JP2006330852A - Compound organic substance quality comparing device and system using biopattern sensing, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a compound organic substance quality comparing device and system which is useful for comparing the features of a compound organic substance such as a fermentation object, to stabilize and improve product quality through the objective evaluation of the compound organic substance, and to provide a recording medium therefor. <P>SOLUTION: Each of the compound organic substance quality comparing device and system is provided with a detection means, a storage means, a calling means, an arithmetic means and an output means. Each of the compound organic substance quality comparing device and system is provided with an input means, a storage means, a calling means, an arithmetic means and an output means. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a composite organic matter quality comparison apparatus and system and a recording medium useful for comparing properties of fermented products and composite composition organic matter.

  In general, the means for comparing the properties of complex composition organic substances such as fermented products are roughly classified into sensory discrimination such as odor, viscosity and color tone, and scientific and objective methods based on analysis.

  Traditional evaluation methods, such as judging the evaluation of fermented products based solely on experience and intuition, are limited, and the objectivity of information (numerical and statistical confirmation of manufacturing methods) and universality (with other data) Poor comparison and feedback).

  On the other hand, although physical properties, chemical composition, and property comparisons based on analysis results of microorganisms are objective, many substances and microorganisms coexist in the composite composition organic matter, so analysis of specific targets In many cases, the comparison of fermented products cannot be made only with the results, and the tendency is particularly remarkable for intermediate products during fermentation.

From this point of view, comparative evaluation of fermented products etc. is extremely difficult, so there are very few domestic patents on the subject, and Patent Publication 2000-109386 “Composting method of organic waste and evaluation method of composted product” can be seen However, in the patent publication 2004-208533 “Fermentation Quality Evaluation System and Recording Media” filed prior to the present invention by the inventors of the present invention, individual substance-based / microorganism-based Instead of using analysis as a qualitative method, an innovative system has been devised, in which each substance / microorganism in the analysis results is regarded as a sensor, and fermented materials are classified and evaluated by their combined pattern. The methodology was shown.
JP 2000-109386 A (pages 1 to 11, FIGS. 1 to 2) JP 2004-208533 A (pages 1 to 13, FIGS. 1 to 4)

  The invention described in Patent Document 2 obtains the presence / absence and quantitative value of individual constituent microorganisms or chemical substances in a sample, and treats them as a set pattern of numerical data as a result. Operation that performs fixing, identification, and quantification is necessary, which causes high costs, and comparative analysis that includes fine differences in the entire waveform that are not included in the extraction target of numerical data is impossible .

  In view of the above defects, an object of the present invention is to eliminate such problems.

  The present invention has been proposed to achieve the above object, and comprises a quality comparison apparatus and system for composite organic matter samples, comprising a detection means, a storage means, a calling means, a calculation means, and an output means, And a quality comparison device and system for a composite organic matter sample, each comprising an input means, a storage means, a calling means, a calculation means, and an output means.

According to the process of the present invention, one input biopattern sensing data detected by sample direct or sample extract analysis is stored in a database, and any other biopattern sensing also stored in the database. Calculate affinity with data. Based on these values, the similarity of the evaluation target composite composition organic material is evaluated.

  Of the apparatus and system according to the present invention, the part relating to the biopattern sensing data detection means assumes practical use in a general-purpose analyzer, for example, in addition to a dedicated apparatus.

  Examples of general-purpose analyzers include HPLC (High Performance Liquid Chromatography), capillary electrophoresis devices, FT-IR (Fourier Transform Infrared Spectroscopy) devices, etc., but these examples are bio patterns by other analysis methods. It does not prevent the acquisition of sensing data.

  In the present invention, the practical use of a general-purpose analyzer utilizing a general-purpose electronic computer device (so-called personal computer) as the control means is also assumed.

  In such a general-purpose analyzer, the analysis conditions, data acquisition conditions, etc. are provided to the analyzer by the application execution program according to the present invention or program operation data, thereby causing the apparatus to execute an analysis suitable for the target analysis. It is possible to obtain appropriate data.

