Method for discovering and screening anti-aging plant-derived materials based on AI analysis

Abstract

The present invention relates to a method for discovering and screening anti-aging plant materials based on AI analysis, and to a method for selecting skin anti-aging substances and plant materials, and a system for selecting skin anti-aging plant materials. The skin anti-aging substance and plant material selection method and skin anti-aging plant material selection system of the present invention can be usefully utilized in the development of products related to skin anti-aging by extensively investigating substances or plant materials having skin anti-aging effects.

Description

Technology Field

[0001] The present invention relates to a method for discovering and screening anti-aging plant materials based on AI analysis, and to a method for selecting skin anti-aging substances and plant materials, and a system for selecting skin anti-aging plant materials. Background Technology

[0003] As the body's primary defense barrier, the skin protects internal organs from external environmental stimuli such as changes in temperature and humidity, ultraviolet rays, and pollutants, and plays a crucial role in maintaining biological homeostasis, including body temperature regulation. Just like other organs, this vital skin gradually ages as we get older, resulting in phenomena such as loss of elasticity, keratinization, wrinkle formation, and skin atrophy.

[0004] Furthermore, skin aging is significant not only for health but also from a cosmetic perspective due to visible signs such as loss of elasticity, keratinization, and wrinkle formation; consequently, there has been a growing interest in cosmetics capable of inhibiting skin aging. There is a particular trend of focusing attention on cosmetics utilizing natural materials that offer excellent anti-aging effects with minimal side effects, making it necessary to conduct research on methods to identify natural materials for the development of cosmetics that can suppress skin aging. Prior art literature

[0006] Japanese Published Patent No. 2014-533959 The problem to be solved

[0007] The present invention aims to solve the aforementioned problem and other related problems.

[0008] The object of the present invention is (a) to obtain the expression pattern of an anti-aging marker gene before and after treatment of a candidate substance in secretory-papillary fibroblasts or secretory-reticular fibroblasts obtained from a first age group of 50 to 70 years, and to obtain the expression pattern of an anti-aging marker gene in secretory-papillary fibroblasts or secretory-reticular fibroblasts of a second age group of 20 to 27 years that are not treated with the candidate substance; and

[0009] (b) a step of comparing the expression pattern of an anti-aging marker gene obtained from the first age group with the expression pattern of an anti-aging marker gene obtained from the second age group to select the candidate substance as a skin anti-aging substance; the present invention provides a method for selecting a skin anti-aging substance, comprising: (b) a step of comparing the expression pattern of an anti-aging marker gene obtained from the first age group with the expression pattern of an anti-aging marker gene obtained from the second age group to select the candidate substance as a skin anti-aging substance.

[0010] Another objective of the present invention is a skin anti-aging plant material selection system performed by a computing device comprising at least one processor, wherein

[0011] Candidate substance data collection and normalization module for collecting and normalizing data regarding the names and compound structures of anti-aging candidate substances;

[0012] A candidate plant material data collection and normalization module that collects and normalizes data regarding the name and candidate substance content of anti-aging candidate plant materials containing the above candidate substance; and

[0013] A structural similarity comparison and anti-aging plant material selection module that calculates the degree of structural similarity between the above-mentioned candidate substance and the skin anti-aging substance selected by the aforementioned skin anti-aging substance selection method, and if it is determined that the above-mentioned candidate substance is similar in structure to the skin anti-aging substance, determines a candidate plant material containing the candidate substance determined to be structurally similar as an anti-aging plant material;

[0014] The present invention provides a skin anti-aging plant material selection system characterized by including

[0015] Another object of the present invention is a method for selecting skin anti-aging plant materials, performed by a computing device comprising at least one processor, wherein

[0016] Step 1: Collecting and normalizing data regarding the names and compound structures of anti-aging candidate substances;

[0017] A second step of collecting and normalizing data regarding the name and content of candidate substances of anti-aging candidate plant materials containing the above candidate substances; and

[0018] A third step of calculating the degree of structural similarity between the above candidate substance and the skin anti-aging substance selected by the aforementioned skin anti-aging substance selection method, and if it is determined that the above candidate substance is similar in structure to the skin anti-aging substance, determining a candidate plant material containing the candidate substance determined to be structurally similar as an anti-aging plant material;

[0019] The present invention provides a method for selecting skin anti-aging plant materials, including

[0020] Another objective of the present invention is to provide a non-transient readable recording medium comprising a program for executing a method for selecting skin anti-aging plant materials, wherein the method is executed by a computing device comprising at least one processor as described above.

[0022] The technical problems to be solved according to the technical concept of the invention disclosed in this specification are not limited to those for solving the problems mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0024] This is explained in detail as follows. Meanwhile, each description and embodiment disclosed in this application may be applied to each other description and embodiment. That is, all combinations of the various elements disclosed in this application fall within the scope of this application. Furthermore, the scope of this application should not be considered limited by the specific descriptions provided below.

[0025] As one embodiment for achieving the above objective, the present invention comprises the steps of: (a) obtaining expression patterns of anti-aging marker genes before and after treatment of secretory-papillary fibroblasts or secretory-reticular fibroblasts obtained from a first age group of 50 to 70 years of age, and obtaining expression patterns of anti-aging marker genes in secretory-papillary fibroblasts or secretory-reticular fibroblasts of a second age group of 20 to 27 years of age that are not treated with said candidate substance; and

[0026] (b) a step of comparing the expression pattern of an anti-aging marker gene obtained from the first age group with the expression pattern of an anti-aging marker gene obtained from the second age group to select the candidate substance as a skin anti-aging substance; the present invention provides a method for selecting a skin anti-aging substance, comprising: (b) a step of comparing the expression pattern of an anti-aging marker gene obtained from the first age group with the expression pattern of an anti-aging marker gene obtained from the second age group to select the candidate substance as a skin anti-aging substance.

