Method for estimating enzyme inhibition constant by using single inhibitor concentration

Abstract

A method for estimating an enzyme inhibition constant, according to one embodiment of the present disclosure, may comprise the steps of: measuring initial reaction rate data at a plurality of substrate concentrations and one inhibitor concentration; setting the one inhibitor concentration to be greater than or equal to a predetermined half-maximal inhibitory concentration; and using the measured initial reaction rate data and the predetermined half-maximal inhibitory concentration value so as to estimate an inhibition constant by fitting to an enzyme inhibition model and applying a normalization term that includes the relationship between the half-maximal inhibitory concentration and at least one candidate inhibition constant.

Description

A method for estimating enzyme inhibition constants using single inhibitor concentrations

[0001] The present disclosure relates to enzyme inhibition analysis and to a method for estimating an enzyme inhibition constant using the concentration of a single inhibitor.

[0002] Drugs are metabolized by enzymes in the liver and intestines and then excreted from the body. If two drugs are taken simultaneously, one drug may bind to the enzyme before the other, interfering with the metabolic process, which can delay breakdown and increase the risk of side effects. The phenomenon in which a specific substance inhibits the substrate metabolism of an enzyme is called enzyme inhibition, and its strength and characteristics can be evaluated by an inhibition constant.

[0003] Existing methods are inefficient as they require repeated experiments using inhibitors of various concentrations, demanding significant resources, time, and labor. Nevertheless, precision is sometimes low, as evidenced by discrepancies in estimates across studies. While various optimal experimental design methods have been proposed as mathematical approaches to address this, difficulties remain in designing precise and efficient experiments because, ironically, these design criteria depend on the inhibition constant being estimated. In other words, there is a need not only to improve precision and increase experimental efficiency but also to develop experimentally accessible criteria.

[0004] According to embodiments of the present disclosure, a method for estimating an enzyme inhibition constant using a single inhibitor concentration is provided.

[0005] According to one embodiment of the present disclosure, a method for estimating an enzyme inhibition constant comprises the steps of measuring initial reaction rate data at a plurality of substrate concentrations and a single inhibitor concentration, wherein the single inhibitor concentration is a predetermined half-maximum inhibition concentration ( The method may include the step of setting the value to be greater than or equal to the half-maximal inhibitory concentration, and the step of fitting an enzyme inhibition model using the measured initial reaction rate data and the predetermined half-maximal inhibitory concentration value, and estimating an inhibition constant by applying a normalization term that includes the relationship between the half-maximal inhibitory concentration and at least one candidate inhibition constant.

[0006] In one example, the concentration of the one inhibitor may be at least 1 time and no more than 10 times the half-maximum inhibitor concentration.

[0007] In one example, the enzyme inhibition model may be a mixed inhibition model.

[0008] In one example, the normalization term may be based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated from the at least one candidate inhibition constant.

[0009] In one example, the total error can be calculated based on the sum of the fitting error and the normalization term weighted by the normalization constant, and the inhibition constant can be estimated by minimizing the total error.

[0010] In one example, the plurality of substrate concentrations are the Michaelis-Menten constant ( It may include 0.2 times, 1 time, and 5 times of ).

[0011] According to one embodiment of the present disclosure, a computer-readable storage medium is disclosed that stores a program that causes, when executed, to perform a method for estimating an enzyme inhibition constant (and steps and operations included therein).

[0012] According to one embodiment of the present disclosure, a system for estimating an enzyme inhibition constant may include: an input module configured to receive initial reaction rate data measured at a plurality of substrate concentrations and a single inhibitor concentration, and a predetermined half-max inhibitor concentration value; a verification module configured to verify whether the single inhibitor concentration is set to be greater than or equal to the predetermined half-max inhibitor concentration; a calculation module configured to fit an enzyme inhibition model using the received initial reaction rate data and the predetermined half-max inhibitor concentration value, and to estimate an inhibition constant by applying a normalization term including a relationship between the half-max inhibitor concentration and at least one candidate inhibition constant; and an output display module configured to output the inhibition constant estimated by the calculation module and to provide a confidence interval, an error landscape, and a classified inhibition type for the inhibition constant estimation result.

[0013] In one example, the input module may be configured to receive data in which the concentration of one inhibitor is between 1 and 10 times the half-maximum inhibitor concentration.

[0014] In one example, the calculation module may be configured to use a mixed inhibition model as the enzyme inhibition model.

[0015] In one example, the calculation module may be configured to calculate the normalization term based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated with the at least one candidate inhibition constant.

[0016] In one example, the calculation module may be configured to calculate a total error based on the sum of the normalization term weighted by a normalization constant and the fitting error, and to estimate the inhibition constant by minimizing the total error.

[0017] In one example, the verification module is such that the plurality of substrate concentrations are the Michaelis-Menten constant ( It can be configured to check whether it includes 0.2 times, 1 time, and 5 times of ).

[0018] The embodiments of the present disclosure can significantly improve the efficiency and reliability of enzyme inhibition analysis in the fields of drug development, clinical practice, and food technology.

[0019] The embodiments of the present disclosure can accurately and precisely estimate the enzyme inhibition constant with only one inhibitor concentration, and specifically, reduce the number of experiments by more than 75% compared to existing methods, thereby saving time and resources while maintaining analytical reliability.

[0020] In addition, the embodiments of the present disclosure can improve precision by eliminating bias caused by unnecessary data and narrowing the confidence interval to obtain estimates close to the actual value. These embodiments are applicable to all types of enzyme inhibition, such as competitive, non-competitive, and mixed types, and can be analyzed using the same procedure without prior information on the type of inhibition, thereby reducing result variance among researchers and supporting standardized inhibition evaluation.

[0021] In addition, the embodiments of the present disclosure are effective for high-speed screening and early clinical evaluation due to their low data requirements, and can be easily implemented as a software package.

[0022] FIG. 1 is a block diagram illustrating an apparatus according to one embodiment of the present disclosure.

[0023] FIG. 2 is a flowchart illustrating a method according to one embodiment of the present disclosure.

[0024] FIGS. 3a and FIGS. 3b are drawings illustrating a standard approach to estimating inhibition constants according to one embodiment of the present disclosure.

[0025] FIGS. 4a to 4e are drawings illustrating that, according to one embodiment of the present disclosure, a precise estimation is possible by using a single inhibitor concentration greater than the inhibition constants.

[0026] FIGS. 5a to 5d show inhibitor concentrations according to one embodiment of the present disclosure. Through regularization These are diagrams to explain that precise estimation is possible by using higher inhibitor concentrations.

[0027] FIGS. 6a to 6d are drawings illustrating that experimental mixed inhibitor data is precisely and accurately estimated by a method according to one embodiment of the present disclosure.

[0028] FIGS. 7a to 7e are drawings illustrating that experimental competitive inhibitor data is precisely and accurately estimated by a method according to one embodiment of the present disclosure.

