Peak analysis method and peak analysis device

Abstract

Provided are a peak analysis method and a peak analysis device that are capable of identifying, from a plurality of items of data obtained from a detection device, the number of peaks included in a spectrum required for performing peak separation processing and a target for the peak separation processing. This peak analysis method analyzes spectrum peaks in a plurality of items of data obtained from a detection device. The peak analysis method includes: a step for setting an analysis target spectrum vector and a non-target spectrum vector; a step for dividing the analysis target spectrum vector into components in the direction of the non-target spectrum vector; and a step for transferring prescribed proportions of the spectrum vectors between the non-target spectrum vector and the analysis target spectrum vector on the basis of comparison of components in the direction of the analysis target spectrum vector with the non-target spectrum vector.

Description

Peak analysis method and peak analysis device

[0001] The present disclosure relates to a peak analysis method and a peak analysis device for analyzing spectral peaks in a plurality of data obtained from a detection device.

[0002] When performing analysis using a multichannel detector such as a photodiode array (PDA) detector or a liquid chromatograph (LC) using a mass spectrometer as a detector, the compounds contained in a sample are separated and analyzed quantitatively or qualitatively. However, when a sample contains many compounds or compounds with similar physical properties, the spectrum obtained from the detector may contain multiple compounds in a single peak, or the tails of multiple peaks may overlap.

[0003] In such cases, the obtained spectrum is subjected to peak separation processing, such as peak deconvolution processing and processing in which peaks are separated by regarding the shape of the peak as a mixed Gaussian distribution obtained by linearly combining multiple Gaussian functions (Japanese Patent No. 6260709 (Patent Document 1)).

[0004] Patent No. 6260709

[0005] However, in these peak separation processes, it is necessary to set the number of peaks (number of compounds) contained in the spectrum beforehand, but in actual analysis, it is not possible to know the number of compounds contained in a sample in advance. Therefore, it is necessary to perform peak separation process by assuming the number of peaks contained in the spectrum in advance, compare and evaluate the processing result with the actual data, and repeat the peak separation process by changing the assumed number of peaks until the evaluation value satisfies a predetermined threshold.

[0006] Furthermore, because these peak separation processes require a large amount of memory capacity for calculations, it is difficult to process all of the data obtained from the detection device due to the hardware configuration, and therefore it is necessary to identify in advance the peaks to be subjected to peak separation processing from the data obtained from the detection device.

[0007] The present disclosure has been made to solve such problems, and aims to provide a peak analysis method and a peak analysis device that can identify the number of peaks contained in a spectrum required for peak separation processing and the target for peak separation processing from multiple data obtained from a detection device.

[0008] A peak analysis method according to the present disclosure is a method for analyzing spectral peaks in a plurality of data sets obtained from a detection device, and includes the steps of: selecting one piece of data to be analyzed from the plurality of data sets as a spectral vector to be analyzed; setting data that is temporally different from the spectral vector to be analyzed as a non-target spectral vector; dividing the spectral vector to be analyzed into components in the direction of the non-target spectral vector; and transferring a predetermined proportion of the spectral vector between the non-target spectral vector and the spectral vector to be analyzed based on a comparison between the directional components of the spectral vector to be analyzed and the non-target spectral vector.

[0009] A peak analysis device according to the present disclosure analyzes spectral peaks. The peak analysis device includes an input unit that receives multiple pieces of data from a detection device, and a calculation unit that analyzes spectral peaks for the multiple pieces of data received by the input unit. The calculation unit selects one piece of data to be analyzed from the multiple pieces of data as a spectral vector to be analyzed, sets data that is temporally different from the spectral vector to be analyzed as a non-target spectral vector, separates the spectral vector to be analyzed into components in the direction of the non-target spectral vector, and, based on a comparison of the directional components of the spectral vector to be analyzed and the non-target spectral vector, transfers a predetermined proportion of the spectral vector between the non-target spectral vector and the spectral vector to be analyzed.

[0010] In the present disclosure, it is possible to easily identify the number of peaks contained in the spectrum required for peak separation processing and the targets for peak separation processing from multiple data obtained from a detection device.

[0011] Fig. 1 is a schematic diagram of a liquid chromatograph including a peak analysis device according to an embodiment; Fig. 2 is a schematic diagram of a data processing unit according to an embodiment; Fig. 3 is a flowchart showing a peak analysis method according to an embodiment; Fig. 4 is a diagram showing an example of a spectrum vector according to an embodiment; Fig. 5 is a diagram showing an example of decomposition and comparison of an analysis target spectrum vector; Fig. 6 is a diagram showing an example of transfer of a part of an analysis target spectrum vector to a pre-target spectrum vector; Fig. 7 is a diagram showing an example of transfer of a part of an analysis target spectrum vector to a post-target spectrum vector; Fig. 8 is a diagram for explaining an analysis result of the peak analysis method according to an embodiment; Fig. 9 is a flowchart showing a peak analysis method according to a modified example.

[0012] Hereinafter, embodiments will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference numerals and description thereof will not be repeated.