[0027] The secretory-papillary fibroblasts of step (a) of the present invention may refer to fibroblasts mainly located in the papillary dermis, which is mainly located in the upper layer of the dermis (immediately below the epidermis), and the secretory-reticular fibroblasts may refer to fibroblasts mainly located in the reticular dermis, which is the lower layer of the dermis. The secretory-papillary fibroblasts or secretory-reticular fibroblasts obtained from the first age group may be obtained from elderly humans, specifically from an age group of 50 to 70 years, or may be used after culturing and aging the secretory-papillary fibroblasts or secretory-reticular fibroblasts.

[0028] The candidate substance of step (a) of the present invention may refer to a substance expected to have an anti-aging effect, and may mean a substance for which it is to verify whether there is a skin anti-aging effect when applied to the skin, etc.

[0029] The expression pattern of anti-aging marker genes obtained from secretory-papillary fibroblasts or secretory-reticular fibroblasts obtained from a first age group prior to treatment with the candidate substance of step (a) of the present invention can be used as a comparison group for the expression pattern of anti-aging marker genes obtained from secretory-papillary fibroblasts or secretory-reticular fibroblasts obtained from a first age group treated with the candidate substance.

[0030] The expression pattern of anti-aging marker genes obtained from secretory-papillary fibroblasts or secretory-reticular fibroblasts of a second age group that were not treated with the candidate substance of step (a) of the present invention can be compared with the expression pattern of anti-aging marker genes obtained from secretory-papillary fibroblasts or secretory-reticular fibroblasts of a first age group that was treated with the candidate substance, and can be used to determine the anti-aging effect of the candidate substance.

[0031] The above second age group refers to an age group of young people, and specifically, may refer to an age group of 20 to 27 years old.

[0032] The expression pattern of the anti-aging marker gene in step (a) of the present invention may refer to the expression level pattern of the anti-aging marker genes appearing in fibroblasts, and the anti-aging marker gene in the secretory-papillary fibroblast may be one or more selected from the group consisting of IL8, FST, CDKN1A, SOD2, CEBPB, FOS, XRCC6, and CAT genes, and the anti-aging marker gene in the secretory-reticular fibroblast may be one or more selected from the group consisting of IL8, SOD2, CEBPB, KLF4, XRCC6, CAT, TERF1, and TINF2 genes.

[0033] Step (b) of the present invention is,

[0034] The expression pattern of the anti-aging marker gene in the first age group treated with the above candidate substance shows an expression pattern similar to the expression pattern of the anti-aging marker gene in the second age group,

[0035] In secretory-papillary fibroblasts of the first age group treated with the above candidate substance, if the expression level of the IL8, FST, CDKN1A, SOD2, or CEBPB gene decreases or the expression level of the FOS, XRCC6, or CAT gene increases compared to before treatment with the candidate substance, the above candidate substance may be selected as a skin anti-aging substance.

[0036] Step (b) of the present invention is,

[0037] The expression pattern of the anti-aging marker gene in the first age group treated with the above candidate substance shows an expression pattern similar to the expression pattern of the anti-aging marker gene in the second age group,

[0038] In secretory-reticular fibroblasts of the first age group treated with the above candidate substance, if the expression level of the IL8, SOD2, or CEBPB gene decreases or the expression level of the KLF4, XRCC6, CAT, TERF1, or TINF2 gene increases compared to before treatment with the candidate substance, the above candidate substance may be selected as a skin anti-aging substance.

[0039] Examples of skin anti-aging substances selected by the above method of the present invention include hesperidin, resveratrol, and honokiol, and lowsweet blueberries, hibiscus flowers, bilberries, cranberries, black mulberry fruits, raspberries, or magnolia bark, which are known to contain these, can be selected as skin anti-aging plant materials.

[0040] As another embodiment for achieving the above objective, the present invention relates to a skin anti-aging plant material selection system performed by a computing device comprising at least one processor, wherein

[0041] Candidate substance data collection and normalization module for collecting and normalizing data regarding the names and compound structures of anti-aging candidate substances;

[0042] A candidate plant material data collection and normalization module that collects and normalizes data regarding the name and candidate substance content of anti-aging candidate plant materials containing the above candidate substance; and

[0043] A structural similarity comparison and anti-aging plant material selection module that calculates the degree of structural similarity between the above-mentioned candidate substance and the skin anti-aging substance selected by the aforementioned skin anti-aging substance selection method, and if it is determined that the above-mentioned candidate substance is similar in structure to the skin anti-aging substance, determines a candidate plant material containing the candidate substance determined to be structurally similar as an anti-aging plant material;

[0044] A skin anti-aging plant material selection system is provided, characterized by including

[0045] The above candidate substance of the present invention may refer to a substance found to be expected to have an anti-aging effect.

[0046] The candidate substance data collection and normalization module of the present invention may collect and normalize data by inputting a term related to anti-aging or skin improvement as a search term, and inputting a command and an example together to produce the data collection result in JSON format.

[0047] The above terms regarding anti-aging or skin improvement may include related terms or synonyms, the above terms regarding skin improvement may include skin condition or pore care, and examples of the above search terms may be the search expression shown in FIG. 7.

[0048] The command and example for producing the above data collection results in JSON format may be the command shown in FIGS. 8 and 9.

