Data analysis system, data analysis method, and recording medium of its program

Abstract

The present invention provides a data analysis system, a data analysis method, and a recording medium of a program thereof. The system comprises: a data storage unit (2) for storing a plurality of analysis results obtained from a plurality of analyses performed as responses, a plurality of parameters included in the analysis conditions as factors, and storing the responses and the factors in association with each other; a data processing unit (4) for generating an approximate expression representing the relationship between the variable and the response using at least one of the factors as a variable; and an information input device (6) for inputting information to the data processing unit (4) by a user, and the data processing unit (4) performing the following steps: setting at least one factor to be a variable by the user; arbitrarily setting the structure of a model formula serving as a basis for the approximate expression by the user; determining the model formula based on the structure set by the user; and determining the coefficients of each term constituting the model formula determined in the model formula determination step by regression analysis, thereby determining the approximate expression.

Description

Technical Field

[0001] The present invention relates to a data analysis system, a data analysis method, and a recording medium for a program thereof for analyzing the relationship between analysis results obtained by performing liquid chromatography analysis or the like under a plurality of analysis conditions and the analysis conditions. Background Art

[0002] In the pharmaceutical field, for example, liquid chromatography is used to identify impurities that have been introduced during manufacture. In this case, since it is necessary to detect as many components as possible in the target substance, it is necessary to explore the "optimal analytical conditions" that result in more peaks of components appearing in the chromatogram. In order to explore the "optimal analytical conditions," it is necessary to examine all analytical conditions that change multiple parameters, such as the flow rate of the mobile phase, the temperature of the separation column, and the composition of the mobile phase (the mixing ratio of the solvents that constitute the mobile phase). However, if data are obtained by performing analysis under all analytical conditions, a significant amount of time is required. Therefore, a method is sometimes employed in which the relationship between analytical conditions and analytical results is approximated using regression analysis, and the approximation result is used to explore the "optimal analytical conditions." Summary of the Invention

[0003] [Problems to be solved by the invention]

[0004] In the method using regression analysis, the various parameters of the analysis conditions are defined as factors, and the analysis results under each analysis condition (such as the resolution of peaks in the chromatogram, the number of peaks, and the retention time of each peak) are defined as responses. An approximate equation is generated that approximates the relationship between the factors and the responses. To generate the approximate equation, a model formula, which serves as the basis for the approximate equation, must first be prepared. Once the model formula is established, the coefficients of each term constituting the model formula are determined through regression analysis using methods such as the least squares method.

[0005] Conventional analysis systems generate approximate equations by determining the coefficients of each operand using methods such as least squares, based on a predetermined model formula. However, recent research has revealed that some analysis parameters may have relationships with the analysis results that are not expressed in the predetermined model formula. Consequently, regression analysis based on the predetermined model formula cannot derive a correct approximate equation representing the relationship between analysis conditions containing such parameters and the analysis results.

[0006] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to increase the degree of freedom in constructing a model formula that is the basis of regression analysis and to improve the accuracy of the derived approximate formula.

[0007] [Technical means to solve the problem]

[0008] The data analysis system of the present invention includes: a data storage unit, which sets multiple analysis results obtained by multiple analyses performed under multiple analysis conditions as responses, sets multiple parameters included in the analysis conditions as factors, and stores the responses and the factors in association with each other; a data processing unit, which is configured to use at least one of the factors as a variable to generate an approximate formula representing the relationship between the variable and the response; and an information input device for the user to input information to the data processing unit, and the data processing unit is configured to perform the following steps: a variable setting step, in which the user sets at least one of the factors to be used as the variable; a structure setting step, in which the user arbitrarily sets the structure of the model formula that serves as the basis of the approximate formula using the variables set in the variable setting step; a model formula determination step, in which the model formula is determined based on the structure set by the user in the structure setting step; and an approximate formula determination step, in which the coefficients of each term constituting the model formula determined in the model formula determination step are determined by regression analysis, thereby determining the approximate formula.