In the system according to the present invention, the input means, storage means, calling means, calculation means, and output means are assumed to be practically used in general general-purpose electronic computer devices (so-called personal computers) in addition to dedicated devices.
This electronic computer apparatus does not prevent the electronic computer apparatus from being shared with the above-described electronic computer apparatus for the general-purpose analyzer.

  The execution program of the present invention can be provided from a recording medium to a computer apparatus via a recording medium reader connected to the apparatus. Further, it is assumed that data input / selection and evaluation result output in the system of the present invention are performed on the screen of a monitor display connected to the computer apparatus or integrated with the computer apparatus.

  According to the present invention, for example, in the production management / evaluation method for pigs, it is possible to carry out growth management based on the properties of pig feces.

  For example, specifically, by targeting livestock feces immediately after excretion, which is considered to directly reflect the intestinal microflora, by performing a numerical analysis of the similarity between samples based on the present invention, It is possible to objectively perform grouping according to growth quality and verification of the establishment and effect of microbial materials.

  This makes it possible to easily examine objective quality differences between lots and between farms, and it is a powerful support tool for verifying the effects when abolishing antibacterial agents and introducing live bacteria and fermented feed. Through these, the domestic pig farming industry can be led to a safer and higher quality direction, and as a result, the advantage over imported meat can be increased.

  According to the present invention, for example, quality evaluation of fermented foods and fermented feeds such as fermented dairy products, alcoholic beverages, pickles and miso, and confirmation of the state of each fermentation stage in an energy conversion system such as composting process by recycling and methane fermentation Easy to do.

  In order to promote objective quality confirmation and comparison of fermented products, evaluation of appropriate manufacturing methods, new development and improvement of products, etc., inputs (raw materials) and outputs that have been regarded as black boxes so far are used. It is indispensable to establish a system that catches real-time fluctuations in the fermentation system corresponding to the middle of the product (product) and accurately evaluates the information.

  If the method of the present invention can be used to classify the state of fermentation and quantify the quality control, for example, it will unify and simplify the huge amount of data management that occurs in daily quality control, such as changes in fermentation state over time In addition, fine adjustment of operating conditions for maintaining and improving quality can be performed more accurately based on numerical indices.

  According to the method of the present invention, for example, biopattern sensing data of a sample collected from the normal operating condition of the decomposition / fermentation system of complex organic matter, temperature at the time of fermentation / decomposition, air permeability, type of substrate to be input and Compared with the biopattern sensing data of the sample at the time of operation, adjusting some of the physical decomposition and fermentation conditions such as quantitative ratio, substrate input and product extraction interval, residence time in the system, number of stirring, etc. The usage to verify the quality is listed.

  With this usage method, for example, examination of whether or not raw materials can be added or changed while maintaining quality, and examination of improving processing efficiency by shortening the processing time, etc., verification of quality accompanying changes in the operating conditions of the decomposition and fermentation system Can be implemented.

  In addition, for example, biopattern sensing data of a sample collected under the normal operating conditions of an existing decomposition / fermentation system can be obtained from a sample collected when a small-scale test fermentation apparatus developed for testing is operated under the same conditions. Compared with biopattern sensing data, its reproducibility can be verified.

  This utilization method verifies the reliability of the reproducibility test of the decomposition / fermentation system using the test fermenter, makes it possible to develop a test fermenter with higher accuracy and adjust the conditions for performing the reproducible test.

  By comparing the similarity of biopattern sensing data, a quality control method is provided mainly for the purpose of maintaining a specific quality in the degradation / fermentation system. Biopattern sensing data by adding a process to store various data related to qualitative and quantitative values and various data related to fermented product quality together with similarity data in a database device that can be stored and recalled for each sample. And similarity data can be used as an “index” of fermented product quality to manage and search data such as chemical state, resulting in a quality control method aimed at improving the quality of decomposition and fermentation systems. It becomes possible.

  The present invention aims at constructing an effective system having a plurality of merits so that it can be applied to various sample / fields and various analysis methods by a common method. Characterized in that it comprises a quality comparison apparatus and system for composite organic matter samples, and an input means, a storage means, a calling means, a computing means, and an output means. It was realized by providing a quality comparison device and system for composite organic matter samples, respectively.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a hardware configuration diagram of a composite organic matter quality comparison apparatus and system 1 that executes an embodiment of the present invention. The complex organic matter quality comparison device and system 1 includes an analysis device 3, an electronic computer device 5, an input device 7, a recording media reader 9, a program recording medium 11, a monitor display 13, a database 15, and the like.

