Method and apparatus for manufacturing target substance modulation agent

Abstract

A method and apparatus for manufacturing a target substance modulation agent are provided. The method for manufacturing a target substance modulation agent may comprise the steps of: obtaining structural data on a plurality of target proteins corresponding to a plurality of target substance-active microorganisms; receiving structural data of a plurality of compounds from a compound structure library prepared in advance; predicting binding affinity for target protein-compound pairs generated from combinations of the plurality of target proteins and the plurality of compounds; applying a plurality of weights, each determined for a respective target substance-active microorganism, to the binding affinity to calculate a weighted score; determining, on the basis of the weighted score, a compound to be used as a target substance modulation agent from among the plurality of compounds; and generating formulation data of the target substance modulation agent including the determined compound, and storing the formulation data in a storage device.

Description

Method and apparatus for manufacturing a target substance control agent

[0001] Cross-citation with related applications

[0002] This application claims the benefit of priority based on Korean Patent Application No. 10-2024-0186557 filed on December 13, 2024, and all contents disclosed in the document of said Korean patent application are incorporated herein as part of this specification.

[0003] The disclosure relates to a method and apparatus for manufacturing a target substance control agent.

[0004] Methane (CH4) is one of the major causes of global warming; in particular, the annual amount of methane gas emitted from ruminants is estimated to be between 800,000 and 1.15 million tons, which is reported to account for 15 to 20% of total global methane emissions. These methane emissions are primarily generated by methanogens residing in the rumen of ruminants, and the activity of various enzymes, including methyl-coenzyme M reductase (MCR), which plays a crucial role in methane production, has been confirmed. However, currently developed methane inhibitors are being developed by simply quantifying methane production after administering additives to culture media or animals, without closely considering the mechanisms of these biochemical enzymes. Consequently, existing methane inhibitors are not based on systematic analysis of enzyme mechanisms, making it difficult to ultimately improve the research and development pipeline.

[0005] Recently, molecular docking technology has been primarily utilized in drug development, particularly in the screening of anticancer drugs; however, there are few instances of its use in the livestock industry to regulate the gut microbiome of ruminants or to inhibit methane-producing bacteria. Although the synthetic compound 3-NOP (3-Nitrooxypropanol) has been reported to inhibit methane production by binding to the MCR of methane-producing bacteria, the enzyme used in that study was not derived from the rumen, and no research has been conducted targeting MCRs derived from rumen methane-producing bacteria. Natural compounds have also been studied extensively for methane reduction; however, due to the lack of sustainability of their reduction effects and unclear mechanisms, there have been cases of side effects such as reduced feed intake, decreased productivity, and inefficient changes in the microbiome. In particular, while red algae extract (Asparagopsis taxiformis) has demonstrated methane reduction effects, its inclusion of bromine compounds lowers feed palatability. Furthermore, there are limitations to its practical application due to reduced dry matter intake and the toxicity of bromine compounds. In this regard, there is a growing demand for natural compounds that are safe and can specifically inhibit rumen methane-producing bacteria while maintaining or enhancing animal productivity. Furthermore, technologies for discovering compounds capable of promoting or inhibiting microorganisms active to any target substance, rather than being limited to methane, are also attracting attention.

[0006] The problem to be solved is to provide a method and apparatus for manufacturing a target substance modulator capable of manufacturing a target substance modulator that can promote or inhibit the production of a target substance.

[0007] Another task to be solved is to discover a substance for methane-reducing feed additives that is safe to use while having excellent methane reduction effects, and to provide a method and apparatus for manufacturing a target substance control agent capable of producing a methane-reducing feed additive containing said substance as a target substance control agent.

[0008] A method for manufacturing a target substance regulatory agent according to one embodiment is a method for manufacturing a target substance regulatory agent performed by a computing device comprising a processor, a memory, and a storage device, wherein the processor comprises the steps of: acquiring sequence or structural data regarding a plurality of target proteins corresponding to a plurality of target substance active microorganisms; receiving structural data of a plurality of compounds from a pre-prepared compound structure library; predicting a binding affinity for a target protein-compound pair formed from the plurality of target proteins and the plurality of compounds; calculating a weighted score by applying a plurality of weights determined for each of the plurality of target substance active microorganisms to the binding affinity; determining a compound among the plurality of compounds to be used as a target substance regulatory agent according to the weighted score; and configuring formulation data of the target substance regulatory agent including the determined compound and storing the formulation data in the storage device, wherein the plurality of weights may include values ​​representing relative abundance obtained for each of the plurality of target substance active microorganisms or the plurality of target proteins.

[0009] The above plurality of weights may include values ​​representing the relative abundance of each target substance active microorganism obtained by performing rRNA sequencing (ribosomal ribonucleic acid sequencing) for each of the above plurality of target substance active microorganisms.

[0010] The step of predicting the binding affinity may include: the step of the processor performing docking on the target protein-compound pair to predict the binding pose for the target protein-compound pair; the step of the processor loading a pre-trained binding affinity prediction model into the memory; and the step of the processor inputting the docked target protein-compound pair into the binding affinity prediction model and obtaining the output of the binding affinity prediction model as the binding affinity.

[0011] The above-mentioned bond affinity prediction model may be trained to take a 4-dimensional tensor containing the 3-dimensional coordinates of an atom and one training feature vector as input and output a predicted value of the bond affinity.

[0012] The above binding affinity prediction model may be trained using experimental values ​​regarding the binding strength of proteins and compounds as training data.

[0013] The step of calculating the weighted score may include: the processor loading a plurality of binding affinity values ​​obtained for each of the plurality of target proteins for each of the compounds into the memory as a plurality of first scores; the processor loading the plurality of weights into the memory; the processor applying the plurality of weights to the plurality of first scores loaded into the memory to calculate a plurality of second scores; the processor calculating a final score determined as a single value for each of the compounds from the plurality of second scores; and the processor calculating the final score as the weighted score.

[0014] The above target substance active microorganisms may be selected from methanogen-producing bacteria, hydrogen sulfide-producing bacteria, acetate-producing bacteria, butyrate-producing bacteria, hydrogen-producing bacteria, carbon monoxide-producing bacteria, nitrogen-producing bacteria, or nitrous oxide-producing bacteria.

[0015] The above-mentioned methane-producing bacteria are Methanobrevibacter smithii, Methanobrevibacter millerae, Methanobrevibacter thaurei, Methanobrevibacter ruminantium, Methanosphaera stadtmanae, Methanobrevibacter olleyae, Methanomicrobium mobile, Methanobrevibacter gottschalkii, Methanosarcina barkeri, and Methanobacterium bryanthi It may be one or more selected from Methanobacterium formicicum and Methanocaldococcus jannaschii.

[0016] The above target protein may include methyl-coenzyme M reductase.

[0017] The compound determined above includes a plurality of compounds, and the formulation data may include information regarding the formulation of the target substance control agent such that all of the plurality of compounds are included.

[0018] An apparatus for manufacturing a target substance control agent according to one embodiment comprises: a processor; and a memory in which an instruction executed by the processor is loaded. The instruction is executed by the processor to cause the processor to obtain sequence or structural data regarding a plurality of target proteins corresponding to a plurality of target substance active microorganisms, receive structural data of a plurality of compounds from a pre-prepared compound structure library, predict binding strength for target protein-compound pairs combined from the plurality of target proteins and the plurality of compounds, apply a plurality of weights determined for each of the plurality of target substance active microorganisms to the binding strength to calculate a weighted score, determine a compound to be used as a target substance control agent among the plurality of compounds according to the weighted score, configure formulation data of the target substance control agent including the determined compound, and store the formulation data in a storage device. The plurality of weights may include values ​​representing relative abundance obtained for each of the plurality of target substance active microorganisms or the plurality of target proteins.

[0019] The above plurality of weights may include values ​​representing the relative abundance of each target substance active microorganism obtained by performing rRNA sequencing (ribosomal ribonucleic acid sequencing) for each of the above plurality of target substance active microorganisms.

[0020] Predicting the binding affinity may include performing docking on the target protein-compound pair to predict the binding pose for the target protein-compound pair, loading a pre-trained binding affinity prediction model into the memory, inputting the docked target protein-compound pair into the binding affinity prediction model, and obtaining the output of the binding affinity prediction model as the binding affinity.

