JP2010017085A - Method for determining commodity - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of determination with objective accuracy by improving a determination method in which a determination standard is liable to get fuzzy in commodity determination or production area determination of farm and marine products and raw materials such as processed foods, foodstuffs, feed, etc., using farm and marine products as raw materials. <P>SOLUTION: The method for determining to which group of a plurality of groups a target commodity to be determined belongs includes calculating probabilities in which target commodities to be determined belong to each group and deciding that a group exhibiting the highest probability is the group to which the target commodity belongs to. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a method for discriminating to which group (variety, production area, etc.) a product such as a living organism or a processed product made from a living organism belongs.

  In recent years, many camouflaged displays of varieties and production areas have been discovered for agricultural and marine products such as vegetables and fruit trees, meat and fish, and products made from these. As products, even if they have the same origin and composition, they are treated as different products if they are produced in different places.Through such disguise display, the infringement of breeder rights and distrust of food safety and security increase. Is coming. In order to control such camouflage display, a method for discriminating goods of a living thing or a product derived therefrom is required. In the variety similarity test conducted by the National Center for Seedling and Seedling Management to identify varieties, which is one of the methods for product discrimination, three types of tests, characteristic comparison, comparative cultivation, and DNA analysis, are used as discrimination methods (non- Patent Literature 1), varieties for which DNA analysis techniques have been established are limited, and it takes time and labor to establish new varieties of DNA analysis techniques, so all agricultural and marine products and processed foods using them as raw materials -At present, it is not possible to discriminate varieties and production areas of raw materials such as food and feed by DNA analysis alone. For this reason, at present, varieties and production areas are discriminated mainly by characteristic comparison and comparative cultivation.

Independent Administrative Institution Seedling Management Center, “Summary of Variety Similarity Test”, [online], Internet <URL: http://www.nCss.go.jp/main/gyomu/hinsyuhogo/hinsyuruijiseishiken.html>

  When discriminating varieties and places of origin as described above, the characteristics that serve as the criteria for discrimination are many descriptions regarding the form as seen in the description of the registered varieties database on the variety registration homepage of the Ministry of Agriculture, Forestry and Fisheries (http: / /www.hinsyu.mAff.go.jp/). For example, the length of leaves is long, short, slightly short, the thickness of leaves is thick, thin, etc., and the speed of growth is only an abstract expression such as fast, slow, etc. There is a problem that it is difficult.

  The present invention has been made in view of such problems, and an object thereof is to provide a clear discrimination method having objective accuracy.