  The program recording medium 11 is a recording medium such as a hard disk, a CD-ROM, a DVD-ROM, an MO disk, a flexible disk, or a memory card that stores the program according to the embodiment of the present invention, and is read by the recording medium reader 9. Provided in possible form.

  The database 15 stores sample identification data 31, biopattern sensing data 33, data similarity 35, and the like.

  According to the first embodiment of the present invention, biopattern sensing data reflecting the composition of some chemical substances in the complex organic material is input to the system and apparatus via the input device 7.

  The input device 7 is mainly used when introducing the result data obtained by analysis outside the device and system of the present invention into the system and device. Examples of biopattern sensing data to be input include electrophoresis images of gel electrophoresis, chromatograms, electropherograms, melting curve graphs, and spectral spectra.

  As the input device 7, for example, an image input device such as a scanner or a digital camera, an input device for an electrical signal from a sensor such as a spectrophotometer or various measuring devices, or an image or waveform already converted into electronic data, A device for capturing these corresponds to this.

  According to the first embodiment of the present invention, for example, the biopattern sensing data obtained from an independent analyzer outside the system can be input to the system, and the advantage of being able to determine the similarity of the composite organic matter can be enjoyed. .

  According to the second embodiment of the present invention, bio-pattern sensing data reflecting the composition of some chemical substances in the complex organic material is acquired by the analyzer 3.

  Examples of the analysis device 3 include a device having a mechanism for acquiring any of data such as a chromatogram, an electropherogram, a melting curve, and a spectroscopic spectrum reflecting an organic substance composition in a sample.

  According to the second embodiment of the present invention, by introducing a sample or an extract of the sample into the apparatus, biopattern sensing data reflecting the composition of some chemical substances in the composite organic material is used to obtain the similarity of the composite organic material. It is possible to enjoy the advantage that a series of processes up to sex determination can be performed in a lump.

  Both the bio-pattern sensing data input in the first embodiment of the present invention and the bio-pattern sensing data contributed as analysis data in the second embodiment are recorded in the database 15 in an arbitrarily callable format and necessary. Accordingly, it is used for arithmetic processing.

  Next, processing of biopattern sensing data by the composite organic matter quality comparison system 1 will be described.

FIG. 2 shows a flowchart of the system processing of the composite organic matter quality comparison system 1. The composite organic matter quality comparison system 1 includes an image data conversion process 500 and a series of processing steps of a fermented product selection / similarity calculation process 501.
In the fermented product selection / similarity calculation process 501, a plurality of composite organic matter samples to be evaluated are selected based on the sample identification data 31 stored in the database 15 (step 201).

  Subsequently, biopattern sensing data 33 corresponding to the selected sample identification data 31 stored in the database 15 is acquired (step 202).

  Subsequently, the similarity calculation between the biopattern sensing data is performed (step 203), and the result is classified for each sample and output to the monitor display 13 (step 204) and at the same time the database 15 as the data similarity 35. Stored.

  The similarity calculation result output to the monitor display 13 and the data similarity 35 stored in the database 15 are, for example, a numerical value of 0 or more and 1 or less indicating that the similarity between data is higher as it is closer to 1, or 0. It can be respectively indicated by a numerical value of 0 or more indicating that the similarity between the data is higher as it is closer to.

  The method of using information related to DNA composition as biopattern sensing data has an advantage that data reflecting the dynamics of the complex microbial system itself constituting the complex organic matter can be obtained. For example, as a method to analyze the diversity of unknown bacterial communities using information related to DNA composition, PCR (Polymerase Chain Reaction) amplification of part or all of the 16s-rRNA gene and DGGE (Denaturing Gradient Gel Electrophoresis) And the like, and the like.

  According to the knowledge of the inventors, PCR-DGGE using primers targeting the V3 region (236 bp) of the 16s-rRNA gene is very difficult for complex organic substances such as cheese, pickles, fermented milk, and feces. It is clear that it gives good separation results. FIG. 3 shows an electrophoretic image obtained by this method.