[0013] [Liquid chromatograph apparatus including a peak analysis device] In this embodiment, an example of a configuration including a peak analysis device that analyzes spectral peaks for multiple data obtained from a detection device including a liquid chromatograph will be described. Note that the detection device to which the peak analysis device is applied is not limited to detection devices including a liquid chromatograph, and may be applied to detection devices such as multi-channel detection devices such as photodiode array (PDA) detection devices and mass spectrometers. Figure 1 is a schematic diagram of a liquid chromatograph apparatus including a peak analysis device according to an embodiment.

[0014] The detection device (LC device 1) including a liquid chromatograph includes a mobile phase container 11 in which a mobile phase is stored, a liquid delivery pump 12 that draws in the mobile phase and delivers it at a constant flow rate, an injector 13 that injects a sample into the delivered mobile phase, a column 14 that separates components in the sample in the time direction, a PDA detector 15 that detects components in the eluate from the column 14, and an A / D converter 16 that converts analog data detected by the PDA detector 15 into digital data.

[0015] A peak analysis device 10 that analyzes spectral peaks for multiple pieces of data obtained from the LC device 1 is connected to the LC device 1. The peak analysis device 10 includes a data analysis unit 2, an input unit 3, and a display unit 4. The data analysis unit 2 performs data collection processing, waveform processing, and qualitative and quantitative analysis processing. The waveform processing performs spectrum convergence processing, peak separation processing, and peak determination processing. The peak analysis device 10 is also connected to an input unit 3 that allows an analyst to perform various input operations, and a display unit 4 that displays processing results, etc.

[0016] 2 is a schematic diagram of a data processing unit according to an embodiment. A program executed by the data analysis unit 2 can perform processes such as data collection, waveform processing, and qualitative and quantitative analysis on multiple pieces of data obtained from the LC device 1. The data analysis unit 2 includes a processor 102, a main memory 104, an input I / F 106, an output I / F 108, a storage 110, and an external drive 112. These components are connected via a processor bus 118.

[0017] The processor 102 is configured with a CPU, a GPU, etc., and can read out programs (for example, an OS 1102 and a program 1104) stored in the storage 110, expand them into the main memory 104, and execute them. The processor 102 executes the program 1104, which performs processes such as waveform processing and qualitative and quantitative analysis on a plurality of data obtained from the LC device 1 via the input I / F 106.

[0018] The main memory 104 is configured with a volatile storage device such as a DRAM or an SRAM. The storage 110 is configured with a nonvolatile storage device such as an HDD or an SSD. The storage 110 stores an OS 1102 for implementing basic functions, as well as a program 1104 for performing processes such as waveform processing, qualitative and quantitative analysis, and the like.

[0019] The input I / F 106 is connected not only to the LC apparatus 1 but also to an input unit 3 that allows an analyst to perform various input operations. The output I / F 108 is connected to a display unit 4, and can display the processing results of waveform processing, qualitative and quantitative analysis processing, etc. on the display unit 4.

[0020] In the LC system 1, a liquid delivery pump 12 draws mobile phase from a mobile phase container 11 and delivers it to a column 14 via an injector 13. The injector 13 injects a predetermined amount of sample into the mobile phase at a predetermined timing according to instructions from a control unit (not shown). The injected sample is carried by the flow of the mobile phase and introduced into the column 14, and various components in the sample are separated in the time direction while passing through the column 14. An eluate containing the separated components is discharged from the outlet of the column 14 and introduced into a PDA detector 15. Although not shown, the PDA detector 15 includes a cell through which the eluate flows, a light source, a spectroscope that wavelength-disperses the light that has passed through the cell, and a PDA element that simultaneously detects the wavelength-dispersed light.

[0021] Light emitted from the light source is irradiated onto a cell and, as it passes through the eluate flowing through the cell, is absorbed in a manner specific to the components in the eluate. The absorbed light is then wavelength-dispersed by a spectrometer, and light within a predetermined wavelength range is detected by a PDA element for each wavelength. Therefore, the PDA detector 15 simultaneously obtains detection signals reflecting the absorption spectrum across the predetermined wavelength range. In response to control from the control unit, the PDA detector 15 repeatedly performs this absorption measurement at predetermined time intervals. The A / D converter 16 converts the detection signals output from the PDA detector 15 into digital data and outputs them to the peak analyzer 10. Therefore, LC measurement of one sample yields three-dimensional data having three dimensions: time t, wavelength λ, and absorbance y (signal intensity).

[0022] Before performing qualitative and quantitative analysis on the multiple data sets obtained from the LC device 1, the peak analyzer 10 performs waveform processing to separate overlapping peaks in the chromatogram and absorption spectrum. Furthermore, for each peak determined by waveform processing, the peak analyzer 10 calculates the peak top position (retention time, absorption wavelength) and the peak area (or height) on the chromatogram. In the qualitative and quantitative analysis, the peak analyzer 10 qualifies the components based on the peak position and quantifies the components based on the peak area or height. The peak analyzer 10 then displays the qualitative and quantitative results on the display unit 4.