[0021] The above-mentioned bond affinity prediction model may be trained to take a 4-dimensional tensor containing the 3-dimensional coordinates of an atom and one training feature vector as input and output a predicted value of the bond affinity.

[0022] The above binding affinity prediction model may be trained using experimental values ​​regarding the binding strength of proteins and compounds as training data.

[0023] Calculating the weighted score may include loading a plurality of binding affinity values ​​obtained for each of the plurality of target proteins for each of the compounds into the memory as a plurality of first scores, loading a plurality of weights into the memory, applying the plurality of weights to the plurality of first scores loaded into the memory to calculate a plurality of second scores, calculating a final score determined as a single value for each compound from the plurality of second scores, and calculating the final score as the weighted score.

[0024] The above target substance active microorganisms may be selected from methanogen-producing bacteria, hydrogen sulfide-producing bacteria, acetate-producing bacteria, butyrate-producing bacteria, hydrogen-producing bacteria, carbon monoxide-producing bacteria, nitrogen-producing bacteria, or nitrous oxide-producing bacteria.

[0025] The above-mentioned methane-producing bacteria are Methanobrevibacter smithii, Methanobrevibacter millerae, Methanobrevibacter thaurei, Methanobrevibacter ruminantium, Methanosphaera stadtmanae, Methanobrevibacter olleyae, Methanomicrobium mobile, Methanobrevibacter gottschalkii, Methanosarcina barkeri, and Methanobacterium bryanthi It may be one or more selected from Methanobacterium formicicum and Methanocaldococcus jannaschii.

[0026] The above target protein may include methyl-coenzyme M reductase.

[0027] A computer-readable recording medium according to one embodiment is a computer-readable recording medium that records instructions executed by a computing device including a processor, memory, and a storage device, wherein the instructions are executed by the computing device to cause the computing device to acquire sequence or structural data regarding a plurality of target proteins corresponding to a plurality of target substance active microorganisms, receive structural data of a plurality of compounds from a pre-prepared compound structure library, predict binding strength for target protein-compound pairs combined from the plurality of target proteins and the plurality of compounds, calculate a weighted score by applying a plurality of weights determined for each of the plurality of target substance active microorganisms to the binding strength, determine a compound to be used as a target substance regulatory agent among the plurality of compounds according to the weighted score, construct formulation data of the target substance regulatory agent including the determined compound, and store the formulation data in the storage device, wherein the plurality of weights may include values ​​representing relative abundance obtained for each of the plurality of target substance active microorganisms or the plurality of target proteins.

[0028] According to the embodiments, target substance regulating agents can be discovered and manufactured using binding affinity as an indicator based on molecular docking and artificial intelligence technologies. In particular, by reflecting the interactions between multiple target substance-active microorganisms and multiple target proteins, and based on a score calculated by applying weights to binding affinity, a compound to be used as a target substance regulating agent can be determined by prioritizing target substance-active microorganisms or target proteins that have a significant influence in a specific environment or condition, while also considering the overall diversity of the target substance-active microorganism or target protein community. Similarly, based on molecular docking and artificial intelligence technologies, a substance for methane-reducing feed additives that is safe to use and has excellent methane reduction effects can be discovered using binding affinity as an indicator, and a methane-reducing feed additive containing said substance can be manufactured. In particular, by reflecting the interactions between multiple methane-producing bacteria and multiple target proteins, and based on a score calculated by applying weights to binding affinity, a compound to be used as a methane-reducing feed additive can be determined by prioritizing methane-producing strains that have a significant influence in a specific environment or condition, while also considering the overall diversity of the microbial community.

[0029] FIG. 1 is a drawing for explaining a target substance control formulation manufacturing apparatus according to one embodiment.

[0030] FIGS. 2 and FIGS. 3 are drawings for explaining the operation of a target substance control formulation manufacturing apparatus according to one embodiment.

[0031] FIG. 4 is a flowchart illustrating a method for manufacturing a target substance control agent according to one embodiment.

[0032] FIGS. 5 and 6 are drawings for explaining the operation of a target substance control formulation manufacturing apparatus according to one embodiment.

[0033] FIG. 7 is a flowchart illustrating a method for manufacturing a target substance control agent according to one embodiment.

[0034] FIG. 8 is a diagram illustrating an example of the implementation of a binding affinity prediction model used in the manufacture of a target substance control formulation according to one embodiment.

[0035] FIG. 9 is a block diagram illustrating a computing device according to one embodiment.

[0036] Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0037] Throughout the specification and claims, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another.

[0038] Terms such as "...part," "...unit," and "module" as described in the specification may refer to a unit capable of processing at least one function or operation described in this specification, and may be implemented as hardware or circuit, software, or a combination of hardware or circuit and software. Additionally, at least some components or functions of the method and apparatus for manufacturing a target substance control agent according to the embodiments described below may be implemented as a program or software, and the program or software may be stored on a computer-readable recording medium.

[0039] FIG. 1 is a drawing for explaining a target substance control formulation manufacturing apparatus according to one embodiment.

[0040] Referring to FIG. 1, a target substance control formulation manufacturing apparatus (10) according to one embodiment may execute program code or instructions loaded into one or more memory devices through one or more processors. For example, the target substance control formulation manufacturing apparatus (10) may be implemented as a computing device (50) as described below in relation to FIG. 9. In this case, one or more processors may correspond to a processor (510) of the computing device (50), and one or more memory devices may correspond to a memory (520) of the computing device (50). The program code or instructions may be executed by one or more processors to determine a compound to be used as a target substance control formulation and to derive a formulation of the target substance control formulation therefrom. In this specification, the term "module" has been used to logically distinguish these functions performed by the program code or instructions.

[0041] The target substance control agent manufacturing device (10) can discover a compound suitable as a control agent capable of controlling a target substance (e.g., a reduction agent in the case of reducing a target substance) and manufacture a target substance control agent using the same. That is, formulation data is configured including the compound discovered by the target substance control agent manufacturing device (10), and the formulation data can be used to manufacture an actual target substance control agent. To this end, the target substance control agent manufacturing device (10) may include a structure data acquisition module (101), a binding force prediction module (102), a compound determination module for the control agent (103), and a formulation data storage module (104).

[0042] The structural data acquisition module (101) can acquire sequence or structural data (e.g., three-dimensional structural data) regarding a plurality of target proteins corresponding to a plurality of target substance active microorganisms. The target substance active microorganisms may be microorganisms, bacteria, fungi, archaea, or genetically modified microorganisms that are involved in the production, inhibition, decomposition, and conversion of target substances, such as methane, hydrogen sulfide, acetate, butyrate, hydrogen, carbon monoxide, nitrogen, nitrous oxide, etc. In some embodiments, the target substance active microorganisms may be selected from, for example, methanogens, hydrogen sulfide-producing bacteria, acetate-producing bacteria, butyrate-producing bacteria, hydrogen-producing bacteria, carbon monoxide-producing bacteria, nitrogen-producing bacteria, or nitrous oxide-producing bacteria. In some embodiments, target substance-active microorganisms that may occur in the rumen environment of cattle may include, but are not limited to, hydrogen sulfide (H2S) producing bacteria, acetate (acetic acid) producing bacteria, butyrate (butyric acid) producing bacteria, hydrogen (H2) producing bacteria, and carbon monoxide (CO) producing bacteria. Examples of hydrogen sulfide (H2S) producing bacteria may include Desulfovibrio and Desulfobacter, examples of acetate (acetic acid) producing bacteria may include acetogen, examples of butyrate (butyric acid) producing bacteria may include Butyrivibrio fibrisolvens and Clostridium butyricum, examples of hydrogen (H2) producing bacteria may include Ruminococcus albus, and examples of carbon monoxide (CO) producing bacteria may include, but are not limited to, bacteria of the genus Carboxydothermus.

