WO2012164956A1 - パターン又はfpの特徴量作成方法、作成プログラム、及び作成装置 - Google Patents
パターン又はfpの特徴量作成方法、作成プログラム、及び作成装置 Download PDFInfo
- Publication number
- WO2012164956A1 WO2012164956A1 PCT/JP2012/003618 JP2012003618W WO2012164956A1 WO 2012164956 A1 WO2012164956 A1 WO 2012164956A1 JP 2012003618 W JP2012003618 W JP 2012003618W WO 2012164956 A1 WO2012164956 A1 WO 2012164956A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- peak
- feature
- target
- attribution
- creation
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/62—Detectors specially adapted therefor
- G01N30/74—Optical detectors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
- G01N30/8624—Detection of slopes or peaks; baseline correction
- G01N30/8631—Peaks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
- G01N30/8675—Evaluation, i.e. decoding of the signal into analytical information
- G01N30/8686—Fingerprinting, e.g. without prior knowledge of the sample components
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/88—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/15—Medicinal preparations ; Physical properties thereof, e.g. dissolubility
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B15/00—Systems controlled by a computer
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C99/00—Subject matter not provided for in other groups of this subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N2030/022—Column chromatography characterised by the kind of separation mechanism
- G01N2030/027—Liquid chromatography
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/88—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86
- G01N2030/8886—Analysis of industrial production processes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/86—Signal analysis
- G01N30/8675—Evaluation, i.e. decoding of the signal into analytical information
Definitions
- the present invention relates to a pattern feature value creation method, a multi-component substance, for example, a multi-component substance FP feature value creation method, a creation program, and a creation device for evaluating quality of a Chinese medicine that is a multi-component drug.
- multi-component substance for example, there is a natural product-derived drug such as Chinese medicine, which is a drug composed of multiple components (hereinafter referred to as multi-component drug).
- Chinese medicine which is a drug composed of multiple components (hereinafter referred to as multi-component drug).
- the quantitative and qualitative profiles of these drugs vary due to the geological factors, ecological factors, collection time, collection location, collection age, growing season weather, etc. of the raw material crude drug used.
- the criteria for determining the quality of a multi-component drug and the like are generally set based on the content and the like of one or several characteristic components in the multi-component drug.
- Non-Patent Document 1 when an active ingredient cannot be identified in a multi-component drug, quantitative analysis is possible, it is easily dissolved in water, does not decompose in hot water, does not chemically react with other components, etc. A plurality of components having the above are selected, and the content of those components obtained by chemical analysis is used as a criterion for evaluation.
- Patent Document 1 a multi-component drug is evaluated by selecting a part of the peak in the HPLC chromatogram data and converting it into a barcode.
- the object of evaluation is limited to “content of specific component” or “chromatography peak of specific component”, and some of the components contained in the multi-component drug are to be evaluated. Only. For this reason, since there are many components that are not subject to evaluation for multi-component drugs, they are insufficiently accurate as an evaluation method for multi-component drugs.
- the problem to be solved is that the existing evaluation method has a limit in efficiently evaluating the quality of multi-component substances with high accuracy.
- the present invention divides a pattern in which a peak changes in time series into a plurality of regions and patterns from the presence rate or the amount of peaks existing in each region.
- the feature of the pattern feature quantity creation method is that a pattern area division feature quantity creation step for creating an area division feature quantity is provided.
- the present invention divides an FP composed of a peak detected from a chromatographic analysis of a multi-component substance and its retention time into a plurality of regions, and calculates the FP region dividing feature amount from the abundance rate or abundance of the peaks existing in each region.
- the feature of the FP feature quantity creation method is that it includes an FP region division feature quantity creation step for creating the FP feature quantity.
- the present invention provides a computer with a pattern area division feature quantity creation function for creating a pattern area division feature quantity from the existence rate or quantity of peaks existing in each area by dividing a pattern whose peak changes in time series into a plurality of areas.
- the realization is a feature of the pattern feature creation program.
- the present invention divides an FP composed of a peak detected from a chromatographic analysis of a multi-component substance and its retention time into a plurality of regions, and calculates the FP region dividing feature amount from the abundance rate or abundance of the peaks existing in each region.
- the feature of the FP feature quantity creation program is that it has an FP region division feature quantity creation function.
- the present invention includes a pattern region division feature amount creation unit that divides a pattern whose peak changes in time series into a plurality of regions and creates a pattern region division feature amount from the presence rate or the amount of the peak existing in each region.
- This is the feature of the pattern feature quantity creation device.
- the present invention divides an FP composed of a peak detected from a chromatographic analysis of a multi-component substance and its retention time into a plurality of regions, and calculates the FP region dividing feature amount from the abundance rate or abundance of the peaks existing in each region.
- the feature of the FP feature quantity creation device is that it includes an FP area division feature quantity creation section for creating a FP feature quantity.
- the pattern or FP feature value creation method of the present invention has the above configuration, the pattern or FP feature value can be easily obtained by area division. Therefore, for example, a feature amount can be created by capturing a fine peak.
- the pattern or FP feature value creation program of the present invention has the above-described configuration, it is possible to easily obtain the pattern or FP feature value by causing a computer to realize each function. Since the pattern or FP feature value creation apparatus of the present invention has the above-described configuration, it is possible to easily obtain the pattern or FP feature value by operating each part.
- Example 1 It is a block diagram of the evaluation apparatus of a multicomponent medicine.
- Example 1 It is a block diagram which shows the evaluation procedure of a multicomponent medicine.
- Example 1 It is explanatory drawing of FP created from three-dimensional chromatogram data (henceforth, 3D chromatography).
- Example 1 (A) is the drug A, (B) is the drug B, and (C) is the FP of the drug C.
- Example 1 It is a figure which shows the retention time of object FP and reference
- Example 1 It is a figure which shows the retention * time * appearance pattern of object FP.
- Example 1 It is a figure which shows the retention, time, and appearance pattern of reference
- Example 1 It is a figure which shows the coincidence number of retention, time, and appearance distance of object FP and reference
- Example 1 It is a figure which shows the coincidence degree of the retention time and appearance pattern of the target FP and the reference FP.
- Example 1 It is a figure which shows the attribution object peak of object FP.
- Example 1 It is a peak pattern figure by three peaks including an assignment object peak.
- Example 1 It is a peak pattern figure by five peaks including an assignment object peak.
- Example 1 It is a figure which shows the tolerance
- Example 1 It is a figure which shows the attribution candidate peak of the reference
- Example 1 It is a peak pattern figure by three peaks of the attribution object peak and the attribution candidate peak.
- Example 1 It is a peak pattern figure by three peaks of the attribution candidate peak and another attribution candidate peak.
- Example 1 It is a peak pattern figure by three peaks of the attribution candidate peak and another attribution candidate peak. (Example 1)) It is a peak pattern figure by three peaks of the attribution candidate peak and another attribution candidate peak.
- Example 1 It is a peak pattern figure by five peaks of the attribution object peak and the attribution candidate peak.
- Example 1 It is a peak pattern figure by five peaks of the attribution candidate peak different from the attribution object peak.
- Example 1 It is a peak pattern figure by five peaks of the attribution candidate peak different from the attribution object peak.
- Example 1 It is a peak pattern figure by five peaks of the attribution candidate peak different from the attribution object peak.
- Example 1 It is a figure which shows the peak pattern configuration candidate peak of the attribution object peak and the attribution candidate peak.
- Example 1 It is a figure which shows the number of all the peak patterns of the attribution object peak when a peak pattern structure candidate peak is set to four.
- Example 1 It is a figure which shows the number of all the peak patterns of the attribution candidate peak when a peak pattern structure candidate peak is set to four.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is explanatory drawing of the comprehensive comparison of the peak pattern of the attribution candidate peak with respect to the peak pattern of the attribution object peak.
- Example 1 It is a figure which shows the calculation method of the coincidence degree of the peak pattern by three peaks of the attribution object peak and the attribution candidate peak.
- Example 1 It is a figure which shows the calculation method of the coincidence degree of the peak pattern by three peaks of the attribution object peak and the attribution candidate peak.
- Example 1 It is a figure which shows the calculation method of the coincidence degree of the peak pattern by five peaks of the attribution object peak and the attribution candidate peak.
- Example 1 It is a figure which shows the UV spectrum of the attribution object peak and the attribution candidate peak.
- Example 1 It is explanatory drawing of the coincidence degree of the UV spectrum of an attribution object peak and an attribution candidate peak.
- Example 1 It is explanatory drawing of coincidence degree calculation of the attribution candidate peak by the comparison of both a peak pattern and UV spectrum.
- Example 1 It is explanatory drawing which shows attribution to the reference
- Example 1 It is a figure which shows the condition where object FP was attributed to reference
- Example 1 It is explanatory drawing which shows quantification by area
- Example 1 It is explanatory drawing which shows the relationship with fluctuation
- Example 1 It is explanatory drawing which changes and changes the position of an area
- Example 1 It is a chart which shows the data of FP type 2.
- Example 1 It is explanatory drawing which shows the pattern of FP type 2.
- Example 1 It is explanatory drawing which shows the feature-value conversion for every area
- Example 1 It is explanatory drawing which shows the setting of a vertical division line (1st line).
- Example 1 It is explanatory drawing which shows the setting of a horizontal parting line (1st line).
- Example 1 It is explanatory drawing which shows the area
- Example 1 It is explanatory drawing which shows the number of the area
- FIG. 6 is an explanatory diagram showing identification of a region 1;
- Example 1 It is a graph which shows the height and total of all the peaks.
- Example 1 It is explanatory drawing which shows the sum total of the peak height of the area
- FIG. Example 1 It is a graph which shows the feature-value of all the area
- Example 1 It is a graph which shows the feature-value in each area
- Example 1 It is a graph which shows the feature-value in each area
- Example 1 It is a graph which shows the feature-value in one way which does not change the position of each vertical and horizontal dividing line.
- Example 1 It is a figure which shows various object FP and its evaluation value (MD value).
- Example 1 It is a figure which shows various object FP and its evaluation value (MD value).
- Example 1 It is a figure which shows various object FP and its evaluation value (MD value).
- Example 1 It is a figure which shows various object FP and its evaluation value (MD value).
- Example 1 It is a figure which shows various object FP and its evaluation value (MD value).
- Example 1 It is process drawing which shows the evaluation method of a multicomponent chemical
- Example 1 It is a quality evaluation flowchart of a multi-component medicine.
- Example 1 It is a quality evaluation flowchart of a multi-component medicine.
- Example 1 It is a data processing flowchart in the FP creation function by a single wavelength.
- Example 1 It is a data processing flowchart in the FP creation function by a plurality of wavelengths.
- Example 1 It is a data processing flowchart in the FP creation function by a plurality of wavelengths.
- Example 1 It is a data processing flowchart in peak attribution processing 1 (selection of standard FP).
- Example 1 It is a data processing flowchart in peak attribution processing 2 (calculation of attribution score).
- Example 1 It is a data processing flowchart in the peak attribution process 3 (identification of a corresponding peak).
- Example 1 It is a data processing flowchart in the peak attribution process 4 (attribute to the reference group FP).
- Example 1 It is a data processing flowchart in the peak attribution process 4 (attribute to the reference group FP).
- Example 1 10 is a flowchart of a retention / time / appearance pattern coincidence calculation process in peak attribution process 1 (selection of reference FP).
- Example 1 It is a flowchart of the coincidence degree calculation process of the UV spectrum in the peak attribution process 2 (calculation of the attribution score).
- Example 1 It is a flowchart of the coincidence calculation processing of peak patterns in peak attribution processing 2 (calculation of attribution score).
- Example 1 It is a flowchart which shows the detail of "FP_type2 creation”.
- Example 1 It is a flowchart which shows the detail of "the feature-quantization process of object FP_type2 by area division”.
- Example 1 It is a flowchart which shows the detail of "integration of the peak feature-value and area
- Example 1 It is a flowchart for creating a reference FP feature value integration file.
- Example 1 It is a flowchart for creating a reference FP feature value integration file.
- Example 1 It is a flowchart which shows the detail of "reference
- Example 1 It is a flowchart which shows the detail of “reference
- Example 1 It is a flowchart which shows the detail of “peak feature-quantization process (creation of the reference group FP).”
- Example 1 It is a flowchart which shows the detail of "the production
- Example 1 It is a flowchart which shows the detail of "the feature-quantization process of the reference
- Example 1 It is a flowchart which concerns on the feature-value integration process of reference
- Example 1 It is a chart which shows the example of data of 3D chromatography.
- Example 1 It is a chart which shows the example of data of peak information.
- Example 1 It is a chart which shows the example of data of FP.
- Example 1 It is a graph which shows the attribution score calculation result (judgment result file example) to the reference
- Example 1 It is a graph which shows the collation process of the peak corresponding by object FP and reference
- Example 1 It is a graph which shows the result (collation result file) example which specified the peak corresponding to object FP and standard FP.
- Example 1 It is a chart which shows the example of data of standard group FP.
- Example 1 It is a graph which shows the example of object FP peak feature-value file.
- Example 1 It is a chart which shows the example of data of object and standard FP type 2.
- Example 1 It is a graph which shows the example of object FP area
- Example 1 It is a graph which shows the example of object FP integrated feature-value file.
- Example 1 It is a graph which shows the reference
- Example 1 It is a chart which shows the example of standard group integration data.
- Example 1 It is a flowchart which shows the detail of the modification of the subroutine 2 applied instead of FIG.
- Example 1 It is a graph which shows the example of calculation of a moving average and a moving inclination.
- Example 1 It is a chart which shows the example of calculation of a moving average and a moving inclination.
- the purpose of making it possible to contribute to improving the accuracy and efficiency of evaluation is to divide the FP consisting of the peaks detected from the chromatography of multi-component substances and their retention times into multiple regions and to exist in each region This is realized by creating an FP region segmentation feature amount from the existence rate or the abundance of the peak.
- Example 1 of the present invention is a multicomponent drug evaluation method, an evaluation program, a pattern FP feature value creation method, a creation program, and a creation device for evaluating a multicomponent substance, for example, a multicomponent drug.
- a multi-component drug is defined as a drug containing a plurality of active chemical ingredients, and includes, but is not limited to, herbal medicines, combinations of herbal medicines, extracts thereof, and herbal medicines.
- the dosage form is not particularly limited.
- liquids, extracts, capsules, granules, pills, suspensions / emulsions, powders, spirits, specified in the 15th revised Japanese Pharmacopoeia Tablets, soaking agents, decoction, tinctures, troches, fragrances, fluid extracts and the like are included.
- Multi-component substances include those other than drugs.
- Kampo medicines are described in the industry unification and voluntary revision of the 148 prescription “Precautions for use” of Kampo medicines for medical use, and the general Kampo prescription (1978).
- the three-dimensional chromatogram data (hereinafter referred to as 3D chromatogram) of the drug to be evaluated is used.
- a target FP from which drug-specific information is extracted is created.
- each peak of the target FP is assigned to the peak correspondence data (hereinafter referred to as a reference group FP) of all the reference FPs created by performing the peak assignment process for all the reference FPs to obtain peak feature values.
- FP type 2 is created from the remaining peaks except for the assigned peaks from the target FP, and the FP type 2 is divided into regions to obtain region divided feature values.
- FIG. 1 is a block diagram of an evaluation apparatus for a multi-component drug
- FIG. 2 is a block diagram showing an evaluation procedure for the multi-component drug
- FIG. 3 is an explanatory diagram of an FP created from 3D chromatography
- the drugs A and (B) are drugs B and (C) is the FP of the drug C.
- the multi-component drug evaluation apparatus 1 includes an FP creation unit 3, a target FP peak attribution unit 5, a target FP peak feature creation unit 7, a target FP type 2 creation unit 9, and a target FP.
- the multi-component drug evaluation device 1 includes a FP feature value creation device that is a pattern.
- the FP creation unit 3 includes a target FP creation unit 29 and a reference FP creation unit 31.
- the target FP peak attribution unit 5 includes a reference FP selection unit 33, a peak pattern creation unit 35, and a peak attribution unit 37.
- the multi-component drug evaluation apparatus 1 is composed of, for example, a single computer and includes a CPU, a ROM, a RAM, and the like (not shown).
- the multi-component medicine evaluation apparatus 1 can obtain an FP feature quantity by realizing an FP feature quantity creation program as a pattern feature quantity creation program installed in a computer.
- the FP feature value creation program uses the FP feature value creation program recording medium in which the FP feature value creation program is recorded, and causes the multi-component drug evaluation apparatus 1 configured by a computer to read the FP feature value creation program, thereby obtaining the FP feature value. It can also be achieved.
- Each component of the evaluation apparatus 1 for a multi-component drug can be configured by a separate computer.
- the target FP region division feature quantity creation unit 11, the target FP feature quantity integration unit 13, and the evaluation unit 27 are configured by one computer, and a reference FP creation unit 31, a reference FP peak attribution unit 15, and a reference FP attribution.
- the result integration unit 17, the reference FP peak feature value creation unit 19, the reference FP type 2 creation unit 21, the reference FP region division feature value creation unit 23, and the reference FP feature value integration unit 25 are configured by other computers. You can also.
- the reference FP integrated feature value is created by another computer and input to the evaluation unit 27 of the evaluation apparatus 1.
- the target FP integrated feature value is created by the quantity integration unit 13, the reference FP creation unit 31, the reference FP peak attribution unit 15, the reference FP attribution result integration unit 17, the reference FP peak feature quantity creation unit 19, and the reference
- a reference FP integrated feature value is generated by the FP type 2 generation unit 21, the reference FP region division feature value generation unit 23, and the reference FP feature value integration unit 25.
- the target FP creation unit 29 of the FP creation unit 3 uses a plurality of peaks at a specific detection wavelength from a 3D chromatogram 41 that is three-dimensional chromatogram data as a chromatograph of Chinese medicine 39.
- This is a functional unit that creates a target FP 43 (hereinafter, simply referred to as “FP43”) from which the retention time and UV spectrum are extracted.
- This FP 43 is composed of three-dimensional information (peak, retention time, and UV spectrum) in the same manner as the 3D chromatography 41.
- the FP 43 is data that inherits the drug-specific information as it is. Nevertheless, since the data volume is compressed to about 1/70, the amount of information to be processed can be greatly reduced and the processing speed can be increased compared to the 3D chromatogram 41.
- 3D chromatography 41 is the result of applying high performance liquid chromatography (HPLC) to Chinese medicine 39.
- HPLC high performance liquid chromatography
- This 3D chromatogram 41 is expressed as a moving speed of each component, which is expressed as a moving distance at a specific time, or in a chart that appears in time series from the column end.
- the detector response with respect to the time axis is plotted, and the appearance time of the peak is called the retention time (retention time).
- the detector is not particularly limited, but an absorbance detector using optical properties (Absorbance Detector) is used, and the peak is obtained three-dimensionally as the signal intensity corresponding to the detection wavelength of ultraviolet rays (UV). Is.
- a transmission detector Transmittance Detector
- UV ultraviolet rays
- the detection wavelength is not limited, and is preferably in the range of 150 nm to 900 nm, particularly preferably in the UV-visible absorption region of 200 nm to 400 nm, and more preferably a plurality of wavelengths selected from 200 nm to 300 nm.
- the 3D chromatograph 41 has at least a Chinese medicine number (lot number), a retention time, a detection wavelength, and a peak as data.
- the 3D chromatogram 41 can also be obtained by a commercially available device, and an example of such a commercially available device is the Agilent 1100 system. Further, the chromatography is not limited to HPLC, and various types can be adopted.
- the 3D chromatogram 41 displays the x-axis as the retention time, the y-axis as the detection wavelength, and the z-axis as the signal intensity as shown in FIGS.
- FP43 has at least a Chinese medicine number (lot number), a retention time, a peak at a specific wavelength, and a UV spectrum as data.
- the FP 43 is displayed in two dimensions with the x-axis as a retention time and the y-axis as a peak at a specific detection wavelength as shown in FIGS. 2 and 3, but similar to the UV spectrum 42 shown as one peak as shown in FIG. This is data having UV spectrum information for each peak.