[0023] The peak analysis device 10 performs peak separation processing to separate overlapping peaks in a spectrum through waveform processing. This peak separation processing requires that the number of peaks (number of compounds) contained in the spectrum be set in advance, but in actual analysis, the number of compounds contained in a sample cannot be known in advance. Therefore, it is necessary to perform peak separation processing by assuming the number of peaks contained in the spectrum in advance, compare and evaluate the processing result with actual data, and repeat the peak separation processing by changing the assumed number of peaks until the evaluation value satisfies a predetermined threshold.

[0024] Therefore, in the peak analysis device 10, a spectral convergence process is performed as preprocessing before the peak separation process to identify the number of peaks. The spectral convergence process can process all of the data obtained from the detection device, and it is possible to identify overlapping peaks in the spectrum and obtain the number of overlapping peaks. Therefore, the peak separation process requires a large amount of memory capacity for calculation, which makes it difficult to process all of the data obtained from the detection device due to the hardware configuration. However, by performing the spectral convergence process as preprocessing, this can be resolved.

[0025] [Spectral Convergence Processing] Next, the spectral convergence processing will be specifically described. Fig. 3 is a flowchart showing a peak analysis method according to an embodiment. First, the peak analysis device 10 reads a plurality of data obtained from the LC device 1 into the main memory 104 (step S101). If the plurality of data obtained from the LC device 1 has not been read (NO in step S101), the peak analysis device 10 returns the processing to step S101.

[0026] If multiple data sets obtained from the LC device 1 are successfully read (YES in step S101), the peak analysis device 10 regards the spectra of the multiple data sets at each time t as a single vector (spectral vector). In other words, the peak analysis device 10 treats the chromatogram data from the LC device 1 as time-varying data of the spectral vector. Specifically, FIG. 4 illustrates an example of a spectral vector according to an embodiment. In the example illustrated in FIG. 4, the PDA detector 15 is simplified by measuring the signal intensities at two wavelengths (X nm, Y nm). Therefore, the spectral vector from time t = 60 msec to time t = 63 msec can be plotted on the two-dimensional graph shown in FIG. 4, with the horizontal axis representing the signal intensity at the X nm wavelength and the vertical axis representing the signal intensity at the Y nm wavelength. Of course, in actual measurements, the PDA detector 15 can measure signal intensities at hundreds to thousands of wavelengths, so the number of dimensions of the spectral vector will be hundreds to thousands. It should be noted that not only the PDA detector 15 but also other detectors such as a UV detector and a mass detector can handle a plurality of measured data as a spectral vector.

[0027] Next, the peak analysis device 10 selects one piece of data to be analyzed from the multiple pieces of data and sets it as an analysis target spectral vector (step S102). The peak analysis device 10 sets the data that precedes the analysis target spectral vector in time as a pre-target spectral vector, and the data that follows the analysis target spectral vector in time as a post-target spectral vector (step S103). The peak analysis device 10 separates the analysis target spectral vector into a first-direction component of the pre-target spectral vector and a second-direction component of the post-target spectral vector (step S104). The peak analysis device 10 compares the first-direction component of the analysis target spectral vector with the pre-target spectral vector.

[0028] 5A and 5B are diagrams showing an example of decomposition and comparison of spectral vectors to be analyzed. In Fig. 5A, spectral vector A at time t = 62 msec is selected as the spectral vector to be analyzed. When spectral vector A at time t = 62 msec is selected as the spectral vector to be analyzed, the pre-target spectral vector is set to spectral vector B at time t = 61 msec, and the post-target spectral vector is set to spectral vector C at time t = 63 msec.

[0029] Next, the spectral vector A shown in Fig. 5(a) is decomposed into a component A1 in the direction of the spectral vector B and a component A2 in the direction of the spectral vector C. The peak analysis device 10 compares the decomposed component A1 with the spectral vector B, as shown in Fig. 5(b).

[0030] Returning to FIG. 3 , the peak analysis device 10 determines whether the magnitude of the pre-target spectral vector is greater than the magnitude of the component in the first direction of the analysis-target spectral vector (step S105). If the magnitude of the pre-target spectral vector is greater than the magnitude of the component in the first direction of the analysis-target spectral vector (YES in step S105), the peak analysis device 10 adds a predetermined proportion of the component in the first direction of the analysis-target spectral vector to the pre-target spectral vector (step S106). On the other hand, if the magnitude of the pre-target spectral vector is equal to or less than the magnitude of the component in the first direction of the analysis-target spectral vector (NO in step S105), the peak analysis device 10 adds a predetermined proportion of the pre-target spectral vector to the component in the first direction of the analysis-target spectral vector (step S107). Here, the predetermined proportion is the proportion (transfer ratio) at which the vector is transferred. The predetermined proportion may be a constant value (e.g., 0.1, 0.6, etc.) or may vary for each vector based on information such as the magnitude of the pre-target spectral vector (or the post-target spectral vector) and the magnitudes of the decomposed components.

