugs known to have predicted target side effects.

On the other hand, the second similarity derivation unit 1400 performs the step of deriving a second similarity value for each of a plurality of existing side effects known to be expressed by the predicted target side effect and the predicted target drug (S120). do.

Specifically, in the step of deriving the second similarity value ( S120 ), a second similarity value for a specific category is derived for the predicted side effect and each of the plurality of existing side effects. Accordingly, a plurality of second similarity values corresponding to the number of a plurality of existing side effects may be derived from the specific category.

That is, in the step of deriving the second similarity value ( S120 ), the degree of similarity between the predicted side effect and other side effects may be measured in order to derive information on the possibility of occurrence of the predicted side effect for the predicted drug.

On the other hand, information on a plurality of existing side effects known to the prediction target drug is input in advance into the computing device 1000 by the user, or the information can be accessed or received from an external database such as the reference database 2000 . can

The second similarity representative information deriving unit 1500 performs a step of deriving second similarity representative information based on the plurality of second similarity values derived from the second similarity deriving unit 1400 ( S130 ).

Specifically, in the step of deriving the second similarity representative information ( S130 ), a second similarity value having a maximum value among a plurality of second similarity values is derived as the second similarity representative information, and thus the second similarity representative information The second similarity representative information derived through the derivation unit 1500 may be included in the input information to be input to the inference model unit 1700, and specifically, it is expressed by the side effect to be predicted and the drug to be predicted among the input information. It may correspond to an input factor for the degree of similarity with a plurality of existing side effects known to be possible.

In this way, the first similarity representative information and the second similarity representative information are obtained by measuring the degree of similarity between the predicted target drug and the existing drugs and the similarity between the predicted target side effect and the existing side effects through the steps S100 to S130. and, based on this, information on the likelihood of the predicted side effect for the predicted drug is derived from the possible predictive unit, so compared to the conventional method for estimating side effects, the degree of similarity for various categories is taken into account. It is possible to derive information about the possibility of manifestation.

Meanwhile, through the steps S100 to S130, the computing device 1000 does not derive only the first similarity representative information and the second similarity representative information, but the learning data and the learning data for the inference model unit 1700 to learn. It is also possible to derive verification data for verifying the inference model unit 1700 learned by .

Specifically, the learning data may include first similarity representative learning information and second similarity representative learning information for a plurality of samples pairing a specific drug and a specific side effect. In S110, first similarity representative information may be derived by using the specific drug of the sample as the predicted target drug, and using the specific side effect of the sample as the predicted target side effect, and the derived first similarity representative information is the first It may correspond to similarity representative learning information. Similarly, in the above-described steps S120 to S130, second similarity representative learning information for the specific drug and the specific side effect may be derived, and the derived second similarity representative information may correspond to the second similarity representative learning information. can

In addition, as described above, the verification data may include first similarity representative verification information and second similarity representative verification information for a plurality of samples pairing a specific drug and a specific side effect, which is described above in steps S100 to It can be derived through step S130.

Finally, the likelihood predicting unit 1600 inputs the derived first similarity representative information and the second similarity representative information to the learned inference model unit 1700 to predict the likelihood of a predicted side effect for the predicted drug. A step of deriving information on (S140) is performed.

In the step of deriving the information on the likelihood (S140), the result value derived through the learned inference model unit 1700 is used as the information on the likelihood or a separate possibility based on the result value. information can also be derived.

4 schematically illustrates the interaction between a drug and a drug according to an embodiment of the present invention.

As shown in FIG. 4 , the step of deriving the first similarity value ( S100 ) includes, as a first category, a correlation between one or more other drugs that the drug affects or is affected by, as a first category, and the prediction target A first for the first category between the correlation between one or more other drugs on which the drug affects or is affected, and the correlation between one or more other drugs on which each of the plurality of existing drugs affects or is affected A similarity value can be derived.

Specifically, in the step of deriving the first similarity value (S100), a drug that affects one or more other drugs, or a correlation that is affected by one or more other drugs is a first category, and the first A plurality of first similarity values for each of the drug to be predicted and the plurality of existing drugs for the category are derived.

The first category for the correlation between drugs and other drugs (DDIs-D) corresponds to a category in consideration of the tendency of similar drugs to exhibit similar reactions to other drugs. As shown in Figure 4, the correlation for each drug can be expressed in the form (x, y, z), x means other drugs having a correlation, y means the type of correlation, z indicates whether it affects or is affected by other drugs.

For example, among the correlations for drug A, (B,1,←) means that drug A is affected by type 1 by drug B. As such, the computing device 1000 stores information on the correlation for each drug based on the information of the external database, and the step of deriving the first similarity value (S100) is the correlation with the drug to be predicted. It is possible to derive a plurality of first similarity values for the first category corresponding to the degree of similarity in the correlation with respect to each of a plurality of existing drugs that are known to be capable of exhibiting a predicted target side effect.