  In the present invention, for example, gel-based electrophoresis data can be used as biopattern sensing data.

  The gel-based electrophoresis method can be carried out with a simple apparatus, but requires time for analysis, and has the following technical problems when comparing homology between samples as in the present invention.

  The first is a method for reading the homology of the bands. The inventors have read the homology of each band on the image manually by visual inspection, but this requires considerable labor and human error is unavoidable.

  The second is gel non-uniformity. Due to distortion at both ends of the gel and uneven stretching in the vertical direction, it is extremely difficult to determine the match for each band.

  The third is the large number of detection bands. In the case of fermented food and feces, a band of about 50 to 100 is detected in many cases. The diversity of the band itself means that a large amount of information is included, which is advantageous in later analysis, but at the same time, the distance between the bands is close, and a problem of determination error is likely to occur.

In order to eliminate these problems, the data conversion process 500 does not use, for example, the following method to extract gel electrophoresis information on a band basis after position correction by linear calculation between lanes. Assuming that distortion is included, luminance information in the lane migration axis direction is extracted as two-dimensional or one-dimensional plot information. By capturing all of this information as a series of data and performing similarity analysis, it is possible to perform high-precision analysis that takes advantage of the amount of information contained in the data and is less susceptible to errors. Become.

  FIG. 4 is an electrophoretic image obtained by subjecting a DNA extract from a swine stool sample to a PCR-DGGE method using a primer targeting the V3 region (236 bp) of the 16s-rRNA gene.

  First, in order to independently extract the electrophoresis lane corresponding to the sample, pixel values are detected along a column parallel to the gel migration axis direction (vertical direction in FIG. 4), and this is detected with respect to the gel migration axis. Then, it is sequentially performed in the entire region between both ends in the vertical direction (lateral direction in FIG. 4).

  Next, the pixel value information on the column is sequentially calculated in the horizontal direction. Examples of the calculation method include histogram processing, maximum value extraction, power calculation, and dispersion calculation. Here, an example of calculation by power calculation is shown.

  The power is the sum of squares of the pixel values. If 0 <i <N for the horizontal position i and 0 <j <M for the vertical position j, the power for each column pixel is expressed by Equation 1.

  FIG. 5 shows the result of this calculation plotted with the power for each column on the vertical axis and the horizontal position on the horizontal axis. By applying this to hard threshold processing, information can be completely separated for each lane.

  Since each lane has a width in the horizontal direction, for example, the average, median, mode, etc. of the pixel values in the horizontal direction are obtained, and this value is plotted on the vertical axis and the vertical position is plotted on the horizontal axis. The chart can be provided to the subsequent fermented product selection / similarity calculation process 501. An example of the chart is shown in FIG.

  Examples of the method for calculating similarity between samples (step 203) in the fermented product selection / similarity calculation process 501 include a cross-correlation analysis method and a DP (dynamic programming) matching method. Here, an example of similarity calculation by the DP matching method is shown.

  The DP matching method is a matching method based on dynamic programming, and is a method for obtaining a correlation (distance) while expanding and contracting nonlinear time. This method is mainly used in areas such as speech and character recognition.

  In the DP matching method, an input pattern and a standard pattern are associated with each other, a shortest path having the smallest distance between the input pattern and the standard pattern is obtained, and the correlation is represented by the sum of the distances. FIG. 7 shows a conceptual diagram regarding association between two sequences by DP matching.

  Specifically, for example, the following method is used. Two time series (feature vector series) patterns A and B are shown in Equation 2.

  Considering the plane consisting of A and B as shown in Equation 2, the time axis correspondence between both A and B patterns is the series of lattice points c = (i [k], j [k]) on this plane.

It can be expressed as This F is called a time conversion function. When the vector distance between the two feature vectors a _i and b _j is represented by d (c) = d (i _k , j _k ), the sum of the distances along F is given by Equation 4.

  However, Equation 5 is used as the distance.

  The smaller the value of Equation 5, the better the correspondence between A and B. Is a positive load function associated with F. Let us consider minimizing Equation 4 with respect to F with the following restrictions.

  Monotonicity and continuity conditions: For example, the distortion of / saN / series to / sNa / or / sN / is not allowed.

  boundary condition:

  Condition of matching window: To prevent extreme expansion and contraction, the matching window r is a constant,

  It is easy to simplify the equation by determining that the denominator is a constant independent of F in Equation 4. For simplification, for example, when the load function is symmetric with respect to, Equation 9 is obtained.

  This

  Therefore, the number 4 is

Is converted. The distance D (A, B) between the two time series patterns A and B can be obtained by solving the minimization problem, but it is efficient without examining all the possibilities of F brute force. Can be solved.

In Equation 4, a partial sum g (i, j) of the distance D (A, B) is obtained. The partial sum g when the end point is fixed at (i _l , j _l ) is

  Equation 12 is a formulation of the dynamic calculation method. When this is rewritten using restrictions on F in Equation 6, Equation 7, and Equation 8,

Can be obtained as follows. FIG. 8 shows a path that can reach the coordinates (i, j) in the symmetric DP matching.

  The image is extracted and converted from the biopattern sensing data of FIG. 4 by the above-described exemplary method, and the calculation of similarity is performed on the chart of each lane by the DP matching method and the linear cross-correlation analysis method. Table 1 and Table 2 show part of the data comparing the above.

  In Table 1, the similarity between lanes 1 and 4 that should be the same sample is lower than the similarity between lanes 2 and 3. This means that correct sample self-other identification between lanes by cross-correlation analysis was hindered due to distortion of the electrophoretic image.

  On the other hand, in Table 2, the similarity between lanes 1 and 4 is the smallest among the data in the table, and correct sample self-other identification can be performed based on this data.

  For all the inter-lane data in FIG. 4, the similarity is calculated using the linear cross-correlation analysis method and the DP matching method, and the determination boundary value that gives the highest correct answer rate is obtained by the self-other identification of the sample. -The chart which compared the correct answer rate at the time of identifying others is shown in FIG.

  The number of electrophoresis lanes in the electrophoresis image of FIG. 4 is 20, the number of samples used is 6, and each sample is run in duplicate in 3 to 4 lanes.

  As a result, the correct response rate of the self-other recognition of the sample, which was about 30% in the linear cross-correlation analysis method, improved to about 90% in the correlation analysis by the DP matching method.

  Thus, according to the calculation method by DP matching according to the present invention, the accuracy of sample similarity evaluation can be dramatically improved.

  There have been few reports on the use of genetic multisystems other than gel electrophoresis in the study of complex microbial systems using methods with a high number of theoretical plates for separation. Recently, HPLC-based DHPLC (Denaturing High Performance liquid Chromatography) method, capillary electrophoresis-based CE-SSCP (Capillary electrophoretic single strand conformation polymorphism) method, TGCE (Temperature gradient capillary electrophoresis) method, and DNA melting curve analysis method are used in recent years It has become like this.

  A chromatogram obtained by analyzing a part of the same pig fecal extract sample used in the DGGE analysis of FIG. 4 by the DHPLC method is shown in FIG. 9, an electropherogram analyzed by the TGCE method is shown in FIG. 10, and a DNA melting curve analysis method is used. The analyzed melting analysis curves are shown in FIG.

  Since all of these analysis data can be obtained as a two-dimensional chart as in the example of the processing result of the data conversion process 500 of the image data by the DGGE method described above, the subsequent fermentation product selection / similarity calculation process 501 In this case, the image data can be directly subjected to the same processing as the chart obtained by the data conversion process 500 from the image data by the DGGE method.

  For example, any analytical data provided in a two-dimensional chart such as a spectral spectrum curve obtained by spectral analysis such as FT-IR has an advantage that it can be directly processed by the fermented product selection / similarity calculation process 501. .