[0031] Specifically, Fig. 6 shows an example of transferring a portion of the analysis target spectral vector to a pre-target spectral vector. As shown in Fig. 6(a), when the decomposed component A1 is compared with the spectral vector B, the spectral vector B is larger than the decomposed component A1. Therefore, a predetermined proportion (e.g., 0.1) of the decomposed component A1, i.e., vector A1a, is transferred to and added to the spectral vector B. Fig. 6(b) shows the spectral vector after adding vector A1a to the spectral vector B, where the decomposed component A1 becomes component A1-A1a, and the spectral vector B becomes spectral vector B+A1a.

[0032] 3 illustrates an example in which, in a comparison between the first-direction component of the analysis-target spectral vector and the pre-target spectral vector, it is determined whether the magnitude of the pre-target spectral vector is greater than the magnitude of the first-direction component of the analysis-target spectral vector, and then it is determined whether to proceed to step S106 or step S107. However, this is not a limitation, and, in a comparison between the first-direction component of the analysis-target spectral vector and the pre-target spectral vector, it may be determined whether the magnitude of the pre-target spectral vector is equal to or greater than the magnitude of the first-direction component of the analysis-target spectral vector, and then it is determined whether to proceed to step S106 or step S107. Furthermore, in the process illustrated in FIG. 3, it is an example in which it is determined whether the magnitude of the pre-target spectral vector is greater than the magnitude of the first-direction component of the analysis-target spectral vector, and then it is determined whether to proceed to step S106 or step S107. However, it is also possible to proceed to step S106 if the determination is YES, and not proceed to step S107 if the determination is NO. Conversely, it may be possible to determine whether the magnitude of the pre-target spectral vector is greater than the magnitude of the component in the first direction of the spectral vector to be analyzed, and if the determination is YES, not proceed to step S106, but if the determination is NO, proceed to step S107.

[0033] 3 , the peak analysis device 10 determines whether the magnitude of the target spectral vector is greater than the magnitude of the second-direction component of the analysis-target spectral vector (step S108). If the magnitude of the target spectral vector is greater than the magnitude of the second-direction component of the analysis-target spectral vector (YES in step S108), the peak analysis device 10 adds a predetermined proportion of the second-direction component of the analysis-target spectral vector to the target spectral vector (step S109). On the other hand, if the magnitude of the target spectral vector is equal to or less than the magnitude of the second-direction component of the analysis-target spectral vector (NO in step S108), the peak analysis device 10 adds a predetermined proportion of the target spectral vector to the second-direction component of the analysis-target spectral vector (step S110).

[0034] Specifically, Fig. 7 shows an example of transferring a portion of the analysis target spectral vector to a post-target spectral vector. As shown in Fig. 7(a), when the decomposed component A2 is compared with the spectral vector C, the spectral vector C is larger than the decomposed component A2. Therefore, a predetermined proportion (e.g., 0.1) of the decomposed component A2, i.e., vector A2a, is transferred to and added to the spectral vector C. Fig. 7(b) shows the spectral vector after adding vector A2a to the spectral vector C, where the decomposed component A2 becomes component A2-A2a, and the spectral vector C becomes spectral vector C+A2a.

[0035] 3 illustrates an example in which, in a comparison between the second-direction component of the analysis-target spectral vector and the target posterior spectral vector, it is determined whether the magnitude of the target posterior spectral vector is greater than the magnitude of the second-direction component of the analysis-target spectral vector, and then it is determined whether to proceed to step S109 or step S110. However, this is not a limitation, and, in a comparison between the second-direction component of the analysis-target spectral vector and the target posterior spectral vector, it may be determined whether the magnitude of the target posterior spectral vector is equal to or greater than the magnitude of the second-direction component of the analysis-target spectral vector, and then it is determined whether to proceed to step S109 or step S110. Furthermore, in the process illustrated in FIG. 3, it is an example in which, in a comparison between the second-direction component of the analysis-target spectral vector and the target posterior spectral vector, it is determined whether to proceed to step S109 or step S110, and then it is determined whether the magnitude of the target posterior spectral vector is greater than the magnitude of the second-direction component of the analysis-target spectral vector. However, it is also possible to proceed to step S109 if the determination is YES, and not proceed to step S110 if the determination is NO. Conversely, it may be possible to determine whether the magnitude of the target spectral vector is greater than the magnitude of the second-direction component of the analysis target spectral vector, and if the determination is YES, not proceed to step S109, but if the determination is NO, proceed to step S110.

[0036] 3, the peak analysis device 10 determines whether all of the data read in step S101 has been processed as a spectral vector to be analyzed (step S111). If all of the data has not been processed (NO in step S111), the peak analysis device 10 returns to step S102. For example, after processing the spectral vector at time t=62 msec shown in FIG. 5 as a spectral vector to be analyzed, the peak analysis device 10 then processes the spectral vector at time t=63 msec as a spectral vector to be analyzed.