- the specific detection wavelength for creating the ⁇ ⁇ FP 43 is not particularly limited and can be variously selected. However, it is important to include all the peaks in 3D chromatography in FP43 in that information is inherited. For this reason, in this Example 1, the detection wavelength was set to 203 nm including all peaks in 3D chromatography.
- not all peaks may be included in a single wavelength.
- a plurality of detection wavelengths are used, and an FP including all peaks is created by combining a plurality of wavelengths as will be described later.
- the peak is the maximum value of the signal intensity (peak height), but the area value can also be adopted as the peak. It is also possible to include only two-dimensional information in which the UV spectrum is not included in the FP, the x-axis is the retention time, and the y-axis is the peak at a specific detection wavelength. In this case, FP can also be created from 2D chromatography as a chromatograph having Chinese medicine number (lot number) and retention time as data.
- (A) in FIG. 4 is the drug A
- (B) is the drug B
- (C) is the FP 55, 57, 59 of the drug C.
- the target FP peak attribution unit 5 is a functional unit that compares the peak of the target FP with the reference FP of the multi-component substance corresponding to the target FP and identifies the corresponding peak.
- the target FP peak attribution unit 5 includes a reference FP selection unit 33, a peak pattern creation unit 35, and a peak attribution unit 37.
- the reference FP selection unit 33 is a functional unit that selects a multi-component substance FP suitable for peak assignment of the target FP from a plurality of reference FPs. That is, in order to perform peak attribution of each peak of the target FP with high accuracy, the degree of coincidence of the peak retention, time, and appearance pattern between the target FP and the reference FP is calculated as shown in FIGS. The reference FP that minimizes the degree of coincidence is selected from all the reference FPs.
- the peak pattern creation unit 35 has n peaks existing in at least one before and after the time axis direction with respect to a peak to be attributed in the target FP 61 (hereinafter referred to as an attribute target peak).
- This is a functional unit that creates a peak pattern composed of a total of n + 1 peaks including the peak of the peak as the peak pattern of the attribution target peak.
- n is a natural number.
- FIG. 11 a peak pattern composed of a total of three peaks including two peaks existing at least before and after the time axis direction is illustrated in FIG. 12 (described later).
- FIG. 12 A peak pattern composed of a total of five peaks including four peaks present in at least one of them is shown.
- the peak pattern creation unit 35 sets all the peaks within the range (allowable width) in which the difference from the retention time of the attribution target peak is set in the reference FP 83 (hereinafter, referred to as “below”).
- a function unit that creates a peak pattern consisting of a total of n + 1 peaks including n peaks existing in at least one of before and after the time axis direction as a peak pattern of attribution candidate peaks. is there.
- FIGS. 15 to 18 show a peak pattern composed of a total of three peaks including two peaks existing at least before and after the time axis direction.
- 19 to 22 show a peak pattern composed of a total of five peaks including four peaks existing at least before and after the time axis direction.
- Example 1 There is no limitation on the allowable width, and 0.5 to 2 minutes is preferable from the viewpoint of accuracy and efficiency. In Example 1, it was 1 minute.
- the peak pattern creation unit 35 can flexibly cope with the case where there is a difference in the number of peaks of the target FP 61 and the reference FP 83 (that is, there is a peak that does not exist in either one). Therefore, as shown in FIG. 23 to FIG. 61 (described later), the peaks constituting the peak pattern (hereinafter referred to as peak pattern constituting peaks) are changed comprehensively by both the attribution target peak and the attribution candidate peak. Create a pattern.
- FIGS. 23 to 61 show the case of a peak pattern composed of a total of three peaks including two peaks existing in at least one of the time axis directions.
- the peak attribution unit 37 is a functional unit that compares the peak patterns of the target FP and the reference FP and identifies the corresponding peak. In the embodiment, the degree of coincidence between the peak pattern of the attribution target peak and the peak pattern of the attribution candidate peak and the coincidence degree of the UV spectrum are calculated to identify the corresponding peak.
- the degree of coincidence of the attribution candidate peak obtained by integrating the two coincidence degrees is calculated, and based on the coincidence degree, each peak of the target FP 61 is assigned to each peak of the reference FP 83.
- the degree of coincidence of peak patterns is based on the difference in the corresponding peak and retention time between the peak patterns of the attribution target peak and the attribution candidate peak as shown in FIGS. 62 to 64 (described later).
- the degree of coincidence of the UV spectra is calculated based on the difference in absorbance at each wavelength of the UV spectrum 135 of the attribution target peak 73 and the UV spectrum 139 of the attribution candidate peak 95 as shown in FIGS. 65 and 66 (described later). . Further, as shown in FIG. 67 (described later), the degree of coincidence of attribution candidate peak 95 is calculated by multiplying these two degrees of coincidence.
- the target FP peak feature value creation unit 7 compares and evaluates the peak identified and attributed by the target FP peak attribution unit 5 and the peak of the reference group FP45, which is a plurality of reference FPs, and is converted into a feature value. It is a functional part created as The plurality of reference FPs are created corresponding to a plurality of Chinese medicines that are multi-component substances serving as evaluation criteria, and the plurality of Chinese medicines are regarded as normal products.
- the target FP peak feature quantity creation unit 7 finally assigns each peak of the target FP43 to the reference group FP45 as shown in FIGS. 2, 68, and 69 (described later) based on the attribution result of the target FP61 and the reference FP83.
- This is a functional unit that creates a target FP peak feature value 47 attributed to each peak.
- the target FP type 2 creation unit 9 creates a pattern composed of the remaining peaks by excluding the characteristicized peak from the target pattern as the target pattern type 2.
- the target FP type 2 (49) shown in FIG. 2 is defined as the FP composed of the peak 47 identified by the target FP peak feature quantity creation unit 7 by removing the peak 47 from the original target FP 43 and its retention time. It is a functional part created as a pattern.
- This target FP type 2 (49) is a FP that collects peaks that have not been converted into feature quantities in the target FP peak feature quantity creation unit 7. By making this target FP type 2 (49) a feature quantity and adding it to the evaluation, a more accurate evaluation can be performed.
- the target FP region segmentation feature amount creation unit 11 divides the target pattern type 2 into a plurality of regions, and creates an FP region segmentation feature amount creation unit that creates a target pattern region segmentation feature amount from the presence rate of peaks existing in each region. It is a functional unit configured to divide the target FP type 2 (49) into a plurality of areas and create the target FP area division feature quantity as the target pattern area division feature quantity from the existence rate of peaks existing in each area.
- the target FP region division feature value creation unit 11 can also use the presence amount instead of the presence rate.
- the abundance ratio is a value obtained by dividing the abundance of the peak height in each region by the sum of the whole peak heights (that is, the abundance of the whole peak height). Therefore, it is possible to create a region segmentation feature amount using the existence amount of the peak height of each region itself.
- the target FP region division feature value creation unit 11 converts the target FP type 2 (49) into a plurality of vertical division lines parallel to the signal intensity axis and a plurality of parallel to the time axis as shown in FIG. 70 (described later), for example.
- the target FP area division feature quantity 51 shown in FIG. 2 is created by dividing the area into grid-like areas using the horizontal dividing lines.
- the target FP feature quantity integration unit 13 uses the target FP peak feature quantity 47 created by the target FP peak feature quantity creation unit 7 and the target FP area division feature quantity 51 created by the target FP area division feature quantity creation step 11. It is a functional unit that integrates and creates a target FP integrated feature.
- the reference FP creation unit 31 of the FP creation unit 3 is a functional unit that creates a plurality of reference FPs in the same manner as the target FP creation unit 29.
- a reference obtained by extracting a plurality of peaks at a specific detection wavelength, its retention time, and a UV spectrum from each 3D chromatogram, which is three-dimensional chromatogram data of a plurality of Chinese medicines (reference Chinese medicines) determined to be normal products.
- FP is created for each standard Chinese medicine.
- the reference FP peak attribution unit 15 is also a functional unit that identifies a peak to be attributed by pattern recognition, similar to the target peak attribution unit 5. However, the reference FP peak attribution unit 15 specifies peaks by calculating attribution scores in the selected combination and order for all the reference FPs.
- the reference FP attribution result integration unit 17 is a functional unit that creates a reference peak correspondence table (described later) by integrating the peaks identified and assigned by the reference peak attribution unit 15.
- the reference FP peak feature value creation unit 19 is a functional unit that creates a reference FP peak feature value obtained by characterizing the plurality of reference FPs based on the reference peak correspondence table created by the reference FP attribution result integration unit 17.
- the reference FP type 2 creation unit 21 functions in the same manner as the target FP type 2 creation unit 9, and includes a peak remaining from the plurality of reference FPs excluding the featured peak and its retention time. This is a functional unit that creates an FP as a reference FP type 2 pattern.
- the reference FP area division feature quantity creation unit 23 functions in the same manner as the target FP area division feature quantity creation unit 11 and divides the reference FP type 2 into a plurality of areas as the FP area division feature quantity creation unit. This is a functional unit that creates a reference FP region segmentation feature amount from the peak presence rate.
- the reference FP area division feature value creation unit 23 changes the position of each divided area and creates a reference FP area division feature value before and after the change. That is, the position of each region is changed by changing and setting the position so that each vertical / horizontal dividing line is translated within the set range.
- the reference FP feature value integration unit 25 functions in the same manner as the target FP feature value integration unit 13 and is a functional unit that generates a reference FP integrated feature value by integrating the reference FP peak feature value and the reference FP region division feature value. is there.
- the evaluation unit 27 compares and evaluates the target pattern integrated feature quantity and a reference pattern integrated feature quantity based on a plurality of reference patterns corresponding to the target pattern integrated feature quantity and serving as an evaluation reference. That is, the evaluation unit 27 is a functional unit that compares and evaluates the target FP integrated feature value as the target pattern integrated feature value and the reference FP integrated feature value as the reference pattern integrated feature value. In the embodiment, the equivalence between the target FP integrated feature value and the reference FP integrated feature value is evaluated by the MT method.
- the MT method means a calculation method generally known in quality engineering.
- “Mathematics of Quality Engineering” published by the Japanese Standards Association (2000), pages 136-138, Quality Engineering Application Course “Technology Development of Chemistry, Pharmacy, Biology” (1999), pages 454-456 and Quality Engineering 11 (5), 78-84 (2003), Introductory MT system (2008).
- MT method program software can be used.
- Commercially available MT method program software includes AMTTS of Angle Tri Co., Ltd .; TM-ANOVA of Japan Standards Association; MT Method for Windows of Oken Co., Ltd., and the like.
- the evaluation unit 27 assigns a variable axis in the MT method to the lot number of Chinese herbal medicine and one of the retention time or the UV detection wavelength in the target FP 43, and sets the peak as a feature amount in the MT method.
- variable axis There is no particular limitation on the allocation of the variable axis, but the retention time is allocated to the so-called item axis in the MT method, the number of multi-component drugs is allocated to the so-called number column axis, and the peak is allocated to the so-called feature amount in the MT method Is preferred.
- a reference point and a unit quantity are obtained from the data and feature quantity to which the axis is assigned using the MT method.
- the reference point, the unit amount, and the unit space are defined according to the description of the MT method literature.
- the MD value is obtained as a value representing the degree of difference from the unit space of the drug to be evaluated by the MT method.
- the MD value is defined in the same manner as the description of the MT method document, and the MD value is obtained by a method described in the document.
- the drug to be evaluated can be evaluated by determining the degree of difference from a plurality of drugs determined as normal products.
- the MD value (MD value: 0.26, 2.20, etc.) can be obtained by the above-described MT method by assigning each target FP of FIGS. 87 to 91 as described above.
- the MD value When evaluating this MD value against the MD value of a normal product, the MD value is obtained in the same manner for a plurality of drugs determined as normal products.
- a threshold value is set from the MD value of the normal product, and the MD value of the evaluation target drug can be plotted as shown in the evaluation result 53 of the evaluation unit 27 in FIG. 2 to determine whether the product is a normal product or an abnormal product. .
- an MD value of 10 or less is regarded as a normal product.
- FIGS. 5 to 69 illustrate the operating principles of the reference FP selection unit 33, the peak pattern creation unit 35, the peak attribution unit 37, and the target FP peak feature value creation unit 7.
- FIG. 5 to 69 illustrate the operating principles of the reference FP selection unit 33, the peak pattern creation unit 35, the peak attribution unit 37, and the target FP peak feature value creation unit 7.
- FIG. 5 to 9 are diagrams for explaining the degree of coincidence of retention, time, and appearance pattern between the target FP and the reference FP according to the reference FP selection unit 33.
- FIG. FIG. 5 is a diagram showing the retention time of the target FP and the reference FP
- FIG. 6 is a diagram showing the retention time, appearance pattern of the target FP
- FIG. 7 is a diagram showing the retention time, appearance pattern of the reference FP. It is.
- FIG. 8 is a diagram showing the number of coincidence between retention, time, and appearance distance between the target FP and the reference FP
- FIG. 9 is a diagram showing the degree of coincidence between the retention, time, and appearance pattern between the target FP and the reference FP.
- FIG. 5 shows the retention times of the target FP 61 and the reference FP 83.
- 6 and 7 show the retention times and appearance patterns in which all the retention time intervals are calculated from the retention times of the target FP 61 and the reference FP 83, and the distances are tabulated.
- the number of coincidences of retention / time / appearance distance is calculated from these appearance patterns, and the number of coincidence of retention / time / appearance distance is shown in a tabular form.
- the degree of coincidence of retention / time / appearance pattern is calculated based on the number of coincidence, and the degree of coincidence of retention / time / occurrence pattern is shown in a table format.
- FIG. 10 to FIG. 12 are diagrams for explaining the peak pattern created by the attribution target peak related to the peak pattern creation unit 35 and its surrounding peaks.
- FIG. 10 is a diagram showing peaks to be attributed to the target FP
- FIG. 11 is a peak pattern created with three peaks including two peripheral peaks
- FIG. 12 is five peaks including four peripheral peaks. It is a figure explaining the peak pattern created by.
- FIGS. 13 and 14 describe the relationship between the attribution target peak and the attribution candidate peak according to the peak pattern creation unit 35
- FIG. 13 is a diagram showing the allowable range of the attribution target peak
- FIG. It is a figure which shows the attribution candidate peak of reference
- FIG. 15 to 18 are examples of peak patterns of attribution target peaks and attribution candidate peaks created by three peaks according to the peak pattern creation unit 35.
- FIG. FIG. 15 is a peak pattern diagram of three peaks of attribution target peak and attribution candidate peak
- FIG. 16 is a peak pattern diagram of three attribution candidate peaks and three attribution candidate peaks
- FIG. FIG. 18 is a peak pattern diagram of three peaks of attribution candidate peaks and three attribution candidate peaks different from the attribution target peak.
- FIGS. 19 to 22 are peak pattern diagrams of attribution target peaks and attribution candidate peaks created with five peaks according to the peak pattern creation unit 35.
- 23 to 61 are diagrams illustrating the principle of exhaustive comparison in which the peak patterns of the attribution target peak and the attribution candidate peak according to the peak pattern creation unit 35 are comprehensively created and compared.
- FIG. 62 and 63 are diagrams illustrating a method for calculating the coincidence of peak patterns created with three peaks related to the peak attribution unit 37.
- FIG. 62 and 63 are diagrams illustrating a method for calculating the coincidence of peak patterns created with three peaks related to the peak attribution unit 37.
- FIG. 64 is a diagram for explaining a method for calculating the coincidence degree of peak patterns created with five peaks related to the peak attribution unit 37.
- FIG. 65 is a diagram showing UV spectra 135 and 139 of the assignment target peak 73 and the assignment candidate peak 95 related to the peak assignment portion 37.
- FIG. 66 is a diagram illustrating the degree of coincidence between the UV spectrum 135 of the assignment target peak 73 and the UV spectrum 139 of the assignment candidate peak 95 related to the peak assignment portion 37.
- FIG. 67 is a diagram illustrating the degree of coincidence of attribution candidate peaks calculated from the degree of coincidence of the peak patterns of the attribution target peak 73 and the attribution candidate peak 95 related to the peak attribution unit 37 and the degree of coincidence of the UV spectra.
- FIG. 68 is a diagram for explaining the attribution of each peak to the reference group FP45 in the target FP43 related to the peak attribution unit 37.
- FIG. 68 is a diagram for explaining the attribution of each peak to the reference group FP45 in the target FP43 related to the peak attribution unit 37.
- FIG. 69 is a diagram for explaining the target FP peak feature quantity 47 indicating the situation in which each peak of the target FP 43 related to the peak attribution unit 37 is attributed to the reference group FP45.
- FIG. 5 is a diagram showing the retention time of the target FP and the reference FP
- FIG. 6 is a diagram showing the retention time, appearance pattern of the target FP
- FIG. 7 is a diagram showing the retention time, appearance pattern of the reference FP. It is.
- FIG. 8 is a diagram showing the number of coincidence between retention, time, and appearance distance between the target FP and the reference FP
- FIG. 9 is a diagram showing the degree of coincidence between the retention, time, and appearance pattern between the target FP and the reference FP.
- FIG. 5 shows the retention times of the target FP 61 and the reference FP 83.
- 6 and 7 show the retention times and appearance patterns in which all the retention time intervals are calculated from the retention times of the target FP 61 and the reference FP 83, and the distances are tabulated.
- the number of coincidences of retention / time / appearance distance is calculated from these appearance patterns, and the number of coincidence of retention / time / appearance distance is shown in a tabular form.
- the degree of coincidence of retention / time / appearance pattern is calculated based on the number of coincidence, and the degree of coincidence of retention / time / occurrence pattern is shown in a table format.
- each peak of the target FP 61 is assigned with a reference FP that is as similar as possible to the target FP 61 in the FP pattern. Selecting a reference FP similar to the target FP 61 from a plurality of reference FPs is an important point in making attribution with high accuracy.
- the similarity of the FP pattern is evaluated based on the degree of coincidence of the retention time, the appearance pattern.
- the retention times of the target FP 61 and the reference FP 83 are as shown in FIG. 5
- the retention times and appearance patterns of the target FP 61 and the reference FP 83 are as shown in FIGS.
- the value of each cell is created as a tabular pattern composed of the retention-time distances for the upper target FP 61 and the reference FP 83 as shown in the lower chart.
- the retention times of the peaks (63, 65, 67, 69, 71, 73, 75, 77, 79, 81) of the target FP 61 are (10.2), (10.5), (10 .8), (11.1), (11.6), (12.1), (12.8), (13.1), (13.6), and (14.0).
- the retention times of the peaks (85, 87, 89, 91, 93, 95, 97, 99, 101, 103, 105) of the reference FP83 are (10.1), (10.4), (10.7), (11.1), (11.7), (12.3), (12.7), (13.1), (13.6), (14.1), (14 4).
- the distance between the retention times is the reference FP appearance pattern in the lower chart of FIG.
- the number of matches is seven.
- the seven coincidence numbers are written in the first line of the target and reference FP retention / time / appearance pattern of FIG.
- the target FP retention / time / appearance pattern 1 to 9 and the reference FP retention / time / appearance pattern 1 to 10 are displayed.
- a brute force comparison is made and the number of matches is obtained.
- the numerical value 7 at the left end circled is the result of comparing the first line of the target and the reference FP retention / time / appearance pattern, and the numerical value 7 next to the numerical value is the target FP retention / This is a result of comparing the first line of the time / appearance pattern with the second line of the reference FP retention / time / appearance pattern.
- the setting range There is no limitation on the setting range, and the range is preferably 0.05 minutes to 0.2 minutes. In Example 1, it was 0.1 minutes.
- the degree of coincidence between the retention time is RP
- a is the number of peaks of the target FP 61 (number of target FP peaks)
- b is the number of peaks of the reference FP 83 (number of reference FP peaks)
- m is the number of matches of the retention time, appearance pattern (number of matches of the appearance distance).
- RP_min which is the minimum value of these RPs, is used as the degree of coincidence between the retention time, the appearance pattern of the target FP 61 and the reference FP 83.
- (0.50) is the degree of coincidence of the target FP 61 with respect to the reference FP.