The hardware block diagram of the composite organic substance quality comparison apparatus 1. FIG. The flowchart of the system processing of embodiment of the composite organic matter quality comparison apparatus 1. FIG. PCR-DGGE electrophoresis image using a primer targeting the V3 region (236 bp) of the 16s-rRNA gene for a DNA extract from a complex organic material (fermented milk product). Electrophoresis image of a DNA extract from a swine stool sample subjected to PCR-DGGE method using a primer targeting the V3 region of 16s-rRNA gene. FIG. 5 is a chart in which the vertical axis represents the power for each column and the horizontal axis represents the horizontal position for the image of FIG. An example of a two-dimensional chart in which average values of horizontal pixel values are plotted on the vertical axis and vertical positions are plotted on the horizontal axis for the separated electrophoresis lanes. The conceptual diagram regarding the matching of two series by DP matching. A path that can reach coordinates (i, j) in symmetric DP matching. The chromatogram which analyzed a part of sample common with what was used in FIG. 4 by the DHPLC method. FIG. 5 is an electropherogram obtained by analyzing a part of a sample common to that used in FIG. 4 by the TGCE method. FIG. 5 is a melting analysis curve obtained by analyzing a part of a sample common to that used in FIG. 4 by a DNA melting curve analysis method. The chart which compared the correct answer rate at the time of calculating a similarity using the linear cross correlation analysis method and DP matching method, and performing self-other identification.

Explanation of symbols

1 Complex organic matter quality comparison system (system)
3 Analyzing device 5 Electronic computing device 7 Input device 9 Recording media reader 11 Program recording media 13 Monitor display 15 Database

Claims (9)

A bio-pattern sensing data detection means for detecting a chemical group in or extracted from a complex organic material sample composed of a plurality of types of microorganisms and metabolites thereof;
Storage means for storing detected biopattern sensing data for each sample;
Calling means capable of specifying and calling stored biopattern sensing data for each sample;
Two or more samples obtained by each of the above means, mathematically and statistically analyzing the biopattern sensing data, and calculating the similarity between the data by quantifying the relationship between the data, arithmetic means,
An output means for outputting the similarity between data;
An apparatus and system for comparing the quality of a composite organic material sample, comprising: In complex organic matter samples composed of multiple types of microorganisms and their metabolites, DNA molecules or RNA molecules derived from the genetic information of the microorganisms present in the sample are detected based on the difference in base sequence or base length Detection means for biopattern sensing data;
Storage means for storing the input biopattern sensing data for each sample;
Calling means capable of specifying and calling stored biopattern sensing data for each sample;
Two or more samples obtained by each of the above means, the bio pattern sensing data is mathematically and statistically analyzed, and the arithmetic means for calculating the similarity between the data obtained by quantifying the relationship between the data;
An output means for outputting the similarity between data;
An apparatus and system for comparing the quality of a composite organic material sample, comprising:   The quality comparison apparatus and system for composite organic matter samples according to the preceding claim, characterized in that a chromatogram is obtained as biopattern sensing data by using a liquid chromatograph method as a part of the detection means.   The quality comparison apparatus and system for composite organic matter samples according to the preceding claim, characterized in that electropherograms are obtained as biopattern sensing data by using capillary electrophoresis as part of the detection means.   The quality comparison apparatus and system for composite organic matter samples according to the preceding claim, characterized in that a DNA melting curve is obtained as biopattern sensing data by using a DNA melting curve analysis method as a part of the detection means.   The quality comparison apparatus and system for composite organic matter samples according to the preceding claim, characterized in that a spectral spectrum curve is obtained as biopattern sensing data by using spectral spectrum analysis as part of the detection means. An input means for inputting biopattern sensing data obtained by analyzing a complex organic matter sample composed of a plurality of types of microorganisms and metabolites thereof into the system;
Storage means for storing the input biopattern sensing data for each sample;
Calling means capable of specifying and calling stored biopattern sensing data for each sample;
Mathematical and statistical analysis of biopattern sensing data of two or more samples obtained by each of the above means, and the similarity between the data is calculated by quantifying the relationship between the data,
Computing means;
An output means for outputting the similarity between data;
An apparatus and system for comparing the quality of a composite organic material sample, comprising:   The quality comparison apparatus for composite organic matter samples according to the preceding claim, characterized in that the similarity of biopattern sensing data is calculated by using a matching method based on dynamic programming of biopattern sensing data as a computing means. And system.   A recording medium readable by a computer or a recording medium reader connected to the computer, which records a program and data for causing the computer to function as each means constituting the quality comparison device and system described in the preceding claims.