KR101995511B1 - Method and system for predicting new drug lead compound candidate by applying association rules between pharmacologically active similas compounds and their fragments - Google Patents

Abstract

A novel lead substance predicting method and system is provided based on the fragment property relationship association rules of similar pharmacological active group compounds. The method comprises the steps of: a system operating by at least one processor for predicting a new lead material candidate, comprising the steps of collecting information of the various pharmacological activities and of compounds having a distinct pharmacological activity, statistically Grouping compounds by pharmacological activity, fragmenting the compounds in pseudo-pharmacologically active groups into minimum chemical functional units to derive fragmented structures for each compound, Selecting at least one particular fragment structure derived and predicting the selected at least one particular fragment structure to a new leader material candidate for pharmacological activity of the compounds in the group.

Description

FIELD OF THE INVENTION The present invention relates to a new lead substance predicting method and system based on fragment association property association rules of similar pharmacologically active group compounds. BACKGROUND ART < RTI ID = 0.0 > [0002] <

The present invention relates to a novel lead substance predicting method and system based on associative structure property association rules of similar pharmacologically active group compounds.

Lead generation is an important step in the development of new drugs, from basic research to the development of new drugs, to clinical trials and licensing, and high quality lead compounds are the starting point for building the foundation for clinical success.

New drug development usually begins with selecting the target of the disease to be developed, screening the known compounds, selecting and testing the initial lead substance, and selecting the leading substance through a simple optimization process. The selected leading substances are made into candidate materials through optimization process of more than one year through various methods of pharmaceutical chemistry and used in clinical experiments. For this reason, the selection process of lead compounds is a starting point for the structural characteristics of compounds related to the pharmacological activity and toxicity of candidate compounds in clinical use, and is an important step in the development of new drugs.

In recent years, sculptural structures have become the leading materials, drug discovery and virtual screening based on them have been used, and it is possible to start with small-scale compounds, which will be widely used in the development stage of new drugs.

The conventional technology for finding lead compounds can be found in the process of discovering new drugs using natural products. The process of discovering a new drug through synthesis using a common structure in many active materials derived from such natural materials as a starting material generally starts with a process for screening a large amount of a natural compound extracted from one natural material. A common structure with initial pharmacological activity derived from the screening process is used as a lead compound to synthesize a structurally similar compound or derivative to increase pharmacological activity, reduce toxicity or side effects, and solve problems in the body drug delivery system It goes through.

The beneficial screening process of these natural compounds is a fundamental empirical method, and it is advantageous that a new leading substance can be derived because it proceeds without prior prediction of the result. On the other hand, there is a problem that enormous cost and time are required in terms of the cost of the test for pharmacological activity and the cost for securing and managing the compound.

In addition, there is a process of extracting a lead substance through a multiple screening method. At present, research on technological innovation and diversity is proceeding in order to select good leading substances globally. Particularly, as the number of substances that can be screened through the high-speed multiple screening method has been drastically increased, companies have tried to purchase synthetic compounds at low cost by approaching chemists around the world in order to secure new leading substances. We have invested heavily in order to dramatically improve the diversity of compounds in order to maximize the variety of compound libraries.

However, practical synthesis methods are limited, and thus the variability tends to actually decrease relative to the number of materials obtained, resulting in a situation in which the initial cost has to be paid more.

Further, there is a method of deriving a high-quality fragment structure having a low molecular weight as compared with a high-speed multiple screening method using an X-ray structure crystallography or a nuclear magnetic resonance technique. However, the method of deriving such a lead substance is also very expensive in terms of time and cost.

In addition, there has been an attempt to discover some pharmacologically active fragment structures through a drug-eluting test on fragments obtained by simple fragmentation of the compound, but this has the disadvantage that the process of finding more important fragments is not found.

In fact, in order to develop new innovative drugs, it is important to select leading substances that are likely to be developed as new drugs. However, as described above, most of the prior art techniques are limited to finding the structures of various minimum units having pharmacological activity, including the cost of the pretreatment process for the candidate search, including the large amount of the pharmacologically active compound and the cost of the pharmacological test, There is a problem with this bar.

Disclosure of the Invention Problem to be solved by the present invention is to provide a compound having a pharmacological activity grouped according to the pharmacological activity similarity by fragmenting pharmacologically active compounds in the group into a minimum chemical functional unit using a compound fragmentation technique, The present invention provides a method for identifying an optimal fragment structure exhibiting a pharmacological activity that can be presented as a leading substance in the development of a new drug, and a system for carrying out the method.

According to one aspect of the present invention, a new lead material candidate prediction method is a method for a system operating by at least one processor to predict a new lead material candidate, comprising: Collecting the compounds determined to have statistically significant similarities to the individual pharmacological activities and grouping them according to their pharmacological activity, fragmenting the compounds in the similar pharmacological active group into the minimum chemical functional units, Selecting at least one particular fragment structure that is commonly derived from the in-group compounds, and selecting the at least one particular fragment structure as a new leader material candidate for pharmacological activity of the in-group compounds .