- the degree of coincidence is calculated for all the reference FPs, the reference FP having the smallest degree of coincidence is selected, and the peak assignment of the target FP is performed for the reference FP.
- the reference FP selection unit 5 can also pattern the target FP 61 and the reference FP 83 with a peak height ratio.
- the Tanimoto coefficient is defined as “the number of coincidence of height ratio / (number of target FP peaks + number of reference FP peaks ⁇ number of coincidence of height ratio)”, and the (1-Tanimoto coefficient) is close to zero. The degree of coincidence can be obtained.
- (1-Tanimoto coefficient) is weighted by (number of target FP peaks ⁇ number of occurrence patterns or height ratio coincidence + 1), and “(1-Tanimoto coefficient) ⁇ (number of target FP peaks ⁇ appearance pattern or high It is possible to select a reference FP that matches more peaks (63, 65,%) Of the target FP 61 by weighting.
- an allowable range of retention time deviation is set between each peak of the attribution target peak 73 and the reference FP 83, and the peak of the reference FP 83 existing in the allowable range (hereinafter, attribution candidate).
- the assignment destination is determined by combining all the information, so the accuracy is improved compared to the peak attribution based on the single information.
- the assignment candidate peak includes a plurality of similar components. After all, it becomes attribution only by peak information, and sufficient accuracy cannot be obtained. Therefore, in order to perform peak assignment with higher accuracy, information in addition to these three pieces of information is necessary.
- a peak pattern including peripheral peak information as shown in FIG. 11 and FIG. 12 was created, and a peak was assigned by comparing the peak patterns.
- the surrounding information is added to the previous three pieces of information, and peak attribution based on the four pieces of information becomes possible, and higher attribution accuracy is obtained.
- the constraint conditions (peak definition, etc.) to be set for existing peak attribution are no longer necessary.
- a peak pattern 115 including peaks 71 and 75 existing in both time axis directions is created for the attribution target peak 73.
- a peak pattern 125 including peaks 69, 71, 75, and 77 that exist in both time axis directions is created for the attribution target peak 73.
- an allowable range of retention time deviation is set between each peak of the attribution target peak 73 and the reference FP 83, and the peak of the reference FP 83 existing within the tolerance range corresponds to the attribution target peak 73.
- Candidate peaks hereinafter referred to as attribution candidate peaks) were used.
- Peak patterns 119, 121, and 123 were created.
- the peak patterns to be compared with the peak pattern 125 of the attribution target peak 73 include peaks that exist both before and after the time axis direction with respect to different attribution candidate peaks 95, 97, and 99. Peak patterns 129, 131, and 133 were created.
- a peak that is a candidate for a peak / pattern configuration peak (hereinafter referred to as a peak / pattern configuration candidate peak) is set in advance from the peripheral peaks of the target peak of the target FP, and the peak / pattern configuration candidate peak In order, a peak pattern is created as a peak constituting a peak pattern.
- a peak pattern configuration candidate peak is set for the attribution candidate peak of the reference FP, and a peak pattern is created using the peak pattern configuration candidate peak in order as a peak pattern configuration peak.
- the degree of coincidence of the peak pattern based on the difference in the corresponding peak and retention time between all the peak patterns of the attribution target peak and the attribution candidate peak created by the peak pattern creation unit 35 ( Hereinafter, P_Sim) is calculated.
- the peak attribution unit 37 uses the minimum value of P_Sim (hereinafter, P_Sim_min) as the degree of coincidence between the peak pattern of the attribution target peak and the attribution candidate peak.
- each of the attribution target peak 73 and the attribution candidate peak 93 has four peak pattern configuration candidate peaks around the time axis direction, and the peak pattern configuration peak is set to any two. To do.
- P_Sim_min which is the minimum value of these P_Sim is determined as the degree of coincidence between the attribution target peak 73 and the attribution candidate peak 93.
- This P_Sim is similarly calculated for all attribution candidate peaks of the attribution target peak 73.
- the peak pattern 115 of the attribution target peak 73 and the peak pattern 119 of the attribution candidate peak 95 are taken as examples.
- the peak and retention time of the attribution target peak 73 are p1 and r1
- the peak and retention time of the peak pattern constituting peak 71 are dn1 and cn1
- the peak pattern constituting peak 75 is Let dn2 and cn2 be peak and retention times.
- the peak and retention time of the attribution candidate peak 95 are p2 and r2
- the peak and retention time of the peak pattern constituting peak 93 are fn1 and en1
- the peak pattern constituting peak 97 is Let the peak and retention times be fn2 and en2.
- the degree of coincidence of the peak pattern is P_Sim
- the degree of coincidence of the peak pattern (P_Sim (73-95)) composed of the three peaks of the attribution target peak 73 and the attribution candidate peak 95 is P_Sim (73-95) (
- d in the equation is a value for correcting a retention time shift.
- FIG. 64 explains a method for calculating the coincidence of peak patterns for comparing peak patterns composed of five peaks.
- the peak pattern 125 of the attribution target peak 73 and the peak pattern 129 of the attribution candidate peak 95 are taken as examples.
- the peaks and retention times of the attribution target peak 73 are p1 and r1
- the peaks and retention times of the peak pattern constituting peaks 69, 71, 75, and 77 are dn1 and cn1, respectively.
- the peaks and retention times of the attribution candidate peaks 95 are p2 and r2, the peaks and retention times of the peak pattern constituting peaks 91, 93, 97, and 99 are fn1 and en1, respectively.
- P_Sim (73-95) (
- d in the equation is a value for correcting a retention time shift.
- the peak attribution unit 37 calculates the degree of coincidence of the UV spectrum between the attribution target peak and the attribution candidate peak as shown in FIGS.
- N is the number of diss.
- the waveform of the UV spectrum includes the maximum wavelength and the minimum wavelength, and the degree of coincidence can be calculated by comparing the maximum wavelength and / or the minimum wavelength.
- the maximum wavelength and the minimum wavelength are the same for compounds that have no absorption characteristics or similar absorption characteristics, but the overall waveform may be quite different. In comparison of the maximum and minimum wavelengths, the waveforms match. The degree may not be calculated.
- the degree of coincidence of this UV spectrum is calculated in the same manner for all attribution candidate peaks of the attribution target peak 73.
- the peak attribution unit 37 calculates the degree of coincidence of the attribution candidate peak obtained by integrating the above two coincidences as shown in FIG.
- the degree of coincidence of this attribution candidate peak is calculated in the same manner for all attribution candidate peaks of the attribution target peak 73.
- this SCORE is compared among all attribution candidate peaks, and the attribution candidate peak having the smallest SCORE is determined as the attribution peak of the attribution target peak 73.
- the peak attribution unit 37 determines the peaks to be attributed to the attribution target peak from the two viewpoints, accurate peak attribution can be realized.
- the target peak feature quantity creation unit 7 assigns each peak of the target FP 43 to the reference group FP 45 as shown in FIG. 68 based on the result of the assignment of the target FP to the reference FP.
- Each peak of the target FP 43 is attributed to the reference FP constituting the reference group FP 45 by the attribution process. Based on this attribution result, it finally belongs to the peak of the reference group FP45.
- the reference group FP45 is created by assigning all the reference FPs defined as normal products as described above, and each peak is the average value (black dot) ⁇ standard deviation (vertical length) of the assigned peak. (Dividing line).
- FIG. 69 shows the result of assigning the target FP43 to the reference group FP45, and this result is the target FP peak feature amount 47 of the target FP43.
- FIG. 70 to FIG. 86 show the operation principle of FP region division feature value creation
- FIG. 70 is an explanatory diagram showing quantification by region division
- FIG. 71 is an explanatory diagram showing the relationship with fluctuations in retention time
- etc. 72 is an explanatory diagram for quantifying by changing the position of the area
- FIG. 73 is a chart showing FP type 2 data
- FIG. 74 is an explanatory diagram showing an FP type 2 pattern
- FIG. 76 is an explanatory diagram showing the setting of the vertical dividing line (first line), and FIG. 77 is the setting of the horizontal dividing line (first line).
- 78 is an explanatory diagram showing area division by vertical and horizontal dividing lines
- FIG. 79 is an explanatory diagram showing the number of areas to be featured
- FIG. 80 is an explanatory diagram showing the identification of the area 1.
- FIG. 81 is a chart showing the height and total of all peaks
- FIG. 82 is the peak height of region 1
- FIG. 83 is an explanatory diagram showing the total
- FIG. 83 is a chart showing the feature amounts of all the areas by the first one pattern
- FIG. 84 is a chart showing the feature amounts in each area formed by sequentially changing the position of the first vertical line
- FIG. 85 is a chart showing the feature values in each area obtained by sequentially changing the position of the first horizontal line
- FIG. 86 is a chart showing one feature quantity without changing the position of each vertical / horizontal dividing line.
- the target FP region segmentation feature amount creation unit 11 or the reference FP region segmentation feature amount creation unit 23 calculates the target FP from the presence rate of the peaks existing in each region obtained by dividing the target FP type 2 or the reference FP type 2 as described above. A region division feature amount or a reference FP region division feature amount is created.
- the area is divided as shown in FIG. 70, for example.
- the FP 55 of the medicine A is divided.
- the plurality of horizontal dividing lines 143 are set at equal ratio intervals in the direction in which the signal intensity increases. By this setting, it is possible to subdivide the region division in the dense peak portion and to grasp the peak existence rate more accurately. However, it is also possible to set at equal intervals by increasing the number of the plurality of horizontal dividing lines 143.
- the retention time and peak height fluctuate like FP55A and 55B due to slight variations in analysis conditions. Due to this variation, there is a possibility that the value in each lattice 145 varies greatly.
- FIG. 73 shows reference FP type 2 data d202, d207, d208 as an example.
- This data has only the information of retention time (RT) and peak height (Height).
- This data corresponds to the reference FP type 2 composed of the peaks remaining from the plurality of reference FPs except for the characteristicized peaks and their retention times in the reference FP type 2 creation unit 21, Each UV spectrum of the peak is excluded.
- the patterns of the reference FP type 2 data d202, d207, d208 are as shown in FIG.
- These FP patterns are divided into regions by vertical and horizontal dividing lines 141 and 143, and feature values are made for each region.
- the first vertical position is set at multiple locations under the following conditions.
- the first horizontal position is set at multiple locations under the following conditions.
- X 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4
- the second and subsequent dividing lines are sequentially set by combining all of these 100 ways, and the region is divided.
- the second and subsequent vertical dividing lines are set until the number of lines specified at the specified interval (equal difference) is reached.
- the first vertical line is 0.0
- the vertical interval 10
- the first horizontal dividing line is 7
- the first horizontal dividing line is 0.5
- the horizontal interval 1
- the horizontal line number is 6.
- Vertical dividing line 0, 10, 20, 30, 40, 50, 60
- Horizontal dividing line 0.5, 1.5, 3.5, 7.5, 15.5, 31.5 Set to
- FIG. 78 shows the set vertical and horizontal dividing lines on the FP based on the previous example.
- ⁇ ⁇ FP is featured for each area surrounded by the vertical and horizontal lines.
- Feature amount total peak height in region / total peak height (method of feature amount)
- FIG. 83 shows the calculation result. (Change the first vertical dividing line in order to make it feature quantity) Each region formed by sequentially changing the position of the first vertical dividing line is converted into a feature amount by the above method. The results are shown in FIG.
- FIG. 85 shows the result.
- FIG. 92 is a process diagram showing a method for evaluating a multi-component drug of Example 1 of the present invention, including the FP feature value creation method of Example 1 of the present invention.
- the multi-component drug evaluation method includes an FP creation step 148, a target FP peak attribution step 149, a target FP peak feature creation step 151, a target FP type 2 creation step 153, and a target FP region.
- the FP creation process 148 includes a target FP creation process 173 and a reference FP creation process 175.
- the target FP peak attribution step 149 includes a reference FP selection step 177, a peak pattern creation step 179, and a peak attribution step 181.
- FP creation step 148 the target FP peak attribution step 149, the target FP peak feature creation step 151, the target FP type 2 creation step 153, the target FP region division feature creation step 155, the target FP feature integration step 157, the reference FP peak attribution step 159, reference FP attribution result integration step 161, reference FP peak feature creation step 163, reference FP type 2 creation step 165, reference FP region segmentation feature creation step 167, reference FP feature integration step 169, evaluation Step 171 is performed using the multi-component drug evaluation apparatus 1 in this embodiment.
- the FP creation step 148 is performed by the function of the FP creation unit 3 in FIG. 1. Similarly, the target FP peak attribution step 149, the target FP peak feature creation step 151, the target FP type 2 creation step 153, and the target FP region division feature Amount creation step 155, target FP feature amount integration step 157, reference FP peak attribution step 159, reference FP attribution result integration step 161, reference FP peak feature amount creation step 163, reference FP type 2 creation step 165, reference FP region division feature
- the amount creation step 167, the reference FP feature amount integration step 169, and the evaluation step 171 are performed using the target FP peak attribution unit 5, the target FP peak feature amount creation unit 7, the target FP type 2 creation unit 9, and the target FP region division feature amount creation unit.
- Target FP feature amount integration unit 13 reference FP peak attribution unit 15, reference FP attribution result integration unit 17, reference FP peak feature quantity production Part 19, the reference FP type 2 creating unit 21, the reference FP region division feature quantity preparation unit 23, the reference FP feature value integrating unit 25, to perform the respective functions of the evaluation unit 27.
- each process can be made to function by a separate computer, for example, a target FP creation process 173, a target FP peak attribution process 149, a target FP peak feature value creation process 151, and a target FP type 2 creation process 153.
- the target FP region segmentation feature creation step 155, the target FP feature integration step 157, and the evaluation step 171 are functioned by one computer, and the reference FP creation step 175, the reference FP peak attribution step 159, the reference The FP attribution result integration step 161, the reference FP peak feature creation step 163, the reference FP type 2 creation step 165, the reference FP region division feature creation step 167, and the reference FP feature integration step 169 are performed by another computer. You can also make it work with.
- the reference FP integrated feature value is created by another computer and supplied to the evaluation step 171.
- the target FP type 2 creation step 153 creates the target FP type 2 as a pattern in which the peak changes in time series.
- the target FP region division feature value creation step 155 is a pattern region division feature value creation step of creating the target pattern region division feature value from the presence rate of peaks existing in each region by dividing the target pattern type 2 into a plurality of regions.
- the FP region division feature value creation step is configured.
- FIGS. 93 to 108 are flowcharts relating to the evaluation program for multi-component drugs
- FIGS. 109 to 116 are flowcharts relating to the creation of reference data
- FIG. 117 is a chart showing an example of 3D chromatogram data
- FIG. 119 is a chart showing an example of FP data
- FIG. 120 is a chart showing an example of an attribution score calculation result (determination result file) to the reference FP of the target FP
- FIG. FIG. 122 shows examples of two intermediate files (attribute candidate peak score table and attribute candidate peak number table) created in the matching process of peaks corresponding to FP and reference FP.
- FIG. 122 shows peaks corresponding to target FP and reference FP.
- FIG. 123 is a chart showing an example of the collation result file that is the identified result
- FIG. 123 is a chart showing an example of data of the reference group FP
- FIG. 124 is an object F belonging to the reference group FP.
- FIG. 125 is a chart showing an example of the target and reference FP type 2 data
- FIG. 126 is a chart showing an example of the target FP region segmentation feature quantity file
- FIG. 127 is the target.
- FIG. 128 is a chart showing an example of an FP feature quantity integration file
- FIG. 128 is a chart showing an example of a reference type 2 group FP
- FIG. 129 is a chart showing an example of reference group integration data.
- FIG. 93 and FIG. 94 are flowcharts showing the steps of the entire process for evaluating the evaluation target drug.
- the processing starts when the system is started, and the FP creation function of the FP creation unit 3 and the target FP of the target FP peak attribution unit 5 Peak attribution function, target FP peak feature value creation function of the target FP peak feature value creation unit 7, target FP type 2 creation function of the target FP type 2 creation unit 9, and target of the target FP region division feature value creation unit 11 FP region segmentation feature creation function, target FP feature integration function of the target FP feature integration unit 13, reference FP peak attribution function of the reference FP peak attribution unit 15, and reference FP attribution of the reference FP attribution result integration unit 17 A result integration function, a reference FP peak feature creation function of the reference FP peak feature creation unit 19, and a reference FP type 2 creation function of the reference FP type 2 creation unit 21 A reference FP region division feature quantity generation function of the reference FP region division feature quantity preparation unit 23, and the reference FP feature value
- FP creation function is realized in step S1.
- the target FP peak attribution function is realized in steps S2, S3, and S4.
- the target FP peak feature value creation function is realized in step S5.
- the target FP type 2 creation function is realized in step S6.
- the target FP region division feature value creation function is realized in step S7.
- the target FP feature value integration function is realized in step S8.
- the evaluation function is realized in steps S9 and S10.
- step S1 “FP creation processing” is executed using 3D chromatography and peak information at a specific detection wavelength as input data.
- the 3D chromatogram is data obtained by analyzing the drug to be evaluated by HPLC.
- the data example 183 of the 3D chromatogram in FIG. 117 the three-dimensional information of the retention time, the detection wavelength, and the peak (signal intensity).
- the peak information is data obtained by processing chromatographic data at a specific wavelength obtained by the same HPLC analysis with an HPLC data analysis tool (for example, ChemStation etc.).
- an HPLC data analysis tool for example, ChemStation etc.
- This is data composed of the maximum and area values of all peaks detected as peaks and the retention time at that time.
- step S1 the target FP creation unit 29 (FIG. 1) of the FP creation unit 3 of the computer functions to create the target FP 43 (FIG. 2) from 3D chromatography and peak information, and output the data as a file.
- the target FP 43 is data composed of a retention time, a peak height, and a UV spectrum for each peak height, as shown in the FP data example 187 in FIG.
- step S2 “target FP attribution process 1” is executed with the target FP and all reference FPs output in step S1 as inputs.
- step S2 the reference FP selection unit 33 of the computer functions to calculate the retention / time / appearance pattern coincidence with the target FP 43 for all the reference FPs, and select a reference FP suitable for the attribution of the target FP 43. .
- the reference FP is an FP created by the same processing as in Step S1 from 3D chromatography and peak information of a drug determined as a normal product.
- a normal product is defined as a drug that has been confirmed to be safe and effective (standard Chinese medicine), and includes a plurality of drugs with different product lots.
- the heel reference FP is data configured in the same manner as the FP data example 187 in FIG.
- step S3 the target FP 43 and the reference FP selected in step S2 are input, and the “target FP attribution process 2” is executed.
- step S3 the peak pattern creation unit 35 (FIG. 1) and the peak attribution unit 37 (FIG. 1) of the computer function.
- peak patterns are comprehensively created as shown in FIGS. 23 to 61 for all peaks of the target FP 43 and the reference FP selected in step S2, and then the degree of coincidence of these peak patterns (FIG. 63 or FIG. P_Sim) in FIG. 64 is calculated. Further, the degree of coincidence of UV spectra (UV_Sim in FIG. 66) between the peaks of the target FP and the reference FP is calculated. Further, the degree of coincidence of attribution candidate peaks (SCORE in FIG. 67) is calculated from these two degrees of coincidence. The calculation result is output to a file similar to the determination result file example 189 of FIG.
- step S4 the determination result file 189 output in step S3 is input, and the “target FP attribution process 3” is executed.
- step S4 the peak attribution unit 37 of the computer functions to identify the peak of the reference FP corresponding to each peak of the target FP based on the degree of coincidence (SCORE) of the attribution candidate peak between the target FP 43 and the reference FP. .
- the result is output to a collation result file similar to the collation result file example 195 of FIG.
- step S5 the verification result file output in step S4 and the reference group FP197 are input, and the “target FP attribution process 4” is executed.
- the reference group FP 197 is peak correspondence data between all the reference FPs created from the all reference FPs by the same processing as in the steps S2 to S4.
- step S5 the target FP peak feature amount generating unit 7 of the computer functions, and based on the matching result file of the target FP 43, each peak of the target FP 43 is changed to a peak of the reference group FP 197 as shown in FIGS. Belong.