[0029] FIG. 8 is a block diagram illustrating an apparatus according to an embodiment of the present disclosure.

[0030] FIG. 1 is a drawing showing the best mode for carrying out the present invention.

[0031] In the following, embodiments of the present disclosure will be described clearly and in detail so that a person skilled in the art can easily practice the present disclosure.

[0032] Terms such as "unit" and "module" used below, or functional blocks illustrated in the drawings, may be implemented in the form of software configurations, hardware configurations, or combinations thereof. In order to clearly explain the technical concept of the present invention, detailed descriptions of redundant components are omitted below.

[0033] In this document, each of the phrases such as "A or B", "at least one of A and B", "at least one of A or B", "A, B or C", "at least one of A, B and C", and "at least one of A, B, or C" may include any one of the items listed together in the corresponding phrase, or all possible combinations thereof.

[0034] The present disclosure relates to optimizing enzyme inhibition analysis through precise estimation using a single inhibitor concentration.

[0035] Enzyme inhibition analysis is essential in new drug development and food processing, and for this purpose, it is necessary to precisely estimate inhibition constants. While these constants are estimated through experiments using various substrate and inhibitor concentrations, discrepancies between studies have raised the need for a systematic experimental design approach that encompasses all types of enzyme inhibition.

[0036] FIG. 1 is a block diagram illustrating a system according to one embodiment of the present disclosure. Specific roles and operations for each component module of the system (100) for estimating an enzyme inhibition constant of FIG. 1 are described. The system (100) of FIG. 1 may include an input module (110), a verification module (120), a calculation module (130), and an output display module (140). Each module is for precisely estimating an inhibition constant according to the example of the present embodiment. It can perform the function of implementing a method for estimating enzyme inhibition constants based on a foundational optimal approach. According to the example of the present embodiment, the half-maximum inhibition concentration ( The optimal approach based on half-maximal inhibitory concentration (an inhibitor concentration that inhibits enzyme activity by 50% under specific conditions) is described in detail in the following figures and paragraphs.

[0037] The input module (110) may be configured to receive various parameters required for an enzyme reaction experiment from the user. The input module (110) may be configured to receive initial reaction rate data measured at multiple substrate concentrations and one inhibitor concentration, and a predetermined half-maximum inhibitor concentration value. Substrate concentration ( ), inhibitor concentration ( ), initial reaction rate value, Value, maximum reaction speed( ), Michaelis-Menten constant (MM)( Data required for analysis, such as ) can be provided to the input module (110) by manually loading an external file (e.g., Excel format) from the user. These input values ​​can be managed as basic data for subsequent analysis.

[0038] In one example, the input module (110) is such that the one inhibitor concentration is the half-maximum inhibitor concentration ( It can be configured to receive data that is between 1 and 10 times the value of ).

[0039] The verification module (120) can be configured to automatically check the format and unit of the input data to prevent input errors. In particular, the verification module (120) is configured for experimental conditions (e.g., ≥ It can verify whether ) is satisfied. The verification module (120) may be configured to verify whether the concentration of the one inhibitor is set to be greater than or equal to the predetermined half-maximum inhibitor concentration. That is, the total concentration of the input inhibitor ( ) is predefined By checking whether the value is greater than or equal to, the verification module (120) can determine whether the experimental data was obtained under valid inhibition conditions. In addition, the verification module (120) can determine whether the substrate concentration ( By checking the distribution of ) values, within the provided substrate concentration range It can be checked whether concentrations corresponding to approximately 0.2 times, 1 time, and 5 times of are included. That is, the verification module (120) can check whether multiple substrate concentrations are included at the Michaelis-Menten constant ( It can be configured to check whether it includes 0.2 times, 1 time, and 5 times of ). Through this, the verification module (120) can verify whether the substrate concentration is set over a sufficiently wide range (from a low level to a high level) to satisfy the data distribution required for estimation.

[0040] The computation module (130) may be configured to perform a non-linear regression analysis on the input data based on the enzyme reaction rate equation of the mixed inhibition model. Specifically, the computation module (130) fits the experimental initial reaction rate values ​​according to substrate concentration and inhibitor concentration into the rate equation of mixed inhibition to obtain the inhibition constant values ​​that best fit the data (e.g., those described below). , ) can be estimated.

[0041] The calculation module (130) may be configured to fit an enzyme inhibition model using the received initial reaction rate data and the predetermined half-maximum inhibition concentration value, and to estimate an inhibition constant by applying a normalization term that includes the relationship between the half-maximum inhibition concentration and at least one candidate inhibition constant.

[0042] In addition, the calculation module (130) The total loss function can be defined by including a regularization term that reflects the theoretical relationship between and the inhibition constants. This regularization term is given by the values ​​of the inhibition constants calculated during the estimation process. It can serve the role of imposing constraints to ensure a probable match with the value. The normalization constant, which is the weight of the normalization term. The optimal value can be selected through cross-validation. The calculation module (130) can calculate the final inhibition constants by minimizing the defined loss function using numerical optimization techniques.

[0043] In one example, the calculation module (130) may be configured to use a mixed inhibition model as the enzyme inhibition model.

[0044] In one example, the calculation module (130) may be configured to calculate the normalization term based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated with the at least one candidate inhibition constant.

[0045] In one example, the calculation module (130) is a normalization constant ( It can be configured to calculate the total error based on the sum of the normalization term and the fitting error weighted by ), and to estimate the inhibition constant by minimizing the total error.

[0046] The output display module (140) performs the role of synthesizing the results of the calculation module and providing them to the user in an easy-to-understand manner. The output display module (140) may be configured to output the inhibition constant estimated by the calculation module (130), and to provide a confidence interval, an error landscape, and classified inhibition types for the inhibition constant estimation results. The output display module (140) may output the estimated inhibition constant values ​​as numbers and may display the statistical reliability of the results by presenting a confidence interval (e.g., a 95% confidence interval) for each value. In addition, the output display module (140) may display the optimal normalization constant determined through cross-validation. The value is also displayed so that the user can check the magnitude of the selected normalization strength.

[0047] Additionally, the output display module (140) can provide an error landscape in the form of a heatmap to intuitively evaluate the goodness of fit of the model fitting results. In one example, the heatmap graph represents the change in loss function values ​​(model errors) for combinations of various low-resolution constant values ​​in color, allowing one to see at a glance how the error changes around the estimated optimal point. Through the error landscape, the user can visually check whether the estimated result corresponds to a global optimal solution or is one of multiple local optimal points.