The grouping step may comprise the step of selecting compounds which have similar pharmacological activity on the basis of the odds ratio between the compound and the pharmacological activity calculated on the basis of the frequency with respect to the individual pharmacological activity of all the compounds and the significance probability indicating statistically significant It can be excavated.

The grouping step may include ranking the compounds having pharmacological activity by pharmacological activity on the basis of the multiplication ratio and comparing the compounds having the significance probability of 0.05 or less among the ranked compounds to compounds having similar pharmacological activity Can be excavated.

Wherein the step of selecting comprises calculating a score of association between a specific compound and a particular fragment structure using the frequency of occurrence between all the compounds in the group and fragmented structures derived through fragmentation and if the association score satisfies the reference value The sculpture structure and compound pairs can be selected.

The degree of association refers to a degree of support indicating the probability that a specific fragment structure and a specific compound occur at the same time from all occurrence frequencies of all compounds in the group and fragment structures derived from all the compounds, And a degree of reliability which is a ratio of the degree of support to the degree of reliability, wherein the step of selecting includes selecting a specific piece from among a piece structure and a compound pair satisfying the reference value, It is possible to evaluate at least one fragment structure that is determined to be of significant relevance and at least one fragment structure that is common from the pair of compounds by evaluating whether the structure has a significant association with a particular compound.

The selecting step may be evaluated as having a significant association between the sculpture structure having the degree of improvement of greater than 1 and the compound pair.

The improvement may be used as a predictive score of a new lead substance candidate for a particular pharmacological activity.

According to another aspect of the present invention, a new lead material candidate prediction method is a method for a system operating by at least one processor to predict a new lead material candidate, the method comprising: Fragmentation of the at least one specific fragment structure into a plurality of fragmented structures, deriving fragmented structures for each compound, selecting at least one particular fragmented structure commonly derived from the compounds, And predicting with the lead material candidates.

Compounds having pharmacological activity with statistically significant similarity can be obtained from compounds having various pharmacological activities and individual pharmacological activities from a public database on the on-line database and publicly available data and user-input information, The multiplication ratio calculated based on the frequency of occurrence between the specific compound and the specific compound, and the significance probability according to the fisher exact test.

The selecting may select at least one specific fragment structure that is common to the pair of fragment pairs and the compound whose association score calculated using the occurrence frequency between the specific fragment and the specific fragment structure satisfies the reference value.

Wherein the step of selecting includes: determining whether a specific compound is present in a specific fragmented structure, a degree of support that is a probability that all the compounds having pharmacological activity and all fragmented structures occur at the same time, At least one piece structure common to each piece structure and compound pair that satisfies all of the reference values and satisfies a predefined significant measure of the degree of improvement that is the ratio of the degree of support to the reliability can be selected.

According to another aspect of the present invention, a method for predicting a new lead material candidate is a system that is operated by at least one processor and predicts a new lead material candidate, the system comprising: Compounds that have been determined to have statistically significant similarity to the individual pharmacological activities from the compounds having activity information and individual pharmacological activities are searched to find pharmacologically active compounds that are grouped by pharmacological activity and compounds in pseudo- A fragment scouring unit for fragmenting the compound into a minimum chemical functional unit to derive fragmented structures for each compound, a fragment structure extracting unit for selecting at least one specific fragmented structure commonly derived from the compounds in the group, When a specific fragment structure is formed by substituting Include Li provide new leading candidate lead compounds for predicting a candidate of a material for the active part.

The pharmacologically active analogue compound extracting unit selects a compound having a pharmacological activity for each pharmacological activity based on the odds ratio between the compound and the pharmacological activity calculated based on the frequency of the individual pharmacological activities of all the compounds , Compounds having a p-value of less than a reference value indicating statistically significant among the ranked compounds are selected, and the selected compounds can be grouped by finding compounds having similar pharmacological activity.

The sculptural structure extracting unit can be used as a measure of significance based on the coincidence frequency between all the compounds in the pseudo-pharmacological active group and the sculptural structures derived from all the compounds, reliability used as a measure of reliability, It is possible to find a significant piece structure and at least one piece structure common to the pair of compounds by evaluating the degree of improvement calculated as a ratio of the reliability and evaluating whether the degree of support, the reliability and the degree of improvement are significant.

A fragmented structure storing information of the at least one fragmented structure and the compound that the fragmented structure extraction unit outputs, a compound database connected to the fragmented structure-compound database, and connected to the terminal through a communication network, And a user interface unit for managing access to the database.

According to an embodiment of the present invention, the construction of a database in which information on compounds and pharmacological activity are collected and integrated can eliminate the processes required for large-scale pharmacologically active compounds and pharmacological activity tests. By using a fragment structure derived from compounds having similar pharmacological activity, it is possible to derive the minimum unit fragment structures having a chemical function highly related to pharmacological activity.