- the result is output to a file similar to the peak data feature amount file example 199 in FIG.
- step S6 the peak data feature file output in step S5 and the target FP are input, and the process of “Create FP_type 2” is executed.
- step S6 the target FP type 2 creation unit 9 of the computer functions, and is composed of the remaining peaks after removing the peak 47 specified by the target FP peak feature creation unit 7 from the original target FP 43 and its retention time.
- the created FP is created as the target FP type 2 (49).
- the result is output to an FP type 2 file (see FP type 2 file example 201 in FIG. 125).
- step S7 “feature amount processing of target FP_type2 by area division” is executed.
- the target FP area division feature value creation unit 11 of the computer functions, and the target FP area division feature value is created by the area division of FIG.
- the result is output to a target FP region division feature value file (see target FP region division feature value file example 203 in FIG. 126).
- step S8 processing of “integration of peak data feature quantity and region division feature quantity” is executed.
- the target FP feature value integration unit 13 of the computer functions, and the target FP peak feature value 47 created by the target FP peak feature value creation unit 7 and the target created by the target FP region division feature value creation unit 11.
- the target FP integrated feature value is created by integrating the FP region division feature value 51.
- the result is output to a target FP feature value integration file (see target FP feature value integration file example 205 in FIG. 127).
- step S9 the computer evaluation unit 27 functions to evaluate the equivalence between the target FP integrated feature value output in step S8 and the reference FP integrated feature value by the MT method, and the evaluation results are shown in FIGS.
- Such an MD value is output (FIGS. 87 to 91).
- step S10 “pass / fail judgment” is executed with the MD value output in step S9 as an input.
- step S10 the evaluation unit 27 of the computer functions to compare the MD value output in step S9 with a preset threshold value (upper limit value of MD value) to determine pass / fail (evaluation result 53 in FIG. 2).
- a preset threshold value upper limit value of MD value
- FIG. 95 is a flowchart when the peak information of a single wavelength in step S1 “FP creation processing” in FIG. 93 is used.
- FIG. 95 shows details of steps for creating an evaluation target FP with a single wavelength, for example, 203 nm.
- an FP composed of a retention time and a peak at a peak detected at 203 nm and a UV spectrum of the peak is created from 3D chromatography and peak information at a detection wavelength of 203 nm.
- step S101 a process of “reading peak information” is executed.
- the peak information is read as the first of the two data necessary for creating the FP, and the process proceeds to step S102.
- step S102 a process of “obtaining peak data (P1) corresponding to the peak retention time (R1) in order” is executed.
- the peak retention time (R1) and peak data (P1) are sequentially acquired from the peak information one by one, and the process proceeds to step S103.
- step S103 the process of “read 3D chromatogram” is executed.
- the 3D chromatogram is read as the second of the two data necessary for creating the FP, and the process proceeds to step S104.
- step S104 processing of “acquire UV spectrum (U1) corresponding to peak retention time (R2) in order” is executed.
- the retention time (R2) and UV spectrum (U1) are acquired from the 3D chromatograph for each sampling rate during HPLC analysis, and the process proceeds to step S105.
- step S105 a determination process of “
- ⁇ threshold?” Is executed.
- R1 and R2 read in steps S102 and S104 correspond to each other within a threshold range. If it corresponds (YES), it is determined that the two retention times are the same, the UV spectrum of the peak with the retention time R1 is U1, and the process proceeds to step S106. If it does not correspond (NO), the two retention times are not the same, and the UV spectrum of the peak with the retention time R1 is not U1, and for comparison with the next data of 3D chromatography, The process proceeds to step S104.
- the threshold value in this determination process is “sampling rate / 2” in 3D chromatography.
- step S106 the process of “normalize U1 to the maximum value 1” is executed.
- U1 determined as the UV spectrum of R1 in S105 is normalized with the maximum value 1, and the process proceeds to step S107.
- step S107 a process of “output R1 and P1 and standardized U1 (target FP)” is executed.
- R1 and P1 acquired from the peak information and U1 normalized in S106 are output to the target FP, and the process proceeds to step S108.
- step S108 a determination process of “End of processing for all peaks?” Is executed. In this process, it is determined whether or not the processing has been performed for all the peaks in the peak information. If the processing has not been completed for all the peaks (NO), in order to process the unprocessed peaks, The process proceeds to step S102. The processing from S102 to S108 is repeated until the processing of all peaks is completed. When the processing of all peaks is completed (YES), the FP creation processing is terminated. [S1: FP creation processing (using multiple wavelengths)] FIGS. 96 and 97 are flowcharts when step S1 “FP creation processing” in FIG. 93 uses peak information of a plurality of wavelengths instead of the peak information of the single wavelength. For example, in this case, a plurality of (n) wavelengths are selected in the detection wavelength axis direction including 203 nm to create an FP.
- 96 and 97 show details of a step of creating FPs having a plurality of wavelengths from n FPs after creating n FPs for each wavelength by the FP creation process using only the single wavelength.
- step S110 the process of “Create FP for each wavelength” is executed.
- an FP creation process using only the single wavelength is performed for each wavelength, n FPs are created, and the process proceeds to step S111.
- step S111 a process of “listing FP in peak number (descending order)” is executed.
- n FPs are listed in descending order of the number of peaks, and the process proceeds to step S112.
- step S112 1 is substituted for n (n ⁇ 1) as initialization of a counter for sequentially processing n FPs, and the process proceeds to step S113.
- step S113 a process of “reading the nth FP in the list” is executed. In this process, the nth FP in the list is read, and the process proceeds to step S114.
- step S114 the process of “obtain all retention times (X)” is executed. In this process, all the retention time information of the FP read in S113 is acquired, and the process proceeds to step S115.
- step S115 the process of “n update (n ⁇ n + 1)” is executed.
- n + 1 is substituted for n as an update of n, and the process proceeds to step S116.
- step S116 a process of “reading the nth FP in the list” is executed. In this process, the nth FP in the list is read, and the process proceeds to step S117.
- step S117 the process of “obtain all retention times (Y)” is executed.
- all the retention time information of the FP read in S116 is acquired, and the process proceeds to step S118.
- step S118 the process of “Integrate X and Y without duplication (Z)” is executed.
- the retention time information X acquired in S114 and the retention time information Y acquired in S117 are integrated without duplication, then stored in Z, and the process proceeds to step S119.
- step S119 the process of “X update (X ⁇ Z)” is executed.
- Z stored in S118 is substituted for X as X update, and the process proceeds to step S120.
- step S120 a determination process of “End of all FP processes?” Is executed. In this process, it is determined whether or not all n FPs created in S110 have been processed. If the process has been completed (YES), the process proceeds to step S121. If there is an unprocessed FP (NO), the process proceeds to S115 in order to execute the processes of S115 to S120 for the unprocessed FP. The processes of S115 to S120 are repeated until all the FP processes are completed.
- step S121 1 is substituted for n (n ⁇ 1) as initialization of a counter for sequentially processing n FPs again, and the process proceeds to step S122.
- step S122 a process of “reading the nth FP in the list” is executed. In this process, the nth FP in the list is read, and the process proceeds to step S123.
- step S123 a process of “obtaining each peak's retention time (R1), peak data (P1), and UV spectrum (U1) in order” is executed.
- the retention time (R1), peak data (P1), and UV spectrum (U1) are obtained one by one from the FP read in S122, and the process proceeds to step S124.
- step S124 a process of “obtaining retention times (R2) in order from X” is executed.
- the retention times of all the FPs are acquired sequentially from X stored without duplication one by one retention time (R2), and the process proceeds to step S125.
- step S126 a determination process of “completion of all X retention times?” Is executed. In this process, it is determined whether or not the comparison between R1 acquired in S123 and the total retention time of X has been completed. If completed (YES), it is determined that the peak with the retention time R1 has been processed, and the process proceeds to step S123 to shift the process to the next peak. If not finished (NO), the next retention of X. In order to shift to time, the process proceeds to step S124.
- step S127 the process of “add (n ⁇ 1) ⁇ analysis time (T) to R1 (R1 ⁇ R1 + (n ⁇ 1) ⁇ T)” is executed.
- the retention time of the peak existing in the first FP of the list having the largest number of peaks remains as it is, the retention time of the peak existing in the second FP of the list does not exist in the first FP of the list.
- the analysis time (T) is added to the time R1, and the retention time of the peak existing in the nth FP of the list does not exist in the list 1 to the (n-1) th FP. ) ⁇ T is added, and the process proceeds to step S128.
- step S128 the process of “output R1, P1, and U1 (target FP)” is executed.
- R1 processed in S127, P1 and U1 acquired in S123 are output to the target FP, and the process proceeds to step S129.
- step S129 the process “Delete R2 from X” is executed.
- the process in which the retention time is R1 ( R2) ends in S127 and S128, the processed retention time (R2) is deleted from X, and the process proceeds to S130.
- step S130 a determination process of “End of all peak processing?” Is executed. In this process, it is determined whether or not the process has been completed for all the peaks of the list nth FP. If the process has been completed (YES), the FP creation process in the list nth FP is terminated, The process proceeds to S131. If there is an unprocessed peak (NO), the process proceeds to step S123 in order to process the unprocessed peak. The processes of S123 to S130 are repeated until all the peaks are processed.
- step S131 the process of “n update (n ⁇ n + 1)” is executed.
- n + 1 is substituted for n as an update of n, and the process proceeds to step S132.
- step S132 a determination process of “End of all FP processes?” Is executed. In this process, it is determined whether all n FPs created in S110 have been processed. If the process has been completed (YES), the FP creation process is terminated. If there is an unprocessed FP (NO), the process proceeds to S122 in order to execute the processes of S122 to S132 for the unprocessed FP. The processes in S122 to S132 are repeated until all FP processes are completed.
- S2 Target FP attribution process 1
- FIG. 98 is a flowchart showing details of the “target FP attribution process 1” in step S2 of FIG. This process is a pre-process for attribution, and a reference FP suitable for attribution of the target FP 43 is selected from a plurality of reference FPs that are regarded as normal products.
- step S201 the process of “read target FP” is executed.
- the attribution target FP is read, and the process proceeds to step S202.
- step S202 the process of “obtain all retention time (R1)” is executed.
- R1 the retention time information of the target FP read in S201 is acquired, and the process proceeds to step S203.
- step S203 a process of “listing file names of all reference FPs” is executed.
- the file names of all the reference FPs are listed in advance, and the process proceeds to step S204.
- step S204 1 is substituted for n (n ⁇ 1) as the initial value of the counter for sequentially processing all the reference FPs, and the process proceeds to step S205.
- step S205 a process of “reading the list n-th reference FP (reference FP n )” is executed.
- the n-th FP of the file name list of all reference FPs listed in S203 is read, and the process proceeds to step S206.
- step S206 the process of “obtain all retention time (R2)” is executed.
- R2 the retention time information of the reference FP read in S205 is acquired, and the process proceeds to step S207.
- step S207 the process of “calculate the degree of coincidence of retention, time, and appearance pattern of R1 and R2 (RP n — min)” is executed.
- RP n — min is calculated from the retention time of the target FP acquired in S202 and the retention time of the reference FP acquired in S206, and the process proceeds to step S208.
- the detailed calculation flow of RP n — min will be separately described with reference to subroutine 1 in FIG.
- step S208 the processing of "Saving RP n _min (RP all _min" is performed.
- the RP n _min calculated in S207 and stored in the RP all _min the process proceeds to step S209.
- step S209 the process of “update n (n ⁇ n + 1)” is executed.
- n + 1 is substituted for n as an update of n, and the process proceeds to step S210.
- step S210 a determination process of “End all reference FP processes?” Is executed. In this process, it is determined whether or not all the reference FPs have been processed. If the process has been completed (YES), the process proceeds to step S211. If there is an unprocessed reference FP (NO), the process proceeds to S205 to execute the processes of S205 to S210 for the unprocessed FP. The processes of S205 to S210 are repeated until the process of all the reference FPs is completed.
- step S211 the processing of "selection criteria FP matching degree from RP all _min is minimized" is executed.
- RP n — min is compared with RP 1 — min calculated for all the reference FPs, a reference FP that has the lowest degree of coincidence of retention, time, and appearance pattern with the target FP is selected, and target FP attribution processing 1 is finished.
- FIG. 99 is a flowchart showing details of the “target FP attribution process 2” in step S3 of FIG. This process is the main process of attribution, and the degree of coincidence (SCORE) of each attribution candidate peak between the target FP 43 and the reference FP selected in step S2 based on the coincidence of the peak pattern and the UV spectrum as described above. Is calculated.
- S3 Target FP attribution process 2
- step S301 the process of “read target FP” is executed.
- the attribution target FP is read, and the process proceeds to step S302.
- step S302 a process of “obtaining the retention time (R1), peak data (P1), and UV spectrum (U1) of the assignment target peak in order)” is executed.
- each peak of the target FP read in S301 is set as the attribute target peak in order, R1, P1, and U1 are acquired, and the process proceeds to step S303.
- step S303 the process of “reading the reference FP” is executed.
- the reference FP selected in [Target FP attribution process 1] in FIG. 98 is read, and the process proceeds to step S304.
- step S304 a process of “obtaining the reference FP peak retention time (R2), peak data (P2), and UV spectrum (U2) in order” is executed.
- R2, P2, and U2 are acquired one by one from the reference FP read in S303, and the process proceeds to step S305.
- step S305 a determination process of “
- R1 and R2 read in steps S302 and S304 correspond within the threshold range. If it corresponds (YES), it is determined that the peak with the retention time R2 is the attribution candidate peak of the peak with the retention time R1, and step S306 is performed to calculate the degree of coincidence (SCORE) of the attribution candidate peak.
- S309 a determination process of “
- d in this determination process is a value for correcting the retention time of the peak of the target FP and the reference FP, the initial value is 0, and as the process proceeds, the difference in retention time between the peaks attributed as needed is calculated. Obtain d and update d with that value.
- the threshold value is an allowable range of retention time for determining whether or not to be an attribution candidate peak.
- step S306 a process of “calculate UV spectrum matching degree (UV_Sim)” is executed.
- UV_Sim is calculated from U1 of the attribution target peak acquired in S302 and U2 of the attribution candidate peak acquired in S304, and the process proceeds to step S307.
- the detailed calculation flow of UV_Sim is described separately in subroutine 2 in FIG.
- step S307 the processing of “calculate the degree of coincidence of peak patterns (P_Sim_min)” is executed.
- a peak pattern is comprehensively created for these peaks from R1 and P1 of the attribution target peaks acquired in S302 and R2 and P2 of the attribution candidate peaks acquired in S304.
- P_Sim_min of these peak patterns is calculated, and the process proceeds to step S308.
- the detailed calculation flow of P_Sim_min is described separately in subroutine 3 of FIG.
- step S308 a process of “calculate the degree of coincidence of attribution candidate peaks (SCORE)” is executed.
- step S309 the process of “Substitute 888888 for SCORE (SCORE ⁇ 888888)” is executed.
- the SCORE of the peak that does not correspond to the attribution candidate peak of the attribution target peak is set to 888888, and the process proceeds to step S310.
- step S310 the process of “save SCORE (SCORE_all)” is executed.
- the SCORE obtained in S308 or S309 is stored in SCORE_all, and the process proceeds to step S311.
- step S311 a determination process of “End of processing of all reference peaks?” Is executed. In this process, it is determined whether or not all the peaks of the reference FP have been processed. If the processing has been completed (YES), the process proceeds to step S312. If there is an unprocessed peak (NO), the process proceeds to S304 in order to execute the processes of S304 to S311 for the unprocessed peak. The processes of S304 to S311 are repeated until all the peaks are processed.
- step S312 the process of “output SCORE_all to the determination result file and initialize SCORE_all (empty)” is executed. In this process, after outputting SCORE_all to the determination result file, SCORE_all is initialized (empty), and the process proceeds to step S313.
- step S313 a determination process of “End processing of all target peaks?” Is executed. In this process, it is determined whether or not all the peaks of the target FP have been processed. If the processing has been completed (YES), the target FP attribution process 2 ends. If there is an unprocessed peak (NO), the process proceeds to S302 in order to execute the processes of S302 to S313 for the unprocessed peak. The processes in S302 to S313 are repeated until all the peaks are processed.
- FIG. 100 is a flowchart showing details of the “target FP attribution process 3” in step S4 of FIG. This process is a post-process of attribution, and specifies the peak of the reference FP corresponding to each peak of the target FP from the matching degree (SCORE) of the attribution candidate peak calculated as described above.
- step S401 the process of “reading the determination result file” is executed.
- the determination result file created in the “target FP attribution process 2” in FIG. 81 is read, and the process proceeds to step S402.
- step S402 a process of “create attribution candidate peak score table with data satisfying condition of“ SCORE ⁇ threshold ”” is executed.
- an attribution candidate score table (see attribution candidate score table 191 in the upper diagram of FIG. 121) is created based on SCORE of the determination result file, and the process proceeds to step S403.
- This attribution candidate peak score table is a table in which, for each peak of the reference FP, only SCOREs smaller than the threshold are arranged in ascending order from the SCORE calculated for all the target FP peaks. Incidentally, the smaller the value of this SCORE, the more likely the peak to be assigned.
- the threshold value is an upper limit value of SCORE for determining whether or not to be an attribution candidate.
- step S403 the process of “create attribution candidate peak number table” is executed.
- an attribution candidate peak number table (see attribution candidate peak number table 193 in the lower diagram of FIG. 121) is created based on the attribution candidate peak score table, and the process proceeds to step S404.
- This attribution candidate peak number table is a table in which each score in the attribution candidate peak score table is replaced with the peak number of the target FP corresponding to the score. Therefore, this table is a table in which the peak numbers of the target FP to be associated with each peak of the reference FP are arranged in order.
- step S404 the process of “obtain the peak number of the target FP to be attributed” is executed.
- the peak number of the target FP positioned at the top for each peak of the reference FP is acquired from the attribution candidate peak number table created in S403, and the process proceeds to step S405.
- step S405 a determination process of “the acquired peak numbers are arranged in descending order (no duplication)?” Is executed. In this process, it is determined whether or not the peak numbers of the target FP acquired in S404 are arranged in descending order without duplication. If they are lined up (YES), it is determined that the peak of the target FP corresponding to each peak of the reference FP has been confirmed, and the process proceeds to step S408. If not (NO), the process proceeds to step S406 in order to review the peak of the target FP to be attributed to the problematic reference FP peak.
- step S406 a process of “compare SCORE between problematic peaks and update attribute candidate peak number table” is executed.
- the SCORE corresponding to the peak number of the target FP having a problem is compared in the attribution candidate score table, and the peak number with the larger SCORE is replaced with the peak number located second. Update, and the process proceeds to step S407.
- step S407 the process of “update attribution candidate peak score table” is executed.
- the attribution candidate peak score table is updated in accordance with the update contents of the attribution candidate peak number table in S406, and the process proceeds to step S404.
- the processing from S404 to S407 is repeated until there is no problem in the peak number of the target FP (there is duplication and not arranged in descending order).
- step S408 a process of “save attribution result (TEMP)” is executed.
- the peak number of all the peaks of the reference FP, the retention time, and the peak data of the target FP specified as the peak corresponding to these peaks are stored in TEMP, and the process proceeds to step S409.
- step S409 a determination process of “Are all peaks of the target FP included in TEMP?” Is executed. In this process, it is determined whether or not the peak data of all peaks of the target FP is included in the TEMP stored in S408. If all are included (YES), it is determined that the processing has been completed for all the peaks of the target FP, and the process proceeds to S412. If there is a peak not included (NO), the process proceeds to step S410 in order to add peak data of a peak not included to TEMP.
- step S410 a process of “correct peak retention time of target FP not included in TEMP” is executed.
- k2 Assigned near the peak of the target FP requiring correction Of the two peaks on the reference FP side, the retention time of the peak with the largest retention time t0: the retention time of the peak of the target FP that needs to be corrected t1: 2 assigned near the peak of the target FP that needs to be corrected
- Retention time of the peak with the shortest retention time among the two peaks on the target FP t2 Peak with the long retention time among the peaks on the two target FPs assigned near the
- step S411 a process of “adding the corrected retention time and the peak data of the peak to TEMP and updating TEMP” is executed.