[0037] If all data has not been processed (YES in step S111), the peak analysis device 10 performs a process of updating the data to the added data (step S112). Specifically, FIG. 8 shows an example of the analysis target spectral vector after the transfer. In FIG. 8, components A1-A1a and A2-A2a are combined, and the spectral vector Aa is set as the spectral vector at time t = 62 msec after the update. Note that the spectral vectors for each time are repeatedly transferred in steps S102 to S111. However, rather than updating the spectral vectors for each transfer, the spectral vectors are updated all at once after all data has been processed. Furthermore, before starting the processes of steps S102 to S111, the peak analysis device 10 reads and copies the data onto the main memory 104 in step S101, and performs the addition and subtraction via vector transfer on the copied data. This is to prevent the problem of results varying depending on the spectral vector of the time from which the processing was started when adding or subtracting via vector transfer on the same data.

[0038] 3, the peak analysis device 10 determines whether the processes of steps S102 to S112 have been performed a predetermined number of times (for example, several tens to several hundreds of times) (step S113). If the processes have not been performed the predetermined number of times (NO in step S113), the peak analysis device 10 returns the process to step S102. In this embodiment, the processes of steps S102 to S112 constitute a single spectral convergence process, and by performing this spectral convergence process multiple times, noise components are canceled from the measured spectrum, a target spectrum for peak separation processing is identified, and the number of peaks contained in the spectrum can be determined.

[0039] When the process has been performed a predetermined number of times (YES in step S113), the peak analysis device 10 determines whether or not it is necessary to analyze other data (step S114). If it is necessary to analyze other data (YES in step S114), the peak analysis device 10 returns the process to step S101. If it is not necessary to analyze other data (NO in step S114), the peak analysis device 10 ends the process.

[0040] Next, the analysis results obtained by performing a spectral convergence process using the peak analysis method according to the present embodiment will be described. FIG. 9 is a diagram illustrating the analysis results obtained by the peak analysis method according to the present embodiment. FIG. 9( a) shows chromatogram data obtained from the LC device 1, and each of the three circled peaks I to III appears to have one peak. However, when the spectral convergence process is performed on the data 200 times, as shown in FIG. 9( b), it is clear that each of the three circled peaks I to III actually has two peaks. In FIGS. 9( a) and 9(b), the horizontal axis represents time and the vertical axis represents signal intensity.

[0041] FIG. 9(c) shows the spectra of each peak III identified in FIG. 9(b). In FIG. 9(c), the horizontal axis represents wavelength, and the vertical axis represents signal intensity. In the peak analysis method according to this embodiment, the number of compounds (number of peaks) contained in a chromatogram, the retention time of each compound, and the spectrum of each compound can be identified in advance by performing a spectral convergence process. This enables easy and accurate peak separation (e.g., deconvolution) processing. In particular, performing the spectral convergence process in advance eliminates the need to pre-determine the number of compounds in the peak separation process. This eliminates the need for trial-and-error adjustments, which involve evaluating results based on a pre-determined number of compounds, thereby speeding up processing. Furthermore, performing the spectral convergence process allows the time range of co-eluting chromatograms to be determined in advance. Therefore, by narrowing down the chromatograms to be processed to those within the time range required for peak separation, the amount of memory required for calculation processing can be reduced. Furthermore, the spectral convergence process itself does not require the number of compounds to be pre-determined, requires less memory, and allows the entire chromatogram data to be processed at once. The spectral convergence process can process the entire chromatogram data at once, so there is no need to perform separate processes for each peak.

[0042] [Modifications] (1) In the above-described embodiment, the spectral convergence process is repeated a predetermined number of times (step S113). However, the number of times the spectral convergence process is repeated need not be a fixed value (e.g., 100 or 200 times), but may be the number of times the process is repeated until the ratio of the number of spectral vectors transferred to the number of spectral vectors to be analyzed becomes equal to or less than a predetermined threshold (e.g., 0.3 or 0.5). Repeating the spectral convergence process reduces the number of times that spectral vectors are added or subtracted, and the spectrum converges to a certain state. Therefore, the ratio (transfer ratio) of the number of spectral vectors transferred to the number of spectral vectors to be analyzed may be monitored to determine the degree of spectral convergence, and the number of times the spectral convergence process is repeated may be varied accordingly.

[0043] Specifically, FIG. 10 is a flowchart showing a peak analysis method according to a modified example. Note that, in the process steps shown in FIG. 10, the same process steps as those shown in FIG. 3 are denoted by the same reference numerals, and detailed description thereof will not be repeated. In the process of steps S102 to S112, the peak analysis device 10 determines whether the ratio of the number of transferred spectral vectors to the number of spectral vectors to be analyzed is equal to or less than a predetermined threshold (step S113a). If the ratio of the transferred number is greater than the predetermined threshold (NO in step S113a), the peak analysis device 10 returns the process to step S102. On the other hand, if the ratio of the transferred number is equal to or less than the predetermined threshold (YES in step S113a), the peak analysis device 10 determines whether it is necessary to analyze other data (step S114).