[0049] The candidate substance data collection and normalization module of the present invention may collect data regarding the structure of a compound consisting of SMILES information of a candidate substance.

[0050] The above-mentioned candidate plant material of the present invention may refer to a plant found to be expected to contain the above-mentioned candidate substance.

[0051] The candidate plant material data collection and normalization module of the present invention may collect and normalize data by inputting a command to search for plants containing candidate substances, along with a command and example to produce data collection results in JSON format.

[0052] The command to search for plants containing the above candidate substance and the command and example to produce data collection results in JSON format may be the command shown in FIGS. 10 to 12.

[0053] The structural similarity comparison and anti-aging plant material selection module of the present invention may convert the structures of the candidate substance and the skin anti-aging substance into molecular vector data, calculate the Tanimoto coefficient to obtain a Tanimoto similarity value, and calculate the degree of similarity. If the Tanimoto similarity value is 0.3 or higher, it may be determined that the candidate substance is similar to the structure of the skin anti-aging substance, but is not limited to such a value. A candidate plant material containing a candidate substance determined to have a similar structure may be determined as an anti-aging plant material. Examples of the finally selected skin anti-aging plant materials may be one or more selected from spinach, avocado fruit, carum petroselinum, turnip leaves, asparagus, and peas.

[0054] As another embodiment for achieving the above objective, the present invention relates to a method for selecting skin anti-aging plant materials, performed by a computing device comprising at least one processor, wherein

[0055] Step 1: Collecting and normalizing data regarding the names and compound structures of anti-aging candidate substances;

[0056] A second step of collecting and normalizing data regarding the name and content of candidate substances of anti-aging candidate plant materials containing the above candidate substances; and

[0057] A third step of calculating the degree of structural similarity between the above candidate substance and the skin anti-aging substance selected by the aforementioned skin anti-aging substance selection method, and if it is determined that the above candidate substance is similar in structure to the skin anti-aging substance, determining a candidate plant material containing the candidate substance determined to be structurally similar as an anti-aging plant material;

[0058] A method for selecting skin anti-aging plant materials including

[0059] The above candidate substance of the present invention may refer to a substance found to be expected to have an anti-aging effect.

[0060] The first step of the present invention may be to input a term related to anti-aging or skin improvement as a search term, and to input together a command and an example that generates the data collection result in JSON format to collect and normalize data.

[0061] The above terms regarding anti-aging or skin improvement may include related terms or synonyms, the above terms regarding skin improvement may include skin condition or pore care, and examples of the above search terms may be the search expression shown in FIG. 7.

[0062] The command and example for producing the above data collection results in JSON format may be the command shown in FIGS. 8 and 9.

[0063] The first step of the present invention may be to collect data regarding the structure of a compound consisting of SMILES information of a candidate substance.

[0064] The above-mentioned candidate plant material of the present invention may refer to a plant found to be expected to contain the above-mentioned candidate substance.

[0065] The second step of the present invention may involve collecting and normalizing data by inputting a command to search for plants containing candidate substances, along with a command and example to produce data collection results in JSON format.

[0066] The command to search for plants containing the above candidate substance and the command and example to produce data collection results in JSON format may be the command shown in FIGS. 10 to 12.

[0067] The third step of the present invention may involve converting the structures of the candidate substance and the skin anti-aging substance into molecular vector data, calculating the Tanimoto coefficient to obtain a Tanimoto similarity value, and calculating the degree of similarity. If the Tanimoto similarity value is 0.3 or higher, it may be determined that the candidate substance is similar to the structure of the skin anti-aging substance, but is not limited to such a value. A candidate plant material containing the candidate substance determined to have a similar structure may be determined as an anti-aging plant material. Examples of the finally selected skin anti-aging plant materials may be one or more selected from spinach, avocado fruit, carum petroselinum, turnip leaves, asparagus, and peas.

[0068] In another embodiment for achieving the above objective, the present invention provides a non-transient readable recording medium comprising a program for executing a method for selecting skin anti-aging plant materials, wherein the method is executed by a computing device comprising at least one processor as described above.

[0070] The present invention will be described in more detail below through examples. These examples are solely for illustrating the present invention, and it will be obvious to those skilled in the art that the scope of the present invention is not to be interpreted as being limited by these examples. Effects of the invention

[0072] The skin anti-aging substance and plant material selection method and skin anti-aging plant material selection system of the present invention can be usefully utilized in the development of products related to skin anti-aging by extensively investigating substances or plant materials having skin anti-aging effects. Brief explanation of the drawing

[0074] FIG. 1 is a block diagram illustrating the configuration of a system according to at least one embodiment of the present invention. FIG. 2 is a flowchart illustrating the operation of selecting anti-aging plant materials according to one embodiment of the present disclosure. FIG. 3 is a block diagram of a computing device in which various embodiments of the present disclosure can be implemented. Figure 4 compares the gene expression patterns between the OLD and YOUNG groups in secretory papillary fibroblasts (SP), secretory reticular fibroblasts (SR), inflammatory fibroblasts (INF), and mesenchymal fibroblasts (MES). Figure 5 compares the expression levels of each gene between the OLD group and the YOUNG group in secretory papillary fibroblasts (SP), secretory reticular fibroblasts (SR), inflammatory fibroblasts (INF), and mesenchymal fibroblasts (MES). Figure 6 compares the gene expression patterns between the OLD and YOUNG groups in secretory papillary fibroblasts (SP) and secretory reticular fibroblasts (SR). Figure 7 shows examples of search terms including terms related to anti-aging or skin improvement. Figures 8 and 9 show an example of a command and an example of an input to produce the data collection results of anti-aging candidate substances in JSON format. FIGS. 10 to 12 illustrate a command to search for plants containing candidate substances, a command to produce data collection results in JSON format, and an example of an input example. Figures 13 and 14 show examples of vectorization of the structure of the compound SMILES. Specific details for implementing the invention

[0075] The technology disclosed in this specification is subject to various modifications and may have various embodiments. Specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the technology disclosed in this specification to specific embodiments, and it should be understood that the technology disclosed in this specification includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the technology disclosed in this specification.