[0048] Additionally, the output display module (140) can automatically classify and output the type of inhibition mechanism based on the final calculated inhibition constant values. The system (100) can determine the type of inhibition as mixed inhibition, competitive inhibition, or non-competitive inhibition by analyzing the relative magnitudes of the two constant values. For example, and In cases where similar results are derived, the inhibitor is classified as a non-competitive inhibition that has nearly the same effect on the enzyme and the enzyme-substrate complex (this will be explained in more detail below). In this way, through the output display module (140), the user can obtain comprehensive information from the algorithm results according to one embodiment of the present disclosure, including not only quantitative inhibition constant values ​​but also the reliability of said values, model error analysis, and the expected type of inhibition action.

[0049] The system of FIG. 1 and the components included therein are not limited by the paragraphs described above and can perform input / verification / calculation / output operations included in the embodiments of the present disclosure described in detail below.

[0050] FIG. 2 is a flowchart illustrating a method according to one embodiment of the present disclosure. The method of FIG. 2 may be a method for estimating an enzyme inhibition constant. The method of FIG. 2 for precisely estimating an inhibition constant according to an example of the present embodiment. It may be a method for estimating the enzyme inhibition constant based on a foundational optimal approach. According to the example of the present embodiment, for precisely estimating the inhibition constant The underlying optimal approach is described in detail in the following figures and paragraphs. The method of FIG. 2 may include steps S210, S220, and S230.

[0051] In step S210, multiple substrate concentrations and one inhibitor concentration ( Initial reaction rate data can be measured at ). In one example, the plurality of substrate concentrations ( ) is the Michaelis-Menten constant ( It may include 0.2 times, 1 time, and 5 times of ). In one example, the concentration of one inhibitor may be 1 time or more and 10 times or less of the half-maximum inhibitor concentration.

[0052] In step S220, the concentration of the one inhibitor can be set to be greater than or equal to a predetermined half-maximum inhibition concentration. Through this, precise estimation of at least one inhibition constant may be possible.

[0053] In step S230, an inhibition constant can be estimated by fitting an enzyme inhibition model using the measured initial reaction rate data and the predetermined half-maximum inhibition concentration value, and applying a normalization term that includes the relationship between the half-maximum inhibition concentration and at least one candidate inhibition constant.

[0054] In one example, the enzyme inhibition model may be a mixed inhibition model.

[0055] In one example, the normalization term may be based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated from the at least one candidate inhibition constant.

[0056] In one example, the total error is a normalization constant ( It can be calculated based on the sum of the fitting error and the normalization term weighted by ), and the inhibition constant can be estimated by minimizing the total error. For example, the normalization constant ( ) can be determined through cross-validation to prevent overfitting and underfitting.

[0057] The method of FIG. 2 and the steps included therein are not limited by the paragraphs described above and may further include steps of operations included in the embodiments of the present disclosure described in detail below.

[0058] FIGS. 3a and 3b are drawings illustrating a standard approach for estimating an inhibition constant according to one embodiment of the present disclosure. With reference to FIGS. 3a and 3b, a standard approach for estimating an inhibition constant according to one embodiment of the present disclosure will be described. Specifically, a mathematical model describing enzyme inhibition and a standard approach for estimating an inhibition constant based on this model will be described.

[0059] Referring to FIG. 3a, enzyme-catalyzed reactions may involve a process in which a substrate (S) binds to a free enzyme (E) to form a reversible enzyme-substrate complex (C), and this complex is converted into products (P). These reactions may be inhibited by an inhibitor (I). The inhibitor may bind to E or C to form reversible enzyme-inhibitor complexes (Y) or enzyme-substrate-inhibitor complexes (B), respectively. In this case, the dissociation constants are, respectively or It can be defined as such. Dissociation constants can be referred to as inhibition constants and can represent the strength of the inhibitory effect. The lower the inhibition constants, the higher the binding affinity between I and E or between I and C, resulting in a stronger inhibitory effect.

[0060] In addition, the relative magnitudes of the two inhibition constants can determine the inhibition mechanism (i.e., type of inhibition). Specifically, If so, competitive inhibition may occur because the inhibitor primarily binds to E and competes with S. Conversely If the inhibitor binds mainly to C, it can non-competitively inhibit the enzyme-catalyzed reaction (uncompetitive inhibition). If the two inhibition constants are of similar magnitude, the inhibitor binds to both E and C with similar affinities, exhibiting mixed inhibition, which is a mixture of competitive and non-competitive inhibition.

[0061] In one example, the initial rate of product formation It can be calculated based on the following mathematical formula 1.

[0062]

[0063] In mathematical formula 1, is the total substrate concentration, is the total inhibitor concentration, ε₀ and ε₀ represent the total enzyme concentrations, respectively (S is the substrate concentration, C is the enzyme-substrate complex concentration, E is the free enzyme concentration, B is the enzyme-substrate-inhibitor complex concentration, and Y is the enzyme-inhibitor complex concentration), is the maximum reaction rate, and is the Michaelis-Menten constant. Equation 1 is the mixing inhibition ( ) as well as competitive inhibition( ) and non-competitive inhibition( Since it is a general formula that can also explain ), it allows for the simultaneous estimation of inhibition constants and identification of inhibition types without prior knowledge of inhibition types.

[0064] FIG. 3b. According to one embodiment, a method (standard approach) for estimating the inhibition type and inhibition constant may follow the following steps: First, an appropriate total inhibitor concentration ( To determine the criteria for the ) setting, one (In one example, Various in (identical to) Half-maximum inhibitory concentration ( ) can be estimated ((i) of Fig. 3b). If it is estimated, Is , , by, is 0, , , An experimental design can be established by setting it as follows ((ii) in Fig. 3b). For each combination of concentrations, the initial rate can be measured, and inhibition constants can be estimated by fitting Equation 1 to the data ((iii) in Fig. 3b). However, it remains uncertain whether such an empirically established standard approach is sufficient, and whether all of this data is necessarily required for accurate and precise estimation.

[0065] FIGS. 4a through 4e are drawings illustrating that, according to one embodiment of the present disclosure, precise estimation is possible by using a single inhibitor concentration greater than the inhibition constants. Specifically, in FIGS. 4a through 4e, a low total inhibitor concentration ( It can be explained that experimental data in ) does not provide the information necessary for estimating the inhibition constant.

[0066] In one example, to determine which experimental conditions enable accurate and precise estimation based on the above mathematical formula 1, , and initial velocity data can be generated by simulation by setting the actual inhibition constant. Referring to Fig. 4a, one Values ​​and various Using the actual value and Initial velocity data obtained based on pairs (points on the graph in Fig. 4a) can be compared with the results of Equation 1 for different parameter pairs (e.g., solid and dotted lines in Fig. 4a). Subsequently, the mean squared relative error (fitting error) between the simulation data and the mixed inhibition model (Equation 1) can be calculated for candidate pairs of candidate inhibition constants. Referring to Fig. 4a, a heatmap of the fitting error terrain can be created by displaying areas with low fitting error in dark and areas with high fitting error in bright. By utilizing the heatmap, it can be analyzed which experimental conditions enable precise estimation.