5 schematically shows an interaction pathway between a plurality of target genes targeted by a drug according to an embodiment of the present invention.

As shown in FIG. 5 , in the step of deriving the first similarity value (S100), node information on genes and proteins included in an interaction path between a plurality of target genes targeted by a drug is added to the second category. including, and it is possible to derive a first similarity value for the second category between node information for the prediction target drug and node information for each of the plurality of existing drugs.

Specifically, in the step of deriving the first similarity value (S100), the drug is administered and node information on the protein, gene, or metabolite included in the interaction path between a plurality of target genes targeted in the human body is a second category, and a plurality of first similarity values for each of a drug to be predicted and a plurality of existing drugs for the second category are derived.

The second category for node information (DDIs-N) of the interaction path between target genes of drugs corresponds to a category that considers the functional similarity in the human body between two drugs. As shown in Figure 5, each drug has one or more target genes (TG _A and TG _B ) that have an initial effect, and the target gene (TG _A and TG _B ) that is finally affected from the first target gene (TG A and TG B) The effect of the drug can propagate along certain network pathways up to TG _C and TG _D ).

Meanwhile, in the network path between target genes, proteins, genes, or metabolites (P1 to P7) corresponding to nodes exist, and the first similarity value for the second category is the first target gene ( It may correspond to the degree of similarity of information on nodes (P1 to P7) included in an interaction path between TG _A and TG _B ) and final target genes (TG _C and TG _D ).

For example, drug A shown in FIG. 5 uses TG _A and TG _B as initial target genes, and the initial target gene TG _A can interact with the final target gene, TG _C , through nodes P1, P4, and P6. , the initial target gene TG _B can interact with the final target gene TG _D via nodes P3, P4, P5, and P7. Therefore, the node information of drug A includes information on proteins, genes, or metabolites constituting the pathway between the initial target gene and the final target gene as described above.

Therefore, node information for each drug is stored in the computing device 1000 based on information from an external database, and in the step of deriving the first similarity value (S100), node information of the predicted drug and the predicted target side effect are expressed It is possible to derive a plurality of first similarity values for the second category corresponding to the similarity of node information for each of a plurality of existing drugs known to be available.

In another embodiment of the present invention, the node information further includes information on a protein, gene, or metabolite corresponding to a path between the initial target gene and the final target gene, and information on the first target gene or the final target gene. can do.

In another embodiment of the present invention, a plurality of paths may exist from the initial target gene to the final target gene, and in this case, the shortest path among the paths from the initial target gene to the final target gene may be used as node information.

The target gene in the present invention may refer to a specific gene in the human body, but is not limited thereto, and may include the concept of a specific protein or a specific metabolite.

On the other hand, as shown in FIG. 5, the step of deriving the first similarity value (S100) includes target gene information for a plurality of target genes targeted by the drug as a fifth category, and the prediction target drug It is possible to derive a first similarity value for the fifth category between the target gene information for the drug and the target gene information for each of the plurality of existing drugs.

Specifically, in the step of deriving the first similarity value (S100), target gene information (Target) for a plurality of target genes targeted by the drug is a fifth category, and a drug to be predicted for the fifth category and deriving a plurality of first similarity values for each of the plurality of existing drugs.

The fifth category for a plurality of target genes of a drug corresponds to a category considering the biological properties of the drug, that is, when two drugs are related to the same target gene, the two drugs undergo similar biological processes and thus tend to have the same side effects. .

Accordingly, information on a plurality of target genes for each drug is stored in the computing device 1000 based on information from an external database, and the step of deriving the first similarity value (S100) is target gene information of the drug to be predicted. and a plurality of first similarity values for the fifth category corresponding to the degree of similarity of target gene information for each of a plurality of existing drugs known to be capable of expressing and predictable side effects may be derived.

Meanwhile, as described above, the target gene information may include information on the initial target genes (TG _A and TG _B ) and the final target genes (TG _C and TG _D ), but preferably, the target gene information It may include only information about the original target genes (TG _A and TG _B ).

6 schematically shows expression gene information for a drug according to an embodiment of the present invention.

As shown in Figure 6, the step of deriving the first similarity value (S100), SNP (Single Nucleotide Polymorphism) related to each of the one or more diseases related to the drug SNP (Single Nucleotide Polymorphism) is expressed for one or more genes expressed gene information It is included in three categories, and a first similarity value for the third category can be derived between the expression gene information for the prediction target drug and the expression gene information for each of the plurality of existing drugs.

Specifically, in the step of deriving the first similarity value (S100), the expression gene information for one or more genes in which SNPs related to each of one or more diseases related to drugs are expressed as a third category, and in the third category A plurality of first similarity values for each of the prediction target drug and the plurality of existing drugs are derived.