- the retention time of the peak of the target FP not included in the TEMP corrected in S410 is compared with the retention time of the reference FP during the TEMP, and the target FP not included in the TEMP at a proper position in the TEMP.
- the corrected retention time and peak data of the peak are added, TEMP is updated, and the process proceeds to S409.
- the processing from S409 to S411 is repeated until all the peaks of the target FP are added.
- step S412 the process of “output TEMP to collation result file” is executed.
- TEMP that specifies the correspondence between all the peaks of the reference FP and all the peaks of the target FP is output as a verification result file, and the target FP attribution process 3 ends.
- [S5: Target FP attribution process 4] 101 and 102 are flowcharts showing details of the “target FP attribution process 4” in step S5 of FIG. This process is a final process of attribution, and each peak of the target FP is determined based on the collation result file (see collation result file example 195 in FIG. 122) created in step S4 in FIG. It belongs to the peak of data example 197 of group FP).
- the reference group FP197 is an FP that specifies the peak correspondence among all the reference FPs as described above. Like the reference group FP data example 197 in FIG. 123, the reference group FP peak number and the reference group retention are used. -Data composed of time and peak height. As shown by the reference group FP45 in FIG. 2, each peak can be represented by an average value (black dot) ⁇ standard deviation (vertical dividing line).
- step S501 the process of “reading the collation result file” is executed.
- the collation result file output in S412 of FIG. 100 is read, and the process proceeds to step S502.
- step S502 the process of “reading the reference group FP” is executed.
- the reference group FP197 that is the final attribution partner of each peak of the target FP is read, and the process proceeds to step S503.
- step S503 a process of “integrating and storing the target FP and the reference group FP (TEMP)” is executed.
- the two files are integrated based on the peak data of the reference FP that exists in common in the collation result file and the reference group FP197, the result is stored as TEMP, and the process proceeds to step S504.
- step S504 a process of “correcting the retention times of all peaks of the target FP having no peak corresponding to the reference FP” is executed.
- the retention times of all the peaks of the target FP having no peak corresponding to the reference FP in the collation result file are corrected to the TEMP retention times stored in S503, and the process proceeds to step S505.
- the retention time is corrected by the same method as in step S410 of the “target FP attribution process 3” in step S4.
- step S505 a process of “obtaining peak data (P1) corresponding to the corrected retention times (R1, R3) in order” is executed.
- the retention times corrected in S504 are sequentially acquired as R1 and R3, and the peak data of the corresponding peaks as P1, and the process proceeds to step S506.
- step S506 a process of “obtaining peak data (P2) corresponding to the retention time (R2) of the candidate candidate peak of the target FP in order from TEMP” is executed.
- the retention time to which the peak of the target FP is not attributed is stored as R2 from the TEMP stored in S503, and the corresponding peak data is acquired as P2, and the process proceeds to step S507.
- step S507 a determination process of “
- step S508 a process of “acquiring UV spectra corresponding to R1 and R2 (U1, U2)” is executed.
- the UV spectra corresponding to the peaks of the retention times R1 and R2 determined to be possible in S507 are acquired from the respective FPs, and the process proceeds to step S509.
- step S509 a process of “calculate UV spectrum coincidence (UV_Sim)” is executed.
- UV_Sim is calculated from the UV spectra U1 and U2 acquired in S508 by the same method as in step S306 of the “target FP attribution process 2” in step S3, and the process proceeds to step S510.
- the detailed calculation flow of UV_Sim will be described separately in subroutine 2 of FIG.
- step S510 a determination process of “UV_Sim ⁇ threshold 2?” Is executed.
- UV_Sim calculated in S509 is smaller than the threshold value 2. If it is small (YES), it is determined that the peak of U1 corresponds to the peak of U2 in the UV spectrum, and the process proceeds to step S511. If UV_Sim is greater than or equal to the threshold value 2 (NO), it is determined that it is not supported, and the process proceeds to step S507.
- step S511 the processing of “R3 ⁇ R2, threshold 2 ⁇ UV_Sim” is executed.
- the threshold 2 is updated to the value of UV_Sim, and the process proceeds to S507.
- step S512 a determination process of “comparison of retention times of all attribution candidate peaks?” Is executed. In this process, it is determined whether or not the comparison of the retention times of R1 and all attribution candidate peaks has been completed. If the comparison has been completed (YES), the process proceeds to step S513. If not completed (NO), the process proceeds to step S507.
- step S513 the processing of “save R1, R3 and P1 and threshold 2 (TEMP2)” is executed.
- the retention time (R1) determined to correspond in S510, the peak (P1) corresponding to R3 updated to the corresponding partner's retention time (R2), and the current threshold value 2 are stored (TEMP2), and S507 is performed.
- step S514 a determination process of “completion of retention times of all non-corresponding peaks?” Is executed. In this process, it is determined whether or not the comparison with the retention time of the attribution candidate peak is completed for the retention times of all the non-corresponding peaks. If it has been completed (YES), it is determined that all non-corresponding peak attribution processing has been completed, and the process proceeds to step S516. If not completed (NO), it is determined that an unprocessed non-corresponding peak remains, and the process proceeds to step S515.
- step S515 the process of “threshold 2 ⁇ initial value” is executed.
- the threshold value 2 updated to UV_Sim in S511 is returned to the initial value, and the process proceeds to step S505.
- step S516 a determination process of “is there a peak having the same value of R3 in TEMP2?” Is executed. In this process, it is determined whether or not a plurality of non-corresponding peaks belong to the same peak in TEMP. If there is a non-corresponding peak attributed to the same peak (YES), the process proceeds to step S517. If it does not exist (NO), the process proceeds to step S518.
- step S517 a process of “comparing threshold values 2 of peaks having the same R3 value and returning R3 of a peak having a large value to the original value (R1)” is executed.
- the threshold value 2 of the peak having the same value of R3 in TEMP2 is compared, the value of R3 of the peak having a large value is returned to the original value (that is, R1), and the process proceeds to step S518.
- step S518 the processing of “add TEMP2 peak to TEMP (only peaks where TEMP retention time and R3 match)” is executed.
- a peak corresponding to R3 is added to TEMP only for the peak in which the retention time of TEMP and R3 match, and the process proceeds to step S519.
- a peak in which R3 does not coincide with the retention time of TEMP is not added because there is no peak that becomes an belonging partner in the reference group FP.
- step S519 the process of “output the peak of the target FP during TEMP (peak feature file)” is executed.
- the peak data of the target FP attributed to the reference group FP197 is output as a peak data feature amount file, and the target FP attribution process 4 ends.
- FIG. 124 shows an example file 199 of peak data feature values output as described above.
- FIG. 103 is a flowchart showing details of “subroutine 1” in the “reference FP selection process” of FIG. In this process, the degree of coincidence of the retention time, the appearance pattern between the FPs (for example, the target FP and the reference FP) is calculated.
- step S1001 the process of “x ⁇ R1, y ⁇ R2” is executed.
- R1 and R2 acquired in S202 and S206 of FIG. 98 are substituted for x and y, respectively, and the process proceeds to step S1002.
- step S1002 the process of “obtaining the number of data of x and y (a, b)” is executed.
- the numbers of data of x and y are acquired as a and b, respectively, and the process proceeds to step S1003.
- step S1003 1 is substituted for i (i ⁇ 1) as the initial value of the counter for sequentially calling the retention times of x (i ⁇ 1), and the process proceeds to step S1004.
- step S1004 the process of “obtain all distance from xi-th retention time (f)” is executed. In this process, the distance between the xith retention time and all subsequent retention times is acquired as f, and the process proceeds to step S1005.
- step S1005 1 is substituted for j (j ⁇ 1) as the initial value of the counter for sequentially calling the retention times of y, and the process proceeds to step S1006.
- step S1006 the process of “obtain all distance from yj-th retention time (g)” is executed.
- the distance between the yj-th retention time and all subsequent retention times is acquired as g, and the process proceeds to step S1007.
- step S1007 the processing of “"
- the retention time distances f and g acquired in S1004 and S1006 are compared in a round-robin manner, and the condition of “
- step S1008 the processing of “calculate the degree of coincidence of retention and time / appearance pattern of f and g (RP fg )” is executed.
- RP fg is obtained from a and b acquired in S1002 and m acquired in S1007.
- RP fg (1 ⁇ (m / (a + b ⁇ m))) ⁇ (a ⁇ m + 1)
- step S1009 the processing of “calculate the degree of coincidence of retention and time / appearance pattern of f and g (RP fg )” is executed.
- RP fg is obtained from a and b acquired in S1002 and m acquired in S1007.
- RP fg (1 ⁇ (m / (a + b ⁇ m))) ⁇ (a ⁇ m + 1)
- step S1009 the processing of “calculate the degree of coincidence of retention and time / appearance pattern of f and g (RP fg )” is executed.
- RP fg is obtained from a and b
- step S1009 the process of “save RP fg (RP_all)” is executed.
- the degree of coincidence calculated in S1008 is stored in RP_all, and the process proceeds to step S1010.
- step S1010 the process of “update j (j ⁇ j + 1)” is executed.
- j + 1 is substituted for j as an update of j, and the process proceeds to step S1011.
- step S1011 a determination process of “Processing complete with all retention times of y?” Is executed. In this process, it is determined whether or not all the retention times for y have been processed. If it has been completed (YES), it is determined that the processing of all retention times of y has been completed, and the process proceeds to step S1012. If not completed (NO), it is determined that an unprocessed retention time remains in y, and the process proceeds to step S1006. That is, the processing from S1006 to S1011 is repeated until all the retention times of y are processed.
- step S1012 the process of “update i (i ⁇ i + 1)” is executed.
- i + 1 is substituted for i as an update of i, and the process proceeds to step S1013.
- step S1013 a determination process of “Processing completed with all retention times of x?” Is executed. In this process, it is determined whether or not all the retention times for x have been processed. If it has been completed (YES), it is determined that all the retention times for x have been processed, and the process proceeds to step S1014. If not completed (NO), it is determined that an unprocessed retention time remains in x, and the process proceeds to step S1004. That is, the processing from S1004 to S1013 is repeated until all the retention times of x are processed.
- step S1014 the process of “obtain minimum value from RP_all (RP_min)” is executed.
- FIG. 104 is a flowchart showing details of “subroutine 2” in “target FP attribution process 2” of FIG. This process calculates the degree of coincidence of the UV spectra.
- step S2001 the process of “x ⁇ U1, y ⁇ U2, z ⁇ 0” is executed.
- the UV spectra U1 and U2 acquired in S302 and S304 of FIG. 99 are substituted into x and y, respectively, and 0 is substituted as the initial value of the square sum (z) of the distance between the UV spectra, step S2002.
- step S2002 the process of “acquire the number of x data (a)” is executed.
- the number of x data is acquired as a, and the process proceeds to step S2003.
- step S2003 1 is substituted into i as an initial value for sequentially calling the absorbance at each detection wavelength constituting the UV spectrum U1 from x, and the process proceeds to step S2004.
- step S2004 the process of “obtain xi-th data (b)” is executed.
- the i-th absorbance data of x to which the UV spectrum U1 is substituted is acquired as b, and the process proceeds to step S2005.
- step S2005 the “acquire yi-th data (c)” process is executed.
- the i-th absorbance data of y into which the UV spectrum U2 is substituted is acquired as c, and the process proceeds to step S2006.
- step S2006 a process of “calculate the sum of squares (z) of the distance between UV spectra (d) and the distance between UV spectra” is executed.
- the process proceeds to step S2007.
- step S2007 the “i update (i ⁇ i + 1)” process is executed.
- i + 1 is substituted for i as an update of i, and the process proceeds to step S2008.
- step S2008 a determination process of “process complete with all data of x?” Is executed. In this process, it is determined whether or not the processing of all data of x and y has been completed. If it has been completed (YES), it is determined that the processing of all data of x and y has been completed, and the process proceeds to step S2009. If not completed (NO), it is determined that unprocessed data remains in x and y, and the process proceeds to step S2004. That is, the processes from S2004 to S2008 are repeated until all the absorbance data of x and y are processed.
- step S2009 the process of “calculate the degree of coincidence between the x and y UV spectra (UV_Sim)” is executed.
- UV_Sim is calculated from the sum of squares z of the distance between the UV spectra and the number of data a.
- UV_Sim ⁇ (z / a)
- This UV_Sim is passed to step S306 in FIG. 99, and the UV spectrum coincidence degree calculation process is terminated.
- FIG. 105 is a flowchart showing details of “subroutine 3” in “target FP attribution process 2” of FIG. This process calculates the coincidence of peak patterns.
- step S3001 a process of “setting the number of peak pattern configuration candidates (m) and the number of peak pattern configuration peaks (n)” is executed.
- the peak pattern configuration candidate number (m) and the peak pattern configuration peak number (n) are set as settings for comprehensively creating peak patterns, and the process proceeds to step S3002.
- step S3002 the process of “x ⁇ target FP name, r1 ⁇ R1, p1 ⁇ P1, y ⁇ reference FP name, r2 ⁇ R2, p2 ⁇ P2” is executed.
- the file names of the target FP and reference FP necessary for the process, and the retention time and peak data acquired in S302 and S304 in FIG. 99 are substituted into x, r1, p1, y, r2, and p2, respectively.
- the process proceeds to step S3003.
- step S3003 the process of “obtain all retention times of x (a)” is executed.
- the file (target FP) with the name assigned to x in S3002 is read, the entire retention time of the file is acquired as a, and the process proceeds to step S3004.
- step S3004 the process of “obtain all retention times of y (b)” is executed.
- the file (reference FP) with the name assigned to y in S3002 is read, the entire retention time of the file is acquired as b, and the process proceeds to step S3005.
- step S3005 the process of “obtain retention time (cm) and peak data (dm) of m peak pattern configuration candidate peaks from a to r1” is executed.
- the retention time of m peak pattern configuration candidate peaks of r1 which is the retention time of the attribution target peak, is obtained from a as cm, and the peak data is obtained as dm, and the process proceeds to step S3006.
- the m peak pattern configuration candidate peaks are m having a retention time close to r1.
- step S3006 the processing of “obtain m retention time (em) and peak data (fm) of m peak pattern configuration candidate peaks from b to r2” is executed.
- the retention time of m peak pattern configuration candidate peaks of r2 which is the retention time of the attribution candidate peak, is acquired from b as em, and the peak data is acquired as fm, and the process proceeds to step S3007.
- the m peak pattern configuration candidate peaks are m having a retention time close to r2.
- step S3007 a process of “arranging cm and dm in retention time order (ascending order)” is executed.
- the cm and dm acquired in S3005 are rearranged so that the retention times are in ascending order, and the process proceeds to step S3008.
- step S3008 a process of “arranging em and fm in the order of retention time (ascending order)” is executed.
- em and fm acquired in S3006 are rearranged so that the retention times are in ascending order, and the process proceeds to step S3009.
- step S3009 the processing of “obtaining peak pattern configuration peak n retention times (cn) and peak data (dn) in order from cm and dm” is executed.
- the retention time of n peak pattern configuration peaks is set to cn and the peak data is set to dn in order, and the process proceeds to step S3010.
- step S3010 a process of “obtaining n retention times (en) and peak data (fn) of n peak pattern constituent peaks in order from em and fm” is executed.
- the retention time of n peak pattern configuration peaks is en and the peak data is acquired in order as fn, and the process proceeds to step S3011.
- step S3011 the processing of “calculate peak pattern coincidence (P_Sim)” is executed.
- r1 and p1 of the attribution target peak obtained so far, cn and dn of the peak pattern constituting peak n, and r2 and p2 of the attribution candidate peak and en of n peaks and peaks constituting the attribution pattern are obtained.
- step S3012 the process of “save P_Sim (P_Sim_all)” is executed.
- P_Sim calculated in S3011 is sequentially stored in P_Sim_all, and the process proceeds to step S3013.
- step S3013 a determination process of “all combinations out of m out of em taken out?” Is executed. In this process, it is determined whether or not the process has been completed for all combinations of extracting n peak pattern configuration peaks from m peak pattern configuration candidate peaks of attribution candidate peaks. If completed (YES), it is determined that the creation of an exhaustive peak pattern and the calculation of the degree of coincidence in the attribution candidate peak have been completed, and the process proceeds to step S3014. If not completed (NO), it is determined that the combination of extracting n from m is not completed, and the process proceeds to step S3010. That is, the processing from S3010 to S3013 is repeated until the processing is completed for all combinations in which n are extracted from m.
- step S3014 a determination process of “all combinations out of m out of cm is completed?” Is executed. In this process, it is determined whether or not the process has been completed for all combinations that extract n peak pattern configuration peaks from m peak pattern configuration candidate peaks of the attribution target peak. If completed (YES), it is determined that the creation of an exhaustive peak pattern and calculation of the degree of coincidence in the attribution target peak have been completed, and the process proceeds to step S3015. If not completed (NO), it is determined that the combination of taking n out of m is not completed, and the process proceeds to step S3009. That is, the processing from S3009 to S3014 is repeated until the processing is completed for all combinations in which n are extracted from m.
- step S3015 the process of “obtain minimum value from P_Sim_all (P_Sim_min)” is executed.
- the minimum value of P_Sim_all stored in S3012 is acquired as P_Sim_min, and this P_Sim_min is passed to step S307 in FIG. 99, and the peak pattern matching degree calculation process is terminated.
- FIG. 106 is a flowchart showing details of “FP_type2 creation” in step S6 of FIG.
- step S601 the process of “read target FP” is executed.
- the file of the target FP 43 (see the FP data example 187 in FIG. 119) is read, and the process proceeds to step S602.
- step S602 a process of “reading a peak data feature file” is executed.
- a peak data feature amount file (see the peak data feature amount file example 199 in FIG. 124) is read, and the process proceeds to step S603.
- the example of the peak data feature file includes the peak information of the target FP 43 attributed to the peak of the reference group FP 45 by the target FP peak feature generating unit 7.
- step S603 a process of “compare the target FP with the peak data feature file” is executed.
- the file of the target FP 43 is compared with the peak data feature file.
- the residual peak of the target FP43 that has not been assigned to the peak of the reference group FP45 is specified, and the process proceeds to step S604.
- step S604 the processing of “output retention time and peak data of peak existing only in target FP” is executed.
- the retention time and peak data of the remaining peak of the target FP 43 are output to the target FP type 2 data file (see the reference and target FP type 2 data example 201 in FIG. 125).
- FIG. 107 is a flowchart showing details of the “feature value processing for target FP_type2 by area division” in step S7 of FIG.
- step S701 the processing of “setting the FP space area division condition” is executed.
- the processing of “setting the FP space area division condition” is executed.
- one position is set for each of the first vertical and horizontal lines (division lines).
- vertical and horizontal dividing lines (first line) are set as dividing lines in the FP space as shown in FIGS. 76 and 77, for example.
- the amplitude is not related.
- step S702 a process of “creating an FP space area division pattern” is executed.
- the positions of the second and subsequent dividing lines are set for all combinations of the first vertical and horizontal dividing lines, and a divided pattern (one) is created.
- the FP space is divided into regions by vertical and horizontal dividing lines as shown in FIG. 78, for example.
- region division the process proceeds to step S703.
- step S703 a process of “reading a file of the target FP_type2” is executed. With this process, the target FP type 2 file is read, and the process proceeds to step S704.
- step S704 the process of “calculate total peak data of entire FP space” is executed.
- the total height of the peaks present in all the divided grids 145 as shown in FIG. 79 is calculated (FIG. 81), and the process proceeds to step S705.
- step S705 the process of “divide the FP space by the division pattern” is executed.
- the target FP type 2 read in S703 is divided into regions as shown in FIG. 79 using the region division pattern set in S702, and the process proceeds to step S706.
- step S706 the processing of “calculate the existence ratio of peak data in the divided area” is executed.
- the calculation result is as shown in FIG.
- the process proceeds to step S707.