In order to achieve the above object, the present invention provides the following [1] to [10].
[1] A method for determining which of a plurality of groups a product to be identified belongs to, wherein the probability that the product to be identified belongs to each group is calculated, and the group having the highest probability is A determination method characterized by determining that a group to which a determination target product belongs. Like this method, the reliability of the discrimination method can be compared and evaluated by displaying the certainty of the discrimination result as a probability.
[2] The discrimination method according to [1], wherein the discrimination is performed comprehensively by combining the results of a plurality of discrimination methods. Like this method, the accuracy of discrimination can be improved by comprehensively discriminating the results of a plurality of discrimination methods.
[3] The discrimination method according to [2], wherein discrimination is performed by weighting each result of the discrimination method. Like this method, it is possible to reflect the discrimination accuracy for each discrimination method in the discrimination result by comprehensively combining the results of multiple discrimination methods weighted for each discrimination method. Can improve the accuracy.
[4] The discrimination method according to any one of [1] to [3], wherein the probability that a product to be discriminated belongs to each group is obtained by the following method:
(1) Select one characteristic of the product to be discriminated and set multiple categories for that characteristic. (2) Examine the characteristics of multiple products that are confirmed to belong to each group, and classify each product as a category. Classifying and determining the appearance probability in each category for each group (3) Examining which category the product to be identified belongs to belongs to the characteristics (4) Probability A belonging to each group: A = (B / C).
B: Appearance probability of a group trying to calculate the probability in the category to which the discrimination target product belongs C: Total sum of appearance probabilities of each group in the category to which the discrimination target product belongs [5] Characteristic represented by a numerical value, The characteristics of the products determined to belong to each group are examined, and the products are classified by excluding the products that showed abnormally high values and the products that showed abnormally low values [4 ] The discriminating method described in the above. As in this method, the accuracy of discrimination can be improved by classifying products using only effective measurement values excluding outliers of both high and low values.
[6] The discrimination method according to any one of [1] to [3], wherein the probability that a product to be discriminated belongs to each group is obtained by the following method:
(1) Set n judgment items for the product to be discriminated. (2) For a plurality of products that have been confirmed to belong to each group, check whether they are applicable to the judgment item. Find the applicable rate and the non-applicable rate. (3) Check whether the product to be judged falls under each judgment item, and if it falls under the item, select the relevant rate of the group to be calculated and fall under the item If not, the non-corresponding rate of the group to be calculated is selected, the sum of the corresponding rates or non-corresponding rates of each item is obtained, and the value divided by n is set as the probability of belonging to the group.
[7] The determination method according to [6], wherein the determination item is detection of DNA serving as a marker.
[8] The discrimination method according to any one of [1] to [7], wherein the product is a living thing or a processed product using a living thing as a raw material.
[9] The discrimination method according to any one of [1] to [7], wherein the product is a living thing and the group is a variety.
[10] A discrimination method characterized by comprehensively discriminating the result of the discrimination method according to [4] or [5] and the result of the discrimination method according to [6] or [7].

  The present invention provides a novel method for discriminating products. By this method, it is possible to accurately discriminate varieties, production areas, etc., such as agricultural products, livestock products, marine products and processed products thereof.

  Hereinafter, the present invention will be described in detail.

  The present invention is a method for discriminating which of a plurality of groups a product to be discriminated belongs to, and calculating the probability that the product to be discriminated belongs to each group, and the group showing the highest probability is , It is determined that the product belongs to the group to which the discrimination target product belongs.

  Hereinafter, in a preferred embodiment of the product discrimination method of the present invention, when determining which of the three varieties A, B, and C a plant belongs to, the traits that are the criteria for discrimination are plant height and genomic DNA. The case where there are two SSR markers will be described as an example with reference to the drawings. The plant height is continuous numerical data, and the SSR marker is alternative data for determining the presence or absence of a marker.

First, a discrimination method based on plant height, which is a trait that serves as a first discrimination criterion, will be described with reference to FIG. In the case of normal cultivar registration based on form, cultivar A is distinguished in three ways: plant height is low, plant B is slightly low, and variety C is high.

  In general, the product discrimination is composed of two steps, that is, basic data creation work necessary for discrimination and work for discriminating the type of discrimination target based on this data. In the case of product discrimination based on plant height, there are two stages: basic data creation process 16 for product discrimination based on plant height and product discrimination process 17 for discrimination based on plant height. First, creation of basic data for product discrimination based on plant height Therefore, the width and distribution of the measured values of the plant height are obtained for the standard products of each variety.

For this purpose, a group 1 of standard product individuals of a variety A, a group 2 of standard product individuals of a product B, and a group 3 of standard product individuals of a product C are prepared. As the number of individuals in the group increases, the width and distribution of measurement values become more reliable, and the reliability of discrimination based on this increases.

Measurement 4 of the height of the standard product is performed for each of the prepared standard products. Any measurement method can be used, but always use the conditions determined first.

Next, the numerical values obtained by this measurement are divided into a plurality of numerical regions step by step. There is no particular limitation on the way of dividing the numerical area, but a way of dividing such that the distribution of the numerical area pattern is likely to be different for each product type is preferable. Here, the plant height displayed in centimeters is divided into five regions above 20 cm, 21 cm to 40 cm, 41 cm to 60 cm, 61 cm to 80 cm, and 80 cm.