[0044] (2) In the above embodiment, the pre-target spectral vector is described as a spectral vector at a time before the analysis target spectral vector, but it may be an average spectral vector at multiple times before the analysis target spectral vector. Similarly, the post-target spectral vector is described as a spectral vector at a time after the analysis target spectral vector, but it may be an average spectral vector at multiple times after the analysis target spectral vector.

[0045] (3) In the above-described embodiment, all of the multiple data sets obtained by the detection device are analyzed. However, thinned data from the multiple data sets obtained by the detection device may also be analyzed. For example, instead of analyzing all of the spectral vectors from time t = 60 msec to time t = 63 msec shown in FIG. 4, the spectral vectors from time t = 60 msec and time t = 62 msec may be analyzed. This reduces the processing time for the spectral convergence process.

[0046] (4) Increasing the specified ratio (the ratio of vectors transferred (transfer rate)) described in the above embodiment can shorten the processing time of the spectral convergence process, but reduces the accuracy of identifying the number of peaks. Conversely, decreasing the ratio increases the processing time of the spectral convergence process, but improves the accuracy of identifying the number of peaks.

[0047] 3 described in the embodiment, the peak analysis device 10 performs both the process of determining whether the magnitude of the pre-target spectral vector is greater than the magnitude of the component of the analysis-target spectral vector in the first direction (step S105) and the process of determining whether the magnitude of the post-target spectral vector is greater than the magnitude of the component of the analysis-target spectral vector in the second direction (step S108). However, the present invention is not limited to this, and the peak analysis device 10 may perform only one of the processes of step S105 or step S108.

[0048] (6) The pre-target spectral vector and post-target spectral vector described in the above embodiment can be generically conceptualized as a non-target spectral vector, where data that is temporally different from the analysis target spectral vector is referred to as a non-target spectral vector. Therefore, a peak analysis method for analyzing spectral peaks in a plurality of data sets obtained from a detection device can be generically conceptualized as including the steps of: selecting one piece of data to be analyzed from the plurality of data sets as the analysis target spectral vector; setting data that is temporally different from the analysis target spectral vector as the non-target spectral vector; dividing the analysis target spectral vector into directional components of the non-target spectral vector; and transferring a predetermined proportion of the spectral vector between the non-target spectral vector and the analysis target spectral vector based on a comparison between the directional components of the analysis target spectral vector and the non-target spectral vector.

[0049] Aspects It will be understood by those skilled in the art that the above-described embodiments are specific examples of the following aspects.

[0050] (Item 1) A peak analysis method according to one aspect is a peak analysis method for analyzing spectral peaks of a plurality of data obtained from a detection device, and includes the steps of selecting one piece of data to be analyzed from the plurality of data as a spectral vector to be analyzed, and setting data that is temporally different from the spectral vector to be analyzed as a non-target spectral vector; dividing the spectral vector to be analyzed into directional components of the non-target spectral vector; and transferring a predetermined proportion of the spectral vector between the non-target spectral vector and the spectral vector to be analyzed based on a comparison between the directional components of the spectral vector to be analyzed and the non-target spectral vector.

[0051] According to the peak analysis method described in paragraph 1, it is possible to easily identify the number of peaks contained in the spectrum required for peak separation processing and the target for peak separation processing from multiple data obtained from the detection device.

[0052] (Clause 2) A peak analysis method according to paragraph 1, wherein the non-target spectral vectors include a pre-target spectral vector that is temporally earlier than the analyzed target spectral vector and a posterior target spectral vector that is temporally later than the analyzed target spectral vector, and includes the steps of: dividing the analyzed target spectral vector into a first-direction component of the pre-target spectral vector and a second-direction component of the posterior target spectral vector; and comparing the first-direction component of the analyzed target spectral vector with the pre-target spectral vector, and if the magnitude of the pre-target spectral vector is greater than the magnitude of the first-direction component of the analyzed target spectral vector, adding a predetermined proportion of the first-direction component of the analyzed target spectral vector to the pre-target spectral vector.

[0053] (Clause 3) The peak analysis method described in Clause 2 further includes a step of comparing the second-direction component of the analysis target spectral vector with the target posterior spectral vector, and if the magnitude of the target posterior spectral vector is greater than the magnitude of the second-direction component of the analysis target spectral vector, adding a predetermined proportion of the second-direction component of the analysis target spectral vector to the target posterior spectral vector.

[0054] (Clause 4) The peak analysis method described in Clause 3 further includes the steps of comparing the first-direction component of the analysis target spectral vector with the pre-target spectral vector, and adding a predetermined proportion of the pre-target spectral vector to the first-direction component of the analysis target spectral vector if the magnitude of the pre-target spectral vector is equal to or smaller than the magnitude of the first-direction component of the analysis target spectral vector; and comparing the second-direction component of the analysis target spectral vector with the post-target spectral vector, and adding a predetermined proportion of the post-target spectral vector to the second-direction component of the analysis target spectral vector if the magnitude of the second-direction component of the analysis target spectral vector is equal to or smaller than the magnitude of the second-direction component of the analysis target spectral vector.