- step S707 a process of “output the existence ratio of each region as a feature amount” is executed.
- one target FP region division feature value file (see one target FP region division feature value file example 203 shown in FIG. 126) is output.
- FIG. 108 is a flowchart showing details of the “integration of peak data feature quantity and area division feature quantity” in step S8 of FIG.
- step S801 a process of “reading a peak data feature value file” is executed. By this processing, the same file as the peak data feature amount file example 199 shown in FIG. 124 is read, and the process proceeds to step S802.
- step S802 a process of “reading an area division feature value file” is executed. With this processing, the target FP region division feature value file 203 shown in FIG. 126 is read, and the process proceeds to step S803.
- step S803 a process of “integrating two feature data as one horizontal line of data” is executed.
- the peak data feature file see the peak data feature file example 199 shown in FIG. 12
- the target FP region segmentation feature file see target FP region segmentation feature file example 203 shown in FIG. 126.
- Is integrated as a target FP feature value integration file see target FP feature value integration file example 205 in FIG. 127
- the process proceeds to step S804.
- step S804 a process of “output integrated data” is executed.
- the target FP feature amount integration file 205 of FIG. 127 is output.
- [Creation of standard FP attribution result feature file] A reference FP feature value integration file for comparing the target FP feature value integration data with the reference FP feature value integration data is created as shown in FIGS.
- the FP creation function of the reference FP creation unit 31 the reference FP peak attribution function of the reference FP peak attribution unit 15, and the reference FP attribution.
- the computer realizes the reference FP region division feature value creation function of the division feature value creation unit 23 and the reference FP feature value integration function of the reference FP feature value integration unit 25.
- the reference FP creation function is realized in step S10001.
- the reference FP peak attribution function is realized in steps S10002, S10003, and S10004.
- the reference FP attribution result integration function is realized in step S10005.
- the reference FP peak feature creation function is realized in step S10006.
- the reference FP type 2 creation function is realized in step S10007.
- the reference FP region division feature value creation function is realized in step S10008.
- the reference FP feature value integration function is realized in step S10009.
- S10001 to S10004 correspond to S1 to S4 related to the creation of the target FP feature amount integrated file of FIGS. 93 and 94, and S1007 to S10009 correspond to S6 to S8.
- step S1000 “FP creation processing” is executed using 3D chromatography and peak information at a specific detection wavelength as input data.
- Both 3D chromatograms and peak data are provided for each of a plurality of evaluation standard drugs (standard Chinese medicines) as an evaluation standard.
- step S10001 the reference FP creation unit 31 (FIG. 1) of the FP creation unit 3 of the computer functions to create a reference FP from the 3D chromatogram and peak information in the same manner as the target FP 43 (FIG. 2). Output as a file.
- step S10002 “reference FP attribution process 1” is executed with all the reference FPs output in step S10001 as inputs.
- step S10002 the reference FP peak attribution unit 15 of the computer functions to select all combinations from all the reference FPs in order to calculate the attribution score in the selected combination and order for all the reference FPs, and the process proceeds to step S10003. .
- step S10003 the combination of the selected reference FP is input, and “reference FP attribution process 2” is executed.
- step S10003 a peak pattern is comprehensively created as shown in FIGS. 23 to 61 for all the peaks of the combination of the reference FP selected in step S2, and then the degree of coincidence of these peak patterns (FIG. 63 or FIG. 64 P_Sim) is calculated. Further, the degree of coincidence of UV spectra (UV_Sim in FIG. 66) is calculated between the peaks of the selected combination of reference FPs. Further, the degree of coincidence of attribution candidate peaks (SCORE in FIG. 67) is calculated from these two degrees of coincidence. The calculation result is output as a determination result file (see determination result file example 189 in FIG. 120).
- step S10004 the determination result file output in step S10003 is input, and the “reference FP attribution process 3” is executed.
- step S10004 corresponding peaks are identified between combinations of selected reference FPs based on the matching degree (SCORE) of attribution candidate peaks among selected combinations of reference FPs.
- the result is output as reference FP attribution data for each reference FP.
- step S10005 all reference FP attribution data output in step S10004 is input, and “reference FP attribution result integration processing” is executed.
- step S10005 the reference FP attribution result integration unit 17 of the computer functions, refers to the peak correspondence relationship of each reference FP identified by the reference FP peak attribution unit 15, and integrates all reference FP attribution data to correspond to the reference FP.
- a table is created and the process proceeds to step S10006.
- the reference FP peak feature value creation unit 19 of the computer functions to create a peak feature value (reference group FP) for all the reference FPs based on the reference FP correspondence table created by the reference FP attribution result integration unit 17. To do.
- the processing in the reference FP peak feature quantity creation unit 19 calculates a statistic (maximum value, minimum value, median value, average value, etc) for each peak (column) in the reference FP correspondence table, and uses that information as a basis. Select the peak (column). The selected peak (row) is output as a reference group FP (see reference group FP example 197 in FIG. 123).
- step S10007 the reference group FP and all reference FPs output in step S10006 are input, and the process of “Create FP_type2” is executed.
- step S10007 the reference FP type 2 creation unit 21 of the computer functions in the same manner as the target FP type 2 creation unit 9, and the featured peaks are obtained from a plurality of reference FPs in the same manner as in step S6 in FIG. An FP composed of the remaining peaks and their retention times is created as a reference FP type 2 (see FP type 2 file example 201 in FIG. 125).
- step S10008 “reference FP_type2 feature value conversion processing” is executed.
- the reference FP region division feature value creation unit 23 of the computer functions, and a reference FP region division feature value is created by the region division of FIGS. 73 to 85.
- the result is output as the reference type 2 group FP (see the reference type 2 group FP example 207 in FIG. 128).
- step S10009 the “reference data creation process” is executed.
- the reference FP feature amount integration unit 25 of the computer functions, and the reference type FP group feature creation unit 23 creates the reference group FP created by the reference FP peak feature creation unit 19 and the reference type 2 group FP created by the reference FP region division feature creation unit 23.
- the result is output as reference group integration data (see reference group integration data example 209 in FIG. 129).
- [S10005: Creation of standard FP correspondence table] 111 and 112 are flowcharts showing details of the “reference FP attribution result integration process (creation of reference FP correspondence table)” in step S10005 of FIG.
- step S10101 a process of “reading attribution data with attribution number 1 as integrated data” is executed.
- the reference FP attribution data that has been assigned for the first time in S10004 and identified the peak correspondence is read as integrated data, and the process proceeds to step S10102.
- step S10102 a process of “reading attribute data second and later in order” is executed.
- the reference FP attribution data that has been subjected to the second attribution process in S10004 and identified the peak correspondence is read as integrated data, and the process proceeds to step S10103.
- step S10103 a process of “integrating integrated data and attribution data with common peak data” is executed.
- the two files are integrated based on the peak data of the reference FP that exists in common in the integrated data and the attribution data, and the integrated data is updated as a result, and the process proceeds to step S10104.
- step S10104 a determination process of “Add all peaks in attribution data to integrated data?” Is executed. In this process, it is determined whether all the peaks of the attribution data have been added to the integrated data. If it has been added (YES), the process proceeds to step S10105. If there is a peak that has not been added (missing peak) (NO), the process proceeds to step S10107 in order to add this missing peak to the integrated data. Note that the processing for adding missing peaks to the integrated data (S10107 to S10120) is the same as steps S504 to S517 in S5 (target FP attribution processing 4).
- step S10121 processing of “add TEMP2 data to integrated data (all retention times and peaks)” is executed.
- all retention times (R3) and peaks (P1) of TEMP2 are added to the corresponding locations of the integrated data, and the process proceeds to step S10122.
- step S10122 processing of “threshold 2 ⁇ initial value, delete all data in TEMP2” is executed.
- the threshold value 2 updated to UV_Sim is returned to the initial value, all data is deleted from TEMP2 containing data such as the retention time and peak of all missing peaks, and the process returns to step S10104.
- step S10105 which is shifted from step S10104, a determination process of “all attribute data processing completed?” Is executed. In this process, it is determined whether or not the processing of all reference data has been completed. If the process is finished (YES), the process proceeds to step S10106 in order to output the reference FP correspondence table that is the result of the integration of all attribution data. If all the processes have not been completed (NO), the process returns to step S10102 to sequentially process the remaining attribution data.
- step S10106 the process of “output integrated data (reference FP correspondence table)” is executed.
- the result of integrating all attribution data is output as a standard FP correspondence table, and the standard FP correspondence table creation process ends.
- FIG. 113 is a flowchart showing details of the “peak feature value conversion process (creation of reference group FP)” in step S10006 of FIG.
- step S10201 the process of “read the reference FP correspondence table” is executed.
- the reference FP correspondence table created in S10005 is read, and the process proceeds to step S10202.
- step S10202 the processing of “calculate statistics for each peak (column)” is executed.
- a statistic maximum value, minimum value, median value, average value, variance, standard deviation, number of existence, presence rate
- step S10203 a process of “selecting a peak (column) with reference to the calculated statistic” is executed.
- a peak is selected with reference to the statistic calculated in S10102, and the process proceeds to step S10204.
- step S10204 the process of “output selected peak (row) (reference group FP)” is executed.
- the peak (column) selection result is output as the reference group FP based on the statistics, and the process of creating the reference group FP is terminated.
- FIG. 123 shows a reference FP correspondence table example 197 that is output as described above.
- FIG. 114 shows “reference FP editing process (reference FP_type2) in step S10007 of FIG. Is a flowchart showing the details of “
- step S10301 a process of “reading the reference FP in order” is executed.
- a plurality of reference FP files (see FP data example 187 in FIG. 119) are read, and the process proceeds to step S10302.
- step S10302 a process of “reading the reference group FP” is executed.
- the data file of the reference group FP (see the data example 197 of the reference group FP in FIG. 123) is read, and the process proceeds to step S10303.
- step S10303 a process of “extracting a peak data feature amount of the reference FP from the reference group FP” is executed.
- the peak data feature value subjected to the attribution process of the reference FP is extracted from the file of the reference group FP45, and the process proceeds to step S10304.
- step S10304 a process of “compare the reference FP with the extracted peak data feature file” is executed, the reference FP is compared with the peak data feature file, and the process proceeds to step S10305.
- step S10305 the processing of “output the retention time and peak data of the peak existing only in the reference FP” is executed, the peak of the peak data feature file is removed from the reference FP, and the process proceeds to step S10306. .
- step S10306 a determination process of “Processing completed with all reference FPs?” Is executed.
- S10007 is finished, and when the process is not finished with all the reference FPs (NO), S10301 to S10305 are repeated.
- a plurality of reference FPs are processed in order, and the peak of the peak data feature file is removed from each reference FP, and the file of the reference FP type 2 (see the target and reference FP type 2 data example 201 shown in FIG. 125). Is created.
- FIG. 115 is a flowchart showing details of the “feature value processing of reference FP_type2 by area division” in step S10008 of FIG.
- step S10401 the processing of “setting the FP space area division condition” is executed.
- a plurality of positions of the first vertical and horizontal lines (division lines) are set.
- a plurality of vertical / horizontal dividing lines (first) 141 and 143 are set as dividing lines in the FP space as shown in FIGS. 76 and 77, for example.
- the process proceeds to step S10402.
- step S 10402 the processing of “setting the FP space area division pattern” is executed.
- the positions of the second and subsequent dividing lines are set for all combinations of the first vertical and horizontal dividing lines, and divided patterns (m ⁇ n) are created.
- this setting for example, as shown in FIG. 78, a plurality of area division patterns by the vertical and horizontal division lines 141 and 143 are set for the FP space.
- region division is performed, the process proceeds to step S10403.
- step S10403 a process of “reading files of reference FP_type2 in order” is executed. With this process, the reference FP type 2 file is read, and the process proceeds to step S10404.
- step S10404 a process of “calculate total peak data of entire FP space” is executed.
- the total height of the peaks present in all the lattices 145 divided as shown in FIG. 79 is calculated (FIG. 81), and the process proceeds to step S10405.
- step S10405 a process of “divide the FP space in order by each division pattern” is executed.
- the FP space is sequentially divided by the plurality of area division patterns set in S10402, and the process proceeds to step S10406.
- step S10406 the processing of “calculate the existence ratio of peak data in the divided area” is executed.
- the calculation results are as shown in FIGS. 83 to 85, for example.
- the process proceeds to step S10408.
- step S10408 the process of “end division with all division patterns” is executed. In this processing, it is determined whether or not the feature amount processing for the plurality of all region division patterns set in S10402 is completed. If the feature amount process is completed (YES), the process proceeds to step S10409. If the feature value process is not completed (NO), the process proceeds to step S10405. Steps S10405 to S10408 are repeated until the feature amount processing for the entire area division pattern is completed.
- step S10409 a determination process of “processing complete with all reference FP_type2?” Is executed. In this process, it is determined whether or not the feature amount process has been completed for all of the plurality of reference FP types 2 created for each of the plurality of reference FPs. If all the reference FP types 2 are finished (YES), S10008 is finished. If all the reference FP types 2 are not finished (NO), the process proceeds to step S10403. Steps S10403 to S10409 are repeated until the feature amount processing in the reference FP type 2 is completed.
- FIG. 128 shows a reference type 2 group FP example 207.
- S10009 Reference data creation processing
- FIG. 116 is a flowchart showing details of the “reference data creation process” in step S10009 of FIG.
- step S10501 a process of “reading an area division feature file” is executed.
- the reference FP region division feature value file (see the reference type 2 group FP example 207 shown in FIG. 128) is read, and the process proceeds to step S10502.
- step S10502 a process of “calculate the number of division patterns when the area is divided” is executed.
- the number of division patterns for area division is calculated.
- the number of division patterns is calculated as 100, for example, as described with reference to FIGS. After this calculation, the process proceeds to step S10503.
- step S10503 a process of “reading the reference group FP” is executed, the reference group FP is read, and the process proceeds to step S10504.
- step S10504 a process of “creating a file (reference group FP2) in which each row of the reference group FP is duplicated by the number of division patterns” is executed.
- the rows of the reference group FP are duplicated according to the number of division patterns to create the reference group FP2.
- the file example 197 of the reference group FP in FIG. 123 is duplicated so as to correspond to the peak data feature amount (reference group FP2) in the reference group integrated data example 209 in FIG. After this duplication, the process proceeds to step S10505.
- step S10505 a process of “integrating the reference group FP2 and the area division feature file for each line” is executed.
- the data of the reference group FP2 and the data of the area division feature amount file copied in S10504 are integrated for each row, and the process proceeds to step S10506.
- step S10506 a process of “output integrated data” is executed.
- a reference FP feature value integration file (see reference group integration data example 209 in FIG. 129) based on the integration result is output.
- a step 165, a reference FP region division feature value creation step 167, a reference FP feature value integration step 169, and an evaluation step 171 are provided.
- the FP creation step 148 includes the target FP creation step 173 and a reference FP creation step 175.
- the target FP peak attribution step 149 includes the reference FP selection step 177, a peak pattern creation step 179, and a peak attribution step 181.
- a target FP peak feature value that has been characterized based on the target FP 43 and a plurality of reference FPs is created, and a target FP type 2 is created as a residual peak of the target FP 43 that has been leaked from this feature quantification, and this target FP type 2 Is divided into a plurality of regions, the target FP region segmentation feature amount is created from the existence rate of the peak existing in each region, and the target FP peak feature amount and the target FP region segmentation feature amount are integrated to create the target FP integrated feature amount
- the target FP integrated feature quantity and the reference FP integrated feature quantity based on a plurality of reference FPs of multi-component substances corresponding to the target FP integrated feature quantity, It is possible to evaluate including the peak of the target peak that has not been made, and it is possible to reliably improve the quality evaluation accuracy of the evaluation target drug.
- the target FP 43 created in the target FP creation step 173 is composed of three-dimensional information (peak, retention time, and UV spectrum) as in the 3D chromatography 41. Therefore, it is data that inherits information specific to the drug as it is. Nevertheless, since the data volume is compressed to about 1/70, the amount of information to be processed can be greatly reduced and the processing speed can be increased compared to the 3D chromatogram 41.
- the target FP creation step 173 creates an FP that combines a plurality of FPs with different detection wavelengths. Thereby, even if it is a multi-component medicine in which components that cannot detect all components at one wavelength are combined, quality evaluation including all components can be performed by synthesizing FPs having a plurality of detection wavelengths. .
- Target FP creation step 173 creates an FP that includes all peaks detected by 3D chromatography. For this reason, it is suitable for the quality evaluation of the Chinese medicine which is a multi-component medicine.
- the reference FP suitable for the attribution of the target FP is compared with the retention / time / occurrence pattern between the FPs, and the reference FP having a good pattern matching degree is selected.
- the peak attribution step 181 attribution processing can be performed between FPs having similar patterns, so that attribution with high accuracy is possible.
- a peak pattern is comprehensively created using a plurality of peripheral peaks for each of the attribution target peak and the attribution candidate peak. As a result, even if the pattern of the entire FP is slightly different between the target FP and the reference FP, high-accuracy attribution is possible in the peak attribution step 181.
- the peak to be assigned is specified by taking into account the degree of coincidence of the UV spectrum of the attribution target peak and the attribution candidate peak in addition to the degree of coincidence of the peak pattern created in the peak pattern creation step 179. ing. Therefore, attribution with high accuracy is possible.
- the peak assignment step 181 all the peaks of the target FP are assigned all at once to the peaks of the reference FP. Therefore, efficient attribution processing is possible.
- FPs composed of multicomponents that are multidimensional data are aggregated in one dimension as MD values by the MT method, and a plurality of evaluation target lots are easily compared and evaluated. For this reason, it is suitable for the evaluation of multi-component drugs composed of a plurality of components.
- the region is divided by a plurality of vertical division lines 141 parallel to the signal intensity axis and a plurality of horizontal division lines 143 parallel to the time axis.
- the plurality of horizontal dividing lines 143 were set at equal ratio intervals in the direction in which the signal intensity increases.
- the region can be subdivided at a portion where the peak density is high, and the peak existence rate can be calculated efficiently by dividing the region.
- the multi-component substance evaluation method further includes the reference FP creation step 175, the reference FP peak attribution step 159, the reference FP attribution result integration step 161, the reference FP peak feature creation step 163, and the reference FP type 2 creation step. 165, a reference FP region division feature value creation step 167, and a reference FP feature value integration step 169.
- a reference FP integrated feature value obtained by integrating the reference FP peak feature value and the reference FP region segmentation feature value can be created and compared with the target FP integrated feature value which is the evaluation step 171, and the quality evaluation of the evaluation target drug Accuracy and efficiency can be further improved.
- the reference FP region division feature value creation step 167 can change the position of each region and create a reference FP region division feature value before and after the change.
- the retention time and peak height fluctuate due to slight variations in analysis conditions, and even if the value in each grid 145 fluctuates greatly in a single pattern, regardless of this fluctuation, The abundance of the peak can be captured, and the accuracy and efficiency of the quality evaluation of the evaluation target drug can be further improved.
- the region is divided by a plurality of vertical division lines 141 parallel to the signal intensity axis and a plurality of horizontal division lines 143 parallel to the time axis.
- the plurality of horizontal dividing lines 143 were set at equal ratio intervals in the direction in which the signal intensity increases.
- the region can be subdivided at a portion where the peak density is high, and the peak existence rate can be calculated efficiently by dividing the region.
- each region 145 is changed by changing the position so that the vertical and horizontal division lines 141 and 143 are translated within the set range.
- the evaluation program for a multi-component drug according to the embodiment of the present invention can realize each function on a computer and improve the accuracy and efficiency of the evaluation.
- the evaluation apparatus for a multi-component drug causes each part 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27 to act to improve the accuracy and efficiency of the evaluation. Can be improved.
- P_Sim Modification of Peak Pattern Matching Calculation
- the peak pattern coincidence calculation (P_Sim) in FIG. 63, FIG. 64, and FIG. 105 is applied to the case of the above embodiment in which the FP is created at the peak height, and based on the difference in the peak heights to be compared. Calculated.