Measurement of plant height measured in standard product 4 The plant height measurement values obtained by classification are applied to the above five numerical areas, and the distribution of numerical values for each product type is shown in a graph. Plant height distribution graph 5 for each plant, plant height distribution graph 6 for each individual of the standard product of variety B, and plant height distribution graph 7 for each individual of the standard product of variety C. Comparing the distribution graphs, it can be seen that the characteristics of each cultivar such as cultivar A having a low plant height, cultivar B having a slightly low plant height, and cultivar C having a high plant height can be represented by a numerical distribution. These graphs are useful for determining whether the distribution pattern of the numerical range differs for each variety and, as a result, whether the trait of plant height is desirable as a criterion for distinguishing the three varieties A, B, and C. However, it is not always necessary to create a graph, and the process may proceed to creation of table 8 of accumulated data of individual plant heights and appearance probabilities for the following varieties.

Next, for each variety, divide the number of individuals belonging to each numerical area by the total number of individuals, and obtain the appearance probability of the measured value in each numerical area. Create a table. In the example of this table, for example, when measuring the plant height of 200 individuals of the standard product of cultivar B and classifying them in the numerical classification of these 5 stages, 82 individuals of plant height included in the numerical range of 41 cm to 60 cm are seen. In the cultivar B, the appearance probability of an individual with a plant height included in this numerical range is 82/200, that is, 0.41.

Now that basic data has been created, any time when these types are discriminated by this method, it is possible to work from the next step 17 for discriminating a product to be discriminated by plant height.

Next, the process 17 for discriminating the product to be discriminated by the plant height will be described by taking as an example the case of discriminating the product of the individual to be discriminated (b) 10 and the individual to be discriminated (b) 11.

First, the measurement target plant height measurement 12 is performed on the discrimination target individual (b) 10 and the discrimination target individual (b) 11, and the plant height measurement result 13 of the discrimination target individual (b) is determined. The plant height measurement result 14 of the individual (b) is obtained. The measuring method of the plant height is the same as that used in the basic data creation process 16 for product discrimination by plant height.

Next, by comparing the numerical values of these measurement results with the accumulated data of individual plant heights for each variety and Table 8 of the appearance rate, the plant height measurement result 13 of the discrimination target individual (a) and the discrimination target individual (b) ) To identify the numerical region included in the plant height measurement result 14. Since the plant height measurement result 13 of the discrimination target individual (I) is 32 cm, it is included in the numerical region of the plant height 21 cm to 40 cm, and the plant height measurement result 14 of the discrimination target individual (B) is 85 cm, so the plant height is 81 cm. Included in the upper numerical range.

For each individual to be identified, identify the appearance probability for each variety in the numerical area containing the measured value, divide the appearance probability of that numerical area for each breed by the appearance probability of that numerical area for all varieties, Appearance probabilities of the respective varieties in the region are obtained, and a product discrimination probability table 15 based on the plant height to be discriminated is created. That is, the probability of appearance of a numerical region of 21 cm to 40 cm including 32 cm, which is the plant height measurement result 13 of the individual (i) to be discriminated, from the accumulated data of the plant height and the appearance rate of each plant for each variety and appearing in the variety A Is 0.31, the appearance probability that appears in the breed B is 0.29, and the appearance probability that appears in the breed C is 0.00. Similarly, the appearance probability of the numerical region above 80 cm including 85 cm which is the plant height measurement result 14 of the individual (b) to be discriminated appears in the variety A is 0.00, the appearance probability that appears in the variety B is 0.04, and appears in the variety C The appearance probability is 0.50.