[0076] In describing the technology disclosed in this specification, detailed descriptions of related prior art are omitted if it is determined that such descriptions would unnecessarily obscure the essence of the technology disclosed in this specification. Furthermore, numbers used in the description of this specification (e.g., First, Second, etc.) are merely identifiers to distinguish one component from another.

[0077] In relation to the description of the drawings, similar reference numerals may be used for similar or related components.

[0078] As used herein, terms such as “computer program,” “component,” “module,” “system,” etc., may be used interchangeably and refer to computer-related entities, hardware, firmware, software, combinations of software and hardware, or executions of software. For example, a component may be, but is not limited to, a procedure executed on a processor, a processor, an object, an execution thread, a program, and / or a computer. For example, both an application executed on a computing device and the computing device itself may be components. One or more components may reside within a processor and / or an execution thread. A component may be localized within a single computer. A component may be distributed among two or more computers.

[0079] The terms used in this specification are for describing the embodiments and are not intended to limit the disclosure. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. The terms “comprises” and / or “comprising” as used in this specification do not exclude the presence or addition of one or more other components in addition to the mentioned components. Although terms such as “first,” “second,” etc. are used to describe various elements or components, these elements or components are not limited by these terms. These terms are used merely to distinguish one element or component from another. Therefore, the first element or component mentioned below may be the second element or component within the technical scope of this disclosure. Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning that is commonly understood by those skilled in the art to which this disclosure pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0080] Furthermore, the terms "information" and "data" as used in this specification may often be used interchangeably. The suffixes "module" and "part" for components used in the following description are assigned or used interchangeably solely for the ease of drafting the specification and do not inherently possess distinct meanings or roles. The purpose and effects of this disclosure, as well as the technical configurations for achieving them, will become clear by referring to the embodiments described in detail below in conjunction with the accompanying drawings. In describing this disclosure, if it is determined that a detailed description of known functions or configurations might unnecessarily obscure the essence of this disclosure, such detailed description will be omitted. Additionally, the terms described below are defined considering their functions in this disclosure, and these may vary depending on the intentions or conventions of the user or operator.

[0081] The present invention will be explained in more detail below through the following examples. However, these examples are intended to illustrate the invention and the scope of the invention is not limited to these examples.

[0082] The scope of rights for the steps in the claims of the present disclosure is determined by the functions and features described in each step, and is not affected by the order in which the steps are described in the claims unless the order of precedence is specified in each step. For example, in a claim describing steps including Step A and Step B, even if Step A is described before Step B, the scope of rights is not limited by the requirement that Step A must precede Step B.

[0083] In the present disclosure, a "module" or "part" performs at least one function or operation and may be implemented in hardware or software, or a combination of hardware and software. Additionally, a plurality of "modules" or a plurality of "parts" may be integrated into at least one module and implemented by at least one processor, except for a "module" or "part" that needs to be implemented in specific hardware.

[0084] Meanwhile, various elements and areas in the drawings are depicted schematically. Accordingly, the technical concept of the present disclosure is not limited by the relative sizes or spacing depicted in the attached drawings.

[0085] An embodiment of the present disclosure will be described in more detail below with reference to the attached drawings.

[0086] FIG. 1 is a block diagram illustrating the configuration of a system according to at least one embodiment of the present invention.

[0087] FIG. 2 is a flowchart illustrating the operation of selecting anti-aging plant materials according to one embodiment of the present disclosure.

[0088] Referring to FIG. 1, the system (100) includes a candidate material data collection and normalization module (110), a candidate plant material data collection and normalization module (120), and a structural similarity comparison and anti-aging plant material selection module (130).

[0089] According to one embodiment, the candidate substance data collection and normalization module (110) is configured to collect and normalize data regarding the name and compound structure of the anti-aging candidate substance (S210).

[0090] For example, the candidate substance data collection and normalization module (110) may collect and normalize data by inputting a term related to anti-aging or skin improvement as shown in FIG. 7 as a search term to collect candidate substance data from a database containing accumulated information on compounds such as CIR REPORT or PubMed, or by inputting commands and examples such as FIG. 8 and 9 together to output the data collection results in JSON format to normalize the data collection results into a consistent form and improve search accuracy by applying prompt engineering in literature analysis based on such a large language model (LLM). This may allow for the extraction of content in a consistent format from the abstracts of literature searched through the search expression, and may collect data regarding the structure of compounds consisting of SMILES information of candidate substances, but is not limited thereto, and various forms of information acquisition methods may be utilized. Additionally, preprocessing such as performing normalization to represent the collected data in a consistent standard expression may be performed, but is not limited thereto, and various data preprocessing may be performed together.

[0091] According to one embodiment, the candidate plant material data collection and normalization module (120) is configured to collect and normalize data regarding the name and candidate material content of an anti-aging candidate plant material containing the candidate material based on the candidate material data collected by the candidate material data collection and normalization module (110) (S220).