The third category for the expressed gene information (SNP) of a drug corresponds to a category considering that two drugs may have similar side effects when the genetic characteristics of diseases related to them are similar. SNP, that is, single nucleotide polymorphism (SNP) corresponds to a factor that determines individual diversity of humans, and SNP appears approximately every 1000 bases in the human genome, thereby expressing individual genetic diversity. Specifically, a person with C (cytosine) may exist at a specific position in the DNA chain and a person with A (adenine) may exist, and individual diversity may be expressed by this difference. The genetic diversity expressed by SNPs in this way can also be utilized in terms of susceptibility to diseases. The susceptibility to a specific disease may be different depending on the genetic diversity of each person.

Accordingly, as shown in FIG. 6 , for each drug (Drug A), one or more diseases (disease 1 to disease N) associated with the corresponding drug (Drug A) exist, and SNPs related to sensitivity may exist, and genes in which each SNP can be expressed (Gene1 to Gene2), that is, an expression gene for each drug may exist.

Accordingly, information on a plurality of expressed genes for each drug is stored in the computing device 1000 based on information from an external database, and the step of deriving the first similarity value (S100) is the expression gene of the predicted drug. It is possible to derive a plurality of first similarity values for the third category corresponding to the degree of similarity of expression gene information for each of a plurality of existing drugs known to be capable of expressing information and a predicted target side effect.

7 schematically shows symptom information related to a drug according to an embodiment of the present invention.

7, the step of deriving the first similarity value (S100) includes symptom information on one or more symptoms or diseases related to the drug as a fourth category, and symptom information on the predicted drug and a first similarity value for the fourth category between symptom information for each of the plurality of existing drugs may be derived.

Specifically, in the step of deriving the first similarity value (S100), symptom information on an indication corresponding to one or more symptoms or diseases related to a drug as a fourth category, and prediction of the fourth category A plurality of first similarity values for each of the target drug and the plurality of existing drugs are derived.

The fourth category for symptom information of drugs corresponds to a category that considers the similarity of phenotypic characteristics between drugs, and indication means a disease or symptom for which a therapeutic effect is expected by a drug.

Accordingly, as shown in FIG. 7 , symptom information on one or more symptoms or diseases related to the corresponding drug may exist for each drug. For example, symptom information for Drug A shown in FIG. 7 may include disease A, disease B, symptom A, and symptom D. As such, the computing device 1000 stores symptom information for each drug based on the information of the external database, and the step of deriving the first similarity value (S100) includes the symptom information on the drug to be predicted and the side effect to be predicted. It is possible to derive a plurality of first similarity values for the fourth category corresponding to the similarity of symptom information for each of a plurality of existing drugs known to be expressed.

In the step of deriving the first similarity value ( S100 ) as in the above-described FIGS. 4 to 7 , the first to fifth categories are included in the two or more categories, and the first to fifth categories A plurality of first similarity values may be derived for each. In this way, the first similarity representative information can be derived based on the plurality of first similarity values derived for each category, and the step of deriving the first similarity representative information ( S110 ) will be described in detail with reference to FIG. 9 . .

On the other hand, in another embodiment of the present invention, the step of deriving the first similarity value (S100) includes information on the chemical structure of the drug as a sixth category, information on the chemical structure of the predicted drug, and A first similarity value for the sixth category between information on the chemical structure of each of the plurality of existing drugs may be additionally derived.

Accordingly, in an embodiment of the present invention, since the degree of similarity between a drug and a drug is measured for each of a plurality of categories, it is possible to exhibit the effect of more accurately predicting the side effects of the drug.

8 schematically illustrates hierarchical elements in which side effects are expressed according to a plurality of anatomically divided hierarchies according to an embodiment of the present invention.

As shown in FIG. 8 , in the step of deriving the second similarity value ( S120 ), the type of the hierarchical element in which the predicted target side effect is expressed and each of the existing side effects is determined for each of a plurality of anatomically divided hierarchies. The second similarity value may be derived based on the type of the hierarchical element to be expressed.

Specifically, as shown in FIG. 8A , each side effect may be classified according to an anatomically classified predetermined hierarchy. In Figure 8 (A), it was divided into three layers, organs (Organs), sub-systems (Subsystems), and systems (Systems), (B) of Figure 8 (B) is anatomically classified side effects are expressed for each predetermined layer. The types of hierarchical elements are shown.

A plurality of hierarchical elements may exist according to an anatomically classified predetermined hierarchy. For example, hierarchical elements such as heart, liver, and lungs may exist in the Organs hierarchy, and hierarchical elements such as respiratory system, muscular system, and digestive system may exist in the Systems hierarchy. On the other hand, each side effect may affect specific hierarchical elements for each anatomically classified layer. For example, as shown in (B) of FIG. 8, side effect 1 may affect the hierarchical elements A1, A3, and A4 of the corresponding layer in the Organs layer, and the Subsystems layer may affect hierarchical elements B2 and B3 of the corresponding layer, and may affect C1, C4, and C5, which are hierarchical elements of the corresponding layer, in the Systems layer. In this way, the classification of side effects into hierarchical elements for each hierarchical classification that is anatomically classified considers that similar side effects are related to similar anatomical levels.