A product discrimination probability table 15 based on the plant height of the discrimination target is a table showing the discrimination probabilities for each discrimination target of the product discrimination from the numerical values of the appearance probabilities. That is, the probability that the individual to be identified (a) 10 whose plant height measurement result is 32 cm is the variety A is the appearance probability 0.31 in which the numerical region of 21 cm to 40 cm appears in the variety A, the appearance probability 0.31 in the variety A, and the variety Appearance probability 0.29 appearing in B and appearance probability 0.00 appearing in variety C is 0.52 divided by the total value of the three numerical values, or 52%. Similarly, the probability that the individual (a) 10 to be discriminated is the breed B is the appearance probability 0.29 in which the numerical region of 21 cm to 40 cm appears in the breed B, the appearance probability 0.31 in the breed A, and the appearance probability 0.29 in the breed B. The probability of 0.48 divided by the total value of the three occurrences of the appearance probability 0.00 in cultivar C, that is 48%, the probability that the individual to be discriminated (b) 10 is cultivar C, Appearance probability 0.00 is 0.00 divided by the total value of three numerical values of appearance probability 0.31 that appears in variety A, appearance probability 0.29 that appears in variety B, and appearance probability 0.00 that appears in variety C, that is, 0%. On the other hand, the probability that the individual (b) 11 to be discriminated whose plant height measurement result is 85 cm is the variety A, the occurrence probability 0 that the numerical area above 81 cm appears in the variety A, the appearance probability 0.00 that appears in the variety A, the variety 0.00 divided by the sum of three numerical values of appearance probability 0.04 appearing in B and appearance probability 0.50 appearing in variety C, that is, 0%, the probability that the individual (b) 11 to be discriminated is variety B is above 81 cm Appearance probability 0.04, which appears in variety B in variety B, is 0.07, or 7%, divided by the sum of the three numerical values of appearance probability 0.00, occurrence probability 0.04, occurrence probability 0.04 in variety B, and appearance rate 0.50 in variety C The probability that the individual (b) 11 to be discriminated is the variety C is the appearance probability 0.50 in which the numerical area above 81 cm appears in the variety C, the appearance probability 0.00 in the variety A, and the appearance probability 0.04 in the variety B. , To breed C 0.93 divided by the sum of the three numbers of the probability of occurrence 0.50 that appears you are, that is, 93%.

Next, a discrimination method using an SSR marker, which is a trait serving as a second discrimination criterion, will be described with reference to FIG.

The SSR stands for S imple S equence R epeat (simple sequence repeats), for example, as CACACA · · · and GATGAT · · ·, in the region where the number base of approximately nucleotide sequence is present as repeat units in an organism of DNA There is also referred to as STR (abbreviation of S hort T andem R epeat) and microsatellite. There are many SSRs in the genome, but there are sites where the number of repeats of the base sequence differs depending on the individual or variety, and the base sequence that can distinguish the individual or variety by the number of repeats in that region is an SSR marker. Call it. For product discrimination by SSR marker, PCR is performed only on the site containing the SSR of the marker site using a primer set complementary to a specific base sequence adjacent to both sides of the SSR, and the DNA of that portion is amplified. This is a technique for determining whether or not an SSR marker fragment having a chain length peculiar to a specific variety can be detected by electrophoresis. In order to discriminate all individuals and varieties to be discriminated from each other, it may be necessary to discriminate them using a combination pattern using a plurality of SSR markers.

As described in the section of product discrimination based on plant height, in general, product discrimination is composed of two steps, that is, basic data creation work necessary for discrimination and work for discriminating the type of discrimination target based on this data. In the case of product discrimination using an SSR marker, as shown in FIG. 2, there are two stages: a basic data creation process 18 for product discrimination using an SSR marker, and a product discrimination process 19 for discrimination using an SSR marker.

First, in order to create basic data for product discrimination by SSR marker, find an SSR marker or set of SSR markers that can discriminate all the varieties you want to discriminate, and create a table of SSR marker chain length specific to each varieties. The basic data creation process 18 for product discrimination using the SSR marker will be described.

As in the case of product discrimination based on the plant height in FIG. 1, a group 1 of standard product individuals of cultivar A, a group 2 of standard product individuals of cultivar B, and a group 3 of standard product individuals of product C are prepared. As long as it is certain that each standard product is a standard product of each type, the standard product may not be the same standard product used for creating basic data for product discrimination by plant height. Using the prepared standard product individuals of the three varieties, the DNA extraction 20 for each individual of the standard product of the product A, the DNA extraction 21 for each individual of the standard product of the product B, and the individual of the standard product of the product C DNA extraction 22 is performed. The DNA extraction method can be any method as long as it can be performed with high reproducibility without inhibiting the PCR and electrophoresis processes performed after the extraction, but once the method has been determined, Always use the same reagent concentration, processing time, temperature and other conditions as the first.