[0092] For example, the candidate plant material data collection and normalization module (120) may collect and normalize data by inputting a command to search for plants containing candidate substances such as those in FIG. 10 to 12, and inputting a command and example to produce the data collection results in JSON format, in order to collect data on the names and candidate substance content of plant materials containing each candidate substance using a large language model (LLM) (e.g., using APIs of perplexity and Open AI).

[0093] According to one embodiment, the structural similarity comparison and anti-aging plant material selection module (130) calculates the degree of structural similarity between the candidate substance and a substance already identified as an anti-aging substance, and if the candidate substance is determined to be similar to the structure of the skin anti-aging substance, the candidate plant material containing the candidate substance determined to be structurally similar is configured to be an anti-aging plant material (S230 and S240).

[0094] For example, the structural similarity comparison and anti-aging plant material selection module (130) converts the structure of the candidate substance and the skin anti-aging substance into molecular vector data, calculates the Tanimoto coefficient to obtain a Tanimoto similarity value, calculates the degree of similarity, and if the Tanimoto similarity value is 0.3 or higher, it can determine that the candidate substance is similar to the structure of the skin anti-aging substance and select the skin anti-aging plant material, and may determine the candidate plant material containing the candidate substance determined to have a similar structure as the anti-aging plant material.

[0095] FIG. 3 is a block diagram of a computing device in which various embodiments of the present disclosure can be implemented.

[0096] Referring to FIG. 3, a computing device (300) according to one embodiment of the present invention may include a memory (310) and a processor (320).

[0097] The memory (310) is configured to store collected data and can store and execute program data. The processor may include each module of the skin anti-aging plant material selection system, specifically, it can collect and normalize candidate substance data and candidate plant material data, and use the candidate substance structure data to compare structural similarity with the anti-aging substance and calculate a structural similarity value. Then, based on the structural similarity value, it can select and output anti-aging plant materials containing the anti-aging substance.

[0098] However, the components described above are not essential for implementing the computing device (300), so the computing device (300) may have more or fewer components than the components listed above.

[0099] The computing device (300) may be included within the system or may refer to the system itself. The computing device (300) may include any type of computer system or computer device, such as, for example, a microprocessor, a mainframe computer, a digital processor, a portable device, and a device controller. However, it is not limited thereto. The processor (320) of the computing device (300) typically controls the overall operation of the computing device (300). The processor (320) can provide or process appropriate information or functions to the user by processing signals, data, information, etc. that are input or output through the components included in the computing device (300) or by running an application program stored in memory (310). Additionally, the processor (320) may control at least some of the components of the computing device (300) to run an application program stored in memory (310). Furthermore, the processor (320) may operate at least two or more of the components included in the computing device (300) in combination with each other to run the application program. The processor (320) of the computing device (300) may be composed of one or more cores and may include a processor for data analysis and deep learning, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU) of the computing device. The processor (320) may read a computer program stored in memory (310) and perform data processing for machine learning according to some embodiments of the present disclosure. According to one embodiment of the present disclosure, the processor (320) may perform operations for learning collected data.The processor (320) can perform calculations for training a neural network, such as processing input data for training in deep learning (DL), extracting features from input data, calculating errors, and updating the weights of the neural network using backpropagation. At least one of the CPU, GPGPU, and TPU of the processor (320) can process the training of the network function. For example, the CPU and GPGPU can together process the training of the network function and data classification using the network function. In addition, in some embodiments of the present disclosure, processors of a plurality of computing devices can be used together to process the training of the network function and data classification using the network function. In addition, a computer program executed in a computing device according to some embodiments of the present disclosure may be a CPU, GPGPU, or TPU executable program.

[0101] Example 1. Selection of skin anti-aging substances through analysis of fibroblast gene expression patterns

[0102] 1.1. Selection of Fibroblasts for Analysis

[0103] To select substances that induce anti-aging or reverse aging of the skin, research was conducted to explore substances that restore the gene expression patterns of aged skin cells to a state close to that of young skin cells.

[0104] First, skin cells were collected from three human donors in the OLD group aged 53 to 70 years and two human donors in the YOUNG group aged 25 to 27 years, and single-cell transcriptome data were analyzed. Among the skin cells, four fibroblast subtypes were analyzed: secretory papillary fibroblasts (SP), secretory reticular fibroblasts (SR), inflammatory fibroblasts (INF), and mesenchymal fibroblasts (MES).

[0105] As a result, it was confirmed that the greatest difference in gene expression patterns was observed between the OLD and YOUNG groups in secretory papillary fibroblasts (SP) and secretory reticular fibroblasts (SR), and subsequently, substances with skin anti-aging effects were investigated based on the gene expression patterns of the two types of fibroblasts (Figs. 4 and 5).

[0107] 1.2. Selection of Target Genes for Analysis of Secretory-Papillary Fibroblasts (SP) and Secretory-Reticular Fibroblasts (SR)

[0108] As a result of analyzing the gene expression patterns of secretory papillary fibroblasts (SP) and secretory reticular fibroblasts (SR) of the OLD group, genes requiring decreased expression or increased expression to normalize to the gene expression pattern of the YOUNG group were identified, as shown in Figure 6. Specifically, among the gene expression patterns of the two groups, we intended to focus on analyzing the expression patterns of genes involved in six major pathways related to anti-aging / reverse aging: cell cycle arrest, epigenetic alteration, mitochondrial dysfunction, proliferation reduction, aging-related secretory phenotype (SASP), and telomere shortening (Table 1).