[0043] In some embodiments, target substance-active microorganisms that may occur in various environments other than the rumen of cattle may include, but are not limited to, hydrogen sulfide (H2S) producing bacteria present in marine sediments, methane producing bacteria present in wetlands and swamps, nitrogen gas (N2) producing bacteria present in soil and sediments, hydrogen (H2) producing bacteria present in the intestines of humans and animals, hydrogen sulfide (H2S) producing bacteria present in acidic mine effluent, nitrous oxide (N2O) producing bacteria present in sewage and wastewater treatment facilities, and acetate (acetic acid) producing bacteria present in lakes or ponds. Examples of such hydrogen sulfide (H2S) producing bacteria may include Desulfovibrio and Desulfobacter, examples of such methane-producing bacteria may include Methanobacterium or Methanosarcina, examples of such nitrogen gas (N2) producing microorganisms may include Anammox bacteria, examples of such hydrogen (H2) producing bacteria may include bacteria of the genus Bacteroides, examples of such hydrogen sulfide (H2S) producing bacteria may include sulfur-oxidizing bacteria such as Thiobacillus, examples of such nitrous oxide (N2O) producing bacteria present in sewage and wastewater treatment facilities may include denitrifying bacteria, and examples of acetate (acetic acid) producing bacteria present in lakes or ponds It may include, but is not limited to, acetogens, etc.

[0044] The structure data acquisition module (101) may also receive structure data (e.g., three-dimensional structure data) of a plurality of compounds from a pre-prepared compound structure library. In some embodiments, the compounds may be various natural compounds derived from nature. The compound structure library may store and manage information regarding the structures of these compounds.

[0045] The binding strength prediction module (102) can predict the binding strength for a target protein-compound pair composed of a plurality of target proteins and a plurality of compounds. In some embodiments, the binding strength prediction module (102) can predict the binding affinity for a target protein-compound pair composed of a plurality of target proteins and a plurality of compounds. Here, binding affinity may be an indicator of how strongly a target protein (i.e., the active site of the target protein) and a compound (i.e., a ligand) bind. In this regard, Kd (Dissociation Constant) is a dissociation constant representing the strength of the ligand binding to the target protein, and may represent the concentration at which the ligand binds to about half of the protein. A smaller Kd value may indicate a stronger binding between the ligand and the protein. That is, a low Kd implies a high binding affinity and may indicate that the ligand binds stably to the protein. Ki (Inhibition Constant) is an inhibition constant representing the ability of an inhibitor (both the terms 'inhibitor' and 'inhibitor' are used herein) to bind to an enzyme or receptor and inhibit its activity; a lower Ki value may indicate that the inhibitor binds more strongly to the enzyme and inhibits its activity more strongly. IC50 (Half Maximal Inhibitory Concentration) may indicate the concentration required for the inhibitor to inhibit the reaction by 50%. A lower IC50 value indicates that the reaction can be inhibited by more than half with a lower concentration, and may indicate that the inhibitor acts efficiently.

[0046] The binding force prediction module (102) can predict the binding pose of a target protein-compound pair by performing docking on the target protein-compound pair. Here, docking (or molecular docking) may be a computer simulation method that predicts the optimal binding method and pose when a ligand and a target protein bind. Here, the pose may represent the position and orientation in which the ligand binds to the active site (or pocket) of the target protein. The binding force prediction module (102) can define the active site on the target protein where the ligand, i.e., the compound, can bind, and calculate the binding pose and energy by placing the compound at various angles and orientations on the active site of the target protein. The binding force prediction module (102) can select the pose with the lowest binding energy among the binding poses derived from the docking simulation results.

[0047] The binding strength prediction module (102) can load a pre-trained binding affinity prediction model into memory. Here, the binding affinity prediction model may be trained using experimental values ​​regarding the binding strength of proteins and compounds (e.g., Kd, ​​Ki, IC50) as training data, along with protein structures and compound binding structures included in, for example, the PDBbind dataset. The binding affinity prediction model can provide concentration-based kinetic prediction values ​​for a pair in units of uM when a new target protein-compound pair is input. Accordingly, the binding strength prediction module (102) can input a docked target protein-compound pair into the binding affinity prediction model and obtain the output of the binding affinity prediction model as binding affinity. That is, the binding strength prediction module (102) can predict the binding affinity between a target protein and a compound after the active site of the target protein has been selected and the binding of the compound to the active site of the target protein has been satisfied.

[0048] In discovering candidate substances, the binding energy for a specific ligand is calculated as a physical constant in kcal / mol, which is a value that predicts the physical binding strength between a single protein and a single ligand. However, without concentration-based kinetic values ​​that explain how strongly multiple target protein molecules bind to and detach from multiple ligand molecules, experimenters must perform separate experiments at various concentrations ranging from low to high concentrations. In particular, it is difficult to observe the reaction below the effective concentration, and cells may die if the concentration exceeds a certain level, making the process of finding the optimal concentration very complex. Since the binding strength prediction module (102) provides concentration-based kinetic prediction values ​​for target protein-compound pairs in uM units, the user can conveniently estimate how much to dissolve and conduct the experiment based on the molecular weight of the compound.

[0049] The compound determination module (103) for the control agent can determine a compound to be used as a target substance control agent among a plurality of compounds based on binding affinity. Specifically, the compound determination module (103) for the control agent can determine a compound that shows high binding performance using binding affinity as an indicator as a compound to be used as a target substance control agent. Thus, a binding affinity-based compound determination method may be suitable when the number of target proteins is one or a small number.

[0050] When there are multiple target proteins, it is necessary to reflect not only binding affinity but also interactions with multiple target proteins. Therefore, when the number of target proteins exceeds a preset number, the compound determination module (103) for the control agent can determine a compound to be used as a target substance control agent among multiple compounds based on a score calculated by applying weights to the binding affinity.

[0051] Specifically, the compound determination module (103) for the regulatory agent can calculate a weighted score by applying a plurality of weights, which are determined for each of the plurality of target substance active microorganisms or the plurality of target proteins, to the binding affinity. That is, if the plurality of target substance active microorganisms include, for example, a first strain, a second strain, and a third strain, a first weight may be determined for the first strain, a second weight may be determined for the second strain, and a third weight may be determined for the third strain. The compound determination module (103) for the regulatory agent can calculate a first binding affinity with the first target protein of the first strain for the first compound, and calculate a first weighted score by applying a first weight to the first binding affinity. Additionally, the compound determination module (103) for the regulatory agent can calculate a second binding affinity with the second target protein of the second strain for the first compound, and calculate a second weighted score by applying a second weight to the second binding affinity. Additionally, the compound determination module (103) for the control agent can calculate the third binding affinity with the third target protein of the third strain for the first compound, and apply a third weight to the third binding affinity to calculate the third weight applied score. The compound determination module (103) for the control agent can calculate a final score for the first compound by combining the first weight applied score, the second weight applied score, and the third weight applied score.

[0052] The compound determination module (103) for the regulatory agent can perform the operation performed on the first compound on the second compound, which is another candidate substance. Specifically, the compound determination module (103) for the regulatory agent can calculate the first binding affinity with the first target protein of the first strain for the second compound, and apply a first weight to the first binding affinity to calculate the first weight applied score. In addition, the compound determination module (103) for the regulatory agent can calculate the second binding affinity with the second target protein of the second strain for the second compound, and apply a second weight to the second binding affinity to calculate the second weight applied score. In addition, the compound determination module (103) for the regulatory agent can calculate the third binding affinity with the third target protein of the third strain for the second compound, and apply a third weight to the third binding affinity to calculate the third weight applied score. The compound determination module (103) for the control agent can calculate a final score for the second compound by combining the first weighted score, the second weighted score, and the third weighted score. In the same way, the compound determination module (103) for the control agent can calculate a final score for the third to Nth compounds (N is a natural number). Here, for clarity and convenience of explanation, the microorganism active to the target substance is exemplified as a strain, but it is obvious that the scope of the present invention is not limited to strains and includes any microorganism active to the target substance.

[0053] The compound determination module (103) for controlling agents can determine which compound to use as a target substance controlling agent among a plurality of compounds by considering the final score for the first compound up to the final score for the Nth compound. For example, if the final score among the first to Nth compounds is lowest for the second compound, followed by the third compound, the first compound, the fourth compound, and the fifth compound in order of increasing, the compound determination module (103) for controlling agents can determine the second compound as the compound to use as a target substance controlling agent. Sometimes, a plurality of compounds with low final scores may be determined as compounds to use as target substance controlling agents. For example, the compound determination module (103) for controlling agents may determine three compounds with low final scores, namely the second compound, the third compound, and the first compound, as compounds to use as target substance controlling agents.