Next, PCR amplification 23 of the SSR marker is performed using the extracted DNA for each individual and a primer set for SSR marker amplification. In order to improve reproducibility and sensitivity, the amount of DNA to be used, the primer set base sequence, PCR conditions, etc. are optimized through a call experiment, and these conditions are always used.

After PCR amplification, the amplified product is subjected to gel electrophoresis for each individual, SSR marker electrophoretic image 24 for each individual of the standard product of cultivar A, SSR marker electrophoretic image 25 for each individual of the standard product of cultivar B, and cultivar C An SSR marker electrophoresis image 26 of each standard product is obtained. For gels, reagents, and electrophoresis conditions used for electrophoresis, conditions that can distinguish differences in SSR marker chain length and ensure sufficient sensitivity and reproducibility are set in advance, and electrophoresis is always performed under those conditions thereafter. .

Next, the position of the band detected in the electrophoretic image of each individual is compared with a size marker or another individual, and it is determined whether or not a band of a size serving as an SSR marker is detected. The determination is made with no band as 0. For example, in the SSR marker electrophoretic image 24 for each individual of the standard product of breed A, in the individual of breed A that has been electrophoresed in the leftmost lane, the three types of SSR markers a, b, and c used in this product discrimination are Detection at 3 locations is 1 (with band) for a and c, 0 (no band) for b, and in the SSR marker electrophoretic image 25 for each individual of the standard product of breed B, it is the third from the left end. In the individual of breed B that was electrophoresed in the lane, the a and c were 0 (no band), and the b was 1 (with band). Here, an SSR marker electrophoretic image 24 for each individual of the standard product of cultivar A, an SSR marker electrophoretic image 25 for each individual of the standard product of cultivar B, and an SSR marker electrophoretic image 26 for each individual of the standard product of cultivar C are shown. Comparing, it can be seen that the SSR markers a and c in the breed A, the SSR markers b and c in the breed B, and the SSR markers a and b in the breed C are detected with a high probability. That is, when this test is performed on an individual whose SSR marker is unknown, breed A is detected if a and c are detected, breed B is detected if b and c are detected, and breed C is detected if a and b are detected. Each probability is expected to be high.

The detection results and detection rates of each SSR marker are obtained by collecting the detection results of such SSR markers for each individual for each type, and a table 27 of the accumulated data and appearance probability of the individual SSR markers for each type is created. In this table, for example, in the case of cultivar A, among 6 standard products, 5 of 6 individuals detected the SSR marker of a, 1 SSR marker of b was detected, 1 of 6 individuals, and SSR of c Markers were detected in 4 out of 6 individuals, and the detection rate of each marker in variety A was 0.83 for A, 0.17 for B, and 0.67 for C, respectively. Furthermore, based on the detection rate data for each SSR marker of each product type, the probability of product discrimination combining the determination results of all SSR markers can be calculated as the average value of the detection results of each marker. . For example, in breed A, the probability that SSR marker a is 1 (with band), SSR marker b is 0 (without band), and SSR marker c is 1 (with band) is the accumulated data of individual SSR markers for each breed. Referring to the column of product type A in Table 27 of the appearance probability, {0.83+ (1-0.17) +0.67} /3=0.78, and this value is the probability of product discrimination when such a detection result is obtained. Can be considered. On the other hand, in the breed B, the probability that the SSR marker a is 1 (with band), the SSR marker b is 1 (with band), and the SSR marker c is 0 (no band). Since it is 0.00, the probability of + is 0.00, in B, the detection rate is 1.00, so the probability of + is 1.00, and in C, the detection rate is 0.67, so the probability of-is 1-0.67 = 0.33. The probability of product discrimination when such a detection result appears can be considered as {0.00 + 1.00 + (1-0.67)} / 3 = 0.78. In the rightmost column of Table 27, the accumulated data of individual SSR markers and appearance probabilities for each type, the maximum probability that each type can be discriminated by these three types of SSR markers is shown.