- the signal intensity (height) maximum value and the signal intensity area value ( Any of the cases where the peak area is expressed in terms of height can be included.
- the FP when the FP is created with the peak area, the FP is created by expressing the area value with the height, and therefore, the FP has the same expression as that when creating with the peak height in the above-described embodiment. For this reason, it can evaluate similarly by the process of the said Example similar to the case where FP is created by peak height.
- FIG. 130 is a flowchart showing details of a modification example of “subroutine 2” in “target FP attribution process 2” in FIG. 99, according to a modification example of subroutine 2 applied instead of FIG. The degree of coincidence of the UV spectra is calculated by the processing according to this modification.
- DNS slope information
- steps S2001 to S2008 are substantially the same as subroutine 2 of FIG.
- the initial setting of the section 1 ⁇ w1 and the section 2 ⁇ w2 is additionally performed and used for a section for calculating a moving average and a moving slope, which will be described later.
- steps S2010 to S2013 are added for DNS addition, and the degree of coincidence can be calculated in consideration of DNS in step S2009A.
- step S2010 a determination process of “Does DNS take into account?” Is executed, and when it is determined that DNS is taken into account (YES), the process proceeds to step S2011, and when it is determined that DNS is not taken into account (NO). The process proceeds to step S2009A.
- the cause of whether or not DNS is taken into account is, for example, the initial setting. For example, when FP is created with a peak area, DNS is added, and when FP is created with a peak height, DNS is not taken into account.
- the UV pattern coincidence can be calculated by the processing that takes the DNS into account, and even when the FP is created with the peak area, the DNS is not taken into account.
- the UV pattern matching degree can be calculated by the processing of the above embodiment.
- step S2011 the process of “calculate moving average of x and y in section 1 (w1)” is executed, and the moving average in section 1 (w1) is obtained.
- step S2012 the processing of “calculate moving slope of x and y in section 2 (w2)” is executed, and the moving slope in section 2 (w2) is obtained.
- step S2013 a process of “calculate the number of mismatched signs of the x and y movement slopes (DNS)” is executed, and the number of coincidence of slopes ( ⁇ ) is calculated from the movement slope calculated in step S2012.
- DAS x and y movement slopes
- ⁇ the number of coincidence of slopes
- step S2013 When the process proceeds from step S2013 to step S2009A, the degree of coincidence is calculated in consideration of DNS in the process of step S2009A.
- step S2010 to step S2009A Note that the processing in the case of shifting from step S2010 to step S2009A is the same as step S2009 in FIG.
- FIG. 131 is a chart showing a calculation example of a moving average and a moving slope.
- UV intensity 131 shows an example of UV data
- the middle shows an example of moving average calculation
- the lower shows an example of calculation of moving slope.
- the UV intensity is expressed as a1 to a7 instead of specific numerical values.
- the UV intensity at 220 nm is a1
- the UV intensity at 221 nm is a2.
- the moving average calculation example and the moving inclination calculation example also use UV intensities a1 to a7 instead of specific numerical values.
- the moving slope is also the section 2 (3) as an example, and in step S2013 (FIG. 130), the values calculated for the sections (m1, m2, m3), sections (m2, m3, m4),. ⁇ ⁇ Is calculated.
- the moving average difference m3 ⁇ m1 is the moving slope, and the ( ⁇ ) is taken out.
- the UV pattern matching degree can be calculated by the process including DNS in the attribution process to the reference group FP and the standard FP attribution result integration process. With this calculation, even if the distance (dis) between the two corresponding points shown in FIG. 66 is larger than the FP created at the peak height, it is easy to handle and the UV pattern matching degree can be calculated accurately. it can. [Others]
- the signal intensity axis is applied as an area value axis, and the signal intensity is applied as an area value. be able to.
- the examples of the present invention are applied to the evaluation of Chinese medicine as a multi-component drug, but can also be applied to the evaluation of other multi-component substances.
- the region division feature amount is created for the target FP type 2 or the reference FP type 2, but it is also possible to create the region division feature amount for the target FP and the reference FP.
- the method includes a target pattern region dividing feature amount creating step of creating a pattern region dividing feature amount from a presence rate or an existing amount of a peak existing in each region by dividing a pattern whose peak changes in time series into a plurality of regions. If present, it can be widely applied.
- FP of the above-mentioned example is intended for all peaks on 3D chromatography, fine data, for example, FP can be created except for peaks whose peak area is less than 5% on 3D chromatography.
- the FP of the above example was created based on the peak height, and the evaluations of FIGS. 87 to 91 were obtained. However, the FP was created based on the peak height even when the FP was created based on the peak area.
- the MD value is obtained by the MT method in the same procedure as in the example, and the evaluation can be obtained in the same manner as in FIGS.
- Chromatography is not limited to 3D chromatography, and a FP composed of a peak excluding the UV spectrum and its retention time can also be used. In this case, it can be performed in the same manner as in the above embodiment except for the degree of coincidence of UV spectra.
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Automation & Control Theory (AREA)
- General Engineering & Computer Science (AREA)
- Library & Information Science (AREA)
- Crystallography & Structural Chemistry (AREA)
- Medicinal Chemistry (AREA)
- Pharmacology & Pharmacy (AREA)
- Food Science & Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Epidemiology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
Description
[多成分薬剤の評価装置]
図1は、多成分薬剤の評価装置のブロック図、図2は、多成分薬剤の評価手順を示すブロック図、図3は、3Dクロマトから作成したFPの説明図、図4(A)は、薬剤A、(B)は、薬剤B、(C)は、薬剤CのFPである。
[ピーク・パターン処理の動作原理]
図5~図69は、前記基準FP選定部33、ピーク・パターン作成部35、ピーク帰属部37、及び対象FPピーク特徴量作成部7の動作原理を説明するものである。
前記基準FP選定部33の機能を、図5~図9を用いてさらに説明する。
RPfg = {1-(m/(a+b-m))}×(a-m+1)
として算出する。
前記ピーク・パターン作成部35の機能を、図10~図67を用いてさらに説明する。
パターンのピーク・パターンが作成される。
P_Sim(73-95) = (|p1-p2|+1)×(|(r1-(r2+d)|+1)
+(|dn1-fn1|+1)×(|(cn1-r1)-(en1-r2)|+1)
+(|dn2-fn2|+1)×(|(cn2-r1)-(en2-r2)|+1)
として算出する。
P_Sim(73-95) = (|p1-p2|+1)×(|(r1-(r2+d)|+1)
+(|dn1-fn1|+1)×(|(cn1-r1)-(en1-r2)|+1)
+(|dn2-fn2|+1)×(|(cn2-r1)-(en2-r2)|+1)
+(|dn3-fn3|+1)×(|(cn3-r1)-(en3-r2)|+1)
+(|dn4-fn4|+1)×(|(cn4-r1)-(en4-r2)|+1)
として算出する。
UV_Sim(73-95)= RMSD(135 vs 139)
として算出する。
RMSD = √{ Σdis2 / n }
として算出する。
SCORE(73-95)
= P_Sim_min(73-95)×UV_Sim(73-95)
として算出する。
[FP領域分割特徴量作成の動作原理]
図70~図86は、FP領域分割特徴量作成の動作原理を示し、図70は、領域分割による数量化を示す説明図、図71は、リテンション・タイム等の変動との関係を示す説明図、図72は、領域の位置を変更して数量化する説明図、図73は、FPタイプ2のデータを示す図表、図74は、FPタイプ2のパターンを示す説明図、図75は、縦・横分割線での領域分割による領域ごとの特徴量化を示す説明図、図76は、縦分割線(1本目)の設定を示す説明図、図77は、横分割線(1本目)の設定を示す説明図、図78は、縦・横分割線による領域分割を示す説明図、図79は、特徴量化する領域の数を示す説明図、図80は、領域1の特定を示す説明図、図81は、全ピークの高さ及び合計を示す図表、図82は、領域1のピーク高さの合計を示す説明図、図83は、最初の1パターンによる全領域の特徴量を示す図表、図84は、縦1本目の位置を順次変更してできた各領域での特徴量を示す図表、図85は、横1本目の位置を順次変更してできた各領域での特徴量を示す図表、図86は、各縦・横分割線の位置を変更しない1通りでの特徴量を示す図表である。
そこで、基準FPタイプ2の場合には、図72のように、各格子145の位置を変更(シフト)し変更前後で数量化する。この操作により基準FP領域分割特徴量を正確に作成することが可能となる。各縦・横分割線141、143を設定範囲内で平行移動させるように位置を変更設定することで前記各格子145の位置を変更する。ここで、各格子145の位置を変更した数量化について、さらに説明する。
縦分割線(1本目)の位置を設定するため、図76のように、1本目のリテンション・タイム(RT)、振幅、刻みを指定する。
(i=0、1、2、・・・・、刻み-1)
例えば、RT=1、振幅=1、刻み回数=10 に指定すると、
縦分割線(1本目)= 0.0、 0.2、 0.4、 0.6、 0.8、 1.0、 1.2、 1.4、 1.6、 1.8
が設定される。
横(1本目)の位置を設定するため、図77のように、1本目の高さ、振幅、刻みを指定する。
横分割線(1本目)=高さ-振幅+(振幅×2/刻み)×i
(i=0、1、2、・・・・、刻み-1)
例えば、高さ=1、振幅=0.5、刻み回数=10 に指定すると、
横分割線(1本目)= 0.5、 0.6、 0.7、 0.8、 0.9、 1.0、 1.1、 1.2、 1.3、 1.4
が設定される。
設定された縦・横分割線(1本目)の全ての組み合わせで、順次2本目以降の標本線を設定し、領域分割する。
(0.0、 0.2、 0.4、 0.6、 0.8、 1.0、 1.2、 1.4、 1.6、 1.8)
×(0.5、 0.6、 0.7、 0.8、 0.9、 1.0、 1.1、 1.2、 1.3、 1.4)=100通り
この100通り全て組み合わせで順次2本目以降の分割線を設定し、領域を分割する。
縦分割線2本目以降は指定した間隔(等差)で指定した本数になるまで設定する。
横分割線2本目以降は指定した間隔(等比)で指定した本数になるまで設定する。
(i=2、・・・・、指定した本数)
例えば、縦1本目=0.0、縦間隔=10、縦本数=7、横分割線1本目=0.5、横間隔=1、横本数=6 の場合、
縦分割線=0、10、20、30、40、50、60
横分割線=0.5、1.5、3.5、7.5、15.5、31.5
に設定される。
設定された縦と横の分割線を先程の例をもとFP上に表記すると、図78のようになる。
各領域は、次の式により特徴量化する。
(特徴量化の方法)
以下に、上の式により、図80に示すd202の領域1の特徴量を求める。
特徴量 = 2 / 15.545472=0.128655
となる。
上記特徴量化方法により、最初の1パターンによる全領域の特徴量を算出する。図83に算出結果を示す。
(縦分割線1本目を順次変更して特徴量化)
縦分割線1本目の位置を順次変更してできた各領域を上記方法で特徴量化する。図84に結果を示す。
横分割線1本目の位置を変更するたびに縦1本目を1通り変更。できた各領域を上記方法で特徴量化する。図85に結果を示す。
100行(100通り)×31列(ファイル名+30特徴量)
のデータとなる。
これまでの処理を全基準データで実施する。例えば、基準データがd202、 d207、 d208の3データであった場合は、
300行(100通り×3データ)×31列(ファイル名+30特徴量)
となる。
対象FPタイプ2では、縦・横分割線(1本目)の組み合わせは、1通り(縦(RT)=1、横(高さ)=1)となるので、この1通りでの特徴量を算出する。
[MD値]
図87~図91は、前記のように、評価部27に係る各種対象FPとその評価値(MD値)を示した図であり、前記のように各対象FPを前記のように帰属処理することで、評価部27にて、上記MT法によりMD値(MD値:0.26、2.20等)を求めることができる。
[多成分薬剤の評価方法]
図92は、本発明実施例1のFPの特徴量作成方法を含む本発明実施例1の多成分薬剤の評価方法を示す工程図である。
[多成分薬剤の評価プログラム]
図93~図108は、多成分薬剤の評価プログラムに係るフローチャート、図109~図116は、基準データの作成に係るフローチャート、図117は、3Dクロマトのデータ例を示す図表、図118は、ピーク情報のデータ例を示す図表、図119は、FPのデータ例を示す図表、図120は、対象FPの基準FPへの帰属スコア計算結果(判定結果ファイル)例を示す図表、図121は、対象FPと基準FPで対応するピークの照合過程で作成する2つの中間ファイル(帰属候補ピークスコア表、帰属候補ピーク番号表)例を示す図表、図122は、対象FPと基準FPで対応するピークを特定した結果である照合結果ファイル例を示す図表、図123は、基準群FPのデータ例を示す図表、図124は、基準群FPに帰属した対象FPのピーク特徴量データのファイル例を示す図表、図125は、対象及び基準FPタイプ2のデータ例を示す図表、図126は、対象FP領域分割特徴量ファイル例を示す図表、図127は、対象FP特徴量統合ファイル例を示す図表、図128は、基準type2群FP例を示す図表、図129は、基準群統合データ例を示す図表である。
[S1:FP作成処理(単一波長のみ利用)]
図95は、図93ステップS1「FP作成処理」の単一波長のピーク情報を利用した場合のフローチャートである。
[S1:FP作成処理(複数波長利用)]
図96、図97は、図93のステップS1「FP作成処理」において、前記単一波長のピーク情報に代え、複数波長のピーク情報を利用した場合のフローチャートである。例えば203nmを含めて、検出波長軸方向に複数(n個)の波長を選択し、FPを作成する場合である。
[S2:対象FP帰属処理1]
図98は、図93ステップS2の「対象FP帰属処理1」の詳細を示すフローチャートである。この処理は、帰属の前処理であり、正常品とされた複数の基準FPから対象FP43の帰属に適した基準FPを選定する。
[S3:対象FP帰属処理2]
図99は、図93ステップS3の「対象FP帰属処理2」の詳細を示すフローチャートである。この処理は、帰属の本処理であり、対象FP43とステップS2で選定した基準FPとの間で、前記のようなピーク・パターン及びUVスペクトルの一致度から各帰属候補ピークの一致度(SCORE)を算出する。
SCORE = UV_Sim × P_Sim_min
として算出し、ステップS310へ移行する。
[S4:対象FP帰属処理3]
図100は、図93ステップS4の「対象FP帰属処理3」の詳細を示すフローチャートである。この処理は、帰属の後処理であり、前記のように算出した帰属候補ピークの一致度(SCORE)から対象FPの各ピークに対応する基準FPのピークを特定する。
補正値 = k1+(k2-k1)*(t0-t1)/(t2-t1)
k1:補正が必要な対象FPのピーク近傍で帰属された2つの基準FP側のピークのうちのリテンション・タイムが小さいピークのリテンション・タイム
k2:補正が必要な対象FPのピーク近傍で帰属された2つの基準FP側のピークのうちのリテンション・タイムが大きいピークのリテンション・タイム
t0:補正が必要な対象FPのピークのリテンション・タイム
t1:補正が必要な対象FPのピーク近傍で帰属された2つの対象FP側のピークのうちのリテンション・タイムが小さいピークのリテンション・タイム
t2:補正が必要な対象FPのピーク近傍で帰属された2つの対象FP側のピークのうちのリテンション・タイムが大きいピークのリテンション・タイム
として基準FPにおけるリテンション・タイムに補正し、ステップS411へ移行する。
[S5:対象FP帰属処理4]
図101、図102は、図93ステップS5の「対象FP帰属処理4」の詳細を示すフローチャートである。この処理は、帰属の最終処理であり、図93ステップS4で作成した照合結果ファイル(図122の照合結果ファイル例195参照)をもとに対象FPの各ピークを基準群FP(図123の基準群FPのデータ例197参照)のピークに帰属する。
図103は、図98の「基準FP選定処理」における「サブルーチン1」の詳細を示すフローチャートである。この処理は、FP間(例えば、対象FPと基準FP)のリテンション・タイム・出現パターンの一致度を計算する。
RPfg = (1-(m/(a+b-m)))×(a-m+1)
として算出し、ステップS1009へ移行する。
[サブルーチン2]
図104は、図99の「対象FP帰属処理2」における「サブルーチン2」の詳細を示すフローチャートである。この処理は、UVスペクトルの一致度を計算する。
d = b-c
z = z+d2
として算出し、ステップS2007へ移行する。
UV_Sim = √(z/a)
として算出し、このUV_Simを図99のステップS306に渡し、UVスペクトルの一致度計算処理を終了する。
[サブルーチン3]
図105は、図99の「対象FP帰属処理2」における「サブルーチン3」の詳細を示すフローチャートである。この処理は、ピーク・パターンの一致度を計算する。
P_Sim= (|p1-p2|+1)×(|(r1-(r2+d)|+1)
+(|dn1-fn1|+1)×(|(cn1-r1)-(en1-r2)|+1)
+(|dn2-fn2|+1)×(|(cn2-r1)-(en2-r2)|+1)
+(|dn3-fn3|+1)×(|(cn3-r1)-(en3-r2)|+1)
+(|dn4-fn4|+1)×(|(cn4-r1)-(en4-r2)|+1)
として算出し、ステップS3012へ移行する。
[S6:対象FPタイプ2の作成処理]
図106は、図93ステップS6の「FP_type2作成」の詳細を示すフローチャートである。
[S7:領域分割による対象FP_type2の特徴量化処理]
図107は、図94ステップS7の「領域分割による対象FP_type2の特徴量化処理」の詳細を示すフローチャートである。
る。この処理により、対象FPタイプ2のファイルが読み込まれ、ステップS704へ移行する。
[S8:ピーク・データ特徴量と領域分割特徴量の統合]
図108は、図94ステップS8の「ピーク・データ特徴量と領域分割特徴量の統合」の詳細を示すフローチャートである。
[基準FP帰属結果特徴量統合ファイルの作成]
対象FP特徴量統合データを基準FP特徴量統合データと比較するための基準FP特徴量統合ファイルは、図109~図116のように作成される。
[S10005:基準FP対応表の作成]
図111、図112は、図110ステップS10005の「基準FP帰属結果統合処理(基準FP対応表の作成)」の詳細を示すフローチャートである。
[S10006:ピーク特徴量化処理]
図113は、図109ステップS10006の「ピーク特徴量化処理(基準群FPの作成)」の詳細を示すフローチャートである。
[S10007:基準FPタイプ2の作成処理]
図114は、図110ステップS10007の「基準FP編集処理(基準FP_type2
の作成)」の詳細を示すフローチャートである。
[S10008:領域分割による基準FP_type2の特徴量化処理]