[0054] That is, the compound determination module (103) for the regulatory agent can load multiple binding affinity values ​​obtained for each of the multiple target substance active microorganisms for each compound into memory as multiple first scores. Additionally, the compound determination module (103) for the regulatory agent can load multiple predetermined weights into memory. The compound determination module (103) for the regulatory agent can calculate multiple second scores by applying multiple weights to the multiple first scores loaded in memory, and can calculate a final score determined as a single value for each compound from the multiple second scores. For example, the final score may be determined as a value corresponding to the average value of the multiple second scores, or as a value to which additional calculations are processed based on the average value. The compound determination module (103) for the regulatory agent can calculate the final score calculated in this way as a weighted score.

[0055] In some embodiments, a plurality of weights may include values ​​representing relative abundance obtained for each of a plurality of target substance-active microorganisms or a plurality of target proteins. Here, relative abundance may represent a ratio comparing how much of different target substance-active microorganisms or different target proteins exist in a specific environment. This value represents the proportion occupied by each target substance-active microorganism or target protein in an environment where a plurality of target substance-active microorganisms or a plurality of target proteins exist, and may reflect the influence on the production of target substances under specific conditions.

[0056] In some embodiments, a plurality of weights may include values ​​representing the relative abundance of target substance-active microorganisms obtained by performing rRNA sequencing (ribosomal ribonucleic acid sequencing) for each of the plurality of target substance-active microorganisms. Here, the relative abundance of target substance-active microorganisms may represent a ratio comparing how much of each different target substance-active microorganism exists in a specific environment. This value represents the proportion occupied by each target substance-active microorganism in an environment where multiple target substance-active microorganisms exist, and may reflect the influence on the production of the target substance under specific conditions. Such relative abundance may be calculated as a value representing the distribution of target substance-active microorganisms identified by determining the composition of the target substance-active microorganism community through the analysis of the rRNA (ribosomal RNA) of the target substance-active microorganisms.

[0057] The formulation data storage module (104) can configure formulation data of a target substance control agent including a determined compound and store the formulation data in a storage device. Specifically, the formulation data storage module (104) can generate formulation data that can function as design data for manufacturing a target substance control agent by using information about the compound determined by the compound determination module (103) for the control agent. For example, the formulation data generated by the formulation data storage module (104) can be transmitted to and utilized by another electronic device or computing device that controls the process of actually producing the control agent.

[0058] In some embodiments, when there is only one type of compound determined by the compound determination module (103) for the control agent, the formulation data may be configured to include that compound. In other embodiments, when there are multiple types of compounds determined by the compound determination module (103) for the control agent, the formulation data may include information regarding the formulation of the target substance control agent so that all of the multiple corresponding compounds are included.

[0059] In some embodiments, the target substance active microorganism includes methanogen, and the target substance regulating agent may include a methane-reducing feed additive.

[0060] In the rumen of ruminants, methane gas can be produced by various methane-producing microorganisms utilizing diverse substrates such as hydrogen (H2), carbon dioxide (CO2), and acetic acid (CH3COOH). This methane production process can be broadly divided into three main stages. The first is the hydrolysis stage, in which anaerobic microorganisms break down high-molecular-weight organic matter into low-molecular-weight organic matter. The second is the acid production stage, in which low-molecular-weight organic matter is converted into acetic acid, propionic acid, carbon dioxide, and hydrogen. Finally, in the methane production stage, methane-producing microorganisms primarily utilize carbon dioxide and hydrogen to generate methane. During this process, methane-producing microorganisms can efficiently produce methane through interactions.

[0061] Hydrogenotrophic methanogenesis is a methane production pathway that uses hydrogen (H2) and carbon dioxide (CO2) as substrates. In this process, hydrogen can be used as an electron to reduce carbon dioxide to methane. Carbon dioxide first combines with methanofuran and can then be reduced to a methyl group with the help of tetrahydromethanopterin. Subsequently, methyl-coenzyme M (Methyl-CoM) is synthesized through the action of transferase, and finally, methane can be produced by coenzyme B. This pathway is the most commonly found methane production pathway in anaerobic environments, particularly in the rumen, and is a pathway that produces methane using hydrogen and carbon dioxide as the main substrates.

[0062] Acetoclastic methanogenesis is a pathway that produces methane using acetic acid (CH3COOH) as a substrate. In this pathway, acetic acid is broken down into methane and carbon dioxide, and acetic acid is converted to acetyl-coenzyme A (acetyl-CoA) by enzymes such as acetyl-coenzyme A synthetase, acetate kinase, and phosphotransacetylase. Subsequently, methyl-coenzyme M (Methyl-CoM) is synthesized through the action of carbon monoxide dehydrogenase, and methane is produced with the help of coenzymes. This pathway is a process that directly converts acetic acid into methane and consumes relatively little energy; however, due to its low efficiency in environments such as the rumen, it may be less common compared to pathways that primarily use hydrogen and carbon dioxide.

[0063] Methylotrophic methanogenesis is a pathway that utilizes methylated compounds, such as methanol and methylamine, as substrates. In this process, methylated compounds, such as methanol or methylamine, can be directly converted into methane or converted into methane through various metabolic pathways. The methyl group of a methylated compound combines with Coenzyme M to form Methyl-Coenzyme M (Methyl-CoM), which is ultimately converted into methane by Coenzyme B. This pathway can primarily be carried out by microorganisms that use methanol or methylamine as substrates in environments such as oceans, wetlands, and anaerobic digesters.

[0064] The target substance control formulation manufacturing device (10) can discover a compound suitable as a feed additive capable of reducing methane gas generated in the rumen of ruminants, and can manufacture a methane-reducing feed additive using the same. That is, formulation data is composed including the compound discovered by the target substance control formulation manufacturing device (10), and the formulation data can actually be used in the manufacture of the methane-reducing feed additive.

[0065] The structural data acquisition module (101) can acquire sequence or structural data (e.g., three-dimensional structural data) regarding multiple target proteins corresponding to multiple methanogenic bacteria. Here, methanogenic bacteria are a type of archaea that produces methane (CH4) during the process of decomposing organic matter in an anaerobic environment, as previously described, and can be found in an oxygen-free environment such as the digestive organ (rumen) of a ruminant animal.

[0066] In some embodiments, a plurality of methanogenic bacteria include Methanobrevibacter smithii, Methanobrevibacter millerae, Methanobrevibacter thaurei, Methanobrevibacter ruminantium, Methanosphaera stadtmanae, Methanobrevibacter olleyae, Methanomicrobium mobile, Methanobrevibacter gottschalkii, Methanosarcina barkeri, and Methanobacterium bryanthi It may be one or more selected from Methanobacterium bryantii, Methanobacterium formicicum, and Methanocaldococcus jannaschii. In this case, the target protein may include methyl-coenzyme M reductase (MCR).

[0067] MCR can be a key enzyme that catalyzes the final step of the methane production process in methanogenic bacteria. MCR can catalyze the reaction that produces methane (CH4) using methyl-coenzyme M (CH3-S-CoM) and coenzyme B (HS-CoB) during the methane production process. Specifically, as methyl-coenzyme M (CH3-S-CoM) is reduced, the methyl group is activated by nickel (F430), and the methyl group is converted into methane (CH4) through a reaction with coenzyme B (HS-CoB). Finally, a mixed disulfide called CoM-SS-CoB (Coenzyme M-Coenzyme B heterodisulfide) is produced, through which methane can be released. In other words, since MCR plays a decisive role in the methane production process, inhibiting the activity of MCR can reduce methane production.

[0068] The structure data acquisition module (101) may also receive structure data of a plurality of compounds (e.g., three-dimensional structure data) from a pre-prepared compound structure library. In some embodiments, the compounds may be various natural compounds derived from nature (e.g., tannin, flavonoid, phenolic compounds, etc.). The compound structure library may store and manage information regarding the structure of these compounds.

[0069] The binding strength prediction module (102) can predict the binding strength for target protein-compound pairs combined from a plurality of target proteins and a plurality of compounds, and for details regarding this, the previously described content regarding the binding strength prediction module (102) above may be referenced.