On the other hand, for example, if a specific side effect affects the lung, which is a hierarchical element of the organ hierarchy, and also affects the respiratory system to which the lung belongs, it is not classified as affecting only the respiratory system, but each is classified independently. Therefore, specific side effects can be classified as those that affect the lungs at the Organs layer and the respiratory system at the Systems layer, respectively.

Accordingly, types of hierarchical elements in which the corresponding side effects can be expressed are stored in the computing device 1000 for each of a plurality of anatomically divided layers for each side effect based on information of an external database, and the second similarity value The step of deriving (S120) is a plurality of second similarity values corresponding to the type of the hierarchical element of the predicted side effect and the similarity of the type of the hierarchical element for each of a plurality of existing side effects known to be expressed by the predicted drug. can be derived.

Meanwhile, although three anatomically divided layers are illustrated in FIG. 8 , the present invention is not limited thereto, and two or more anatomically divided layers may be used in another embodiment of the present invention.

9 schematically illustrates a process of deriving first similarity representative information and second similarity representative information from a plurality of first similarity values and a plurality of second similarity values for each category according to an embodiment of the present invention.

FIG. 9A corresponds to a diagram showing first similarity values for each of two or more categories for a predicted drug and each of the plurality of existing drugs. Specifically, as shown in the upper left of Figure 9 (A), in order to predict whether the predicted target side effect (green diamond) can be expressed by the predicted target drug (red drug), as shown at the bottom, It is possible to derive a first similarity value for each of two or more categories between the four existing drugs that are known to be capable of expressing the predicted side effect (green diamond) and the predicted drug (red-based drug).

To this end, the step of deriving the first similarity value (S100) is, for each category, the number of elements of intersection between the information on the corresponding category of the predicted drug and the information on the corresponding category of each of the existing drugs. A value obtained by dividing the number of elements in the union may be calculated as the first similarity value.

Specifically, in the step of deriving the first similarity value ( S100 ), the first similarity value for each category between the predicted drug and the existing drug may be calculated using Equation (1) below.

- 식 (1)

- Equation (1)

Equation (1) above is to calculate the first similarity value for drug D _A and drug D _B. For a specific category, the drug D _A information set S _A and the drug D _B information set S _B For a given category, the number of elements in the intersection of sets S _A and S _B divided by the number of elements in the union of sets S _A and S _B corresponds to the first similarity value between drug D _A and drug D _B for that specific category. The range of the first similarity value has a value of 0 when the two drugs are not completely similar, and a value of 1 when the two drugs are completely similar.

Preferably, the above formula (1) can be used to calculate the first similarity value for the first to fifth categories described above, and for the sixth category for the chemical structure of the drug, Tanimoto coefficient (Tanimoto coefficient) ) can also be calculated through a separate formula for calculating

Therefore, as shown in FIG. 9A , the first similarity values of the predicted drug and existing drugs for the first category (DDIs-D) are 0.23, 0.87, 0.79, 0.44 in order from the top. was calculated, and the first similarity values of the predicted drug and existing drugs for the fifth category (Target) were calculated to be 0.45, 0.71, 0.24, and 0.95 in order from the top.

Meanwhile, in the step of deriving the first similarity representative information ( S110 ), the first similarity representative information is derived by extracting the maximum value from among the plurality of first similarity values derived for each category as described above. In the first similarity representative information shown in FIG. 9A , the maximum value of 0.87 among a plurality of first similarity values for a first category (DDIs-D) is a plurality of first similarity values for a fifth category (Target). A maximum value of 0.95 among the similarity values may be included, and a maximum value of 0.86 among the plurality of first similarity values for the sixth category may be included.

FIG. 9B corresponds to a diagram illustrating a second similarity value for each of a predicted side effect and a plurality of existing side effects. Specifically, as shown on the left side of FIG. 9 (B), in order to predict whether the predicted target side effect (green diamond) can be expressed by the predicted target drug (red drug), the predicted target side effect (green diamond) ) and a plurality of existing side effects (blue, purple, yellow, pale yellow diamonds) known to be expressed by the predicted drug (red drug) can be derived from a second similarity value.

To this end, the step of deriving the second similarity value ( S120 ) includes a set of hierarchical elements in which the predicted target side effect is expressed and a hierarchical element in which each of the existing side effects is expressed for each of a plurality of anatomically divided hierarchies. calculating a hierarchical similarity value obtained by dividing the number of elements in the intersection between sets by the number of elements in the union; and calculating a value obtained by dividing the sum of the layer similarity values calculated for each of the plurality of layers by the number of the plurality of layers as the second similarity value.