In addition, since the compounds are structurally well distributed to enable qualitative analysis and validation of the structure and activity, once the active substance is derived, the results of subsequent studies can be rapidly performed.

1 is a schematic block diagram of a novel lead material prediction system in accordance with an embodiment of the present invention.
2 is a flowchart of a new lead material predicting method according to an embodiment of the present invention.
3 is a schematic block diagram of a novel lead material prediction system in accordance with another embodiment of the present invention.
FIG. 4 is a flowchart illustrating a process for discovering a pharmacologically active analogous compound according to an embodiment of the present invention.
5 is a flowchart illustrating a fragmentation process according to an embodiment of the present invention.
6 is a conceptual diagram illustrating a fragmentation process according to an embodiment of the present invention.
7 is a flowchart illustrating a process of extracting a piece structure to be provided as a new lead material candidate according to an embodiment of the present invention.
8 is a block diagram illustrating a hardware configuration of a new lead material predicting system according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification, when an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise.

Also, the terms " part, "" ... "," module ", and the like described in the specification mean a unit for processing at least one function or operation and may be implemented by hardware or software or a combination of hardware and software .

In this specification, a pharmacological effect means that a substance enters the organism in vivo, alone or in combination with any substance in vivo, to produce the desired pharmacological effect.

Lead compounds are substances that are thought to be pharmacologically or biologically effective and may require some improvement in order to achieve the desired effect.

In addition, the term "chemical fragmentation" refers to the generation of a compound as a minimum chemical functional unit, a "fragmentation structure".

FIG. 1 is a schematic block diagram of a novel lead material predicting system according to an embodiment of the present invention, FIG. 2 is a flowchart of a new lead material predicting method according to an embodiment of the present invention, Figure 2 is a schematic block diagram of a novel lead material prediction system according to an embodiment.

1, the novel lead substance prediction system 100 includes a pharmacological activity-compound database 101, a pharmacologically active compound-like compound discovery unit 103, a compound fragmentation unit 105, a fragment structure discovery unit 107, A lead material candidate providing unit 109, and a sculptural structure-compound database 111. Operations of these structures will be described with reference to FIG.

Referring to FIG. 2, the pharmacologically active compound-based compound search unit 103 searches pharmacological activity-compound database (pharmacological activity-compound database) in which compounds having pharmacological activity and pharmacological activity are mapped based on information collected via the Internet and / (S101).

The pharmacologically active analogue compound locating unit 103 performs a significance evaluation on the information stored in the pharmacological activity-compound database 101 to identify compounds having similar pharmacological activity and group them according to the pharmacological activity similarity of the pharmacologically active compounds. (S103).

The compound fragmentation unit 105 slices the pharmacologically active compound in the group into the minimum chemical functional units using the compound fragmentation technique (S105).

The fragment structure extracting unit 107 calculates the degree of association score indicating the degree of association between the fragmented structure fragmented from the pharmacologically active compound and the pharmacologically active compound (S107).

Based on the association score, the lead substance candidate providing unit 109 finds an optimal fragment structure exhibiting a pharmacological activity that can be presented as a lead substance in the development of a new drug, and integrates information of the pharmacologically active analogous compounds associated with the extracted fragment structure To construct a fragment structure-compound database 111 (S109).

Thereafter, the lead material candidate providing unit 109 provides the fragment structure itself, which is stored in the fragment structure-compound database 111, as a new lead material candidate (S111). Also, the lead material candidate providing unit 109 provides a combination of fragmented structures (S111) so that it can be utilized in the development of new drugs having various activities.

Referring to FIG. 3, the lead material candidate providing unit 109 may be connected to the display device 200 to output a new lead material candidate to the screen of the display device 200.

In addition, the new leading material prediction system 100 may further include a user interface unit 113. The user interface unit 113 is connected to the terminal 400 connected through the communication network 300. The user interface unit 113 is a means for providing access to the fragmented structure-compound database 111 and can be configured to be connected to the terminal 400 through the communication network 300 to exchange various information. For example, the user interface unit 113 can provide an environment in which the terminal 400 can search for the fragment structure as a new lead material candidate. The communication network 300 may be a network such as a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN) 4G mobile communication network, Wi-Fi, WiBro, etc. The communication method does not depend on wired or wireless, and any communication method may be used. The terminal 400 is equipped with a memory means such as a smart phone, a personal computer (PC), a tablet PC, a personal digital assistant (PDA), a web pad and the like, .

Now, with reference to FIGS. 4 to 7, the detailed operation of the new leading material prediction system 100 will be described.

FIG. 4 is a flow chart showing a process for searching for a pharmacologically active compound, according to an embodiment of the present invention, and shows the operation of the pharmacologically active compound-searching unit 103.