図115は、図110ステップS10008の「領域分割による基準FP_type2の特徴量化処理」の詳細を示すフローチャートである。
[S10009:基準データ作成処理]
図116は、図110ステップS10009の「基準データの作成処理」の詳細を示すフローチャートである。
[実施例1の効果]
本発明実施例1の多成分物質の評価方法では、前記FP作成工程148と、対象FPピーク帰属工程149と、対象FPピーク特徴量作成工程151と、対象FPタイプ2作成工程153と、対象FP領域分割特徴量作成工程155と、対象FP特徴量統合工程157と、基準FPピーク帰属工程159と、基準FP帰属結果統合工程161と、基準FPピーク特徴量作成工程163と、基準FPタイプ2作成工程165と、基準FP領域分割特徴量作成工程167と、基準FP特徴量統合工程169と、評価工程171とを備えている。
[ピーク・パターンの一致度計算(P_Sim)の変形例]
図63、図64、図105でのピーク・パターンの一致度計算(P_Sim)は、FPをピーク高さで作成した上記実施例の場合について適用し、比較対象のピーク高さの差に基づいて計算した。
n=2の場合
P_Sim= (p1/p2♯1)×(|(r1-(r2+d)|+1)
+(dn1/fn1♯1)×(|(cn1-r1)-(en1-r2)|+1)
+(dn2/fn2♯1)×(|(cn2-r1)-(en2-r2)|+1)
n=4の場合
P_Sim= (p1/p2♯1)×(|(r1-(r2+d)|+1)
+(dn1/fn1♯1)×(|(cn1-r1)-(en1-r2)|+1)
+(dn2/fn2♯1)×(|(cn2-r1)-(en2-r2)|+1)
+(dn3/fn3♯1)×(|(cn3-r1)-(en3-r2)|+1)
+(dn4/fn4♯1)×(|(cn4-r1)-(en4-r2)|+1)
ここに、#1は、比較対象の2つの値の比(大きい値/小さい値) であることを示している。
[サブルーチン2の変形例]
図130は、図104に代えて適用するサブルーチン2の変形例に係り、図99の「対象FP帰属処理2」における「サブルーチン2」の変形例の詳細を示すフローチャートである。この変形例に係る処理により、UVスペクトルの一致度を計算する。
UV_Sim = √(z/a)×1.1DNS
として算出し、このUV_Simを図81のステップS306に渡し、UVスペクトルの一致度計算処理を終了する。
[その他]
本発明実施例のパターン又はFPの特徴量作成方法、作成プログラム、及び作成装置では、FPをピーク面積で作成するときは、シグナル強度軸を面積値軸、シグナル強度を面積値として同様に適用することができる。
3 FP作成部
5 対象FPピーク帰属部
7 対象FPピーク特徴量作成部
9 対象FPタイプ2作成部
11 対象FP領域分割特徴量作成部
13 対象FP特徴量統合部
15 基準FPピーク帰属部
17 基準FP帰属結果統合部
19 基準FPピーク特徴量作成部
21 基準FPタイプ2作成部
23 基準FP領域分割特徴量作成部
25 基準FP特徴量統合部
27 評価部
31 基準FP作成部
33 基準FP選定部
35 ピーク・パターン作成部
37 ピーク帰属部
39 漢方薬
41 3Dクロマト
42 対象FPに含まれるピークのUVスペクトル
43 対象FP
45 基準群FP
47 基準群FPに帰属した対象FP
49 対象FPタイプ2(FP)
51 対象FP領域分割特徴量
53 対象FPの評価結果
55 薬剤AのFP
57 薬剤BのFP
59 薬剤CのFP
61 対象FP(リテンション・タイム10.0-14.5分)
63、65、67、69、71、73、75、77、79、81 対象FP(リテンション・タイム10.0-14.5分)中の各ピーク
83 基準FP(リテンション・タイム10.0-14.5分)
85、87、89、91、93、95、97、99、101、103、105 基準FP(リテンション・タイム10.0-14.5分)中の各ピーク
107 対象FPリテンション・タイム・出現パターン
109 基準FPリテンション・タイム・出現パターン
111 リテンション・タイム・出現距離の一致数
113 リテンション・タイム・出現パターンの一致度
115 対象FP帰属対象ピークのピーク・パターン(3本)
117、119、121、123 基準FP帰属候補ピークのピーク・パターン(3本)125 対象FP帰属対象ピークのピーク・パターン(5本)
127、129、131、133 基準FP帰属候補ピークのピーク・パターン(5本)135 帰属対象ピークのUVスペクトル
139 帰属候補ピークのUVスペクトル
141 縦領域分割線
143 横領域分割線
145 縦・横領域分割線により分割した各領域(格子)
147 各領域をピーク高さ
148 FP作成工程
149 対象FPピーク帰属工程
151 対象FPピーク特徴量作成工程
153 対象FPタイプ2作成工程
155 対象FP領域分割特徴量作成工程(パターン領域分割特徴量作成工程、FP領域分割特徴量作成工程)
157 対象FP特徴量統合工程
159 基準FPピーク帰属工程
161 基準FP帰属結果統合工程
163 基準FPピーク特徴量作成工程
165 基準FPタイプ2作成工程
167 基準FP領域分割特徴量作成工程(パターン領域分割特徴量作成工程、FP領域分割特徴量作成工程)
169 基準FP特徴量統合工程
171 評価工程
173 対象FP作成工程
175 基準FP作成工程
177 基準FP選定工程
179 ピーク・パターン作成工程
181 ピーク帰属工程
183 3Dクロマト・データ例
185 ピーク情報データ例
187 FPデータ例
189 判定結果ファイル例
191 帰属候補ピークスコア表例
193 帰属候補ピーク番号表例
195 照合結果ファイル例
197 基準群FPデータ例
199 対象FPピーク特徴量ファイル例
201 FPタイプ2データ例
203 対象FP領域分割特徴量ファイル例
205 対象FP統合特徴量ファイル例
207 基準type2群FPデータ例
209 基準群統合データ例
Claims (24)
- 時系列でピークが変化するパターンを複数の領域に分割し各領域に存在するピークの存在率又は存在量からパターン領域分割特徴量を作成するパターン領域分割特徴量作成工程、
を備えたことを特徴とするパターンの特徴量作成方法。 - 多成分物質のクロマトから検出されたピークとそのリテンション・タイムとで構成されるFPを複数の領域に分割し各領域に存在するピークの存在率又は存在量からFP領域分割特徴量を作成するFP領域分割特徴量作成工程、
を備えたことを特徴とするFPの特徴量作成方法。 - 請求項2記載のFPの特徴量作成方法であって、
前記多成分物質は、多成分薬剤である、
ことを特徴とするFPの特徴量作成方法。 - 請求項3記載のFPの特徴量作成方法であって、
前記多成分薬剤は、生薬、生薬の組合せ、それらの抽出物、漢方薬の何れかである、
ことを特徴とするFPの特徴量作成方法。 - 請求項2~4の何れかに記載のFPの特徴量作成方法であって、
前記FP領域分割特徴量作成工程は、シグナル強度軸又は面積値軸に平行な複数の縦分割線と時間軸に平行な複数の横分割線とにより前記領域の分割を行う、
ことを特徴とするFPの特徴量作成方法。 - 請求項5記載のFPの特徴量作成方法であって、
前記複数の横分割線は、シグナル強度又は面積値が増大する方向へ等比間隔で設定された、
ことを特徴とするFPの特徴量作成方法。 - 請求項2~6の何れかに記載のFPの特徴量作成方法であって、
前記FP領域分割特徴量作成工程は、前記各領域の位置を変更し変更前後で前記FP領域分割特徴量を作成する、
ことを特徴とするFPの特徴量作成方法。 - 請求項7記載のFPの特徴量作成方法であって、
前記FP領域分割特徴量作成工程は、前記各縦・横分割線を設定範囲内で平行移動させるように位置を変更設定することで前記各領域の位置を変更する、
ことを特徴とするFPの特徴量作成方法。 - 時系列でピークが変化するパターンを複数の領域に分割し各領域に存在するピークの存在率又は存在量からパターン領域分割特徴量を作成するパターン領域分割特徴量作成機能、
をコンピュータに実現させることを特徴とするパターンの特徴量作成プログラム。 - 多成分物質のクロマトから検出されたピークとそのリテンション・タイムとで構成されるFPを複数の領域に分割し各領域に存在するピークの存在率又は存在量からFP領域分割特徴量を作成するFP領域分割特徴量作成機能、
を備えたことを特徴とするFPの特徴量作成プログラム。 - 請求項10記載のFPの特徴量作成プログラムであって、
前記多成分物質は、多成分薬剤である、
ことを特徴とするFPの特徴量作成プログラム。 - 請求項11記載のFPの特徴量作成プログラムであって、
前記多成分薬剤は、生薬、生薬の組合せ、それらの抽出物、漢方薬の何れかである、
ことを特徴とするFPの特徴量作成プログラム。 - 請求項10~12の何れかに記載のFPの特徴量作成プログラムであって、
前記FP領域分割特徴量作成機能は、シグナル強度軸又は面積値軸に平行な複数の縦分割線と時間軸に平行な複数の横分割線とにより前記領域の分割を行う、
ことを特徴とするFPの特徴量作成プログラム。 - 請求項13記載のFPの特徴量作成プログラムであって、
前記複数の横分割線は、シグナル強度又は面積値が増大する方向へ等比間隔で設定された、
ことを特徴とするFPの特徴量作成プログラム。 - 請求項10~14の何れかに記載のFPの特徴量作成プログラムであって、
前記FP領域分割特徴量作成機能は、前記各領域の位置を変更し変更前後で前記FP領域分割特徴量を作成する、
ことを特徴とするFPの特徴量作成プログラム。 - 請求項15記載のFPの特徴量作成プログラムであって、
前記FP領域分割特徴量作成機能は、前記各縦・横分割線を設定範囲内で平行移動させるように位置を変更設定することで前記各領域の位置を変更する、
ことを特徴とするFPの特徴量作成プログラム。 - 時系列でピークが変化するパターンを複数の領域に分割し各領域に存在するピークの存在率又は存在量からパターン領域分割特徴量を作成するパターン領域分割特徴量作成部、
を備えたことを特徴とするパターンの特徴量作成装置。 - 多成分物質のクロマトから検出されたピークとそのリテンション・タイムとで構成されるFPを複数の領域に分割し各領域に存在するピークの存在率又は存在量からFP領域分割特徴量を作成するFP領域分割特徴量作成部、
を備えたことを特徴とするFPの特徴量作成装置。 - 請求項18記載のFPの特徴量作成装置であって、
前記多成分物質は、多成分薬剤である、
ことを特徴とするFPの特徴量作成装置。 - 請求項19記載のFPの特徴量作成装置であって、
前記多成分薬剤は、生薬、生薬の組合せ、それらの抽出物、漢方薬の何れかである、
ことを特徴とするFPの特徴量作成装置。 - 請求項18~20の何れかに記載のFPの特徴量作成装置であって、
前記FP領域分割特徴量作成部は、シグナル強度軸又は面積値軸に平行な複数の縦分割線と時間軸に平行な複数の横分割線とにより前記領域の分割を行う、
ことを特徴とするFPの特徴量作成装置。 - 請求項21記載のFPの特徴量作成装置であって、
前記複数の横分割線は、シグナル強度又は面積値が増大する方向へ等比間隔で設定された、
ことを特徴とするFPの特徴量作成装置。 - 請求項18~22の何れかに記載のFPの特徴量作成装置であって、
前記FP領域分割特徴量作成部は、前記各領域の位置を変更し変更前後で前記FP領域分割特徴量を作成する、
ことを特徴とするFPの特徴量作成装置。 - 請求項23記載のFPの特徴量作成装置であって、
前記FP領域分割特徴量作成部は、前記各縦・横分割線を設定範囲内で平行移動させるように位置を変更設定することで前記各領域の位置を変更する、
ことを特徴とするFPの特徴量作成装置。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP12791963.7A EP2717045B1 (en) | 2011-06-01 | 2012-05-31 | Creation method, creation program, and creation device for characteristic amount of pattern or fingerprint |
KR1020127032279A KR101442117B1 (ko) | 2011-06-01 | 2012-05-31 | 패턴 또는 fp의 특징량작성방법, 작성프로그램, 및 작성장치 |
CN201280001661.5A CN102959395B (zh) | 2011-06-01 | 2012-05-31 | Fp的特征值作成方法以及作成装置 |
US13/806,725 US20130204539A1 (en) | 2011-06-01 | 2012-05-31 | Feature value preparing method, feature value preparing program, and feature value preparing device for pattern or fp |
HK13108497.6A HK1181119A1 (zh) | 2011-06-01 | 2013-07-19 | 的特徵值作成方法以及作成裝置 |
US15/261,462 US10605792B2 (en) | 2011-06-01 | 2016-09-09 | Method of and apparatus for formulating multicomponent drug |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011-123849 | 2011-06-01 | ||
JP2011123849 | 2011-06-01 |
Related Child Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US13/806,725 A-371-Of-International US20130204539A1 (en) | 2011-06-01 | 2012-05-31 | Feature value preparing method, feature value preparing program, and feature value preparing device for pattern or fp |
US15/261,462 Continuation-In-Part US10605792B2 (en) | 2011-06-01 | 2016-09-09 | Method of and apparatus for formulating multicomponent drug |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012164956A1 true WO2012164956A1 (ja) | 2012-12-06 |
Family
ID=47258825
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/003618 WO2012164956A1 (ja) | 2011-06-01 | 2012-05-31 | パターン又はfpの特徴量作成方法、作成プログラム、及び作成装置 |
Country Status (8)
Country | Link |
---|---|
US (2) | US20130204539A1 (ja) |
EP (1) | EP2717045B1 (ja) |
JP (1) | JP5912880B2 (ja) |
KR (1) | KR101442117B1 (ja) |
CN (1) | CN102959395B (ja) |
HK (1) | HK1181119A1 (ja) |
TW (1) | TW201312109A (ja) |
WO (1) | WO2012164956A1 (ja) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6891459B2 (ja) * | 2016-11-16 | 2021-06-18 | 富士フイルムビジネスイノベーション株式会社 | 通信装置 |
CN109001354B (zh) * | 2018-05-30 | 2020-09-04 | 迈克医疗电子有限公司 | 波峰识别方法和装置、色谱分析仪及存储介质 |
US11977085B1 (en) | 2023-09-05 | 2024-05-07 | Elan Ehrlich | Date rape drug detection device and method of using same |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004264051A (ja) * | 2003-02-07 | 2004-09-24 | Shinichi Usui | リポタンパク質の分析方法及び分析プログラム |
JP2006170647A (ja) * | 2004-12-13 | 2006-06-29 | Intec Web & Genome Informatics Corp | ピーク抽出方法およびピーク抽出装置 |
JP2007085868A (ja) * | 2005-09-21 | 2007-04-05 | Asahi Breweries Ltd | 飲食品中のポリフェノールの定量分析方法、定量分析装置および飲食品の設計方法 |
JP2008541095A (ja) * | 2005-05-12 | 2008-11-20 | ウオーターズ・インベストメンツ・リミテツド | 化学分析データの視覚化 |
JP2010118053A (ja) * | 2008-10-23 | 2010-05-27 | Microsoft Corp | 反復処理 |
Family Cites Families (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3023664A (en) * | 1959-12-03 | 1962-03-06 | Coleman Instr Inc | Chromatographic reader |
JPH01178864A (ja) * | 1988-01-07 | 1989-07-17 | Takara Shuzo Co Ltd | クロマトグラム又はスペクトルの解析方法 |
GB0016459D0 (en) * | 2000-07-04 | 2000-08-23 | Pattern Recognition Systems As | Method |
JP4886933B2 (ja) * | 2001-01-12 | 2012-02-29 | カウンセル オブ サイエンティフィック アンド インダストリアル リサーチ | クロマトグラフフィンガープリントならびに単一の医薬および処方物の標準化のための新規な方法 |
CN100356380C (zh) * | 2001-02-13 | 2007-12-19 | 科学与工业研究会 | 一种色谱指纹图谱和单一药物和制剂标准化的新方法 |
JP4185933B2 (ja) | 2003-03-31 | 2008-11-26 | 株式会社メディカル・プロテオスコープ | 試料解析方法及び試料解析プログラム |
US7178386B1 (en) * | 2003-04-10 | 2007-02-20 | Nanostream, Inc. | Parallel fluid processing systems and methods |
WO2005073713A2 (en) * | 2004-01-28 | 2005-08-11 | Council Of Scientific And Industrial Research | A method for standardization of chemical and therapeutic values of foods & medicines using animated chromatographic fingerprinting |
US8428883B2 (en) * | 2004-06-07 | 2013-04-23 | Tsumura & Co. | Multi-component medicine evaluation method |
EP1812079A4 (en) * | 2004-10-15 | 2012-07-25 | Spectral Dimensions Inc | EVALUATION OF PHARMACEUTICAL MIXTURES |
JP2007315941A (ja) * | 2006-05-26 | 2007-12-06 | Univ Of Miyazaki | 植物品種判定装置、植物品種判定方法及び植物品種判定用プログラム |
JP4837520B2 (ja) * | 2006-10-16 | 2011-12-14 | アングルトライ株式会社 | スペクトル波形パターンの領域分割方法およびプログラム |
JP4905265B2 (ja) * | 2007-06-18 | 2012-03-28 | 株式会社島津製作所 | クロマトグラフ質量分析装置用データ処理装置 |
SG186302A1 (en) * | 2010-06-16 | 2013-02-28 | Abbott Lab | Comparison of protein samples |
-
2012
- 2012-05-31 EP EP12791963.7A patent/EP2717045B1/en active Active
- 2012-05-31 WO PCT/JP2012/003618 patent/WO2012164956A1/ja active Application Filing
- 2012-05-31 JP JP2012125382A patent/JP5912880B2/ja active Active
- 2012-05-31 CN CN201280001661.5A patent/CN102959395B/zh active Active
- 2012-05-31 KR KR1020127032279A patent/KR101442117B1/ko active IP Right Grant
- 2012-05-31 US US13/806,725 patent/US20130204539A1/en not_active Abandoned
- 2012-06-01 TW TW101119807A patent/TW201312109A/zh unknown
-
2013
- 2013-07-19 HK HK13108497.6A patent/HK1181119A1/zh unknown
-
2016
- 2016-09-09 US US15/261,462 patent/US10605792B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004264051A (ja) * | 2003-02-07 | 2004-09-24 | Shinichi Usui | リポタンパク質の分析方法及び分析プログラム |
JP2006170647A (ja) * | 2004-12-13 | 2006-06-29 | Intec Web & Genome Informatics Corp | ピーク抽出方法およびピーク抽出装置 |
JP2008541095A (ja) * | 2005-05-12 | 2008-11-20 | ウオーターズ・インベストメンツ・リミテツド | 化学分析データの視覚化 |
JP2007085868A (ja) * | 2005-09-21 | 2007-04-05 | Asahi Breweries Ltd | 飲食品中のポリフェノールの定量分析方法、定量分析装置および飲食品の設計方法 |
JP2010118053A (ja) * | 2008-10-23 | 2010-05-27 | Microsoft Corp | 反復処理 |
Non-Patent Citations (6)
Title |
---|
"Mathematics for Quality Engineering", 2000, JAPANESE STANDARDS ASSOCIATION, pages: 136 - 138 |
"Quality Engineering of Application Course", 1999, JAPANESE STANDARDS ASSOCIATION, article "Technical Developments in Chemistry, Pharmacy and Biology", pages: 454 - 456 |
INTRODUCTION TO MT SYSTEM, 2008 |
PHARMACEUTICALS MONTHLY, vol. 28, no. 3, 1986, pages 67 - 71 |
QUALITY ENGINEERING, vol. 11, no. 5, 2003, pages 78 - 84 |
See also references of EP2717045A4 |
Also Published As
Publication number | Publication date |
---|---|
EP2717045B1 (en) | 2021-08-11 |
EP2717045A1 (en) | 2014-04-09 |
CN102959395A (zh) | 2013-03-06 |
JP5912880B2 (ja) | 2016-04-27 |
US10605792B2 (en) | 2020-03-31 |
HK1181119A1 (zh) | 2013-11-01 |
KR101442117B1 (ko) | 2014-09-18 |
US20170007502A1 (en) | 2017-01-12 |
KR20130029406A (ko) | 2013-03-22 |
TW201312109A (zh) | 2013-03-16 |
CN102959395B (zh) | 2016-03-30 |
EP2717045A4 (en) | 2014-11-26 |
JP2013011598A (ja) | 2013-01-17 |
TWI561818B (ja) | 2016-12-11 |
US20130204539A1 (en) | 2013-08-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5915540B2 (ja) | ピーク帰属方法、帰属プログラム、及び帰属装置 | |
JP5912880B2 (ja) | パターン又はfpの特徴量作成方法、作成プログラム、及び作成装置 | |
JP5915539B2 (ja) | パターンの評価方法、多成分物質の評価方法、評価プログラム、及び評価装置 | |
JP5954180B2 (ja) | パターンの評価方法、多成分物質の評価方法、評価プログラム、及び評価装置 | |
JP5910506B2 (ja) | 集合データの類似性評価方法、類似性評価プログラム、及び類似性評価装置 | |
JP5895847B2 (ja) | Fp作成方法、作成プログラム、作成装置、及びfp |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201280001661.5 Country of ref document: CN |
|
ENP | Entry into the national phase |
Ref document number: 20127032279 Country of ref document: KR Kind code of ref document: A |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12791963 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13806725 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012791963 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |