Data processing device and data processing method

Abstract

The present invention provides a technique capable of accurately obtaining a correction coefficient for correcting a peak area of a chromatogram when obtaining a component ratio on the basis of the chromatogram of components contained in a sample. A data processing device according to the present invention uses a peak area of a first chromatogram obtained by measuring a calibration standard sample having a known concentration ratio between two components to obtain a correction coefficient for correcting the peak area, and calculates, as a component ratio in an unknown sample, a component ratio obtained by correcting a peak area of a second chromatogram obtained by measuring the unknown sample on the basis of the correction coefficient (see fig. 3).

Description

Data processing apparatus, data processing method

[0001] The present invention relates to a technique for calculating the abundance ratio of components contained in a sample.

[0002] For gene therapy drugs, as a vector for introducing a therapeutic gene into cells, for example, adeno-associated virus (AAV) is used. Structurally, AAV is an icosahedral capsid protein that encapsulates a DNA strand. This is called a complete particle. On the other hand, a state without a DNA strand, that is, an icosahedron composed only of capsid proteins, is called a hollow particle. Gene therapy drugs use complete particles as AAV vectors, but hollow particles are also generated in the process of drug production. For quality control, it is necessary to measure the contamination rate of hollow particles in the therapeutic drug.

[0003] When detecting complete particles and hollow particles by ion exchange chromatography, two peaks appearing on the chromatogram are detected, and the contamination rate of hollow particles can be calculated from the area ratio of each peak.

[0004] The following Patent Document 1 describes a technique for calculating the ratio of HbA1c contained in a hemoglobin sample using a chromatogram. In this document, the ratio of the peak area of HbA1c to the sum of the peak areas of all hemoglobins is determined as the ratio of HbA1c in the sample. This method will be called the simple percentage method.

[0005] The following Patent Document 2 describes a method for correcting the relationship between the peak area and the component ratio by multiplying a correction factor by the peak area when using a chromatogram in the same manner. This method will be called the modified percentage method.

[0006] Japanese Patent Application Laid-Open No. 2012-108118, Japanese Patent Application Laid-Open No. 1-214754

[0007] If the peak area in a chromatogram directly represents the ratio of its components, the component ratio can be determined using the simple percentage method. On the other hand, if, for example, the peak area of ​​one component is detected with twice the sensitivity of another component, it is necessary to correct the peak area to account for that sensitivity. For example, if 1 million particles A are detected with a peak area of ​​1 million counts, and 1 million particles B are detected with a peak area of ​​2 million counts, the formula for calculating the ratio of particle A is A / {A + 0.5 × B}. The modified percentage method is based on this idea.

[0008] When using the modified percentage method, it is necessary to determine the correction factor (0.5 in the above example) in advance. However, even if the correction factor is determined, its actual value may change over time. If the correction factor is not accurate, the accuracy of the measurement results of the component ratios (calculation results using the modified percentage method) will decrease.

[0009] This invention has been made in view of the above-mentioned problems, and aims to provide a technique that can accurately determine a correction coefficient for correcting the peak area of ​​a chromatogram when determining the component ratio based on the chromatogram of components contained in a sample.

[0010] The data processing device according to the present invention determines a correction coefficient to correct the peak area of ​​a first chromatogram obtained by measuring a calibration standard sample in which the concentration ratio between two components is known, and calculates the component ratio obtained by correcting the peak area of ​​a second chromatogram obtained by measuring an unknown sample based on the correction coefficient, as the component ratio in the unknown sample.

[0011] According to the data processing device of the present invention, when determining the component ratio based on the chromatogram of components contained in a sample, it is possible to accurately determine the correction coefficient for correcting the peak area of ​​the chromatogram. Other problems, configurations, advantages, etc. of the present invention will become clear from the following description of embodiments.

[0012] This is an example of a chromatogram of components contained in a sample. It is a diagram plotting the relationship between R and Rc. This is a configuration diagram of the data processing device 1 according to Embodiment 5.

[0013] <Embodiment 1> Figure 1 shows an example of a chromatogram of components contained in a sample. Here, we assume that components A and B are contained in the sample. Component A is, for example, HbA1c, and component B is, for example, HbA0. Using the peak areas of component A and component B, the ratio R of component A in the sample is expressed by the following formula 1. This is called the simple percentage method. For the ratio R of component B, simply replace the numerator with B: R = A / (A + B) (Formula 1)

[0014] On the other hand, the peak area ratio may not always correspond to the component ratio. For example, in a sample containing particle A and particle B, particle B may be detected with twice the sensitivity of particle A. In such cases, it is necessary to correct the peak area of ​​particle B by multiplying it by a correction factor x. In this case, the ratio Rc of particle A in the sample is expressed by the following equation 2. In this example, x = 0.5. This is called the corrected percentage method. For the ratio Rc of particle B, simply replace the numerator with x × B: Rc = A / {A + x × B} (Equation 2)

[0015] The correction coefficient x needs to be determined in advance, for example, as a literature value or experimental value. However, even if the correction coefficient x is obtained, its value may change over time. If the correction coefficient x is not accurate, the component ratio cannot be accurately determined. Therefore, in Embodiment 1 of the present invention, a method is proposed that can determine the correction coefficient x using a simple and accurate method.

[0016] In calibration standard samples where the relative abundance of each component is known in advance, Rc in the above example is already known, so it is thought that the correction coefficient x can be determined by solving an equation with the correction coefficient x as an unknown variable. Therefore, in Embodiment 1, the chromatogram of the calibration standard sample was obtained by actual measurement, and the correction coefficient x was determined using the peak area of ​​that chromatogram.

[0017] Figure 2 plots the relationship between R and Rc. By removing AB from equations 1 and 2, we obtain equation 3 below. A plot can be created for each value of the correction coefficient x: Rc = 1 / {1 + x(-1 + 1 / R)} (Equation 3)

[0018] The value of Rc is known in advance for the calibration standard sample. The value of R can be calculated by obtaining the chromatogram of the calibration standard sample through actual measurement. By applying these values ​​to the graph in Figure 2, x can be determined. For example, if the measured R is 0.4 and Rc is 0.34, then x = 1.3. It is not necessary to create plots in Figure 2 for all values ​​of x; x can be determined by interpolation for the gaps.

[0019] The following describes the specific steps for finding x.

[0020] Step 1: Prepare a calibration standard sample with a known Rc. For example, by mixing the preparation samples of component A and component B, a mixed sample with a component A ratio of Rc = 10% is prepared and used as the calibration standard sample.

[0021] Step 2: Place the calibration standard sample into the chromatography apparatus and obtain a chromatogram. Measure the peak areas of component A and component B from the chromatogram.

[0022] Step 3: Determine the ratio R of component A using the simple percentage method from the measured peak area.

[0023] Step 4: Apply Rc = 10% and the R obtained in Step 3 to the graph in Figure 2, and find the correction coefficient x so that it matches Rc = 10% and the R obtained in Step 3. This completes Equation 3.

[0024] Step 5: Place the unknown sample, whose Rc is unknown, into the chromatography apparatus and determine the R of the unknown sample in the same manner as in Steps 3 and 4.

[0025] Step 6: The Rc of the unknown sample is calculated by substituting the R obtained in Step 5 into Equation 3. This Rc is based on actual measurements and is calculated based on the known Rc of the calibration standard sample, so sufficient accuracy is considered to be ensured.

[0026] <Embodiment 1: Summary> The method according to Embodiment 1 determines a correction coefficient x based on the peak area ratio of the first chromatogram of a calibration standard sample whose component concentration ratios are known. Furthermore, the component ratio Rc (component ratio Rc obtained by corrected percentage) obtained by correcting the peak area ratio of the second chromatogram of an unknown sample with the correction coefficient x is determined based on a relationship between the component ratio R and Rc using simple percentages. This makes it possible to obtain a highly accurate correction coefficient x and to accurately determine the component ratio based on that correction coefficient x.

[0027] <Embodiment 2> Embodiment 2 of the present invention describes an example in which the method of Embodiment 1 is applied to a gene therapy drug using an adeno-associated virus vector. In this case, component A is a hollow particle, component B is a complete particle, and it is necessary to multiply component B by a correction coefficient x. The procedure for determining the percentage of hollow particles (Rc) is described below.

[0028] Step 1: A calibration standard sample with a hollow particle content (Rc) of the reference value Eref is placed in the chromatography apparatus to obtain a chromatogram. The peak areas of component A and component B are measured from the chromatogram.

[0029] Step 2: Determine the ratio Emeas of component A using the simple percentage method from the measured peak area.

[0030] Step 3: By applying Eref and Emeas to the graph in Figure 2, we find a conversion formula that converts Emeas to Eref. This conversion formula does not necessarily have to be formula 3; any conversion formula that can convert R to Rc (i.e., Emeas to Eref) can be used.

[0031] Step 3: Supplement: When deriving the transformation formula, ensure that if R is 0%, then Rc is also 0%. Similarly, ensure that if R is 100%, then Rc is also 100%. These are called boundary conditions.

[0032] Step 4: The Rc of the unknown sample is calculated by substituting the R obtained in Step 5 into Equation 3. This Rc has sufficient accuracy, similar to Embodiment 1. It also satisfies the origin coincidence requirements for both 0% and 100%.

[0033] <Embodiment 3> The methods of Embodiments 1 and 2 can be used to determine the ratio of hemoglobin components in a blood sample. For example, by using the methods of Embodiments 1 and 2, with component A being HbA1c and component B being HbA0, the ratio of HbA1c can be determined.

[0034] On the other hand, the Japan Diabetes Society recommends calculating the HbA1c ratio by obtaining a linear regression equation (y = ax + b) from two calibration samples. This calculation formula does not satisfy the boundary conditions described in step 3 of Embodiment 2. In contrast, by applying the method described in Embodiment 2 to the hemoglobin component, the accurate component ratio can be determined using a single calibration sample while satisfying the boundary conditions.

[0035] The specific procedure is the same as in Embodiment 2. That is, x is determined by substituting the component ratio of HbA1c in the calibration standard sample (Rc) and the component ratio of the measured HbA1c (R) into Equation 3, respectively. Rc is then determined by substituting the R of the unknown sample into the determined Equation 3. This allows for the accurate determination of the component ratio of HbA1c.

[0036] Hemoglobin has four components in addition to HbA1c (HbA1a, HbA1b, HbF, and HbA0), but the peak of HbA0 is dominant among these four components. Therefore, by defining components A and HbA1c, and component B as HbA0 (or the sum of HbA1a, HbA1b, HbF, and HbA0), the proportion of HbA1c can be easily determined using only the peaks of two components.

[0037] <Embodiment 4> The methods of Embodiments 1 and 2 can be used to determine the ratio of each component in a sample containing 20 amino acid components that make up a protein. For example, 17 of the 20 components, excluding asparagine, glutamine, and tryptophan, can be measured. A detection wavelength of 570 nm is used for ordinary amino acids (primary amines), and a detection wavelength of 440 nm is used for secondary amines such as proline. However, to simplify the procedure, proline may also be detected using the same 570 nm wavelength as the other amino acids.

[0038] Index numbers 0 to 16 are assigned to each of the 17 components. Component A is defined as proline, and components B1 to B16 are the other 16 components. Sensitivity coefficients x in the modified percentage method are set to x1 to x16, corresponding to each component. Since the ratios of each component in the calibration standard sample (Rc1 to Rc16) are known, 16 equations can be obtained by substituting the measured values ​​of each component (R1 to R16) into equation 3. By solving each equation, x1 to x16 can be determined. The ratios Rc1 to Rc16 (i = 1 to 16) of components B1 to B16 can be determined by the following equation 4: Rci = (xi × Bi) / {A + Σ ｊ＝１ １６ (xj×Bj)} (Formula 4)

[0039] <Embodiment 5> Figure 3 is a configuration diagram of a data processing device 1 according to Embodiment 5 of the present invention. The data processing device 1 is a device that implements the methods described in Embodiments 1 to 4. The data processing device 1 may be configured as part of a chromatography apparatus, as a device that controls a chromatography apparatus, or as an independent device.

[0040] The data processing device 1 comprises a correction coefficient calculation unit 11 and a component calculation unit 12. The correction coefficient calculation unit 11 calculates a correction coefficient x. The component calculation unit 12 uses the calculated correction coefficient x to determine the component ratios within the sample.

[0041] The correction coefficient calculation unit 11 performs, for example, steps 3 to 4 in Embodiment 1. The peak areas of component A and component B may be obtained from a chromatography device, or the correction coefficient calculation unit 11 may calculate using a chromatogram output by the chromatography device. The component calculation unit 12 performs, for example, steps 5 to 6 in Embodiment 1. The R of the unknown sample may be obtained from a chromatography device, or the component calculation unit 12 may calculate using a chromatogram output by the chromatography device.

[0042] <Regarding Modifications of the Present Invention> In the above embodiments, the correction coefficient calculation unit 11 and the component calculation unit 12 can be configured by hardware such as a circuit device that implements these functions, or can be configured by a computing device such as a CPU (Central Processing Unit) executing software that implements these functions.

[0043] In the above embodiments, the conversion formula for converting R to Rc does not necessarily have to be Formula 3. Even when using Formula 3, instead of algebraically obtaining x, for example, x that satisfies R and Rc on a graph as shown in FIG. 2 may be obtained by numerical interpolation. As long as it satisfies the boundary conditions, any conversion formula other than Formula 3 may be used. For example, functions such as sigmoid curves, tanh, sinh, arcsinh, etc. can be used. On the other hand, if it is not necessary to satisfy the boundary conditions, an appropriate monotonically increasing function (e.g., quadratic curve, polynomial curve, etc.) can also be used.

[0044] In the above embodiments, when plotting R and Rc on a graph as shown in FIG. 2 and obtaining x by numerical interpolation based on the plot, for example, spline interpolation can be used. The mathematical formula obtained by spline interpolation is not necessarily the same as Formula 3, but it is considered that the accuracy of the correction coefficient x can be sufficiently ensured. For example, when plotting a plurality of pairs of R and Rc, a mathematical formula that satisfies these plots can be obtained by spline interpolation or the like. On the other hand, when there is only one pair of R and Rc, the correction coefficient x may be obtained by substituting them into Formula 3.

[0045] In the above embodiments, the correction coefficient x may be recalculated at predetermined intervals (e.g., once a day). Thereby, even when daily differences occur, the accuracy of the correction coefficient x can be ensured. Further, when the correction coefficient x changes significantly from the previous value, it can be regarded as an abnormality and used as an abnormality detection means.

[0046] In the above embodiments, even for a sample in which three or more components appear on the chromatogram, the above embodiments can be used as a method for accurately obtaining the area ratio between two of the peaks.

[0047] Supplement the calculation formula of the correction percentage method. Let the correction coefficient of component A be a and the correction coefficient of component B be b, then the ratio of component A is aA / (aA + bB). Dividing the numerator and denominator of this formula by a gives A / {A + (b / a)B}. As a result, the correction coefficient of component A is no longer necessary, and only the correction coefficient of component B needs to be obtained. Therefore, the correction percentage for two components is as shown in Equation 2. Similarly, even in the case of three or more components, the correction coefficient of the first component is not necessary. Therefore, in Embodiment 4, only 16 correction coefficients of components B1 to B16 are obtained. Further, on the premise that the total of all components is 100% as a boundary condition, if the total of components B1 to B16 is subtracted from 100%, the ratio of component A can be obtained. Therefore, even without obtaining the correction coefficient of component A, the ratio of component A can be calculated.

[0048] 1: Data processing device 11: Correction coefficient calculation unit 12: Component calculation unit

Claims

1. A data processing device for calculating the relative abundance of components contained in a sample, comprising: a correction coefficient calculation unit that determines a correction coefficient for correcting the peak area based on the peak area of ​​a first chromatogram obtained by measuring a calibration standard sample in which at least two components are mixed and the concentration ratio between the at least two components is known; and a component calculation unit that calculates the component ratio obtained by correcting the peak area of ​​a second chromatogram obtained by measuring an unknown sample based on the correction coefficient as the relative abundance of the component corresponding to each peak in the second chromatogram.

2. The data processing apparatus according to claim 1, wherein the calibration standard sample includes a first component and a second component, and the correction coefficient calculation unit calculates the correction coefficient such that the correction percentage of the components in the calibration standard sample is expressed by summing the peak area of ​​the first component appearing in the first chromatogram and the peak area of ​​the second component appearing in the first chromatogram multiplied by the correction coefficient.

3. The data processing apparatus according to claim 2, characterized in that the correction coefficient calculation unit sums the peak areas of each component in the calibration standard sample that appear in the first chromatogram; the correction coefficient calculation unit identifies one of the components in the calibration standard sample that appear in the first chromatogram as a target component and obtains the peak area of ​​the target component; the correction coefficient calculation unit calculates the simple percentage of the proportion of the target component in the calibration standard sample by dividing the peak area of ​​the target component by the sum; and the correction coefficient calculation unit obtains the correction coefficient by substituting the known proportion of the target component in the calibration standard sample into a conversion formula that converts the simple percentage to the corrected percentage.

4. The data processing apparatus according to claim 3, wherein the correction coefficient calculation unit plots the simple percentage and the corrected percentage for each candidate value of the correction coefficient, and the correction coefficient calculation unit identifies the conversion formula in the plot from which the known ratio of the target component is obtained.

5. The data processing apparatus according to claim 3, characterized in that the component calculation unit calculates the relative abundance of the component in the unknown sample by converting the simple percentage obtained for the unknown sample into the corrected percentage using the correction coefficient.

6. The calibration standard sample is a sample composed of complete particles made of adeno-associated virus and hollow particles that do not contain DNA strands, wherein the mixing ratio of the hollow particles is known, and the component calculation unit calculates the abundance ratio of the hollow particles in the unknown sample based on the second chromatogram, as described in claim 1.

7. The calibration standard sample is a sample composed of complete particles made of adeno-associated virus and hollow particles that do not contain DNA strands, and the proportion of the hollow particles is known; the correction coefficient calculation unit specifies the conversion formula such that when the simple percentage is 0, the corrected percentage is also 0; the correction coefficient calculation unit specifies the conversion formula such that when the simple percentage is 100, the corrected percentage is also 100; and the component calculation unit calculates the proportion of the hollow particles in the unknown sample based on the second chromatogram, as described in claim 4.

8. The data processing apparatus according to claim 1, wherein the calibration standard sample is composed of HbA1c and other hemoglobin in the blood, and the component calculation unit calculates the relative abundance of HbA1c in the unknown sample based on the second chromatogram.

9. The data processing apparatus according to claim 1, characterized in that the calibration standard sample comprises three or more components, the correction coefficient calculation unit calculates a correction coefficient for each component contained in the calibration standard sample, and the component calculation unit calculates the relative abundance of each component in the unknown sample based on the second chromatogram using the correction coefficient obtained for each component contained in the calibration standard sample.

10. The data processing apparatus according to claim 3, wherein the calibration standard sample contains amino acids, the correction coefficient calculation unit calculates the correction coefficient for each of the target components using the amino acids as the target components, and the component calculation unit calculates the relative abundance of each of the amino acids in the unknown sample based on the second chromatogram using the correction coefficient obtained for each of the target components.

11. The data processing device according to claim 1, characterized in that the correction coefficient calculation unit recalculates the correction coefficient at predetermined intervals.

12. A data processing method for calculating the relative abundance of components contained in a sample, comprising the steps of: determining a correction coefficient for correcting the peak area based on the peak area of ​​a first chromatogram obtained by measuring a calibration standard sample in which at least two components are mixed and the concentration ratio between the at least two components is known; and calculating the component ratio obtained by correcting the peak area of ​​a second chromatogram obtained by measuring an unknown sample based on the correction coefficient as the relative abundance of the component corresponding to each peak in the second chromatogram.

Images