4, the pharmacologically active compound-identifying unit 103 collects pharmacological activity information, compound information including natural products, disease information and the like (S201), and integrates them to construct a pharmacological activity-compound database 101 S203).

At this time, the pharmacologically active compound-identifying unit 103 may collect information necessary for constructing the pharmacological activity-compound database 101 on the Internet or input from the user.

The pharmacologically active compound-based compound searching unit 103 can acquire pharmacologically active compounds such as DrugBank, Korean Traditional Knowledge Portal (KTKP), Traditional Chinese Medicine System Pharmacology Database (TCMSP), Traditional Chinese Medicine Integrated Database (TCMID) Information collected through UMLS, Icd10, ClinicalTrial, SIDER Side Effect Resource, Comparative Toxicogenomics Database (CTD), Therapeutic Target Database (TTD) and the like can be stored in the pharmacologically active compound database 101. In addition, the pharmacologically active compound-identifying unit 103 may collect the official information with high reliability on-line and store it in the pharmacological activity-compound database 101 even though the pharmacologically active compound-identifying unit 103 is not configured as a public database.

The pharmacological activity-compound database 101 can store individual pharmacological activity information and information of a compound having pharmacological activity. For example, Nintedanib is a drug used to treat idiopathic pulmonary fibrosis (IPF) targeting multiple receptor tyrosine kinases (RTKs) and nonreceptor tyrosine kinases (nRTKs). Nintedanib has been shown to inhibit platelet derived growth factor receptor (PDGFR) and adenosine triphosphate (ATP) receptors such as fibroblast growth factor receptor (FGFR), vascular endothelial growth factor receptor (VEGFR) and Fms-like tyrosine kinase- Binds competitively to binding pouches and blocks intracellular signaling, which is critical for the proliferation, migration and transformation of fibroblasts, which represents the essential mechanism of IPF pathology. In addition, nintedanib inhibits nRTKs, Lck, Lyn and Src kinases, and this content can be stored in the pharmacological activity-compound database 101. Of course, although only one pharmacological activity is described here, one compound may have a plurality of pharmacological activities.

The pharmacologically active compound-based discovering unit 103 generates a frequency table by deriving frequency numbers corresponding to individual pharmacological activities of the compound using the pharmacological activity-compound database 101 (S203).

For example, the compound is a compound corresponding to a pharmacologically active N ₁ corresponds to M1, M2, M3, pharmacologically active N ₂ the compound M1, M3, M4, M5, pharmacological activity of the N ₃ M1, that the M2 Assuming, the frequency of the pharmacologically active N ₁ of the M1 is to be 1, because M1 is other pharmacological activity of N _2, N ₃ appear, the frequency of the other pharmacologically active is the second. Since the compounds exhibiting pharmacological activity N ₁ are M ₁ , M 2 and M 3, the frequency of pharmacological activity N ₁ for other compounds is 2. At this time, one compound (M) may have n pharmacological activities (N ₁ , ..., N _K , _K ? ₁ ). In this way, the pharmacological activity-compound frequency table reflecting the frequency of the corresponding compound for each pharmacological activity of the natural product can be summarized as shown in Table 1 below. The frequency table is implemented as 2x2.

Pharmacologically active N ₁ Other pharmacological activity system Compound M1 Frequency a Frequency c a + c Other compounds Frequency b Frequency d b + d system a + b c + d a + b + c + d

Here, a represents the frequency recorded in the first column of the first row of Table 1. That is, it shows a variable (variable) indicating the number of occurrences for a particular compound M1 of specific pharmacologically active N _1.

b " means the frequency recorded in the first column of the second row of Table 1. That is, it means a frequency for the specific pharmacological activity N ₁ of the other compounds except the specific compound M1.

c means the frequency recorded in the second column of the first row of Table 1. < tb >< TABLE > That is, it means a frequency for the pharmacological activity other than the specific pharmacological activity N ₁ of the specific compound M1.

d means the frequency recorded in the second column of the second row of Table 1. < tb >< TABLE > That is, it means a frequency with respect to pharmacological activities other than the specific pharmacological activity N ₁ of the remaining compounds other than the specific compound M1.

In this way, the pharmacologically active analogue compound locating unit 130 generates a frequency table as shown in Table 1 for all pharmacological activities and compounds stored in the pharmacological activity-compound database 101 (S205).

The pharmacologically active compound-based discovering unit 130 derives the multiplication ratio of the pharmacological activity-related compound contained in the frequency table (S207) through the significance evaluation (S209) and evaluates the association based on the multiplication ratio (S209).

The pharmacologically active analogue compound locating unit 130 may utilize the fisher exact test as a criterion for determining whether the compound actually has a significant pharmacological activity on the frequency tables generated in step S205.

The pharmacologically active analogue compound locating unit 130 selects compounds on the basis of multiplication ratios and groups the compounds exhibiting a specific pharmacological activity based on the significance (p-value) of the selected compounds (S211) A compound judged to be significant in terms of similarity, i.e., a pharmacologically active compound, is searched (S213).

Herein, the hypergeometric distribution that the pharmacologically active analogue compound extracting unit 130 should draw in order to perform the significance evaluation between the pharmacological activity and the compound has a frequency of the compound M1 ₁ corresponding to the pharmacological activity N ₁ as shown in Table 1 Assuming that the compounds related to pharmacological activity are variously distributed according to various cases, each probability value constituting the pharmacological activity can be obtained by the following equation (1).

Here, a, b, c, and d represent the corresponding frequencies in Table 1, respectively.

The pharmacologically active analogue compound locating unit 130 can derive the compounding ratio for each pharmacological activity through Equation (1). These multiplication ratios are respectively generated for a plurality of frequency tables.

In addition, the pharmacologically active analogue compound locating unit 130 shows the significance between the compound and the pharmacological activity constituting the plurality of pharmacological activity-compound frequency tables, and the probability (p-value) of the right-hand side of the frequency table for the initial sub- (2) < / RTI >

In Equation (2), n may have a value from 0 to a + b + c + d.

At this time, when the significance value (p-value) obtained in Equation (2) has a value of 0.05 or less, the pharmacologically active analogue compound finding unit 130 determines that the compound is significant for the pharmacological activity. For example, if the significance test value for Table 1 is 0.05 or less, the pharmacologically active compound-like compound finding unit 130 judges that the compound (M1) is significant for the pharmacological activity (N ₁ ).

The pharmacologically active analogue compound locating unit 130 selects one or more compounds for a specific pharmacological activity based on the odds ratio obtained in Equation (1). For example, the higher the odds ratio, the higher the ranking.

The pharmacologically active analogue compound locating unit 130 selects compounds having a p-value of 0.05 or less among the selected compounds, i.e., compounds for which similarity is judged to be significant for a specific pharmacological activity. These select compounds are grouped into a compound exhibiting the specific pharmacological activity (S211), and compounds in the group are excluded as pharmacologically active compounds (S213).

FIG. 5 is a flowchart illustrating a fragmentation process according to an embodiment of the present invention. FIG. 6 is a conceptual diagram illustrating a fragmentation process according to an embodiment of the present invention, and shows the operation of the compound fragmentation unit 105.

5, the compound fragmentation unit 105 receives the pharmacological activity-similar compounds determined to be significant through the evaluation of the pharmacological activity between the pharmacological activity and the compound from the pharmacologically active compound-based drug discovery unit 130 (S301)

The compound fragmentation unit 105 fragmentizes the pharmacologically active compound into a minimum chemical functional unit using a compound fragmentation technique, which can be schematically illustrated in FIG. Referring to FIG. 6, the pharmacologically active compound 1 can be fragmented into fragmented structure 1 and fragmented structure 2. This diagram is a conceptual view of fragmentation, and may be somewhat different from the actual compound and fragment structure.

Typically, as a druggable compound in which some of the dielectrics are drug targets conforms to the rule of 5, the Fragmentation Structure has the rule of 3 (molecular weight <300, ClogP < 3, the number of hydrogen bond donors and acceptors each should be <3 and the number of rotatable bonds should be <3. In an embodiment of the present invention, a piece structure is created using Discovery Studio software. The compound segmentation unit 105 performs fragmentation (S305, S309, S313, S317, and S321) by selecting one of five items (Rings, RingAssemblies, BridgeAssemblies, Chains, BemisMurckoAssemblies) (S303, S307, S311, S315, , Various sculptural structures are formed according to the structure and role that can act as functional groups (S323). Here, the five items, Rings, RingAssemblies, BridgeAssemblies, Chains, and BemisMurckoAssemblies, are the fragmentation units that can be used in the fragmentation technique. For example, RingAssemblies segmentation can mean that the compound performs segmentation in RingAssemblies units.

 FIG. 7 is a flowchart illustrating a process of extracting a piece structure to be provided as a candidate for a new lead material according to an embodiment of the present invention, and shows operations of the piece structure searching unit 107 and the lead material candidate providing unit 109.

The fragment structure extracting unit 107 maps the fragment structure derived through fragmentation and the corresponding pharmacologically active compound (S401), and maps the fragment structure as shown in Table 2.

Table 2 illustrates a list for analysis of association rules for pharmacologically active compound fragment structures and pharmacologically active compound relationships.

Number Items One Fragment structure 1, compound 2 2 Fragment structure 1, fragment structure 2, compound 1 3 Fragment structure 1, compound 3 4 Fragment structure 1, compound 2, compound 3 5 Fragment structure 2, compound 2

In Table 2, the order number is a fragmentation sequence number, for example, sequence number 1 indicates fragmentation performed first, and item number 1 corresponds to fragmentation structure 1 derived through fragmentation with respect to compound 2. In this case, the segmentation in each order may be different from that in FIG. For example, in Sequence 1, segmentation may be done in units of Rings, and Sequence 2 may be segmented in BridgeAssemblies.

The fragment structure extracting unit 107 derives the association rule between the pharmacologically active compound and the fragment structure (S403), and calculates the relevance score according to the association rule (S405).

At this time, in order to derive the association rule, the fragment structure search unit 107 displays the occurrence frequencies of the compounds in the fragment structure on the basis of Table 2 as a concurrent occurrence table as shown in Table 3.

According to Table 2, the co-occurrence frequency of fragmented structure 1 and compound 1 is 1, and the co-occurrence frequency of compound 3 and fragmented structure 1 is 2. Table 3 shows the frequency of such coincidence. Table 3 shows a model of a symmetric matrix.

Sculpture structure 1 Sculpture structure 2 Compound 1 Compound 2 Compound 3 Sculpture structure 1 4 One One 2 2 Sculpture structure 2 One 2 One One 0 Compound 1 One One One 0 0 Compound 2 2 One 0 3 One Compound 3 2 0 0 One 2

According to Table 3, it is possible to know how many times each piece structure and compound were derived together. For example, if we look at the intersection of fragmented structure 1 and compound 2, we can see that two items occurred together.

In addition, the data value on the diagonal in Table 3 represents the total number of occurrences including the item. For example, in the items where the fragment structure 1 and the fragment structure 1 intersect, the frequency of occurrence of the fragment structure 1 in Table 2 is recorded. Since four occurrences occurred in the order numbers 1, 2, 3 and 4 in Table 2, The frequency is 4. In the items where the fragmented structure 2 and the fragmented structure 1 intersect, the frequency at which the fragmented structure 2 and the fragmented structure 1 occur simultaneously is recorded. Since the occurrence occurs once in the order 2 in Table 2, the coincidence frequency is 1. Through these Table 3, association rules are calculated as Support, Confidence, and Improvement.

Here, the degree of support represents a probability for an event including the fragment structure (A) and the pharmacologically active compound (component) (B) together. In this case, the larger the probability value is, the more frequent the rule is used as a measure of the importance of the rule.

The support of two items A and B represents the proportion of occurrences that both item A and item B simultaneously include, and can be calculated through equation (3).

In addition, the reliability means the probability that there is a pharmacologically active compound (component) (B) when there is a fragment structure (A). At this time, the higher the probability value, the higher the probability that the pharmacologically active compound (component) (B) exists when the fragment structure (A) is present, and is used as a measure for the reliability of the rule.

Further, the degree of improvement is a value derived from the ratio of the reliability of the association rule to the degree of support, and can be calculated through Equation (5).

The degree of improvement represents the rate of increase of the probability of item B when item A is given, compared to the probability of item B when item A is not given. The larger this value is, the greater whether the occurrence of item A affects the occurrence of item B.

)와 P(B)가 같게 되어 향상도가 1이 된다. 어떤 규칙의 향상도가 1보다 크면, 이 규칙은 결과를 예측하는데 있어서 우연적 기회보다 우수하다는 것을 의미하고, 향상도가 1보다 작으면 이러한 규칙은 결과를 예측하는데 있어서 우연적 기회보다 나쁠 것이다. If the occurrence of item A and item B are not related, P (

) And P (B) are the same, and the degree of improvement becomes 1. If the improvement of any rule is greater than 1, this rule means that it is better than the chance chance in predicting the outcome, and if the improvement is less than 1, this rule will be worse than the chance chance in predicting the outcome.

Thus, the fragment structure search unit 107 generates association rules from the concurrent occurrence frequency table in Table 3, and generates association rules between the fragment structures, between the compounds, between the fragment structures and the compounds. However, since the present invention proposes a fragment structure having a high correlation with the compounds having pharmacological activity as a leading substance, the association rule of the compound and the fragment structure is important, but the relation rule relationship between the fragment structures is meaningless.

The piece structure searching unit 107 finally selects a piece structure and a compound having a high degree of association score based on the degree of support, reliability, and degree of improvement for the generated association rules.

At this time, if the degree of support and the reliability are equal to or more than a predetermined reference value, the piece structure searching unit 107 judges a meaningful association rule. That is, the fragmentation structure (A) derived from the association rule and the pharmacologically active compound (B) are determined to be related. And finally evaluation is made by the score of improvement in the determined association rule. At this time, when the degree of improvement is larger than 1, it is judged that the degree of prediction is excellent.