[0009] The data analysis method of the present invention includes: an analysis data preparation step, in which a plurality of analysis results respectively obtained by a plurality of analyses performed under a plurality of analysis conditions are respectively set as responses, a plurality of parameters included in the analysis conditions are respectively set as factors, and preparation is performed in a state where the responses and the factors are correlated with each other; a variable setting step, in which at least one of the factors is arbitrarily set as a variable; a structure setting step, in which the variables set in the variable setting step are used to arbitrarily set the structure of a model formula as a basis for an approximate formula representing the relationship between the variables and the responses; a model formula determination step, in which the model formula is determined based on the structure set in the structure setting step; and an approximate formula determination step, in which the coefficients of each term constituting the model formula determined in the model formula determination step are determined by regression analysis, thereby determining the approximate formula.

[0010] [Effects of the Invention]

[0011] In the data analysis system of the present invention, the user is configured to set at least one parameter as a variable among multiple parameters included in the analysis conditions, and the user arbitrarily sets the structure of the model formula that serves as the basis of the approximate formula using the set variable. Therefore, the degree of freedom in constructing the model formula that serves as the basis of regression analysis is improved, and the accuracy of the derived approximate formula is improved.

[0012] In the data analysis method of the present invention, at least one parameter among the multiple parameters included in the analysis conditions is arbitrarily set as a variable, the set variable is used, the analysis result is used as a response, and the structure of the model formula serving as the basis of the approximate formula is arbitrarily set. Therefore, the degree of freedom in constructing the model formula serving as the basis of regression analysis is improved, and the accuracy of the derived approximate formula is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flowchart illustrating an embodiment of a data analysis method.

[0014] Figure 2 This is a block diagram showing an example of the structure of a data analysis system that executes the data analysis method.

[0015] Figure 3 This is an example of the model formula setting screen.

[0016] Figure 4 This is an example of the model formula detailed setting screen.

[0017] [Explanation of Symbols]

[0018] 1: Data Analysis System

[0019] 2: Data storage unit

[0020] 4: Data Processing Department

[0021] 6: Information input device

[0022] 8: Display DETAILED DESCRIPTION

[0023] Hereinafter, an embodiment of a data analysis system and a data analysis method will be described with reference to the accompanying drawings.

[0024] First, use Figure 1 The data analysis method of this embodiment is described with reference to the flowchart of FIG.

[0025] Initially, analytical data to be used in regression analysis is prepared (step 101). Analytical data refers to the analytical results (e.g., peak resolution, number of peaks, retention time of each peak, etc.) obtained by performing analysis on the same sample while varying multiple analytical condition parameters (e.g., mobile phase flow rate, column oven temperature, mobile phase solvent composition, mobile phase solvent mixing ratio, gradient method, sample injection volume, etc.) for each analysis. These data are then mapped to the various analytical condition parameters. In regression analysis, the analytical results obtained through analysis are defined as "responses," and the various analytical condition parameters are defined as "factors."

[0026] Next, based on the prepared analysis data, factors that should be used as variables in the generated approximate formula are selected and set from the various parameters of the analysis conditions (step 102). In the setting of the variables, any one or more parameters among the multiple parameters included in the analysis conditions can be set as variables. After setting the variables, the structure of the model formula that is the basis of the approximate formula representing the relationship between the variables and the response is set (step 103). The structure of the model formula can also be set arbitrarily. For example, an existing structure that only uses four arithmetic operations such as linear equations and quadratic equations can be used as the structure of the model formula, or a novel structure that incorporates arbitrary operations such as square roots, power of variables, exponential functions, and logarithmic functions can be used as the structure of the model formula.

[0027] After the structure of the model formula is set, the model formula is determined based on the set structure (step 104). The model formula is a formula that includes the sum of terms with indefinite coefficients of variables. After the model formula is determined, the coefficients of each term in the model formula are determined by regression analysis, thereby determining the approximate formula (steps 105 and 106). As a regression analysis for determining the coefficients of each term, in addition to the least squares method, Bayesian inference can also be used. The method of regression analysis can be arbitrarily set by the user. After the approximate formula is determined by regression analysis, processing such as drawing an approximate curve can be performed.

[0028] Figure 2 An example of the structure of a data analysis system for executing the data analysis method is shown in FIG.

[0029] The data analysis system 1 is implemented by an electronic computer installed with a computer program for implementing the data analysis method, and includes a data storage unit 2 , a data processing unit 4 , an information input device 6 , and a display 8 .

[0030] The data storage unit 2 is a storage area for storing analytical data obtained by the analysis device 100 and is implemented as a portion of an information storage element such as a hard disk drive. The analysis device 100 is, for example, a liquid chromatograph. The data processing unit 4 performs analysis based on the data analysis method described above on the analytical data stored in the data storage unit 2. The data processing unit 4 is connected to an information input device 6 and a display 8. The information input device 6 is implemented by a keyboard, a mouse, etc., and a user can input information to the data processing unit 4 through the information input device 6. Information corresponding to user prompts is output from the data processing unit 4 to the display 8 as needed and displayed on the display 8.

[0031] When performing regression analysis using the data analysis system 1, the user specifies analytical data to be analyzed from the data stored in the data storage unit 2 (preparing analytical data). Once the analytical data to be analyzed is specified, the data processing unit 4 displays a model formula setting screen on the display 8 for setting the model formula. The model formula setting screen also displays information necessary for setting the model formula.

[0032] Figure 3 This is an example of a model formula setting screen. In this example, the model formula setting screen includes a variable setting field, a model formula type setting field, and a preview field.

[0033] The variable settings section lists the factors (parameters included in the analysis conditions) that can be set as variables. The user can arbitrarily select one or more factors to be used as variables in the approximate equation. In this example, two factors are selected and set as variables X1 and X2, respectively.

[0034] In the Model Formula Type Settings section, you can set the structure of the model formula. In this example, you can choose between simple and detailed settings. In the simple setting, the maximum degree 1 and maximum degree 2 of the model formula are provided as options for the model formula structure. By selecting either maximum degree, the user can easily set the structure of the model formula for either a linear or quadratic equation. Furthermore, in the simple setting, you can select whether to incorporate interaction terms in the model formula, allowing the user to easily set the structure of the model formula using interaction terms.

[0035] When the user selects detailed setting in the model formula type setting column, the data processing unit 4 executes the model formula arbitrary setting mode and displays the following on the display 8: Figure 4 The model formula details setting screen shown. In addition to the four arithmetic operations, the model formula details setting screen also provides options for operations such as square root, coefficient power, exponential function, and logarithmic function. Users can use any operation from these options to set the structure of any model formula.

[0036] The data processing unit 4 generates a model formula of the structure set by the simple setting or the detailed setting, and configures it to be displayed in the preview column. Figure 3 In the example, the model formula "Y = aX1" is shown when the maximum degree is set to 2 in the simple setting and the interaction term is used. 2 +bX2+cX1+dX2+f". In the model formula, X1 and X2 are variables (factors), Y is the response, and a~f are the coefficients of each item.

[0037] When the user confirms the model formula displayed in the preview column and executes regression analysis using this model formula, an instruction to determine the model formula is input to the data processing unit 4. In this way, the model formula serving as the basis for regression analysis is determined. Figure 3 In the example, a "determine" button is configured at the lower right corner. By pressing this "determine" button (for example, using a mouse to point the cursor at the determine button and click it), an instruction to determine the model formula is input to the data processing unit 4.

[0038] When determining the model formula, the data processing unit 4 uses a regression analysis method such as the least squares method to determine the coefficients of each term constituting the model formula ( Figure 3 and Figure 4 In regression analysis, the coefficients are repeatedly fine-tuned so that the response value Y calculated by applying each factor to each variable in the model formula is close to the actual response value. The coefficients that best approximate the calculated response value Y to the actual response value are determined. By determining the coefficients of each term in the model formula, an approximate formula is determined that approximates the relationship between the factor and the response.

[0039] After determining the approximate formula, the data processing unit 4 may have a function of drawing an approximate line based on the approximate formula and displaying it on the display 8. Based on the approximate line and other information displayed on the display 8, the user can determine the optimal analysis conditions for the target sample.

[0040] The above-described embodiment is merely an example of implementation of the data analysis system, data analysis method, and computer program of the present invention. Implementation of the data analysis system, data analysis method, and computer program of the present invention is as follows.

[0041] In one embodiment of the data analysis system of the present invention, it includes: a data storage unit, which sets multiple analysis results obtained by multiple analyses performed under multiple analysis conditions as responses, sets multiple parameters included in the analysis conditions as factors, and stores the responses and factors in association with each other; a data processing unit, which is configured to use at least one of the factors as a variable to generate an approximate formula representing the relationship between the variable and the response; and an information input device for the user to input information to the data processing unit, and the data processing unit is configured to perform the following steps: a variable setting step, in which the user sets at least one of the factors to be used as the variable; a structure setting step, in which the user arbitrarily sets the structure of the model formula that serves as the basis of the approximate formula using the variables set in the variable setting step; a model formula determination step, in which the model formula is determined based on the structure set by the user in the structure setting step; and an approximate formula determination step, in which the coefficients of each term constituting the model formula determined in the model formula determination step are determined by regression analysis, thereby determining the approximate formula.

[0042] In a first aspect of the one embodiment of the data analysis system, a display electrically connected to the data processing unit is included, and the data processing unit is configured to display, on the display, options for the structure of the model formula and / or options for items to be incorporated into the model formula during the structure setting step, allowing the user to arbitrarily select, thereby allowing the user to set the structure of the model formula. This aspect allows the user to easily set a model formula of any structure.

[0043] In the first aspect, the data processing unit is configured to execute a model formula arbitrary setting mode in which a user can input an arbitrary model formula structure in the structure setting step. This allows the structure of the model formula to be unlimited, and even when a new relationship between a factor and a response is identified, an approximate formula that takes this relationship into account can be generated.

[0044] Furthermore, in the first aspect, the data processing unit is configured to display a preview of the structure model formula set by the user on the display during the structure setting step. This aspect allows the user to confirm the structure model formula set by the user, thereby preventing the generation of an incorrect structure model formula.

[0045] In a second aspect of the one embodiment of the data analysis system, the analysis is liquid chromatography analysis, the analysis result is any one of the number of peaks in the chromatogram, the resolution of the peaks in the chromatogram, and the retention time of the peaks appearing in the chromatogram, and the analysis conditions include at least one of the types of one or more solvents constituting the mobile phase, the flow rates of each of the one or more solvents, the temperature of the separation column, and the sample injection volume as the parameters. The second aspect can be combined with the first aspect.

[0046] In a third aspect of the embodiment of the data analysis system, the regression analysis is a least squares method. The third aspect can be combined with the first aspect and / or the second aspect.

[0047] In a fourth aspect of the embodiment of the data analysis system, the regression analysis is Bayesian inference. The fourth aspect can be combined with the first aspect and / or the second aspect.

[0048] In one embodiment of the data analysis method of the present invention, it includes: an analysis data preparation step, setting multiple analysis results obtained by multiple analyses performed under multiple analysis conditions as responses, setting multiple parameters included in the analysis conditions as factors, and preparing in a state where the responses and the factors are correlated with each other; a variable setting step, arbitrarily setting at least one of the factors as a variable; a structure setting step, using the variables set in the variable setting step, arbitrarily setting the structure of a model formula that serves as the basis of an approximate formula representing the relationship between the variables and the responses; a model formula determination step, determining the model formula based on the structure set in the structure setting step; and an approximate formula determination step, determining the coefficients of each term constituting the model formula determined in the model formula determination step through regression analysis, thereby determining the approximate formula.

[0049] In a first aspect of the embodiment of the data analysis method, in the structure setting step, the structure of the model formula is set using a structure and / or item selected from a plurality of pre-prepared options regarding the structure of the model formula and / or a plurality of pre-prepared options regarding items to be incorporated into the model formula. This aspect makes it easy to set a model formula of any structure.

[0050] In the first aspect, the structure of the model formula is generated in the structure setting step. This allows the structure of the model formula to be unlimited, and even when a new relationship between a factor and a response is identified, an approximate formula that takes this relationship into account can be generated.

[0051] In a second aspect of the one embodiment of the data analysis method, the analysis is liquid chromatography analysis, the analysis result is any one of the number of peaks in the chromatogram, the resolution of the peaks in the chromatogram, and the retention time of the peaks appearing in the chromatogram, and the analysis conditions include at least one of the types of one or more solvents constituting the mobile phase, the flow rates of each of the one or more solvents, the temperature of the separation column, and the sample injection volume as the parameters. This second aspect can be combined with the first aspect.

[0052] In a third aspect of the one embodiment of the data analysis method, the regression analysis is a least squares method. This third aspect can be combined with the first aspect and / or the second aspect.

[0053] In a fourth aspect of the one embodiment of the data analysis method, the regression analysis is Bayesian inference. This fourth aspect can be combined with the first aspect and / or the second aspect.

[0054] In one embodiment of the computer program of the present invention, the computer program is configured to be executed on a computer to thereby execute the above-described data analysis method.

Claims

1. A data analysis system, characterized in that: include: a data storage unit that stores, as responses, peak separations in a plurality of chromatograms obtained by a plurality of liquid chromatography analyses performed on the same sample under a plurality of analysis conditions, a plurality of parameters included in the analysis conditions as factors, and associates the responses with the factors; a data processing unit configured to use at least one of the factors as a variable and generate an approximate expression representing a relationship between the variable and the response; and an information input device for a user to input information to the data processing unit, and The data processing unit is configured to perform the following steps: a display step of displaying the factors and a plurality of operands; a variable setting step, wherein the user sets at least one of the factors to be used as the variable; a structure setting step of setting the structure of a model formula serving as a basis for the approximate formula using the variable by selecting at least one of the plurality of operands by a user; a model formula determination step of determining the model formula based on the structure set by the user in the structure setting step; as well as The approximate formula determination step determines the coefficients of each term constituting the model formula determined in the model formula determination step by regression analysis, thereby determining the approximate formula.

2. The data analysis system according to claim 1, wherein: including a display electrically connected to the data processing unit, and The data processing unit is configured to display, on the display, options for the structure of the model formula and / or options for items to be incorporated into the model formula in the structure setting step, for the user to select arbitrarily, thereby setting the structure of the model formula by the user.

3. The data analysis system according to claim 2, wherein: The structure of the model formula includes at least one arithmetic operation term.

4. The data analysis system according to claim 2 or 3, wherein: The structure of the model formula includes at least one of a square root, a power, an exponential function, and a logarithmic function.

5. The data analysis system according to claim 2, wherein: The data processing unit is configured to be able to execute a model formula arbitrary setting mode for a user to input an arbitrary structure of the model formula in the structure setting step.

6. The data analysis system according to claim 2 or 3, wherein: The data processing unit is configured to display a preview of a model formula of the structure set by the user on the display in the structure setting step.

7. The data analysis system according to any one of claims 1 to 3, wherein: The analysis is liquid chromatography analysis, The analysis conditions include at least one of the types of one or more solvents constituting the mobile phase, flow rates of each of the one or more solvents, temperature of the separation column, and sample injection volume as the parameters.

8. The data analysis system according to any one of claims 1 to 3, wherein: The regression analysis is the least squares method.

9. The data analysis system according to any one of claims 1 to 3, wherein: The regression analysis is Bayesian inference.

10. A data analysis method, characterized in that: include: an analysis data preparation step, wherein the peak separations in a plurality of chromatograms obtained by a plurality of liquid chromatography analyses performed on the same sample under a plurality of analysis conditions are respectively set as responses, a plurality of parameters included in the analysis conditions are respectively set as factors, and the responses and the factors are prepared in a state in which they are correlated with each other; a display step of displaying the factors and a plurality of operands; a variable setting step of arbitrarily setting at least one of the factors as a variable; a structure setting step of selecting at least one of the plurality of operands to set a structure of a model formula serving as a basis for an approximate formula representing a relationship between the variable and the response; a model formula determination step of determining the model formula based on the structure set in the structure setting step; and The approximate formula determination step determines the coefficients of each term constituting the model formula determined in the model formula determination step by regression analysis, thereby determining the approximate formula.

11. The data analysis method according to claim 10, wherein: In the structure setting step, the structure of the model formula is set using a structure and / or item selected from a plurality of pre-prepared options regarding the structure of the model formula and / or a plurality of pre-prepared options regarding items to be incorporated into the model formula.

12. The data analysis method according to claim 10 or 11, wherein: In the structure setting step, an arbitrary structure of the model formula is generated. The data analysis method according to claim 12 , wherein the structure includes at least one arithmetic operation term.

14. The data analysis method according to claim 12, wherein: The structure includes at least one of a square root, a power, an exponential function, and a logarithmic function.

15. The data analysis method according to claim 10, wherein: The analysis is liquid chromatography analysis, and the analysis conditions include at least one of the types of one or more solvents constituting the mobile phase, the flow rates of the one or more solvents, the temperature of the separation column, and the sample injection amount as the parameters.

16. The data analysis method according to claim 10, wherein: The regression analysis is the least squares method.

17. The data analysis method according to claim 10, wherein: The regression analysis is Bayesian inference.

18. A recording medium storing a computer program, characterized in that: The computer program is configured to be executed on a computer, thereby executing the data analysis method according to claim 10 or 11.

Images