[0110] [Table 1]

[0111]

[0112] Among the genes involved in the six major pathways related to reverse / anti-aging, those that showed a significant difference in gene expression patterns between the OLD group and the YOUNG group were selected as key genes to be used for gene expression pattern analysis. In this case, whether a significant difference in expression was observed was determined statistically based on a p-value (significance probability) of less than 0.05. The selected core genes are as shown in Tables 2 and 3, where SP log2FC values ​​recorded in blue indicate genes that showed higher expression in the secretory-papillary fibroblasts (SP) of the OLD group compared to those of the YOUNG group, SP log2FC values ​​recorded in red indicate genes that showed lower expression in the secretory-papillary fibroblasts (SP) of the OLD group compared to those of the YOUNG group, SR log2FC values ​​recorded in blue indicate genes that showed higher expression in the secretory-reticulofibroblasts (SR) of the OLD group compared to those of the YOUNG group, and SR log2FC values ​​recorded in red indicate genes that showed lower expression in the secretory-reticulofibroblasts (SR) of the OLD group compared to those of the YOUNG group. Specifically, in the secretory-papillary fibroblasts (SP) of the OLD group, the expression of IL8, FST, CDKN1A, SOD2, and CEBPB genes was higher and the expression of FOS, XRCC6, and CAT genes was lower compared to those of the YOUNG group, and in the secretory-reticulofibroblasts (SR) of the OLD group, the expression of IL8, SOD2, and CEBPB genes was higher and the expression of KLF4, XRCC6, CAT, TERF1, and TINF2 genes was lower compared to those of the YOUNG group.

[0114] [Table 2]

[0115]

[0116] [Table 3]

[0117]

[0119] 1.3. Selection of Skin Anti-Aging Substances through the Selection of Substances that Normalize Gene Expression Patterns

[0120] In order to select a skin anti-aging substance by exploring substances that normalize the expression pattern of the core gene selected in Example 1.2 above from the pattern of the OLD group to a pattern of the YOUNG group, substances that change the gene pattern to a pattern of the YOUNG group when treated with secretory papillary fibroblasts (SP) or secretory reticular fibroblasts (SR) were investigated.

[0121] Specifically, using the LINCS L1000 database, we predicted changes in gene expression patterns that occur when a specific substance is applied to secretory papillary fibroblasts (SP) or secretory reticular fibroblasts (SR). The LINCS L1000 database is a database that accumulates expression patterns for various genes classified by substance or cell type, observed when various substances are treated to A375, A549, HCC515, HA1E, HEPG2, HT29, MCF7, PC3, or VCAP cell lines, etc. It is a database capable of predicting expression patterns for genes for which changes in gene expression patterns have not been directly measured from the data. Since it can be predicted that if a specific gene exhibits the same expression pattern in each cell line when the target compound is treated to A375, A549, HCC515, HA1E, HEPG2, HT29, MCF7, PC3, or VCAP cell lines, the same expression pattern will also be observed when the compound is treated to other cells (e.g., SP or SR cells), in this study, regarding the core genes selected in the aforementioned Example 1.2 appearing in A375, A549, HCC515, HA1E, HEPG2, HT29, MCF7, PC3, or VCAP cell lines upon treatment with each substance Predicted the expression pattern.

[0122] The process of determining whether a treated substance is an anti-aging substance that changes the expression pattern to be closer to the pattern of the YOUNG group according to the predicted expression pattern involves converting the expression values ​​of genes before and after treatment into ranks and calculating the correlation to obtain a Zhang score value for each cell line; the closer the Zhang score value is to -1, the more likely it is to be judged as an anti-aging substance. For each substance, the Zhang score value for each cell line is comprehensively considered as the arithmetic mean, and the closer the Zhang score value is to -1, the more likely it is to be selected as an anti-aging substance that changes the expression pattern of key genes to be closer to the YOUNG group pattern when treated in secretory-papillary fibroblasts (SP) or secretory-reticular fibroblasts (SR). A Zhang score value close to -1 is determined as the arithmetic mean of each material being between -1 and -0.35, but it is not limited to these values; even if it is farther from -1 than -0.35, it can be selected if it is relatively closer to -1 compared to other materials. The substances selected for secretory-papillary fibroblasts (SP) or secretory-reticular fibroblasts (SR) are shown in Tables 4 and 5, respectively. The parts marked in red indicate substances selected as anti-aging agents.

[0124] [Table 4]

[0125]

[0126] [Table 5]

[0127]

[0128] Consequently, Honokiol, Hesperidin, Ilomastat, and Resveratrol were selected as anti-aging substances that change the expression patterns of key anti-aging related genes to be close to the patterns of the YOUNG group when treated with secretory papillary fibroblasts (SP) or secretory reticular fibroblasts (SR). Considering that these are substances known as anti-aging substances, it can be seen that the method of selecting anti-aging substances through the normalization of gene expression patterns as described above is appropriate. Plant materials containing large amounts of the selected substances were investigated, and considering the regulations of each country and the potential for commercial use of the materials, the skin anti-aging plant materials shown in Table 6 below were finally selected.

[0129] [Table 6]

[0130]

[0132] Example 2. Utilization of a skin anti-aging plant material selection system

[0133] 2.1. Data Collection for Skin Anti-Aging Candidate Substances

[0134] In order to select a skin anti-aging plant material from the skin anti-aging substance selected by the method of Example 1 described above, name data for candidate substances having effects related to anti-aging, skin condition, and pore management was first collected from a database containing accumulated information on compounds, such as CIR REPORT or PubMed.