[0067] According to exemplary analysis results, the precision of the inhibition constant estimation is and inhibition constants and It varies depending on the relationship (Figs. 4a to 4c), and can be divided into four types.

[0068] first, go and In the case of being much smaller ( and ), dark areas and It can be widely distributed throughout. In Fig. 4b, =0.1uM is the actual parameter value ( = When the value is much smaller than =1uM, wide ranges of parameter pairs (e.g., ▲,▼,●) exhibit low fitting errors, indicating that precise estimation is difficult. This suggests that the given data can be explained and The large number of pairs may lead to identification problems, implying that precise estimation is impossible. In Fig. 4b, the curve fitted with these parameter pairs matches the data well.

[0069] Second, go It is bigger than, but In cases smaller than ( ), dark areas It can be widely distributed only in the direction. In Fig. 4c, =2uM is the actual = Larger than 1uM but actual If =10uM is smaller, the same but different Pairs with values ​​(e.g., ▲, ●) show an accurate fit is precisely estimated, but This indicates that this is not the case. This is similar to the actual inhibition constant. while This indicates that other pairs can also explain the data, and Although precise estimation is possible, This may mean that it is impossible due to identification issues.

[0070] Third, conversely go It is bigger than, but In cases smaller than ( ), dark areas It can be widely distributed only in the direction. In Fig. 4d, =2uM This is the actual = Larger than 1uM but actual If it is much smaller than =10uM, the same but different Pairs with values ​​(e.g., ▼,●) show an accurate fit is precise, but indicates that this is not the case. This is similar to the actual inhibition constant. while can explain other pairs of data, Although precise estimation is possible, This may mean that it is impossible due to identification issues.

[0071] finally, go and When all are larger ( and ), dark areas may be narrowly distributed. In Fig. 4e, =10uM are the two actual parameter values ​​( = If it is greater than or similar to =1uM), two and All can be estimated precisely (●). This is similar to the actual inhibition constant. and This implies that only pairs can explain the data, indicating that precise estimation is possible without identifiability issues.

[0072] For example, in FIGS. 4b through 4e, the actual data = , , ( It was obtained at =1uM, and the initial velocity data It can be normalized to =0.1 uM / min / min / protein. Such analytical results are merely illustrative and do not limit the scope of the embodiments of the present disclosure.

[0073] In the exemplary analysis results above Changes in precision depending on This is because the approximation method of the above mathematical formula 1 varies depending on the settings. and In the case of, and As a result, the two inhibition constants in the above mathematical equation 1 can be ignored. Due to this, and Identification issues may arise for everyone. If it becomes larger than a certain inhibition constant ( or ), The inhibition constant that is much larger than or It is approximated and ignored, and as a result, identification problems may occur only for the ignored inhibition constants.

[0074] Unlike the standard approach of embodiments that changes , this embodiment is smaller than the inhibition constant This may suggest that it is unnecessary for accurate and precise estimation. This implication is consistent with that of the examples regarding experimental designs for the accurate estimation of inhibition constants in competitive and non-competitive inhibition, and said examples (competitive inhibition) or It has been suggested that if the (non-competitive inhibition) ratio is less than 1, the accuracy of estimation and type classification decreases. One embodiment of the present disclosure can generalize this to all inhibition types.

[0075] FIGS. 5a to 5d show inhibitor concentrations according to one embodiment of the present disclosure. Through normalization These are diagrams to explain that precise estimation is possible by using higher inhibitor concentrations.

[0076] In one example, can serve as an experimental criterion to avoid identification issues. In the explanation accompanying the previous figures, appropriate for precisely estimating inhibition constants The criteria were explained. The criteria are It was based on the relationship between and inhibition constants. However, since inhibition constants cannot be known prior to estimation, suitable for the experiment Determining the value in advance still presents difficulties. On the other hand, can be determined before estimating the inhibition constant (see Fig. 3b). Therefore is appropriate It can be used as an experimental criterion for determining. Hereinafter, according to one embodiment of the present disclosure An exemplary relationship between and inhibition constants will be explained.

[0077] In one example, An exemplary work relationship between and inhibition constants can be based on the Cheng-Prusoff equation, which is well known in enzyme reaction kinetics. For example, An exemplary relationship between the inhibition constants can be based on Fig. 5a and Equation 2 below.

[0078]

[0079] The variables and constants included in Equation 2 were explained together with Equation 1, and Measured together with Is and It is the weighted harmonic mean (H()) of, and is the weight.

[0080] According to mathematical formula 2, can satisfy one of the following three cases: , , or . Therefore, if so, in the first case However, in the second case Only precise estimation may be possible. On the other hand, in the third case, the two values and All can be precisely estimated (see Fig. 5a).

[0081] Referring to Fig. 5a, is always or Since it is larger, and If the difference is within 10 times If you use (b), (c), or both values ​​(d) can be estimated with precision.

[0082] Referring to Fig. 5b, If it is known, the unknown and is the weighted harmonic mean formula It must satisfy [this], and this can be included in the estimation process through a normalization term (normalization error calculated based on Equation 3 below).

[0083]

[0084] The heatmap in Fig. 5b represents the normalized error, and the low normalized error area (black area) is satisfying the conditions and It can correspond to the value.

[0085] Referring to Fig. 5c, if the normalized error is included in the fitting error (total error), and It can be estimated more precisely (dotted line), and the lower error area (black area) can be reduced compared to the case without normalization (Figs. 5b to 5d).

[0086] Referring to Fig. 5d, The based optimal approach is a single class Normalization alone can significantly reduce the number of experiments required for precise and accurate estimation compared to the method of the embodiment described with Fig. 3b.

[0087] Next, Examples for improving precision over a wider range, including base normalization, will be described.

[0088] In one example, While using only allows for the precise estimation of at least one inhibition constant, other inhibition constants In cases where it is significantly larger, an issue of identification remains (see Fig. 5a). To address these limitations Precision can be improved by including information in the estimation process. Specifically, Since is known, the inhibition constant is the above mathematical formula 2( ; (See Fig. 5b) must be satisfied. In order to include this constraint in the estimation process and The squared relative error between them can be defined as the normalized error (Fig. 5b).

[0089] In one embodiment, a normalization constant ( A new loss function (total error) can be defined by combining the normalized error weighted by ) with the fitting error: Total Error = Fitting Error + × Normalized error. The value can be selected to minimize the cross-validation error. In one example, the inhibition constant can be estimated by minimizing the total error rather than the fitting error, thereby allowing the initial velocity data and Information can be simultaneously reflected in the estimation process. According to this embodiment, the estimation precision of both inhibition constants can be increased compared to when only the fitting error is used.

[0090] in other words, In the case where (see Fig. 5a), estimation using the total error can further improve precision (see Fig. 5c). In particular, Precise estimation can be achieved even when is lower than one of the inhibition constants. Specifically, In this case, using the total error allows for precise estimation of both constants (see Fig. 5c), which is possible with only the fitting error. It was impossible for (see Fig. 5a).