The fragment structure finding unit 107 extracts a pharmacokinetic association main fragment structure or a common fragment structure and a combination fragment structure in which individual fragment structures are combined based on the association score (S407). For example, when there is a compound M1 having a pharmacologically active N ₁ , a compound M2, and a compound M3, a frequency table is constructed as shown in Table 3 for the fragment structures of each of the compounds M1, M2, and M3, And the degree of improvement score is greater than 1, the fragment structure finding unit 107 judges that the fragment structures are related to the pharmacological activity N ₁ . At this time, the fragment structure extracting unit 107 can extract the fragment structure having the highest degree of improvement score into the main fragment structure or common fragment structure for pharmacological activity N ₁ . In addition, the fragment structure extracting unit 107 can extract fragment structures having an enhancement score of greater than ₁ into a combinatorial fragment structure for pharmacological activity N ₁ .

The lead material provider 109 stores the main fragment structure and the combination fragment structure extracted in step S407 into the fragment structure-compound database 111 together with the corresponding pharmacologically active compound and stores the fragment structures into a new leader material candidate (S409). For example, fragmented structure 1 may be provided as a candidate new starting material that exhibits pharmacological activity N ₁ . In addition, fragmented structure 1 and fragmented structure 2 can be provided as a candidate for a combination new leader material exhibiting pharmacological activity N ₁ .

In addition, the lead material provider 109 can provide the prediction of the specific fragment structure for a specific pharmacological activity, and can be used as a material for evaluating the degree of predictability of the specific pharmacological activity of the fragment structures based on the degree of improvement have. For example, if the degree of enhancement of fragmented structure 1 against pharmacological activity N ₁ is 1.3 and the degree of enhancement of fragmented structure 2 against pharmacological activity N ₁ is 1.5, the predictive value for pharmacological activity N ₁ is Can be estimated to be higher than structure 1.

In this way, the common fragment structure possessed by the compounds having similar pharmacological activity and structure can be discovered through the discovery of the association rule between the fragment structure of the pharmacologically active compounds and the pharmacologically active compound, and a database is constructed by integrating the results, It can be provided as a candidate for a drug substance, and can be provided for use in the development of a new drug having various activities through the combination thereof. In this way, a large amount of costly and time-consuming processes such as securing of a pharmacologically active compound and a process for testing pharmacological activity, which are necessary for a lead substance discovery process, which was a problem and a problem of the prior art, Can be solved by constructing In addition, the disadvantage of missing sculptural structures derived from simple fragmentation of compounds is the lack of a process for finding more important sculptural structures that occur in an attempt to discover some pharmacologically active fragment structures through a limited drug efficacy testing process. Can be solved by using fragmentation structures derived from the compounds and identifying minimal fragment structures with high pharmacological activity and association scores.

8 is a block diagram illustrating a hardware configuration of a new leading material prediction system according to another embodiment of the present invention.

8, the new lead material prediction system 500 includes at least one storage unit 501, at least one output unit 503, at least one input unit 505, at least one communication unit 507, And hardware including one processor 509. The new lead material prediction system 500 includes various software such as an operating system, middleware, and programs that operate in combination with hardware. The hardware and software of the new leading material prediction system 500 have the configuration and capabilities to implement the present invention.

The storage unit 501 stores a program for implementing the configuration and method described with reference to Figs. 1 to 7. The storage unit 501 includes any medium that stores or transmits data in a form readable by an apparatus such as a computer, and includes, for example, a read only memory (ROM), a random access memory (RAM) Storage media, flash memory devices, and other electrical, optical, or acoustic signal transmission media, and the like. The program stored in the storage unit 501 includes instructions that implement the operation of the new leading material prediction system 500 described with reference to FIGS. The processor 509 loads the program and performs the operations of the new leading material prediction system 500 described in FIGS. 1 to 7 described in the present invention.

The output unit 503 outputs information according to the operation of the processor 509 to the screen. For example, a piece structure and / or a piece structure combination is output as a new lead material. At this time, it can be outputted on the screen, printed matter, or output in various ways which can be recognized by the user. The input unit 505 collects the information necessary for constructing the pharmacological activity-compound database 101 on the Internet or inputs it from the user and outputs it to the processor 509.

The communication unit 507 is connected to the communication network 300 and performs a data transmission / reception function.

In this case, each of the configurations 501, 503, 505, and 507 is shown mounted in one processor 509, which means that the configurations 501, 503, 505, and 507 operate on a processor basis , A single processor 509, but may be implemented as separate servers connected to each other via Ethernet or a network.