[0135] Specifically, a literature analysis based on a large language model (LLM) was performed by searching for and normalizing substance names from databases such as CIR REPORT or PubMed to collect name data for candidate substances with effects related to anti-aging, skin condition, and pore management. The search keywords used included terms related to anti-aging or skin improvement as examples, and were expanded to include their synonyms and related terms. Examples of the keywords used and examples of search queries using them are shown in Figure 7.

[0136] For this large language model (LLM)-based literature analysis, prompt engineering was applied, specifically using few-shot prompting. By inputting multiple examples together when querying information via API, the collected data was normalized into a consistent format, and search accuracy was improved.

[0137] During the application of prompt engineering, a command was entered to normalize the data collection results into a consistent form by generating the literature search and learning results in JSON format, a text-based data storage format. This has the effect of extracting content in a consistent format from the abstracts of the literature searched through the above search expression, and example phrases were entered together. Examples of commands applied to prompt engineering are shown in Figures 8 and 9.

[0138] Subsequently, structural information of substances was collected from databases such as PubChem, which have accumulated structural information of compounds (e.g., SMILES structure, etc.), using the collected candidate substance name data. At this time, substances for which structural information was not disclosed were excluded so as not to be included in subsequent steps. Structural information of substances was searched based on the substance names identified through the aforementioned search, and SMILES structure and CID information were collected through this process. Examples of excerpts from the collected results are shown in Tables 7 and 8. Through this method, data on anti-aging candidate substances was collected and normalized for use in subsequent steps.

[0139] [Table 7]

[0140]

[0141] [Table 8]

[0142]

[0144] 2.2. Data Collection on Candidate Plant Materials for Skin Anti-Aging

[0145] After the structural information of each candidate substance collected in the aforementioned Example 2.1 was confirmed, data on plant materials containing each candidate substance was collected using a large language model (LLM) (e.g., using APIs from perplexity and Open AI). At this time, not only the names of the plant materials but also the content of the candidate substances contained within the plant bodies were collected. In this process as well, prompt engineering was applied similarly to the method described above to improve the accuracy of data collection and to normalize the data collection results into a consistent form. Through this, data on anti-aging candidate plant materials was collected and normalized. An example of a search input word with a prompt command applied to search for plant materials containing candidate substances is shown in Fig. 10, and an example of an input word with a prompt command applied to collect information on the content of candidate substances contained in each plant material using the collected plant material names is shown in Figs. 11 and 12. Data on anti-aging candidate plant materials was collected and normalized through this method and utilized in subsequent steps. Examples of excerpts from the collected data on anti-aging candidate plant materials are shown in Tables 9 to 11.

[0147] [Table 9]

[0148]

[0149] [Table 10]

[0150]

[0151] [Table 11]

[0152]

[0154] 2.3. Selection of Skin Anti-Aging Plant Materials

[0155] To select skin anti-aging plant materials, first, a structural similarity analysis was performed between the skin anti-aging substance selected by the method of Example 1 described above and the skin anti-aging candidate substance collected by the method of Example 2.1 described above. The structural similarity analysis was performed using ECFP4-based Tanimoto similarity analysis.

[0156] Specifically, using a molecular vectorization tool called ECFP4 2048 bits, the structural data of the anti-aging substance selected by the method of Example 1 described above and the anti-aging candidate substance collected in Example 2.1 described above were converted into molecular vector data and divided into two groups (anti-aging substance group and candidate substance group), and after inputting each molecular vector data, the Tanimoto coefficient between the two groups was calculated to obtain Tanimoto similarity values.

[0157] Through this, data on anti-aging candidate substances showing high structural similarity to the anti-aging substance selected by the method of Example 1 described above were collected, and by comparing this with the data on anti-aging candidate plant materials collected in Example 2.2 described above, plant materials containing a large amount of anti-aging candidate substances showing high structural similarity were selected. The vectorization of molecular structures for each substance used to determine structural similarity was performed by fragmenting the SMILES structure and converting it into vector values ​​aligned with binary vector column positions, some examples of which are shown in Figures 13 and 14. Table 12 shows an example of excerpts from the analysis results of candidate plant materials containing candidate substances confirmed to have high similarity based on structural similarity calculations (Tanimoto similarity calculated value close to 1) (the values ​​listed in the columns labeled HONOKIOL and HESPERIDIN represent the structural similarity of the candidate substances (Column 1) with HONOKIOL and HESPERIDIN, and similarity was determined to be high when the Tanimoto similarity value was 0.3 or higher, but is not limited thereto).

[0159] [Table 12]

[0160]

[0161] Subsequently, from the above-mentioned selected materials, skin anti-aging plant materials as shown in Table 13 below were finally selected, taking into account regulations by country and commercial applicability.

[0163] [Table 13]

[0164]

[0166] Foregoing, specific parts of the content of the present invention have been described in detail. It will be apparent to those skilled in the art that such specific descriptions are merely preferred embodiments and do not limit the scope of the present invention. Accordingly, the actual scope of the present invention is defined by the appended claims and their equivalents. Explanation of the symbols

[0168] 100: System 110: Candidate Material Data Collection and Normalization Module 120: Candidate Plant Material Data Collection and Normalization Module 130: Structural Similarity Comparison and Anti-Aging Plant Material Screening Module 300: Computing device 310: Memory 320: Processor

Claims

Claim 1 (a) obtaining expression patterns of anti-aging marker genes before and after treatment of secretory papillary fibroblasts or secretory reticular fibroblasts obtained from a first age group of 50 to 70 years, and obtaining expression patterns of anti-aging marker genes in secretory papillary fibroblasts or secretory reticular fibroblasts of a second age group of 20 to 27 years that are not treated with said candidate substance, wherein said anti-aging marker genes include IL8, SOD2, CEBPB, XRCC6, and CAT genes; and (b) a step of comparing the expression pattern of anti-aging marker genes obtained from the first age group with the expression pattern of anti-aging marker genes obtained from the second age group to select the candidate substance as a skin anti-aging substance, wherein the selection is made by selecting the candidate substance as a skin anti-aging substance when the expression levels of IL8, SOD2, and CEBPB genes decrease and the expression levels of XRCC6 and CAT genes increase compared to before treatment with the candidate substance; comprising a method for selecting a skin anti-aging substance. Claim 2 A method for selecting a skin anti-aging substance according to claim 1, wherein the fibroblasts in step (a) are secretory-papillary fibroblasts, and the anti-aging marker genes further include FST, CDKN1A, and FOS genes, and the selection in step (b) is to select the candidate substance as a skin anti-aging substance when the expression level of the FST or CDKN1A gene decreases or the expression level of the FOS gene increases compared to before treatment with the candidate substance. Claim 3 A method for selecting a skin anti-aging substance according to claim 1, wherein the fibroblasts in step (a) are secretory-reticular fibroblasts, the anti-aging marker genes further include KLF4, TERF1, and TINF2 genes, and the selection in step (b) is to select the candidate substance as a skin anti-aging substance when the expression level of the KLF4, TERF1, or TINF2 genes increases compared to before treatment with the candidate substance. Claim 4 delete Claim 5 delete Claim 6 A skin anti-aging plant material selection system performed by a computing device comprising at least one processor, comprising: a candidate material data collection and normalization module for collecting and normalizing data regarding the name and compound structure of an anti-aging candidate substance; a candidate plant material data collection and normalization module for collecting and normalizing data regarding the name and candidate substance content of an anti-aging candidate plant material including said candidate substance; and a structural similarity comparison and anti-aging plant material selection module for calculating the degree of structural similarity between said candidate substance and a skin anti-aging substance selected by the method of any one of claims 1 to 3, and, if said candidate substance is determined to be similar to the structure of said skin anti-aging substance, determining a candidate plant material including the candidate substance determined to be structurally similar as an anti-aging plant material. Claim 7 In claim 6, the above candidate substance data collection and normalization module collects and normalizes data by inputting a term related to anti-aging or skin improvement as a search term, and inputting commands and examples together to produce data collection results in JSON format, thereby forming a skin anti-aging plant material selection system. Claim 8 In claim 6, the above candidate substance data collection and normalization module collects data regarding the structure of a compound consisting of SMILES information of a candidate substance, in a skin anti-aging plant material selection system. Claim 9 In claim 6, the above candidate plant material data collection and normalization module collects and normalizes data by inputting a command to search for plants containing candidate substances, and together inputting a command and example to produce data collection results in JSON format, thereby forming a skin anti-aging plant material selection system. Claim 10 In claim 6, the above-mentioned structural similarity comparison and anti-aging plant material selection module converts the structures of the candidate substance and the skin anti-aging substance into molecular vector data, calculates the Tanimoto coefficient to obtain a Tanimoto similarity value, and calculates the degree of similarity, thereby forming a skin anti-aging plant material selection system. Claim 11 In claim 10, the above-mentioned structural similarity comparison and anti-aging plant material selection module determines that the above-mentioned candidate substance is similar to the structure of the above-mentioned skin anti-aging substance when the above-mentioned Tanimoto similarity value is 0.3 or higher, a skin anti-aging plant material selection system. Claim 12 A method for selecting skin anti-aging plant materials, performed by a computing device comprising at least one processor, comprising: a first step of collecting and normalizing data regarding the name and compound structure of an anti-aging candidate substance; a second step of collecting and normalizing data regarding the name and candidate substance content of an anti-aging candidate plant material containing said candidate substance; and a third step of calculating the degree of structural similarity between said candidate substance and a skin anti-aging substance selected by the method of any one of claims 1 to 3, and if said candidate substance is determined to be similar to the structure of said skin anti-aging substance, determining that a candidate plant material containing the candidate substance determined to be structurally similar is an anti-aging plant material. Claim 13 A method for selecting skin anti-aging plant materials according to claim 12, wherein the first step is to input a term related to anti-aging or skin improvement as a search term, and to input together a command and example to produce data collection results in JSON format to collect and normalize data. Claim 14 A method for selecting skin anti-aging plant materials, wherein the first step of claim 12 is to collect data regarding the structure of a compound consisting of SMILES information of a candidate substance. Claim 15 A method for selecting skin anti-aging plant materials according to claim 12, wherein the second step involves collecting and normalizing data by inputting a command to search for plants containing candidate substances, along with a command and example to produce data collection results in JSON format. Claim 16 A method for selecting skin anti-aging plant materials according to claim 12, wherein the third step is to convert the structures of the candidate substance and the skin anti-aging substance into molecular vector data, calculate the Tanimoto coefficient to obtain a Tanimoto similarity value, and calculate the degree of similarity. Claim 17 A method for selecting skin anti-aging plant materials according to claim 16, wherein the third step is to determine that the candidate substance is similar to the structure of the skin anti-aging substance when the Tanimoto similarity value is 0.3 or higher. Claim 18 A non-transient readable recording medium comprising a program for executing a method for selecting skin anti-aging plant materials, wherein the method comprises the method according to claim 12.

Images