[0091] likewise, Even in this case, using the total error allows for precise estimation of both constants (see Fig. 5c), which is possible with only the fitting error. It was impossible for (see Fig. 5a).

[0092] However, in competitive cases ( go (cases where it is much larger) and non-competitive cases ( this In cases where it is much larger than ) even if normalization is applied, each and Precise estimation of may be impossible. However, in competitive (or non-competitive) cases (or The estimation of ) is not important.

[0093] Based on the above results of the present embodiment, for precisely estimating the inhibition constant according to the present embodiment A basis optimal approach can be proposed (see Fig. 5d).

[0094] first, can be estimated in the same way as the standard approach described in FIGS. 3a and 3b (see FIG. 5d (i)). After was estimated, the experimental design is Is , , Set to Is It can be established by setting it to a single value that satisfies (Fig. 5d (ii)). After measuring the initial velocity for each concentration combination, an inhibition constant that minimizes the total error can be estimated by fitting the above Equation 1, which includes a normalization term (see Equation 3 above), to the data (see Fig. 5d (iii)).

[0095] According to the present embodiment, for precisely estimating the inhibition constant The underlying optimal approach can estimate the inhibition constant precisely, accurately, and efficiently because it uses only the measurements necessary for precise estimation from the data used in the standard approach described in FIGS. 3a and 3b. For precisely estimating the inhibition constant according to the present embodiment The underlying optimal approach can enable precise, accurate, and efficient estimation of mixed inhibition experimental data.

[0096] FIGS. 6a to 6d are drawings illustrating that experimental mixed inhibitor data is precisely and accurately estimated by a method according to one embodiment of the present disclosure.

[0097] In one example, for the case of mixed inhibition using actual experimental data, to accurately estimate the inhibition constant according to the present embodiment In order to evaluate whether the underlying optimal approach can estimate the inhibition constant precisely, accurately, and efficiently, (e.g., 10 to 500 μM) and Data on triazolam (substrate) and ketoconazole (inhibitor) pairs obtained from experiments varying (e.g., 0, 0.01 to 0.5 μM) can be utilized, and these, respectively =72μM and It may include a range of =0.04μM. Referring to Fig. 6a, triazolam and its inhibitor ketoconazole ( =0.040μM, When =50μM; normalized initial velocity of =72 μM ( ) varies triazolam concentrations ( = 10, 25, 50, 100, 250, 500 μM) and ketoconazole concentration ( It was measured in combinations of = 0, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5 μM, and fitted with the above mathematical formula 1 (solid line).

[0098] Referring to FIG. 6b, the whole or a single Heatmaps of total errors before (top) and after (bottom) normalization are plotted under the conditions. Total If you use a range (i.e., standard conditions), precise and It is possible to estimate and can be identified as a mixed inhibition. Single Under conditions (yes: In the case where =0.01 μM), and The estimate is = 0.041 μM, = 0.041 μM (won), = 0.041 μM, =100 μM(triangle), = 100 μM, As indicated by the low error indication at the =0.041μM (square) point, it may not be precise. However go (yes: Passing through (=0.05 μM) (yes: As it increases to (= 0.5 μM), the precision can become similar to standard conditions. Normalization can improve precision to the level of standard conditions.

[0099] Figure 6c illustrates the estimated parameters (dots and asterisks) and 95% confidence intervals (error bars). In Figure 6c, A single lower than If you use and The estimation is inaccurate and not precise (triangle; located outside the range of 1.5 times the standard condition estimate (dotted line)), whereas, according to the example of the present embodiment, for precisely estimating the inhibition constant Based optimal approach (e.g., =0.5 μM; diamond) can estimate the two parameters precisely and accurately at a level similar to standard conditions.

[0100] In one example, the whole and When using a dataset (standard conditions), both inhibition constants can be estimated accurately ( =0.041μM, = 0.041 μM). This can be evidenced by the narrow low-error region in the error landscape (see Fig. 6b) and the confidence intervals (CIs; see Fig. 6c). The estimated and Since the values ​​did not differ significantly, it can be classified as a mixed inhibitor.

[0101] Next, each single The precision, accuracy, and efficiency of the estimation can be evaluated using only the data. In this case, the two inhibition constants may not be accurately estimated even if normalization is applied. This is because the low error (i.e., the black area) appears wide in the error terrain compared to standard conditions (see Fig. 6b; =0.01μM) and confidence intervals (see Fig. 6c; It can be confirmed as =0.01, 0.025μM). This uncertainty may lead to misclassification of inhibition types.

[0102] for example, =0.041μM, =0.041μM (mixed type; circle in Fig. 6b), =0.041μM, =100 μM (competitive type; triangle in Fig. 6b), =100 μM, =0.041 μM (non-competitive; rectangle in Fig. 6b) can all exhibit low errors in the error terrain (see Fig. 6b; =0.01μM). Also estimated (See Fig. 6c; =0.025μM) and (See Fig. 6c; The value of (=0.01) may be inaccurate as it falls outside the range of 1.5 times the value estimated under standard conditions.

[0103] Estimation uncertainty and inaccuracy in is It can be gradually resolved as it increases. In this case, both inhibition constants can be accurately estimated within a 1.5x range (see Fig. 6c; =0.05μM). However, error topography (Fig. 6b; =0.05) and confidence intervals (see Fig. 6c; =0.05μM) can still show a wide low error region compared to standard conditions.

[0104] go In the case where it becomes larger, the two inhibition constants can be estimated precisely and accurately, and the error terrain (see Fig. 6b; =0.5μM) and confidence intervals (Fig. 6c; =0.25, 0.5μM) can exhibit a narrow, low error region similar to the standard approach. This precision can be further enhanced by adding normalization (see Fig. 6c).

[0105] For precisely estimating the inhibition constant according to the example of the present embodiment The base optimal approach can achieve higher efficiency by reducing the number of experiments required to 1 / 7 while providing precise and accurate estimates equivalent to standard conditions (see Fig. 6d).

[0106] FIGS. 7a to 7e are drawings illustrating that experimental competitive inhibitor data is precisely and accurately estimated by a method according to one embodiment of the present disclosure.

[0107] In the following, according to the example of the present embodiment, for precisely estimating the inhibition constant It will be further explained that the underlying optimal approach demonstrates precise, accurate, and efficient estimation even for competitive inhibition experimental data. For precisely estimating the inhibition constant according to the example of this embodiment In order to further evaluate whether the underlying optimal approach can achieve precise, accurate, and efficient estimation even in the case of competitive or non-competitive inhibition using actual experimental data, (12.5 to 100 uM) and Data on chlorzoxazone (substrate) and ethambutol (inhibitor) pairs obtained from experiments varying (0, 0.41 to 11.1 μM) can be utilized, and these, respectively =39.1 uM and It may include a range of =4uM. Referring to Fig. 7a, chlorzozazone and its inhibitor ethambutol ( =50uM; at =39.1 uM Normalized initial velocity of =4uM ( ) various chlorzozone concentrations ( =12.5, 25, 50, 75, and 100 μM) and ethambutol concentration ( It was measured in combinations of =0, 0.41, 1.23, 3.7, and 11.1 uM and fitted with the above mathematical formula 1 (solid line).

[0108] Referring to Fig. 7b, the whole or single Heatmaps of total errors before (top) and after (bottom) normalization are plotted under the conditions. Total If you use (i.e., standard conditions) It was estimated precisely, but, Even with normalization applied, it is estimated imprecisely over a wide range, which may indicate competitive inhibition. A single Under the conditions go (yes: at =0.41μM) (yes: =3.7μM), and (yes: As it increases to =11.1μM) The precision of the estimation can be similar to that of standard conditions.

[0109] Referring to Fig. 7c, the estimated by applying normalization Values ​​(points) and 95% confidence intervals (error bars) are plotted. A single lower than If you use, the estimated This is inaccurate or imprecise, estimated under standard conditions It can be located outside the range of 1.5 times (dotted line) (triangle). On the other hand, for precisely estimating the inhibition constant according to the example of the present embodiment The underlying optimal approach (Diamond) can enable precise and accurate estimation equivalent to standard conditions.

[0110] Referring to Fig. 7d, the estimated under standard conditions The asymptotic distribution of is located at a very large value (~10¹⁶ μM; dotted square), which may reflect competitive inhibition. A single lower than Using may result in an unexpected mode appearing around 10 μM (dotted rectangle), potentially leading to the inhibition type being misclassified as mixed. On the other hand, for precisely estimating the inhibition constant according to the example of this embodiment The based optimal approach is This misclassification can be prevented because the distribution matches standard conditions.

[0111] In one example, the whole and When using the dataset (standard conditions), a narrow vertical area showing low error in the error terrain (see Fig. 7b) and As confirmed by the confidence intervals of (see Fig. 7c) Only can it be estimated with precision. This vertical region is regardless of normalization. Much larger than It can be extended to the scope to represent types of competitive inhibition. Also The estimated value It can be confirmed that it is competitive inhibition as it appears as a unimodal distribution located nearby (see Fig. 7d).

[0112] Next, each single Evaluates the precision, accuracy, and efficiency of the estimation using only data. In the case of, is the error topography (see Fig. 7b; =0.41uM) and confidence intervals (see Fig. 7c; As confirmed by (=1.23uM), a wide low error region appears compared to standard conditions, so the estimation may be imprecise and inaccurate. In addition, the estimated The value may be inaccurate if it deviates from the range of 1.5 times the value under standard conditions (see Fig. 7c; =0.41 μM). In the error topography, the low error region extends to approximately 10 μM (mixed type), so the classification of inhibition types also appears imprecise (see Fig. 7b; =0.41uM). This uncertain type classification is This was also confirmed by the distribution of estimates, and the two modes (Mixed type) and (Competitive) may be located nearby (see Fig. 7d; =0.41, 1.23uM).

[0113] In one example, Estimation uncertainty and inaccuracy in is go As it increases, it can be gradually resolved. In the case of, The error topography (see Fig. 7b; =3.7uM) and Confidence intervals (see Fig. 7c; It appears as a narrow vertical region similar to standard conditions at =3.7uM), allowing for precise and accurate estimation. The estimated The value can also be located within the 1.5-fold range (see Fig. 7c; =3.7uM). However The distribution of the estimated values ​​ranges from approximately 10 μM to 10¹⁶ μM (see Fig. 7d; =3.7uM) Type classification may still be uncertain. go In cases where it has become larger The estimated value It appears as a unimodal distribution in the vicinity (see Fig. 7d; =11.1uM) Type classification can be made more precise. Due to the relationship, this inhibition can be accurately classified as competitive. Referring to FIG. 7e, for precisely estimating the inhibition constant according to the example of the present embodiment The base optimal approach (diamond in Fig. 7e) can achieve higher efficiency by reducing the number of experiments required to one-fifth compared to the standard conditions (triangle in Fig. 7e), while providing precise and accurate estimations equivalent to standard conditions.

[0114] In one embodiment described together with FIGS. 3a and 3b, inhibition constants ( , To estimate the inhibition constant, initial reaction rate data obtained at various substrate and inhibitor concentrations must be fitted to the inhibition model (Equation 1). It is well known that using varying substrate and inhibitor concentrations allows for the accurate and precise estimation of the inhibition constant when the inhibition type (i.e., competitive, non-competitive, or mixed) is known in advance. However, since identifying the inhibition type often requires the accurate and precise estimation of the inhibition constant, the 'chicken and egg' problem may arise.

[0115] This problem can be solved if the inhibition constants of a mixed inhibition model, which encompasses all types of inhibition and can be applied without prior information, can be estimated accurately and precisely. However, the experimental design for successful estimation using a mixed inhibition model has not been clear until now. In one embodiment of the present disclosure, an optimal design is proposed by analyzing the error topography (see FIGS. 4a to 4e). The mixed inhibition model based solely on the initial rate measured at a higher single inhibitor concentration and It can be estimated accurately and precisely (see Figs. 5a to 5d). This was possible with much less data than under conventional experimental conditions using a wide range of inhibitor concentrations, and this , , This is possible because the relationship between them was included as a normalization term (see Equations 3 and 4) and reflected in the estimation process along with the existing data fitting.

[0116] Based on these results, enabling accurate and precise estimation even with limited experimental data A basis-based optimal approach is proposed (see FIGS. 5a to 5d). In addition, if the user inputs only the relevant data, for precisely estimating the inhibition constant according to the example of the present embodiment. A user-friendly package can also be provided to facilitate the easy application of the underlying optimal approach.

[0117] For precisely estimating the inhibition constant according to the example of the present embodiment The underlying optimal approach holds significant potential for enzyme inhibition studies, including the identification of inhibition types. Conventional inhibition type identification has relied on visual methods using linearized versions of inhibition models, such as Lineweaver-Burk plots or Dixon plots. However, these linearization techniques can introduce significant errors.

[0118] More modern approaches include the mixed type (Equation 1) and the competitive type (in Equation 1 Assuming), non-competitive type (from mathematical formula 1 There is a method to select the optimal model using metrics such as R² or AIC (Akaike information criterion) after performing a nonlinear fitting on the model (assuming). However, the choice of which model selection criterion to use can still be somewhat subjective.