[0070] The compound determination module (103) for the control agent can determine a compound to be used as a methane-reducing feed additive among a plurality of compounds based on binding affinity. Specifically, the compound determination module (103) can determine a compound that shows a high binding performance using binding affinity as an indicator as a compound to be used as a methane-reducing feed additive. Thus, a compound determination method based on binding affinity may be suitable when the number of target proteins is one or a small number.

[0071] When there are multiple target proteins, it is necessary to reflect not only binding affinity but also interactions with multiple target proteins. Therefore, when the number of target proteins exceeds a preset number, the compound determination module (103) for the control agent can determine a compound to be used as a methane-reducing feed additive among multiple compounds based on a score calculated by applying weights to the binding affinity.

[0072] Specifically, the compound determination module (103) for the regulatory agent can calculate a weighted score by applying a plurality of weights determined for each of the plurality of methane-producing bacteria to the binding affinity. That is, when the plurality of methane-producing bacteria include a first strain, a second strain, and a third strain, a first weight may be determined for the first strain, a second weight may be determined for the second strain, and a third weight may be determined for the third strain. The compound determination module (103) for the regulatory agent can calculate a first binding affinity with a first target protein of the first strain for the first compound and calculate a first weighted score by applying a first weight to the first binding affinity. Additionally, the compound determination module (103) for the regulatory agent can calculate a second binding affinity with a second target protein of the second strain for the first compound and calculate a second weighted score by applying a second weight to the second binding affinity. Additionally, the compound determination module (103) for the control agent can calculate the third binding affinity with the third target protein of the third strain for the first compound, and apply a third weight to the third binding affinity to calculate the third weight applied score. The compound determination module (103) for the control agent can calculate a final score for the first compound by combining the first weight applied score, the second weight applied score, and the third weight applied score.

[0073] The compound determination module (103) for the regulatory agent can perform the operation performed on the first compound on the second compound, which is another candidate substance. Specifically, the compound determination module (103) for the regulatory agent can calculate the first binding affinity with the first target protein of the first strain for the second compound, and apply a first weight to the first binding affinity to calculate the first weight applied score. In addition, the compound determination module (103) for the regulatory agent can calculate the second binding affinity with the second target protein of the second strain for the second compound, and apply a second weight to the second binding affinity to calculate the second weight applied score. In addition, the compound determination module (103) for the regulatory agent can calculate the third binding affinity with the third target protein of the third strain for the second compound, and apply a third weight to the third binding affinity to calculate the third weight applied score. The compound determination module (103) for the control agent can calculate a final score for the second compound by combining the first weighted score, the second weighted score, and the third weighted score. In the same way, the compound determination module (103) for the control agent can calculate a final score for the third to Nth compounds (N is a natural number).

[0074] The compound determination module (103) for controlling formulations can determine a compound to be used as a methane-reducing feed additive among a plurality of compounds by considering the final score for the first compound up to the final score for the Nth compound. For example, if the final score among the first to Nth compounds is lowest for the second compound, followed by the third compound, the first compound, the fourth compound, and the fifth compound in order of increasing, the compound determination module (103) for controlling formulations can determine the second compound as the compound to be used as a methane-reducing feed additive. Sometimes, a plurality of compounds with low final scores may be determined as compounds to be used as methane-reducing feed additives. For example, the compound determination module (103) for controlling formulations may determine three compounds with low final scores, namely the second compound, the third compound, and the first compound, as compounds to be used as methane-reducing feed additives.

[0075] That is, the compound determination module (103) for the control agent can load multiple binding affinity values ​​obtained for each of the multiple methane-producing bacteria for each compound into memory as multiple first scores. Additionally, the compound determination module (103) for the control agent can load multiple predetermined weights into memory. The compound determination module (103) for the control agent can calculate multiple second scores by applying multiple weights to the multiple first scores loaded in memory, and can calculate a final score determined as a single value for each compound from the multiple second scores. For example, the final score may be determined as a value corresponding to the average value of the multiple second scores, or as a value to which additional calculations are processed based on the average value. The compound determination module (103) for the control agent can calculate the final score calculated in this way as a weighted score.

[0076] In some embodiments, a plurality of weights may include values ​​representing the relative abundance of each methane-producing bacterium obtained by performing rRNA sequencing for each of the plurality of methane-producing bacteria. Here, the relative abundance of each methane-producing bacterium may represent a ratio comparing how much of each different methane-producing bacterium exists in a specific environment. This value represents the proportion occupied by each strain in an environment where methane-producing microorganisms exist, such as the rumen of ruminants, and may reflect the influence on methane production under specific conditions. Such relative abundance can be calculated as a value representing the distribution of methane-producing microorganisms in the rumen, which is identified by determining the composition of the microbial community through the analysis of the rRNA of methane-producing microorganisms. For example, to understand the composition of methanogenic bacteria and bacterial communities in the rumen microbiome, the relative abundance of each methanogenic bacterium can be evaluated by using the PacBio HiFi Long-read sequencing platform to analyze the entire 16S rRNA gene region (about 1.5 kb) used to identify and analyze bacterial and archaea microbial communities.

[0077] For example, relative abundance can be calculated for multiple methane-producing strains in multiple samples (e.g., five samples), and the average of these values, i.e., the average relative abundance, can be obtained as a weight. A higher weight value may indicate that the methane-producing strain is more widespread in a specific environment or condition. That is, being the most abundant in the entire sample means having the highest contribution to methane production, so an inhibitor targeting such high-concentration strains can have a greater impact on reducing overall methane emissions. Conversely, since low-concentration strains are few in number, targeting these strains may have a smaller impact on overall methane production. In other words, the compound determination module (103) for the control agent can determine the compound to be used as a methane-reducing feed additive by giving priority to target substance-active microorganisms or target proteins with a large impact, while simultaneously considering the overall diversity of the microbial community.

[0078] In some embodiments, the compound determination module (103) for the control agent is a minimization function Min such as the following k It can operate based on.

[0079] Min k { S j = ∑(i=0 to n)(ω i * D ij )}

[0080] Here, ω i is the weight for the microbial community and the inhibitor, and D ij can represent the corresponding suppression measure. Minimization function Min k It aims to find inhibitor combinations that can minimize the overall methane production potential by targeting the most widespread and influential methane-producing strains. That is, the minimization function Min kIt can be understood that the lower the value of , the stronger the bond. In some embodiments, the relative abundance and bond affinity described above are ω i and D ij It can correspond to. i and n may represent the item number and total number of items in the weight matrix for microbial communities and inhibitors. For example, if there are a total of 12 items of methane-producing strains, i increases from 1 to n (12). ω1 may represent the relative abundance (weight) of the first methane-producing strain, and ω2 may represent the relative abundance (weight) of the second methane-producing strain. j and m may represent the item number and total number of items of candidate ligands. For example, if there are 30,000 candidate ligands, j may increase from 1 to m (30,000).

[0081] In some embodiments, weights may be applied only to target substance active microorganisms that have high relative abundance in a preset number of samples, and not to other target substance active microorganisms. In some embodiments, the criterion for target substance active microorganisms with high relative abundance may be determined based on whether the average of relative abundance values ​​obtained from multiple samples is above a preset threshold. For example, if the relative abundance of a specific methane-producing strain appears above the threshold in all five samples, that strain may be included in the weight matrix and weights may be applied. Conversely, strains that do not meet the threshold due to low relative abundance in some samples may be excluded from the weight matrix and not reflected in the calculation.

[0082] In addition, in some embodiments, the threshold value may be dynamically calculated based on the relative abundance distribution of the entire microbial community. For example, weights may be applied only to target substance active microorganisms having relative abundance values ​​in the top 30%, while target substance active microorganisms in the bottom 70% may be excluded. This method allows for the more efficient identification of compounds with high actual methane reduction effects during the final score calculation process by selectively reflecting microorganisms that contribute significantly to methane production.

[0083] Furthermore, in some embodiments, whether weights are applied may vary depending on sample data for each environmental condition. For example, even the same methane-producing strain may exhibit high abundance in a rumen environment but low abundance in a wetland environment. In such cases, the strain may be included in formulation data specialized for the rumen environment but excluded from formulation data specialized for the wetland environment.