Since the basic data can now be created, any time when these types are discriminated by this method, it is possible to perform the operation from the next step 19 for discriminating the product to be discriminated by the SSR marker.

Next, in the product discrimination process 19 based on the SSR marker, the product discrimination between two individuals, i.e., the discrimination target individual (A) 10 and the discrimination target individual (B) 11 as in the case of the product discrimination based on the plant height, is performed. An example will be described.

First, with regard to two individuals, that is, the individual to be discriminated (a) 10 and the individual to be discriminated (b) 11, the DNA extraction 28 of the individual to be discriminated (a) and the DNA extraction 29 of the individual to be discriminated (b) 29. And performing PCR amplification 30 of the SSR marker to be discriminated using the extracted DNA for each individual to be discriminated and the primer set for SSR marker amplification. For the amount of DNA to be used, the base sequence of the primer set, the PCR conditions, etc., the conditions used in the basic data creation process 18 for product discrimination using the SSR marker are used as they are.

After PCR amplification, the amplified product is subjected to gel electrophoresis for each individual to be discriminated, and an SSR marker electrophoretic image 31 to be discriminated is obtained. The conditions used in the basic data creation process 18 for product discrimination using the SSR marker are used as they are for gels, reagents, and electrophoresis conditions used for electrophoresis.

Next, it is determined whether or not a band having a size serving as an SSR marker is detected in the electrophoretic image of each individual to be determined, and a table 32 of data on the target SSR marker is created.

The prepared SSR marker data table 32 is compared with the individual SSR marker accumulation data and appearance probability table 27 for each breed, and the probability of detection results for each SSR marker is calculated. A product discrimination probability table 33 using SSR markers is created. For example, as for the detection result of the SSR marker of the individual (a) 10 to be discriminated, A is −, B is +, and C is +. From this detection result, the probability that the individual (a) 10 to be discriminated can be discriminated as the breed A is 1-0.83 = 0.17 for A, 0.17 for B, and 0.67 for C. The discrimination probability that the individual (b) 10 is the breed A is (0.17 + 0.17 + 0.67) /3=0.34, that is, 34%. Similarly, the detection result of the SSR marker of the individual (b) 11 to be discriminated is A for +, B for +, and C for −. From this detection result, the probability that the individual to be discriminated (b) 11 can be discriminated as the breed B is 0.00 for A, 1.00 for B, and 1-0.67 = 0.33 for C. The discrimination probability that the individual (b) 11 is the breed B is (0.00 + 1.00 + 0.33) /3=0.44, that is, 44%.

As described above, the product discrimination probability data of the discrimination target individual (b) 10 and the discrimination target individual (b) 11 by the two discrimination methods of the plant height and the SSR marker are obtained. Next, referring to FIG. 3, a method of weighting two types of discrimination methods, product discrimination by plant height and product discrimination by SSR marker, and performing product discrimination by combining both will be described.

Considering the relative certainty of the two distinct methods of plant height and SSR marker, weight each distinction method. Here, plant height may be affected by environmental conditions when actually cultivated, but SSR markers do not have that possibility, so the discrimination result by plant height and the discrimination result by SSR marker are 3: 7 An example of weighting will be shown.

The product discrimination probability table 15 based on the plant height of the discrimination target obtained in the product discrimination process 17 based on the plant height in FIG. 1 and the discrimination target obtained in the product discrimination process 19 based on the SSR marker in FIG. Based on the product discrimination probability table 33 of SSR markers, the probability that each of the individuals (b) 10 and (b) 11 to be discriminated is each of the varieties A, B, and C, The overall product discrimination probability table 34 is created by multiplying the probability by 30% and multiplying the product discrimination probability by the SSR marker by 70% weight distribution. For example, the product discrimination probability that the individual (a) 10 to be discriminated is cultivar A is 52% in the product discrimination probability based on the plant height and 34% in the product discrimination probability based on the SSR marker. The probability of
0.52 × 0.3 / (0.3 + 0.7) + 0.34 × 0.7 / (0.3 + 0.7) = 0.39, that is, 39%.

Similarly, the probability that the individual (b) 11 to be discriminated is the breed C is
0.93 × 0.3 / (0.3 + 0.7) + 1.00 × 0.7 / (0.3 + 0.7) = 0.98, that is, 98%.

In this way, the probabilities of discriminating each product type are calculated for all discrimination target individuals, and a comprehensive product discrimination probability table 34 is created.

From the obtained numerical value of the product discrimination probability, the discrimination target individual (b) 10 has a high probability of being B, and the discrimination target individual (b) 11 has a fairly high probability of being C. Can do.

  In the example of product discrimination shown here, the first discrimination method, plant height, is a discrimination method based on morphology, and the second discrimination method, SSR marker, is a discrimination method based on DNA. Comprehensive discrimination is performed in combination, but the discrimination method used may be only a single or multiple discrimination method based on form, or may be only a single or multiple discrimination method based on DNA. A method combining a discrimination method based on DNA and a discrimination method based on DNA may be used, or other discrimination methods may be used. As a discrimination method other than a discrimination method based on morphology or a discrimination method based on DNA, a determination method based on a constituent component of a living organism, a living organism attached to or attached to a living organism, or a chemical component can be considered.

Furthermore, in the example of product discrimination shown here, the first discrimination method, plant height, is a continuous numerical value, and the second discrimination method, SSR marker, is an alternative of 0 or 1 indicating the presence or absence of a band. Evaluate and classify different classification methods with different weights one by one and calculate the overall classification probability, but the classification method used can be a single one or more than three types, When a plurality of discrimination methods are used in combination, a plurality of discrimination methods using continuous numerical values may be combined, or only a plurality of discrimination methods of 0 or 1 may be combined. In addition, the choice is not limited to two choices.

Further, in the illustrated example, 3: 7 is used for weighting, but the relative specific gravity of each discrimination method may be any of 0 to 1.

In addition, traits that are standard for discrimination such as color and tree shape such as flower color and leaf color that are not normally expressed numerically can also be expressed numerically. Numerals such as plant height and SSR marker, and presents a color sample according to the JIS conventional color name and Munsell system, visually determines the sample color closest to the comparison target, and the sample color as in the JHS color chart Can be quantified by displaying with the color number attached to it and the red, green, and blue tone values of the sample color. If these values show different distributions depending on the variety and production area, use them. Thus, it is possible to perform discrimination that displays the discrimination probabilities of varieties and production areas.

In the example shown in the figure, two types of discrimination methods, plant height and SSR marker, are used to prepare a group of the same number of standard items for each type and to create both basic data for the two types of discrimination methods. However, as long as it is an individual that is surely a standard product of each product type, the standard product individual may be different for each discrimination method. Further, the number of standard product individuals prepared for each product type may be different as long as the number that can provide sufficient probability for probability calculation can be prepared.

Further, the weighting method is not limited to that described in the present embodiment. In addition, it is needless to say that various other configurations can be adopted without departing from the gist of the present invention.

It is the figure which showed the flow of the goods discrimination by plant height. It is the figure which showed the flow of the product discrimination | determination by an SSR marker. It is the figure which showed the flow which weights the 2 types of discrimination methods, plant height and SSR marker, and performs product discrimination comprehensively combining the results of both.

Explanation of symbols

1 ... A population of standard product individuals of variety A, 2 ... A population of standard product individuals of product type B, 3 ... A population of standard product individuals of product type C, 4 ... Measurement of plant height of standard products, 5 ... Plant height distribution graph for each individual of the standard product of variety A, 6 ... Plant height distribution graph for each individual of the standard product of variety B, 7 ... Plant height distribution graph of each of the standard product of variety C 8 ... Accumulated data of plant height and appearance probability table for each varieties, 9 ... Measurement of discrimination target, 10 ... Discrimination target individual (I), 11 ... Discrimination target individual ( B), 12 ... measurement of the plant height of the discrimination target, 13 ... measurement result of the plant height of the discrimination target individual (b), 14 ... measurement result of the plant height of the discrimination target individual (b), 15 ... Product discrimination probability table based on plant height to be discriminated, 16 ... Basic data creation process for product discrimination by plant height, 17 ... According to plant height The process of discriminating the target product, 18 ... The basic data creation process for discriminating the product by the SSR marker, 19 ... The process of discriminating the target product by the SSR marker, 20 ... The standard product of the product A DNA extraction for each individual, 21 ... DNA extraction for each individual of the standard product of variety B, 22 ... DNA extraction for each individual of the standard product of variety C, 23 ... PCR amplification of the SSR marker, 24 ..SSR marker electrophoretic image for each individual of standard product of cultivar A, 25 ... SSR marker electrophoretic image of individual for standard product of cultivar B, 26 ... SSR for each individual of standard product of cultivar C Marker electrophoresis image, 27... SSR marker accumulation data and appearance probability table by individual, 28... DNA extraction of individual to be discriminated (a), 29. ) DNA extraction, 30 ... PCR amplification of SSR markers to be discriminated, 31 ... SSR markers electrophoresis image of the subject, the table data 32 ... discrimination object of SSR markers, the probability table of items determined by the SSR markers 33 ... determination target

Claims (10)

  This is a method for determining which product to be classified belongs to which of a plurality of groups, the probability that the product to be identified belongs to each group is calculated, and the group showing the highest probability is the product to be identified A determination method characterized by determining that the group belongs to.   The discrimination method according to claim 1, wherein the discrimination is comprehensively performed by combining the results of a plurality of discrimination methods.   The discrimination method according to claim 2, wherein discrimination is performed by weighting each result of the discrimination method. The determination method according to any one of claims 1 to 3, wherein the probability that a product to be determined belongs to each group is obtained by the following method.
(1) Select one characteristic of the product to be discriminated and set multiple categories for that characteristic. (2) Examine the characteristics of multiple products that are confirmed to belong to each group, and classify each product as a category. Classifying and determining the appearance probability in each category for each group (3) Examining which category the product to be identified belongs to belongs to the characteristics (4) Probability A belonging to each group: A = (B / C).
B: Appearance probability of a group to calculate the probability in the category to which the discrimination target product belongs C: Total appearance probability of each group in the category to which the discrimination target product belongs   Check the characteristics of the products that are determined to belong to each group, and exclude the products that showed abnormally high values and the products that showed abnormally low values. The determination method according to claim 4, wherein the classification is performed. The determination method according to any one of claims 1 to 3, wherein the probability that a product to be determined belongs to each group is obtained by the following method.
(1) Set n judgment items for the product to be discriminated. (2) For a plurality of products that have been confirmed to belong to each group, check whether they are applicable to the judgment item. Find the applicable rate and the non-applicable rate. (3) For the product to be discriminated, check whether it corresponds to each judgment item, and if it falls under the item, select the corresponding rate of the group to be calculated and apply to the item If not, the non-corresponding rate of the group to be calculated is selected, the sum of the corresponding rates or non-corresponding rates of each item is obtained, and the value divided by n is set as the probability of belonging to the group.   The determination method according to claim 6, wherein the determination item is detection of DNA serving as a marker.   The discrimination method according to any one of claims 1 to 7, wherein the product is a living thing or a processed product using the living thing as a raw material.   The discrimination method according to any one of claims 1 to 7, wherein the commodity is a living thing and the group is a variety.   A discrimination method characterized by comprehensively discriminating the result of the discrimination method according to claim 4 or 5 and the result of the discrimination method according to claim 6 or 7.

Images