[0119] Due to these limitations, there were cases where reported inhibition types differed even for the same enzyme inhibition. To overcome these limitations, according to the example of the present embodiment, for precisely estimating the inhibition constant The underlying optimal approach uses a mixed inhibition model that integrates all inhibition types into a single equation (Equation 1). In particular, for precisely estimating the inhibition constant according to the example of this embodiment The based optimal approach is and By enabling accurate, precise, and efficient estimation, it is possible to provide a basis for reliably calculating both inhibition constants and inhibition types based on a mixed inhibition model.

[0120] In one example, inhibition constants fit an appropriate inhibition model to the data or It can be estimated using the simplified Chung-Prusov equation (Equation 4 below) based only on information.

[0121]

[0122] The latter approach (the Chung-Prusov equation-based method) is well known for its simplicity and high accuracy. However, because the Chung-Prusov equation varies depending on the type of inhibition, To estimate the inhibition constant based solely on data, the type of inhibition must be known in advance. On the other hand, to precisely estimate the inhibition constant according to the example of the present embodiment The underlying optimal approach can utilize the general Chung-Prusov equation (Equation 2), which does not require a type of inhibition, as a normalization term (Equation 3) in the estimation process. By including the Chung-Prusov equation in the fitting process, more precise estimation may be possible than with the existing standard method based on inhibition models.

[0123] In addition, for precisely estimating the inhibition constant according to the example of the present embodiment The foundational optimal approach significantly reduces the number of experiments required compared to standard methods, thereby decreasing the consumption of reagent and enzyme resources and minimizing unnecessary experimental repetitions. As a result, this optimal approach can enhance the overall efficiency of evaluation and development processes in various fields, such as new drug development and food technology, and substantially reduce costs and manpower consumption.

[0124] As an example, to leverage the advantages of the optimal approach mentioned above, automated optimal approach packages can be provided in MATLAB and R, allowing users to easily and visually verify the accuracy and precision of the estimation results. Along with initial reaction rate data, the user can provide substrate concentration, inhibitor concentration (in Excel format), , , By simply inputting into the package, this package can provide estimates of the inhibition constant and the error landscape. Users can intuitively evaluate the precision of the estimate by checking the width of the error landscape and simultaneously identify the type of inhibition through the shape of the error landscape.

[0125] In one embodiment of the present disclosure, inhibition constants and To estimate , initial velocity data may be required. In one example, this data can be obtained through Ordinary Differential Equation (ODE) simulations. In another example, this data can be derived from previously reported experimental results. The data generated by the simulation is the substrate ( ) and inhibitor( ) It can be used to analyze how concentration settings affect the estimation of the inhibition constant. For example, this data can be derived from time-series product (P) data generated by full model simulations using a function (e.g., ode15s function) embedded in a program contained in memory (e.g., MATLAB R2023a). The full model can be constructed based on a mass-action kinetics model of enzyme inhibition (see Fig. 3a). In one example, the full model can be constructed based on Equations 5 through 11 below.

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132]

[0133] In the above mathematical formulas 5 to 11, S is the concentration of the substrate, C is the concentration of the enzyme-substrate complex, E is the concentration of the free enzyme, B is the concentration of the enzyme-substrate-inhibitor complex, Y is the concentration of the enzyme-inhibitor complexes, I is the concentration of the inhibitor, and P is the concentration of the product. is the amount of change over time, Each of these can represent a forward or reverse reaction rate constant.

[0134] In one example, , , and To represent the ODE based on the parameters, the response rates within the entire model can be modified to reflect those parameters. This modification was based on the definition of each parameter, and The ratio can be determined by detailed equilibrium conditions. The modified result may be as shown in Equations 12 through 18 below, and the variables and constants included in Equations 12 through 18 are as described in the preceding paragraphs.

[0135]

[0136]

[0137]

[0138]

[0139]

[0140]

[0141]

[0142] For the above ODE calculation, the initial condition is and It can be set as, and reverse reaction rate constants , and All can be set to 100. To calculate the initial rate data, the time point (τ) when the product reaches 1% of the total substrate concentration and the product concentration at that time ( ) is recorded and can be used. Afterwards normal distribution error By adding ~N(0,0.01), the observed product value ( ) can be obtained (calculated) as shown in mathematical formula 19 below.

[0143]

[0144] In one example, the initial reaction rate It can be calculated based on the following mathematical formula 20.

[0145]

[0146] In one example, actual experimental data can be used to verify whether the method (approach) according to one embodiment of the present disclosure can provide accurate and precise estimations. The data includes initial velocity data and Includes values, which may be obtained from inhibition between triazolam (substrate) and ketoconazole (inhibitor) corresponding to mixed inhibition, and / or inhibition between chlorzozazone (substrate) and ethambutol (inhibitor) corresponding to competitive inhibition. For example, triazolam-ketoconazole data and / or The value can be extracted from the initial velocity graph using a program stored in memory (e.g., PlotDigitizer). For the chlorzozozone-ethambutol data, the initial velocity data of another embodiment and The value can be utilized.

[0147] In one example, the accuracy and precision of the inhibition constant estimated under various experimental conditions can be evaluated using the error landscape. The error landscape represents the shape of the loss function used in the estimation, and specific and It can be visualized on an xy plane defined by a range. To set this range, the estimation target and Minimum values ​​of the candidates ( , ) and maximum values( , ) can be set. Afterwards, divide this range into 100 equal parts on a logarithmic scale. ( ) and ( Candidate values ​​of ) can be generated as shown in Equation 21 below.

[0148]

[0149] To analyze these candidates, temperament ( ) and inhibitor( Initial rate data obtained under experimental conditions consisting of ) concentration ( ) can be compared with the calculated initial velocity data. The calculated data ( ) is each candidate , It can be derived by substituting the experimental conditions into the inhibition model (Equation 1 above). To compare this, the fitting error for each candidate can be calculated as shown in Equation 22 below.

[0150]

[0151] By plotting the fitting error calculated in this way on the xy plane, an error topography can be generated, and on the x-axis , on the y-axis The fitting error values ​​of each candidate can be plotted by color (see Fig. 4a).

[0152] In one example, regarding normalization, a normalization term may be included in the loss function used for estimation. This normalization term (see Equation 3 above) is experimentally obtained and calculated using the Chung-Prusov equation (Mathematical Equation 2 above) It can be defined as the squared relative error between. In the estimation process with normalization applied, after adding the normalization term to the fitting error, the normalization constant ( The total error multiplied by ) can be used as the loss function. The value can affect the bias-variance trade-off. If ε is too small, overfitting occurs, which can result in low bias but high variance. Conversely If ε is too large, variance may decrease, but bias may arise due to underfitting. Therefore, for effective estimation using regularization, an appropriate ε is necessary. A value must be selected.