[0055] (Clause 5) A peak analysis method according to paragraph 1, wherein the non-target spectral vectors include a pre-target spectral vector that is temporally earlier than the analyzed target spectral vector and a posterior target spectral vector that is temporally later than the analyzed target spectral vector, and the method includes the steps of: dividing the analyzed target spectral vector into a first-direction component of the pre-target spectral vector and a second-direction component of the posterior target spectral vector; and comparing the first-direction component of the analyzed target spectral vector with the pre-target spectral vector, and, if the magnitude of the pre-target spectral vector is smaller than the magnitude of the first-direction component of the analyzed target spectral vector, adding a predetermined proportion of the pre-target spectral vector to the first-direction component of the analyzed target spectral vector.

[0056] (Clause 6) The peak analysis method according to Clause 5 further includes a step of comparing the second-direction component of the analysis target spectral vector with the target posterior spectral vector, and adding a predetermined proportion of the target posterior spectral vector to the second-direction component of the analysis target spectral vector if the magnitude of the target posterior spectral vector is smaller than the magnitude of the second-direction component of the analysis target spectral vector.

[0057] (Clause 7) The peak analysis method according to clause 6 further includes the steps of: comparing the first-direction component of the analysis target spectral vector with the pre-target spectral vector; and, if the magnitude of the pre-target spectral vector is equal to or greater than the magnitude of the first-direction component of the analysis target spectral vector, adding a predetermined proportion of the first-direction component of the analysis target spectral vector to the pre-target spectral vector; and comparing the second-direction component of the analysis target spectral vector with the post-target spectral vector; and, if the magnitude of the post-target spectral vector is equal to or greater than the magnitude of the second-direction component of the analysis target spectral vector, adding a predetermined proportion of the second-direction component of the analysis target spectral vector to the post-target spectral vector.

[0058] (Item 8) In the peak analysis method according to any one of Items 2 to 7, the pre-target spectral vector is a spectral vector at a time before the target spectral vector to be analyzed, or an average spectral vector at multiple times before the target spectral vector to be analyzed.

[0059] (Item 9) In the peak analysis method according to any one of Items 2 to 8, the post-target spectral vector is a spectral vector at a time later than the analysis target spectral vector, or an average spectral vector at multiple times later than the analysis target spectral vector.

[0060] (Item 10) A peak analysis method according to any one of items 1 to 9, wherein not all of the plurality of data obtained from the detection device are analyzed, but thinned-out data are analyzed.

[0061] (Item 11) In the peak analysis method according to any one of items 1 to 10, the process of analyzing the spectral peaks is repeated a predetermined number of times for a plurality of data obtained from the detection device.

[0062] (12) In the peak analysis method according to the 11th aspect, the predetermined number of times is the number of times until the ratio of the number of spectral vectors to be transferred to the number of spectral vectors to be analyzed becomes equal to or smaller than a predetermined threshold.

[0063] (13) In the peak analysis method according to any one of the above items 1 to 12, the detection device is a chromatograph, and the spectral peaks are analyzed for the obtained chromatogram.

[0064] (Clause 14) A peak analysis device according to one aspect is a peak analysis device that analyzes spectral peaks, and includes an input unit that receives multiple pieces of data from a detection device, and a calculation unit that analyzes spectral peaks for the multiple pieces of data received by the input unit. The calculation unit selects one piece of data to be analyzed from the multiple pieces of data as a spectral vector to be analyzed, sets data that is temporally different from the spectral vector to be analyzed as a non-target spectral vector, separates the spectral vector to be analyzed into directional components of the non-target spectral vector, and, based on a comparison of the directional components of the spectral vector to be analyzed with the non-target spectral vector, transfers a predetermined proportion of the spectral vector between the non-target spectral vector and the spectral vector to be analyzed.

[0065] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0066] 1 LC device, 2 data analysis unit, 3 input unit, 4 display unit, 10 peak analysis device, 11 mobile phase container, 12 liquid delivery pump, 13 injector, 14 column, 15 PDA detector, 16 A / D converter, 102 processor, 104 main memory, 106 input I / F, 108 output I / F, 110 storage, 112 external drive, 118 processor bus.

Claims

1. A peak analysis method for analyzing spectral peaks of a plurality of data obtained from a detection device, comprising the steps of: selecting one piece of data to be analyzed from the plurality of data as a spectral vector to be analyzed, and setting a piece of data that is temporally different from the spectral vector to be analyzed as a non-target spectral vector; dividing the spectral vector to be analyzed into directional components of the non-target spectral vector; and transferring a predetermined proportion of spectral vectors between the non-target spectral vector and the spectral vector to be analyzed based on a comparison between the directional components of the spectral vector to be analyzed and the non-target spectral vector.