Specifically, in the step of deriving the second similarity value ( S120 ), the second similarity value between the predicted side effect and the existing side effect may be calculated using Equation (2) below.

- 식 (2)

- Equation (2)

In Equation (2), S _l (SE _a ,SE _b ) may be calculated by Equation (1) described above, and l in Equation (2) corresponds to a specific layer among a plurality of anatomically divided layers. and n corresponds to the number of anatomically divided plurality of layers. In the case of calculating the second similarity value with respect to FIG. 8, n may correspond to 3. Similarly, the range of the second similarity value calculated by Equation (2) may have a value of 0 or more and 1 or less.

Accordingly, as shown in FIG. 9B , the second similarity values of the predicted side effects and the existing side effects were calculated to be 0.33, 0.44, 0.5, and 0.11 in order from the top.

Meanwhile, in the step of deriving the second similarity representative information ( S130 ), the second similarity representative information is derived by extracting a maximum value from among the plurality of second similarity values derived as described above. The second similarity representative information shown in FIG. 9B may include a maximum value of 0.5 among a plurality of second similarity values.

Accordingly, as described in FIGS. 9A and 9B , the first similarity representative information and the second similarity representative between the predicted drug (red drug) and the predicted target side effect (green diamond) The similarity representative information including the information may be derived as shown in FIG. 9C . The similarity representative information may be input to the learned inference model unit 1700, and the inference model unit 1700 determines the possibility of occurrence of the corresponding predicted target drug with respect to the corresponding predicted target drug based on the received similarity representative information. information can be derived.

The 'Class' item described in (C) of FIG. 9 corresponds to a label assigned when the similarity representative information is used as training data or verification data for learning or verification of the inference model unit 1700, and 'Class' The 'False' value of the ' item may correspond to a value labeled when the relationship between the predictive drug and the predicted adverse effect is unknown, and the 'Ture' value of the 'Class' item is the corresponding predictive drug and the corresponding predictive drug. It may correspond to a labeled value if the relationship of the predicted adverse event is known.

In this way, the similarity representative information derived through the steps S100 to S130 may be used as input information for predicting the possibility of the occurrence of the predicted target side effect for the corresponding predicted drug, and derived through the steps S100 to S130 The similarity representative information to be obtained may correspond to similarity representative learning information or similarity representative verification information for learning or verification of the inference model unit 1700, and the similarity representative learning information and the similarity representative verification information include the drug and its side effects Relationships between them may be labeled.

10 schematically shows an internal configuration of an inference model unit 1700 according to an embodiment of the present invention.

10, the inference model unit 1700 includes a plurality of first inference models, each of the plurality of first inference models includes a different inference algorithm, and the inference algorithm is a random Forest (Random forest), Logistic regression (Logistic regression), Naive Bayesian (Naive Bayesian) and may include XGBoost.

Specifically, (A) of FIG. 10 shows the result values derived by inputting similarity representative information for the predicted drug and the predicted target side effect to the plurality of first inference models. The plurality of first inference models are different inference algorithms, specifically, a random forest, logistic regression, and naive Bayesian, each of which is an algorithm for classifying input information, of the plurality of first inference models. ) and XGBoost's inference algorithm.

In FIG. 10(A), four first inference models are shown, and the algorithms of each first inference model are described as random forest, logistic regression, naive Bayesian, and XGBoost. However, in another embodiment of the present invention, the present invention is not limited thereto, and the inference model unit 1700 may include two or more first inference models, and the inference algorithm of the first inference model is a random forest. , logistic regression, naive Bayesian, and XGBoost may include various inference algorithms for classifying input information.

On the other hand, as shown in (B) of FIG. 10, the inference model unit 1700, based on the first similarity representative information and the second similarity representative information, of the predicted side effect for the predicted drug a plurality of first inference models for deriving information on first feasibility; and a second inference model for deriving information on the second likelihood of the predicted side effect for the predicted drug based on the plurality of first likelihood information derived from each of the plurality of first inference models may include

Specifically, the inference model unit 1700 may be implemented as an ensemble learning method, and thus each of the plurality of first inference models may correspond to a classifier, and the second inference model may correspond to a meta-classifier. That is, it includes a plurality of first inference models and a second inference model, and each of the plurality of first inference models receives similarity representative information about the predicted drug and the predicted adverse effect, and receives information on the first likelihood, respectively. derived, and the second inference model may receive information on a plurality of first feasibility derived by the plurality of first inference models, respectively, and derive information on the second feasibility.

The information on the second likelihood may correspond to the information on the likelihood derived from the above-described likelihood predictor 1600 , or the information on the second likelihood in the likelihood predictor 1600 . It is also possible to derive additional information on the possibility of expression based on the above.

Meanwhile, the inference model unit 1700 of the present invention may apply various conventional methods such as majority voting and majority voting to perform ensemble learning.

11 schematically illustrates a process in which the first inference model learns from training data according to an embodiment of the present invention.

As shown in FIG. 11 , the plurality of first inference models include a plurality of first samples and any one of a first drug paired with any one side effect known to the first drug. It can be learned by the first similarity representative learning information and the second similarity representative learning information for each of a plurality of second samples that pair the second drug and any one side effect that is not known with respect to the second drug.

Specifically, the first inference model includes first similarity representative learning information and second similarity representative learning information for a plurality of first samples, and first similarity representative learning information for a plurality of second samples and Learning may be performed using the similarity representative learning information including the second similarity representative learning information as learning data.

The first sample corresponds to a sample in which a first drug, which is a specific drug, and any one specific side effect already known for the first drug are paired, and the similarity representative learning information for the first sample is the above-described step It may be derived in the same way as the method of deriving the similarity representative information through steps S100 to S130. Meanwhile, the similarity representative learning information for the first sample may further include a labeling value such as 'Ture' when the first inference model learns in a supervised learning method.

On the other hand, the second sample corresponds to a sample in which a second drug, which is a specific drug, and any one specific side effect not known to the second drug are paired, and the similarity representative learning information for the second sample is described above. It can be derived in the same way as the method of deriving similarity representative information through steps S100 to S130. Similarly, the similarity representative learning information for the second sample may further include a labeling value such as 'False' when the first inference model learns in a supervised learning method.

In another embodiment of the present invention, the verification data for verifying the first inference model may include similarity representative verification information for the plurality of first samples and similarity representative verification information for the plurality of second samples, and , the similarity representative verification information may be derived in the same way as the method of deriving the similarity representative information through the above-described steps S100 to S130.

In addition, the test data for testing the first inference model may include similarity representative test information for the plurality of first samples and similarity representative test information for the plurality of second samples, and the similarity representative test information can be derived in the same way as the method of deriving the similarity representative information through the above-described steps S100 to S130.

In this way, when the training data, verification data, and test data are derived with respect to the first inference model, each data may be derived at a predetermined ratio, and preferably, a plurality of data included in the training data, verification data, and test data A ratio of the first sample and the plurality of second samples may be 8:1:1, respectively.

12 schematically illustrates a process of deriving similarity representative learning information according to an embodiment of the present invention.

Specifically, in FIG. 12(A), a plurality of drugs already known for a specific side effect (green triangle), a plurality of drugs paired with the specific side effect (green triangle) in the learning data, and corresponding in the test or verification data Multiple drugs paired with specific side effects (green triangles) are shown.

On the other hand, in (B) of FIG. 12, in order to derive similarity representative information between a specific drug (red drug) and the specific side effect (green triangle) included in the learning data, the specific side effect (green triangle) is already known. A plurality of first similarity values are derived for each of a plurality of categories between a plurality of drugs and the specific drug (red drug), and a maximum value among the plurality of first similarity values for each category is extracted as the first similarity representative information of the corresponding category. do.

On the other hand, as shown in (B) of FIG. 12 , if the relationship between the specific drug (red drug) and the specific side effect (green triangle) is already known, first between the specific drug (red drug) A similarity value may not be derived.

In addition, although not shown in (B) of FIG. 12 , a plurality of second similarity values between the specific side effect (green triangle) and a plurality of side effects already known for the specific drug (red drug) can be derived. , a maximum value among the plurality of second similarity values may be extracted as the second similarity representative information. Similarly, when the relationship between the specific drug (red drug) and the specific side effect (green triangle) is already known, the second similarity value between the specific side effect (green triangle) may not be derived.

Similarly, as shown in FIG. 12C , similarity representative information for test data or verification data can also be derived in a similar way.

13 schematically shows Area Under the ROC Curve (AUC) values for various configurations of the inference model unit 1700 according to an embodiment of the present invention.

In FIG. 13A , each inference model unit 1700 according to whether the configuration of a plurality of categories in the step of deriving the first similarity value ( S100 ) and the step of deriving the second similarity value ( S120 ) is included. ) corresponds to a diagram showing Area Under the ROC Curve (AUC) values.

Specifically, 'Base' in FIG. 13(A) corresponds to the inference model unit 1700 considering only the similarity to the fifth category (Target) and the sixth category (chemical structure of the drug) used in the conventional method. and 'Base + SE-AH' corresponds to the inference model unit 1700 that additionally considers the similarity between the predicted side effect and the existing side effect in the configuration for 'Base', and 'Base + DDIs-N' is 'Base' Corresponds to the inference model unit 1700 that additionally considers the similarity to the above-described second category in the configuration for , and 'Base + SNPs' is the inference that additionally considers the similarity to the above-described third category in the configuration for 'Base' Corresponds to the model unit 1700, 'Base + DDIs-D' corresponds to the inference model unit 1700 that additionally considers the similarity to the above-described first category in the configuration for 'Base', and 'All features' It corresponds to the inference model unit 1700 taking into account all the categories and similarities between the predicted side effects and the existing side effects.

As for the AUC value of the inference model unit 1700 according to each configuration shown in FIG. 13A, the highest value was derived in the case of 'All features', and thus compared to the conventional method ('Base'). It can be confirmed that predicting side effects in consideration of various similarity items in the invention can predict side effects more effectively.

13B shows a case in which the inference model unit 1700 includes only a single first inference model (Proposed method (RF)) and the inference model unit 1700 includes a plurality of first inference models and a second Corresponds to the figure showing the AUC value for the case including the inference model (Proposed method (stacking)).

As shown in (B) of FIG. 13 , a plurality of first and second inference models are used in an ensemble learning method than when the inference model unit 1700 predicts side effects using only a single first inference model. It can be confirmed that side effects can be predicted more effectively if included.

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

The computing device 1000 illustrated in FIG. 2 described above may include components of the computing device 11000 illustrated in FIG. 14 .

14, the computing device 11000 includes at least one processor 11100, a memory 11200, a peripheral interface 11300, an input/output subsystem ( It may include at least an I/O subsystem) 11400 , a power circuit 11500 , and a communication circuit 11600 . In this case, the computing device 11000 may correspond to the computing device 1000 illustrated in FIG. 2 .

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

In this case, access to the memory 11200 from other components such as the processor 11100 or the peripheral device interface 11300 may be controlled by the processor 11100 .

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

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

The power circuit 11500 may supply power to all or some 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 a power source. It may include 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, if necessary, the communication circuit 11600 may transmit and receive an RF signal, also known as an electromagnetic signal, including an RF circuit, thereby enabling communication with other computing devices.

This embodiment of FIG. 14 is only an example of the computing device 11000, and the computing device 11000 may omit some components shown in FIG. 14, or further include additional components not shown in FIG. 14, or 2 It may 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 may include various communication methods (WiFi, 3G, LTE) in the communication circuit 11600 . , Bluetooth, NFC, Zigbee, etc.) may include a circuit for RF communication. Components that may be included in the computing device 11000 may be implemented as hardware, software, or a combination of both hardware and software including an integrated circuit specialized for one or more signal processing or applications.

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 configured as a PC-based program or an application dedicated to a mobile terminal. An application to which the present invention is applied may be installed in the computing device 1000 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 in response to a request from the computing device 1000 .

The device described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, devices and components described in the embodiments may include, 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, may be implemented using one or more general purpose or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that can include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computing devices, and may be stored or executed in a distributed manner. Software and data may be stored in 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, etc. 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 available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

According to an embodiment of the present invention, since the first similarity value is derived by using the correlation between the drug and the drug as a category for each of the drug to be predicted and a plurality of existing drugs, the prediction target side effect of the drug to be predicted more accurately It is possible to exert an effect that can predict the possibility of manifestation.

According to an embodiment of the present invention, since the first similarity value is derived by using the node information between the plurality of target genes targeted by the drug as a category for each of the drug to be predicted and the plurality of existing drugs, the drug to be predicted more accurately It can exert the effect of predicting the possibility of the occurrence of the predicted target side effect.

According to an embodiment of the present invention, the first similarity value is derived by using the expression gene information in which the SNP is expressed for each of one or more diseases related to the drug for each of the prediction target drug and the plurality of existing drugs as a category, so that the first similarity value is more accurately It can exert the effect of predicting the possibility of the occurrence of the predicted side effect for the predicted drug.

According to an embodiment of the present invention, since the second similarity value is derived between the predicted side effect and a plurality of existing side effects known for the predicted drug, it is possible to more accurately predict the likelihood of the predicted side effect for the predicted drug can have an effect.

According to an embodiment of the present invention, the learned inference model unit for deriving information on the likelihood of occurrence includes a second inference model implemented by ensemble learning in which the result values derived from the plurality of first inference models are input. , it can exert the effect of improving the performance of the inference model unit.

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible from the above description by those skilled in the art. For example, the described techniques are performed in an order different from the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (13)