[0153] To avoid overfitting and underfitting, a predefined range Using cross-validation within It can be decided. One value inside ...can be selected. Subsequently, one of the initial velocity data is excluded and designated as test data, and the remaining data can be used as training data. Using the training data, the total error loss function is minimized to r to obtain estimates and can be obtained. Estimated and The initial velocity is calculated by substituting it into the inhibition model (Mathematical Equation 1 above), and the squared relative error (test error) between the test data and the calculated initial velocity can be calculated. This process is repeated for all combinations of training and test data, and the test errors are summed to calculate the cross-validation error (CV error) of the corresponding r. This procedure My all r( Iterate over ), and the r with the smallest cv error is It can be selected as.

[0154] In one example, regarding the calculation of asymptotic confidence intervals and distributions, the confidence interval can be a key indicator of the precision of the estimate. A narrower confidence interval implies a more precise estimate. For example, the estimated and The bootstrapping method can be used to obtain the confidence interval. In this method, the data can be resampled 1,000 times to include the same number of data points as the original data. Then, by fitting a hindering model to each resampled dataset class ... can be estimated. By expressing the estimates obtained in this way as a histogram, the asymptotic distribution of the inhibition constant can be derived. By excluding the lower 2.5% and upper 97.5% from this distribution, the 95% confidence interval of the inhibition constant can be calculated.

[0155] FIG. 8 is a block diagram illustrating an apparatus according to an embodiment of the present disclosure.

[0156] The device (800) may include at least one of a processor (810), memory (820), a transceiver (830), an input interface device (840), and an output interface device (850). Each component may be connected by a common bus (860) to communicate with each other. Additionally, each component may be connected via an individual interface or an individual bus centered around the processor (810), rather than via the common bus (860).

[0157] The processor (810) can be implemented in various types such as an Application Processor (AP), a Central Processing Unit (CPU), a Graphic Processing Unit (GPU), etc., and may be any semiconductor device that executes instructions stored in memory (820). The processor (810) may execute program instructions stored in memory (820). The processor (810) may be configured to perform calculation(s) described together with FIGS. 1 through 7e.

[0158] The processor (810) can store program instructions for implementing at least one function for the corresponding modules in memory (820) to control the operation described in conjunction with FIGS. 1 to 7e to be performed.

[0159] The memory (820) may include various forms of volatile or non-volatile storage media. For example, the memory (820) may include ROM (read-only memory) and RAM (random access memory). In an embodiment of the present disclosure, the memory (820) may be located inside or outside the processor (810), and the memory (820) may be connected to the processor (810) through various known means. Additionally, the memory (820) may store one or more program instructions for performing operation(s) according to an embodiment of the present disclosure, and the processor (810) may perform the operation(s) described above by executing said program instructions.

[0160] The transmitting and receiving unit (830) can perform the function of transmitting and receiving data processed / to be processed by the processor (810) to and from an external device and / or external system.

[0161] The input interface device (840) is configured to provide data to the processor (810).

[0162] The output interface device (850) is configured to output data from the processor (810).

[0163] The above description describes specific embodiments for implementing the present invention. The present invention will include not only the embodiments described above, but also embodiments that can be simply modified or easily modified. Furthermore, the present invention will include technologies that can be easily modified and implemented using the embodiments. Accordingly, the scope of the present invention should not be limited to the embodiments described above, but should be defined by the claims set forth below as well as equivalents to the claims of this invention.

[0164] The embodiments of the present disclosure are not implemented only through the systems, devices, and / or methods described above, but may also be implemented through a program that realizes a function corresponding to the configuration of the embodiments of the present disclosure or a recording medium on which such a program is recorded, and such implementation can be easily achieved by a person skilled in the art to which the present disclosure belongs from the description of the embodiments described above.

[0165] Although embodiments of the present disclosure have been described in detail above, the scope of the rights of the present disclosure is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present disclosure also fall within the scope of the rights of the present disclosure.

[0166] The present disclosure relates to enzyme inhibition analysis. More specifically, it can be used in a method for estimating enzyme inhibition constants using the concentration of a single inhibitor.

Claims

1. In a method for estimating enzyme inhibition constants, A step of measuring initial reaction rate data at multiple substrate concentrations and a single inhibitor concentration; A step of setting the concentration of the above-mentioned inhibitor to be greater than or equal to a predetermined half-maximum inhibitory concentration; and A method comprising the step of fitting an enzyme inhibition model using the measured initial reaction rate data and the predetermined half-max inhibitory concentration value, and estimating an inhibition constant by applying a normalization term including the relationship between the half-max inhibitory concentration and at least one candidate inhibition constant.

2. In Paragraph 1, A method in which the concentration of the above-mentioned inhibitor is at least 1 time and no more than 10 times the above-mentioned maximum inhibitor concentration.

3. In Paragraph 1, The above enzyme inhibition model is a mixed inhibition model.

4. In Paragraph 1, The above normalization term is a method based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated from the at least one candidate inhibition constant.

5. In Paragraph 4, The total error can be calculated based on the sum of the fitting error and the above normalization term weighted by a normalization constant, and The above inhibition constant is a method of estimating by minimizing the above total error.

6. In Paragraph 1, The above plurality of substrate concentrations are the Michaelis-Menten constant ( A method including 0.2 times, 1 time, and 5 times of ).

7. A computer-readable storage medium storing a program that causes, when executed, to perform a method for estimating an enzyme inhibition constant according to any one of claims 1 to 6.

8. In a system for estimating enzyme inhibition constants, An input module configured to receive initial reaction rate data measured at multiple substrate concentrations and a single inhibitor concentration, and a predetermined half-maximum inhibitor concentration value; A verification module configured to verify whether the concentration of the one inhibitor is set to be greater than or equal to the predetermined half-maximum inhibitor concentration; A computation module configured to fit an enzyme inhibition model using the received initial reaction rate data and the predetermined half-max inhibitory concentration value, and to estimate an inhibition constant by applying a normalization term including the relationship between the half-max inhibitory concentration and at least one candidate inhibition constant; and A system comprising an output display module configured to output an inhibition constant estimated by the above calculation module, and to provide a confidence interval, an error landscape, and classified inhibition types for the result of the inhibition constant estimation.

9. In Paragraph 8, The above input module is a system configured to receive data in which the concentration of one inhibitor is at least 1 time and no more than 10 times the half-maximum inhibitor concentration.

10. In Paragraph 8, The above calculation module is a system configured to use a mixed inhibition model as the enzyme inhibition model.

11. In Paragraph 8, A system configured such that the calculation module calculates the normalization term based on the relative error between the half-maximum inhibition concentration and the weighted harmonic mean calculated with at least one candidate inhibition constant.

12. In Paragraph 8, The above calculation module is a normalization constant ( A system configured to calculate a total error based on the sum of the normalization term and the fitting error weighted by ), and to estimate the inhibition constant by minimizing the total error.

13. In Paragraph 8, The verification module above is such that the plurality of substrate concentrations are the Michaelis-Menten constant ( A system configured to check whether it includes 0.2 times, 1 time, and 5 times of ).

Images