The embodiments of the present invention described above are not implemented only by the apparatus and method, but may be implemented through a program for realizing the function corresponding to the configuration of the embodiment of the present invention or a recording medium on which the program is recorded.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

Claims (15)

WHAT IS CLAIMED IS: 1. A method for predicting a new lead material candidate by a system operating by at least one processor,
Collecting information of the compounds having various pharmacological activities and individual pharmacological activities,
Identifying the compounds that are found to have statistically significant similarity to the individual pharmacological activities and grouping them by pharmacological activity,
Fragmenting the compounds in the pseudo-pharmacologically active group into minimal chemical functional units to derive fragmented structures for each compound,
The association scores are calculated between the compounds and the fragment structures using the frequency of occurrence between the compounds in the pseudo-pharmacologically active group and the fragment structures derived through fragmentation, and the association score is calculated based on the predefined threshold value Selecting a particular compound to meet and a particular fragment structure, and
Providing said selected fragment structure as a new lead material candidate for pharmacological activity of said particular compound,
Wherein the selecting comprises:
A degree of support indicating a probability that a specific fragment structure and a specific compound occur at the same time based on the occurrence frequency, a reliability indicating a probability of occurrence of the specific compound when the specific fragment structure is generated, And selecting the specific fragment structure and the specific compound that are not less than the respective threshold values. The method of claim 1,
Wherein the grouping comprises:
Based on the odds ratio between the compound and the pharmacological activity calculated on the basis of the frequency for the individual pharmacological activity of all the compounds and the significance probability indicating that the pharmacological activity is statistically significant, Candidate prediction method. 3. The method of claim 2,
Wherein the grouping comprises:
Selecting a compound having a pharmacological activity by pharmacological activity on the basis of the multiplication ratio and finding a compound having a similar probability of 0.05 or less among the ranked compounds as a compound having a similar pharmacological activity; Material candidate prediction method. delete delete The method of claim 1,
Wherein the selecting comprises:
Wherein the improvement is evaluated as having a significant association with a sculptural structure larger than one pair of compounds. The method of claim 1,
The above-
A method for predicting a new lead substance candidate, which is used as a predictive score of a new lead substance candidate for a particular pharmacological activity. WHAT IS CLAIMED IS: 1. A method for predicting a new lead material candidate by a system operating by at least one processor,
Fragmenting the compounds having pharmacological activity having statistically significant similarities into minimum chemical functional units to derive fragment structures for each compound,
Selecting at least one particular fragment structure commonly derived from said compounds, and
Predicting said selected at least one particular fragment structure to a new lead material candidate for said pharmacological activity,
Wherein the selecting comprises:
A support which is a probability that a specific compound and a specific fragment structure occur at the same time using all the compounds having similar pharmacological activity and the occurrence frequency of all fragment structures, reliability that is the probability of occurrence of a specific compound when a specific fragment structure occurs, Calculating a degree of improvement that is a ratio of the degree of support to the reliability; and
Selecting at least one piece structure common to each piece structure and compound pairs in which the degree of support and the reliability both meet a predetermined reference value and the degree of improvement satisfies a predefined significant measure
Wherein the new candidate material candidate prediction method comprises: 9. The method of claim 8,
Compounds having pharmacological activity with statistically significant similarity,
Based on the multiplication ratio between the specific pharmacological activity and the specific compound and the significance probability according to the fisher exact test,
The multiplication ratio may be,
Which is calculated using a specific pharmacological activity calculated from compounds having various pharmacologically active information and individual pharmacological activities collected from a public database on-line or inputted by a user and the occurrence frequency between a specific compound. delete delete A memory for storing a program for predicting a new lead material candidate, and
And at least one processor for executing the program,
The program includes:
Compounds that have been determined to have statistically significant similarity to the individual pharmacological activities from compounds having an individual pharmacological activity and information of various pharmacological activities collected from external disclosure data and user input data are grouped by pharmacological activity ,
Fragments in pseudo-active groups are fragmented into minimal chemical functional units, resulting in fragmented structures for each compound,
Selecting at least one specific fragment structure commonly derived from the compounds in the group,
Providing said selected at least one particular fragment structure as a novel lead material candidate for pharmacological activity of said in-group compounds,
To select at least one particular fragment structure commonly derived from the compounds in the group, to be used as a measure of significance based on the coincidence frequency between all compounds in the pseudo-pharmacologically active group and the fragmented structures derived from all of the compounds Reliability that is used as a measure of reliability, and degree of improvement calculated by the ratio of the degree of support and reliability,
Evaluating whether the degree of support, the reliability and the degree of improvement are significant to identify significant piece structures and at least one piece structure common to the pair of compounds. The method of claim 12,
The program includes:
Ranking the compounds having the pharmacological activity by pharmacological activity on the basis of the odds ratio between the compound and the pharmacological activity calculated based on the frequency of the individual pharmacological activities of all the compounds,
Compounds with a p-value less than or equal to a reference value indicating statistically significant among the ranked compounds are selected,
A candidate agent candidate prediction system comprising instructions for identifying and grouping selected compounds as compounds with similar pharmacological activity. delete The method of claim 12,
A fragmented structure-compound database for storing information between at least one piece structure and the compound derived in common, and
A user interface unit connected to the fragmented structure-compound database and managing access to the fragmented structure-compound database to a terminal connected through a communication network;
Wherein the new candidate material candidate prediction system further comprises:

Images