[0084] The formulation data storage module (104) can configure formulation data for a methane-reducing feed additive including a determined compound and store the formulation data in a storage device. Specifically, the formulation data storage module (104) can generate formulation data that can function as design data for manufacturing a methane-reducing feed additive by using information about the compound determined by the control agent compound determination module (103) and information about the feed. For example, the formulation data generated by the formulation data storage module (104) can be transmitted to and utilized by another electronic device or computing device that controls the process of actually producing feed.

[0085] In some embodiments, when there is only one type of compound determined by the compound determination module (103) for the control agent, the formulation data may be configured to include that compound. In other embodiments, when there are multiple types of compounds determined by the compound determination module (103) for the control agent, the formulation data may include information regarding the formulation of the methane-reducing feed additive so that all of the multiple corresponding compounds are included.

[0086] FIGS. 2 and FIGS. 3 are drawings for explaining the operation of a target substance control agent manufacturing apparatus according to one embodiment. Here, the description is given as an example where the target substance active microorganism includes a methane-producing bacterium and the target substance control agent includes a methane-reducing feed additive, but the description extends to any target substance active microorganism and target substance control agent that are not limited to methane-producing bacteria and methane-reducing feed additives.

[0087] Referring to FIG. 2, a group of methane-producing bacteria (20) comprising a plurality of methane-producing bacteria (201, 202, 203) may be identified. A structural data acquisition module (101) may acquire three-dimensional structural data regarding a plurality of target proteins (211, 212, 213) corresponding to the plurality of methane-producing bacteria (201, 202, 203). Here, the plurality of target proteins (211, 212, 213) may form a target protein group (21). Meanwhile, the structural data acquisition module (101) may receive three-dimensional structural data of a plurality of compounds (221, 222, 223) from a pre-prepared compound structure library (22).

[0088] Next, referring to FIG. 3, the binding force prediction module (102) can predict the binding energy for the target protein-compound pair by performing docking on the target protein-compound pair and predict the binding affinity between the docked target protein and compound using a pre-trained binding affinity prediction model.

[0089] For example, referring to table (T1), the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 for compound (CH0001) can be predicted as 'SC00001_1', 'SC00001_2', 'SC00001_3', ..., 'SC00001_12', respectively. And the average binding affinity 'SC00001_A' for 'SC00001_1', 'SC00001_2', 'SC00001_3', ..., 'SC00001_12' can be calculated. Next, the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 for the compound (CH0002) can be predicted as 'SC00002_1', 'SC00002_2', 'SC00002_3', ..., 'SC00002_12', respectively. And the average binding affinity 'SC00002_A' for 'SC00002_1', 'SC00002_2', 'SC00002_3', ..., 'SC00002_12' can be calculated. By repeating this process, the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 to the compound (CH37576) are predicted as 'SC37576_1', 'SC37576_2', 'SC37576_3', ..., 'SC37576_12', respectively, and the average binding affinity 'SC37576_A' for 'SC37576_1', 'SC37576_2', 'SC37576_3', ..., 'SC37576_12' can be calculated.

[0090] The compound determination module (103) for controlling formulations can determine a compound to be used as a methane-reducing feed additive among a plurality of compounds (CH0001 to CH37576) based on the average binding affinity indicated by region (A1).

[0091] FIG. 4 is a flowchart illustrating a method for manufacturing a target substance control agent according to one embodiment.

[0092] Referring to FIG. 4, a method for manufacturing a target substance regulatory agent according to one embodiment may include the steps of: obtaining sequence or structural data regarding a plurality of target substance active microorganisms (S401); receiving structural data of a plurality of compounds from a pre-prepared compound structure library (S402); predicting a binding force for a target protein-compound pair combined from a plurality of target proteins and a plurality of compounds (S403); determining a compound to be used as a target substance regulatory agent among a plurality of compounds based on the binding force (S404); configuring formulation data of a target substance regulatory agent including the determined compound (S405); and storing the formulation data in a storage device (S406).

[0093] For more detailed information regarding the above method, reference may be made to other embodiments described in this specification; therefore, redundant details are omitted here.

[0094] FIGS. 5 and 6 are drawings for explaining the operation of a target substance control agent manufacturing apparatus according to one embodiment. Here, the description is given as an example where the target substance active microorganism includes a methane-producing bacterium and the target substance control agent includes a methane-reducing feed additive, but the description extends to any target substance active microorganism and target substance control agent that are not limited to methane-producing bacteria and methane-reducing feed additives.

[0095] Referring to FIG. 5, when there are multiple target proteins, it is necessary to reflect not only binding affinity but also interactions with multiple target proteins. Therefore, when the number of target proteins exceeds a preset number, the compound determination module (103) for the control agent can determine a compound to be used as a methane-reducing feed additive among multiple compounds based on a score calculated by applying weights to the binding affinity.

[0096] Referring to the table (T1) shown in FIG. 3, the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 for the compound (CH0001) can be predicted as 'SC00001_1', 'SC00001_2', 'SC00001_3', ..., 'SC00001_12', respectively. And the average binding affinity 'SC00001_A' for 'SC00001_1', 'SC00001_2', 'SC00001_3', ..., 'SC00001_12' can be calculated. Next, the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 for the compound (CH0002) can be predicted as 'SC00002_1', 'SC00002_2', 'SC00002_3', ..., 'SC00002_12', respectively. And the average binding affinity 'SC00002_A' for 'SC00002_1', 'SC00002_2', 'SC00002_3', ..., 'SC00002_12' can be calculated. By repeating this process, the binding affinities for methane-producing strains 1, 2, 3, ..., and 12 to the compound (CH37576) are predicted as 'SC37576_1', 'SC37576_2', 'SC37576_3', ..., 'SC37576_12', respectively, and the average binding affinity 'SC37576_A' for 'SC37576_1', 'SC37576_2', 'SC37576_3', ..., 'SC37576_12' can be calculated.

[0097] Next, referring to the table (T2) shown in FIG. 5, relative abundances 'RA01_1', 'RA01_2', 'RA01_3', 'RA01_4', and 'RA01_5' can be calculated from five samples for strain (MS01). Then, the average relative abundance 'RA01_A' for 'RA01_1', 'RA01_2', 'RA01_3', 'RA01_4', and 'RA01_5' can be calculated. Next, relative abundances 'RA02_1', 'RA02_2', 'RA02_3', 'RA02_4', and 'RA02_5' can be calculated from five samples for strain (MS02). And the average relative abundance 'RA02_A' for 'RA02_1', 'RA02_2', 'RA02_3', 'RA02_4', and 'RA02_5' can be calculated. By repeating this process, relative abundances 'RA12_1', 'RA12_2', 'RA12_3', 'RA12_4', and 'RA12_5' are calculated from five samples for strain (MS12), and the average relative abundance 'RA12_A' for 'RA12_1', 'RA12_2', 'RA12_3', 'RA12_4', and 'RA12_5' can be calculated. That is, the area indicated by region (A2) can represent the average relative abundance per strain.

[0098] Next, referring to the table (T3) shown in FIG. 6, the weighted scores for strains 1, 2, 3, ..., and 12 that produce methane for the compound (CH0001) can be predicted as 'SC00001_1' X 'RA01_A', 'SC00001_2' X 'RA02_A', 'SC00001_3' X 'RA03_A', ..., 'SC00001_12' X 'RA12_A', respectively. And the average weighted score 'SC00001_F' for 'SC00001_1' X 'RA01_A', 'SC00001_2' X 'RA02_A', 'SC00001_3' X 'RA03_A', ..., 'SC00001_12' X 'RA12_A' can be calculated. Next, the weighted scores for methane-producing strains 1, 2, 3, ..., and 12 for compound (CH0002) can be predicted as 'SC00002_1' X 'RA01_A', 'SC00002_2' X 'RA02_A', 'SC00002_3' X 'RA03_A', ..., 'SC00002_12' X 'RA12_A', respectively. And the average weighted score 'SC00002_F' for 'SC00002_1' X 'RA01_A', 'SC00002_2' X 'RA02_A', 'SC00002_3' X 'RA03_A', ..., 'SC00002_12' X 'RA12_A' can be calculated. By repeating this process, the weighted scores for methane-producing strains 1, 2, 3, ..., and 12 for the compound (CH37576) are predicted to be 'SC37576_1' X 'RA01_A', 'SC37576_2' X 'RA02_A', 'SC37576_3' X 'RA03_A', ..., 'SC37576_12' X 'RA12_A', respectively, and 'SC37576_1' X 'RA01_A', 'SC37576_2' X 'RA02_A', 'SC37576_3' X 'RA03_A', ..., the average weighted score 'SC37576_F' for 'SC37576_12' X 'RA12_A' can be calculated. That is, the area (A3) indicates the average weighted score, i.e., the final score.

[0099] The compound determination module (103) for the control agent can determine a compound to be used as a methane-reducing feed additive among a plurality of compounds (CH0001 to CH37576) based on the final score indicated by the area (A3).

[0100] FIG. 7 is a flowchart illustrating a method for manufacturing a target substance control agent according to one embodiment.

[0101] Referring to FIG. 7, a method for manufacturing a target substance control agent according to one embodiment may include the steps of: obtaining sequence or structural data regarding a plurality of target proteins corresponding to a plurality of target substance active microorganisms (S701); receiving structural data of a plurality of compounds from a pre-prepared compound structure library (S702); predicting binding strength for a target protein-compound pair combined from a plurality of target proteins and a plurality of compounds (S703); applying a plurality of weights determined for each of the plurality of target substance active microorganisms or a plurality of target proteins to the binding strength to calculate a weighted application score (S704); determining a compound to be used as a target substance control agent among a plurality of compounds according to the weighted application score (S705); configuring formulation data of a target substance control agent including the determined compound (S706); and storing the formulation data in a storage device (S707).

[0102] For more detailed information regarding the above method, reference may be made to other embodiments described in this specification; therefore, redundant details are omitted here.

[0103] FIG. 8 is a diagram illustrating an example of the implementation of a binding affinity prediction model used in the manufacture of a target substance control formulation according to one embodiment.

[0104] Referring to FIG. 8, the binding affinity prediction model used by the binding strength prediction module (102) to predict binding affinity may be trained to take a 4-dimensional tensor containing the 3-dimensional coordinates of atoms and a single training feature vector as input and output a predicted value of binding affinity. Specifically, the binding affinity prediction model may include a Deep 4D Convolutional Neural Network that predicts the binding affinity between a protein and a ligand using a single output neuron. The neural network may largely include a convolutional layer section and a dense layer section. The convolutional layer section may generate a feature map that highlights the spatial occurrence of a specific pattern within the data by identifying patterns encoded by the filters of each convolutional layer. For example, the model uses three convolutional layers with 64, 128, and 256 filters, respectively, and the output of the last convolutional layer can be flattened and passed as input to the dense layer blocks. Next, the dense layers may include, for example, three layers composed of 1,000, 500, and 200 neurons, respectively. Dropout was applied to each dense layer, and the dropout probability was set to 0.5. Additionally, L2 weighted regularization with λ = 0.001 is applied to prevent overfitting. Since the ReLU activation function is used in both the convolutional layers and the dense layers, non-linearity may be introduced into the model.

[0105] In some embodiments, the 4-dimensional tensor used in the binding affinity prediction model may be constructed by combining the 3-dimensional coordinates of atoms and a training feature vector. Here, the 3-dimensional coordinates provide positional information for each atom and may be an essential element for understanding spatial interactions between proteins and ligands. The training feature vector contains 19 characteristics for each atom, which can enable the model to understand various chemical interactions between atoms. These 19 characteristics may include information such as atomic type (e.g., C, N, O, etc.), bonding state, hybridization state, partial charge, hydrophilicity / hydrophobicity, polarity, atomic radius, electronegativity, polarizability, atomic mass, ionization energy, oxidation state, covalent radius, intra-molecular position, electron affinity, hybrid bond length, rotatability, steric hindrance, and hydrogen bond donor / acceptor ability. By reflecting the different characteristics of each atom, this vector can accurately predict the strength of the binding between a protein and a ligand. Therefore, the 4-dimensional tensor can provide the complex data necessary to predict protein-ligand binding affinity by combining spatial information (3 dimensions) and chemical property information (1 dimension) of atoms.This method may be designed so that the model can predict binding affinity even in new protein-ligand combinations by learning binding data of various protein-ligand complexes, particularly through the PDBBind dataset.

[0106] FIG. 9 is a block diagram illustrating a computing device according to one embodiment.

[0107] Referring to FIG. 9, the method and apparatus for manufacturing a target substance control formulation according to the embodiments can be implemented using a computing device (50). This computing device (50) can be implemented as various types of electronic devices, servers, or similar devices, and its functions can be implemented through a combination of software and hardware.

[0108] The computing device (50) may include at least one of a processor (501) communicating via a bus (509), a memory (502), a storage device (503), a display device (504), a network interface device (505) providing access to a network (40) for communication with other entities, and an input / output interface device (506) providing a user input interface or a user output interface. Of course, the computer device (50) may additionally include any electronic device necessary to implement the technical concept described in this specification, although not shown in FIG. 9.

[0109] The processor (501) can be implemented as various types of computing devices, such as an MCU (Micro Controller Unit), AP (Application Processor), CPU (Central Processing Unit), GPU (Graphic Processing Unit), NPU (Neural Processing Unit), QPU (Quantum Processing Unit), etc. The processor (501) is a semiconductor device that executes instructions stored in memory (502) or storage device (503) and can perform a core role in the system. Program code and data stored in memory (502) or storage device (503) instruct the processor (501) to perform specific tasks, thereby enabling the operation of the entire system. The processor (501) can be configured to implement the functions or methods described above in relation to FIGS. 1 through 8.

[0110] The memory (502) and storage device (503) may include various forms of volatile or non-volatile storage media for storing and accessing data of the system. For example, the memory (502) may include read-only memory (ROM) or random access memory (RAM). In some embodiments, the memory (502) may be embedded inside the processor (501), in which case the data transfer speed between the memory (502) and the processor (501) may be very fast. In some other embodiments, the memory (502) may be located outside the processor (501), in which case the memory (502) may be connected to the processor (501) through various data buses or interfaces. Such connection may be made through various known means, for example, a PCIe (Peripheral Component Interconnect Express) interface for high-speed data transfer or a memory controller. Meanwhile, examples of storage devices (503) include HDD (Hard Disk Drive) or SSD (Solid State Drive), and the scope of the present invention is not limited to the elements listed above for the purpose of explanation.

[0111] In some embodiments, at least some components or functions of the method and apparatus for manufacturing a target substance control agent according to the embodiments may be implemented as a program or software executed on a computing device (50), and the program or software may be stored on a computer-readable recording medium or storage medium. Specifically, a computer-readable recording medium or storage medium according to one embodiment may have a program recorded on a computer that includes a processor (501) that executes a program or instructions stored in a memory (502) or a storage device (503) for executing steps included in the implementation of the method and apparatus for manufacturing a target substance control agent according to the embodiments.

[0112] In some embodiments, at least some of the components or functions of the method and apparatus for manufacturing a target substance control formulation according to the embodiments may be implemented using hardware or circuits of a computing device (50), or may be implemented using separate hardware or circuits that can be electrically connected to the computing device (50).

[0113] According to the embodiments, target substance regulating agents can be discovered and manufactured using binding affinity as an indicator based on molecular docking and artificial intelligence technologies. In particular, by reflecting the interactions between multiple target substance-active microorganisms and multiple target proteins, and based on a score calculated by applying weights to binding affinity, a compound to be used as a target substance regulating agent can be determined by prioritizing target substance-active microorganisms or target proteins that have a significant influence in a specific environment or condition, while also considering the overall diversity of the target substance-active microorganism or target protein community. Similarly, based on molecular docking and artificial intelligence technologies, a substance for methane-reducing feed additives that is safe to use and has excellent methane reduction effects can be discovered using binding affinity as an indicator, and a methane-reducing feed additive containing said substance can be manufactured. In particular, by reflecting the interactions between multiple methane-producing bacteria and multiple target proteins, and based on a score calculated by applying weights to binding affinity, a compound to be used as a methane-reducing feed additive can be determined by prioritizing methane-producing strains that have a significant influence in a specific environment or condition, while also considering the overall diversity of the microbial community.

[0114] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art to which the present invention belongs, using the basic concept of the present invention as defined in the following claims, also fall within the scope of the present invention.