2. The peak analysis method according to claim 1, wherein the non-target spectral vector includes a pre-target spectral vector that is temporally earlier than the analyzed target spectral vector, and a posterior target spectral vector that is temporally later than the analyzed target spectral vector, and the method further comprises the steps of: dividing the analyzed target spectral vector into a first directional component of the pre-target spectral vector and a second directional component of the posterior target spectral vector; and comparing the first directional component of the analyzed target spectral vector with the pre-target spectral vector, and if the magnitude of the pre-target spectral vector is greater than the magnitude of the first directional component of the analyzed target spectral vector, adding a predetermined proportion of the first directional component of the analyzed target spectral vector to the pre-target spectral vector.

3. The peak analysis method according to claim 2, further comprising a step of comparing the second direction component of the analyzed spectral vector with the target posterior spectral vector, and if the magnitude of the target posterior spectral vector is greater than the magnitude of the second direction component of the analyzed spectral vector, adding the predetermined proportion of the second direction component of the analyzed spectral vector to the target posterior spectral vector.

4. The peak analysis method according to claim 3, further comprising the steps of: comparing the first direction component of the analyzed spectral vector with the pre-object spectral vector, and adding the predetermined proportion of the pre-object spectral vector to the first direction component of the analyzed spectral vector if the magnitude of the pre-object spectral vector is equal to or smaller than the magnitude of the first direction component of the analyzed spectral vector; and comparing the second direction component of the analyzed spectral vector with the object posterior spectral vector, and adding the predetermined proportion of the object posterior spectral vector to the second direction component of the analyzed spectral vector if the magnitude of the object posterior spectral vector is equal to or smaller than the magnitude of the second direction component of the analyzed spectral vector.

5. The peak analysis method according to claim 1, wherein the non-target spectral vector includes a pre-target spectral vector that is temporally earlier than the analyzed target spectral vector, and a posterior target spectral vector that is temporally later than the analyzed target spectral vector, and the method further comprises the steps of: dividing the analyzed target spectral vector into a first directional component of the pre-target spectral vector and a second directional component of the posterior target spectral vector; and comparing the first directional component of the analyzed target spectral vector with the pre-target spectral vector, and if the magnitude of the pre-target spectral vector is smaller than the magnitude of the first directional component of the analyzed target spectral vector, adding a predetermined proportion of the pre-target spectral vector to the first directional component of the analyzed target spectral vector.

6. The peak analysis method according to claim 5, further comprising a step of comparing the second direction component of the analyzed spectral vector with the target posterior spectral vector, and if the magnitude of the target posterior spectral vector is smaller than the magnitude of the second direction component of the analyzed spectral vector, adding the predetermined proportion of the target posterior spectral vector to the second direction component of the analyzed spectral vector.

7. The peak analysis method according to claim 6, further comprising the steps of: comparing the first direction component of the analyzed spectral vector with the pre-object spectral vector, and adding a predetermined proportion of the first direction component of the analyzed spectral vector to the pre-object spectral vector if the magnitude of the pre-object spectral vector is equal to or greater than the magnitude of the first direction component of the analyzed spectral vector; and comparing the second direction component of the analyzed spectral vector with the object posterior spectral vector, and adding a predetermined proportion of the second direction component of the analyzed spectral vector to the object posterior spectral vector if the magnitude of the object posterior spectral vector is equal to or greater than the magnitude of the second direction component of the analyzed spectral vector.

8. A peak analysis method according to claim 2 or claim 5, wherein the pre-target spectral vector is a spectral vector at a time prior to the analysis target spectral vector, or an average spectral vector at a plurality of times prior to the analysis target spectral vector.

9. A peak analysis method according to claim 3 or 6, wherein the post-target spectral vector is a spectral vector at a time later than the analysis target spectral vector, or an average spectral vector at a plurality of times later than the analysis target spectral vector.

10. The peak analysis method according to claim 1, wherein not all of the plurality of data obtained from the detection device are analyzed, but thinned-out data are analyzed.

11. The peak analysis method according to claim 1, further comprising repeating a process of analyzing the spectral peaks a predetermined number of times for the plurality of data obtained from the detection device.

12. The peak analysis method according to claim 11, wherein the predetermined number of times is the number of times until a ratio of the number of spectral vectors to be transferred to the number of spectral vectors to be analyzed becomes equal to or smaller than a predetermined threshold value.

13. The peak analysis method according to claim 1, wherein the detection device is a chromatograph and the spectral peaks are analyzed for the obtained chromatogram.

14. A peak analysis device for analyzing spectral peaks, comprising: an input unit for receiving a plurality of data from a detection device; and a calculation unit for analyzing spectral peaks for the plurality of data received by the input unit, wherein the calculation unit performs a process of: selecting one piece of data to be analyzed from the plurality of data as a spectral vector to be analyzed, setting a piece of data that is temporally different from the spectral vector to be analyzed as a spectral vector other than the target spectral vector, dividing the spectral vector to be analyzed into components in the direction of the non-target spectral vector, and transferring a predetermined ratio of spectral vectors between the non-target spectral vector and the spectral vector to be analyzed based on a comparison between the directional components of the spectral vector to be analyzed and the non-target spectral vector.

Images