A method for predicting drug side effects, implemented with a computing device, comprising:
deriving a first similarity value for each of two or more categories between a predictive drug and a plurality of existing drugs known to have a predictable side effect;
extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect;
deriving each second similarity value between the predicted side effect and a plurality of known side effects for the predicted drug;
extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and
Inputting the first similarity representative information and the second similarity representative information to a learned inference model unit to derive information on the possibility of occurrence of the predicted target side effect for the predicted drug; How to predict side effects.
The method according to claim 1,
The step of deriving the first similarity value comprises:
The first category includes the interaction between one or more other drugs that the drug affects or is affected,
In the first category between the correlation between one or more other drugs that the predicted target drug affects or is affected, and the correlation between one or more other drugs that each of the plurality of existing drugs affects or is affected A method of predicting a side effect of a drug, deriving a first similarity value for the drug.
The method according to claim 1,
The step of deriving the first similarity value comprises:
Contains node information on genes and proteins included in the interaction pathway between a plurality of target genes targeted by the drug as a second category,
A method of predicting a side effect of a drug by deriving a first similarity value for the second category between node information for the prediction target drug and node information for each of the plurality of existing drugs.
The method according to claim 1,
The step of deriving the first similarity value comprises:
Including expression gene information for one or more genes in which SNP (Single Nucleotide Polymorphism) related to each of one or more diseases related to drugs is expressed as a third category,
A method of predicting side effects of a drug by deriving a first similarity value for the third category between the expressed gene information for the prediction target drug and the expressed gene information for each of the plurality of existing drugs.
The method according to claim 1,
The step of deriving the first similarity value comprises:
The 4th category and symptom information about one or more symptoms or diseases related to the drug
Contains target gene information for a plurality of target genes targeted by the drug as a fifth category,
A first similarity value for the fourth category between the symptom information for the prediction target drug and the symptom information for each of the plurality of existing drugs, and
A method of predicting a side effect of a drug by deriving a first similarity value for the fifth category between the target gene information for the prediction target drug and the target gene information for each of the plurality of existing drugs.
The method according to claim 1,
The step of deriving the first similarity value comprises:
For each category, a value obtained by dividing the number of elements of the intersection between the information on the corresponding category of the prediction target drug and the information on the corresponding category of each of the existing drugs by the number of elements in the union is calculated as the first similarity value Drug, How to predict the side effects of
The method according to claim 1,
The step of deriving the second similarity value comprises:
For each of a plurality of anatomically divided hierarchies, the side effects of drugs that derive the second similarity value based on the type of hierarchical element in which the predicted target side effect is expressed and the type of the hierarchical element in which each of the existing side effects is expressed How to predict.
The method according to claim 1,
The step of deriving the second similarity value comprises:
For each anatomically divided plurality of hierarchies, the hierarchical similarity value obtained by dividing the number of elements of the intersection between the set of hierarchical elements in which the predicted target side effect is expressed and the set of hierarchical elements in which each of the existing side effects is expressed by the number of elements in the union calculating; and
Calculating a value obtained by dividing the sum of the hierarchical similarity values calculated for each of the plurality of layers by the number of the plurality of layers as the second similarity value;
The method according to claim 1,
The inference model unit,
a plurality of first inference models for deriving information on a first likelihood of occurrence of the predicted target drug with respect to the predicted drug based on the first similarity representative information and the second similarity representative information; and
A second inference model for deriving information on the second likelihood of the predicted side effect with respect to the predicted drug based on the plurality of information on the first likelihood derived from each of the plurality of first inference models A method of predicting side effects of a drug, comprising:
10. The method of claim 9,
Each of the plurality of first inference models includes a different inference algorithm,
The reasoning algorithm is
Methods of predicting side effects of drugs, including Random forest, Logistic regression, Naive Bayesian and XGBoost.
10. The method of claim 9,
The plurality of first inference models,
A plurality of first samples pairing any one first drug with any one side effect known to the first drug, and any one second drug and any one side effect unknown to the second drug It is learned by the first similarity representative learning information and the second similarity representative learning information for each of a plurality of paired second samples,
The first similarity representative learning information and the second similarity representative learning information are,
A method for predicting a side effect of a drug, derived by the steps of claim 1 .
As a system for predicting drug side effects,
a first similarity derivation unit for deriving a first similarity value for each of two or more categories between a predicted drug and a plurality of existing drugs known to have a predictable side effect;
a first similarity representative information derivation unit for extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect;
a second similarity derivation unit for deriving each second similarity value between the predicted side effect and a plurality of existing side effects known to the predicted drug;
a second similarity representative information derivation unit for extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and
An expression possibility prediction unit for inputting the first similarity representative information and the second similarity representative information to the learned inference model unit to derive information on the possibility of the occurrence of the predicted side effect for the prediction target drug; , a system for predicting side effects of drugs.
A computer-readable medium for implementing a method of predicting drug side effects performed on a computing device having one or more processors and one or more memories, comprising:
deriving a first similarity value for each of two or more categories between a predictive drug and a plurality of existing drugs known to have a predictable side effect;
extracting a maximum value from a plurality of the first similarity values for each of the categories to derive first similarity representative information between the predicted drug and the predicted side effect;
deriving each second similarity value between the predicted side effect and a plurality of known side effects for the predicted drug;
extracting a maximum value from a plurality of the second similarity values to derive second similarity representative information between the predicted drug and the predicted side effect; and
Inputting the first similarity representative information and the second similarity representative information to a learned inference model unit, and deriving information on the possibility of occurrence of the predicted target side effect for the predicted drug; Computer including; readable medium.

Images