Claims

1. A method for manufacturing a target substance control agent, performed by a computing device including a processor, memory, and storage device, wherein The above processor acquires sequence or structural data regarding a plurality of target proteins corresponding to a plurality of target substance-active microorganisms; The above processor receives structural data of a plurality of compounds from a pre-prepared compound structure library; The above processor predicts the binding affinity for a target protein-compound pair combined from the plurality of target proteins and the plurality of compounds; The above processor calculates a weighted score by applying a plurality of weights, determined for each of the plurality of target substance active microorganisms, to the binding force; The above processor determines, according to the weighted score, a compound to be used as a target substance control agent among the plurality of compounds; and The processor comprises the step of configuring formulation data of the target substance control agent including the determined compound, and storing the formulation data in the storage device. The above plurality of weights are, A value representing the relative abundance obtained for each of the plurality of target substance-active microorganisms or the plurality of target proteins, Method for manufacturing a target substance control agent.

2. In Paragraph 1, A method for manufacturing a target substance regulating agent, wherein the plurality of weights include values ​​representing the relative abundance of each target substance active microorganism obtained by performing rRNA sequencing (ribosomal ribonucleic acid sequencing) for each of the plurality of target substance active microorganisms.

3. In Paragraph 1, The step of predicting the bonding force above is, The above processor performs docking on the target protein-compound pair to predict the binding pose for the target protein-compound pair; The processor loads a pre-trained coupling affinity prediction model into the memory; and A method for manufacturing a target substance regulatory agent, comprising the step of the processor inputting the docked target protein-compound pair into the binding affinity prediction model and obtaining the output of the binding affinity prediction model as the binding affinity.

4. In Paragraph 3, A method for manufacturing a target substance control agent, wherein the above-described binding affinity prediction model is trained to take a 4-dimensional tensor containing the 3-dimensional coordinates of atoms and one training feature vector as input and output a predicted value of the binding affinity.

5. In Paragraph 3, A method for manufacturing a target substance regulating agent, wherein the above-mentioned binding affinity prediction model is trained using experimental values ​​regarding the binding strength of a protein and a compound as training data.

6. In Paragraph 1, The step of calculating the above-mentioned weighted score is, The processor loads a plurality of binding affinity values ​​obtained for each of the plurality of target proteins for each of the compounds into the memory as a plurality of first scores; The above processor loads the plurality of weights into the memory; The above processor calculates a plurality of second scores by applying a plurality of weights to a plurality of first scores loaded in the memory; The above processor calculates a final score determined as a single value for each compound from the plurality of second scores; and A method for manufacturing a target substance control agent, comprising the step of the processor calculating the final score as the weighted score.

7. In Paragraph 1, A method for preparing a target substance-regulating agent, wherein the target substance-active microorganism is selected from methanogen, hydrogen sulfide-producing, acetate-producing, butyrate-producing, hydrogen-producing, carbon monoxide-producing, nitrogen-producing, or nitrous oxide-producing bacteria.

8. In Paragraph 7, The above-mentioned methane-producing bacteria are Methanobrevibacter smithii, Methanobrevibacter millerae, Methanobrevibacter thaurei, Methanobrevibacter ruminantium, Methanosphaera stadtmanae, Methanobrevibacter olleyae, Methanomicrobium mobile, Methanobrevibacter gottschalkii, Methanosarcina barkeri, and Methanobacterium bryanthi A method for manufacturing a target substance modulator, comprising one or more selected from Methanobacterium bryantii, Methanobacterium formicicum, and Methanocaldococcus jannaschii.

9. In Paragraph 8, A method for preparing a target substance regulating agent, wherein the above target protein comprises methyl-coenzyme M reductase.

10. In Paragraph 1, The compound determined above comprises a plurality of compounds, and A method for manufacturing a target substance control agent, wherein the above formulation data includes information regarding the formulation of the target substance control agent such that all of the above plurality of compounds are included.

11. As a device for manufacturing a target substance control formulation, processor; and It includes a memory in which instructions executed by the above processor are loaded, and The above instruction is executed by the processor, causing the processor, Acquiring sequence or structural data regarding multiple target proteins corresponding to multiple target substance-active microorganisms, and Structural data of multiple compounds is provided from a pre-prepared compound structure library, and Predicting binding strength for target protein-compound pairs combined from the plurality of target proteins and the plurality of compounds, and A weighted score is calculated by applying a plurality of weights, determined for each of the plurality of target substance active microorganisms, to the above binding force, and Based on the above weighted score, a compound to be used as a target substance control agent among the plurality of compounds is determined, and Formulating the formulation data of the target substance control agent including the compound determined above, and The above formulation data is stored in a storage device, and The above plurality of weights are, A value representing the relative abundance obtained for each of the plurality of target substance-active microorganisms or the plurality of target proteins, Target substance control formulation manufacturing device.

12. In Paragraph 11, A device for manufacturing a target substance regulatory agent, wherein the plurality of weights above include values ​​representing the relative abundance of each target substance active microorganism obtained by performing rRNA sequencing (ribosomal ribonucleic acid sequencing) for each of the plurality of target substance active microorganisms.

13. In Paragraph 11, Predicting the above bonding force is, Docking is performed on the above target protein-compound pair to predict the binding pose for the above target protein-compound pair, and Load a pre-trained coupling affinity prediction model into the memory, and An apparatus for manufacturing a target substance control agent, comprising inputting the docked target protein-compound pair into the binding affinity prediction model and obtaining the output of the binding affinity prediction model as the binding affinity.

14. In Paragraph 13, A device for manufacturing a target substance control agent, wherein the above-described binding affinity prediction model is trained to take a 4-dimensional tensor containing the 3-dimensional coordinates of atoms and one training feature vector as input and output a predicted value of the binding affinity.

15. In Paragraph 13, A device for manufacturing a target substance control agent, wherein the above-mentioned binding affinity prediction model is trained using experimental values ​​regarding the binding strength of a protein and a compound as training data.

16. In Paragraph 11, Calculating the above weighted score is, For each of the above compounds, a plurality of binding affinity values ​​obtained for each of the plurality of target proteins are loaded into the memory as a plurality of first scores, and Loading the above plurality of weights into the memory, Calculate a plurality of second scores by applying the plurality of weights to the plurality of first scores loaded in the memory, and Calculate a final score determined as a single value for each compound from the plurality of second scores above, and A device for manufacturing a target substance control agent, comprising calculating the above final score as the above weighted score.

17. In Paragraph 11, A device for manufacturing a target substance control agent, wherein the target substance active microorganism is selected from methanogen, hydrogen sulfide-producing, acetate-producing, butyrate-producing, hydrogen-producing, carbon monoxide-producing, nitrogen-producing, or nitrous oxide-producing bacteria.

18. In Paragraph 17, The above-mentioned methane-producing bacteria are Methanobrevibacter smithii), Methanobrevibacter millerae, Methanobrevibacter thaurei, Methanobrevibacter ruminantium, Methanosphaera stadtmanae, Methanobrevibacter olleyae, Methanomicrobium mobile, Methanobrevibacter gottschalkii, Methanosarcina barkeri, and Methanobacterium bryanthi A device for manufacturing a target substance control agent, comprising one or more selected from Methanobacterium bryantii, Methanobacterium formicicum, and Methanocaldococcus jannaschii.

19. In Paragraph 18, The above target protein is a device for manufacturing a target substance regulatory agent comprising methyl-coenzyme M reductase.

20. A computer-readable recording medium that records instructions executed by a computing device including a processor, memory and storage device, The above instruction is executed by the computing device, causing the computing device, Acquiring sequence or structural data regarding multiple target proteins corresponding to multiple target substance-active microorganisms, and Structural data of multiple compounds is provided from a pre-prepared compound structure library, and Predicting binding strength for target protein-compound pairs combined from the plurality of target proteins and the plurality of compounds, and A weighted score is calculated by applying a plurality of weights, determined for each of the plurality of target substance active microorganisms, to the above binding force, and Based on the above weighted score, a compound to be used as a target substance control agent among the plurality of compounds is determined, and Formulating the formulation data of the target substance control agent including the compound determined above, and The above formulation data is stored in the storage device, and The above plurality of weights are, A value representing the relative abundance obtained for each of the plurality of target substance-active microorganisms or the plurality of target proteins, Computer-readable recording medium.

Images