Numerical value identification method of Listeria monocytogenes

Abstract

The invention belongs to the field of microbial identification, and discloses a numerical identification method of Listeria monocytogenes. By counting the positive probability data of 12 biochemical experiments such as hemolysis test and xylose of different known listeria, the mode frequency, the total occurrence frequency, the multi-item total occurrence frequency, the identification percentage and the mode likelihood ratio of different listeria are determined. 12 corresponding biochemical experiments are carried out on the to-be-identified listeria monocytogenes, the species of the listeria monocytogenes are determined based on the identification percentage and the mode likelihood ratio, and the negative or positive identification results of the related biochemical reactions are clear in boundary and easy to distinguish. According to the method, the 17 strains of the listeria are effectively distinguished, the identification accuracy is as high as 99.23%, the identification cost is low, and the result is accurate.

Description

Technical Field

[0001] The present invention belongs to the field of microbial identification, and particularly relates to a numerical identification method for Listeria Background Art

[0002] The contamination of Listeria monocytogenes, a foodborne pathogen, has become a major threat that cannot be ignored in the global public health field. Its harm not only deeply affects the physical health and life safety of consumers, but also leads to a huge consumption of medical resources and incalculable economic losses to food producers. Bacteria of the genus Listeria are facultative anaerobic Gram-positive short bacilli. So far, only L. monocytogenes and L. ivanovii in this genus have been reported to be pathogenic, but generally only L. monocytogenes is considered to be able to cause human diseases. Compared with other foodborne pathogens that infect the gastrointestinal tract, L. monocytogenes can penetrate the blood-brain and placental barriers, increasing the severity of the disease. In recent years, Listeria contamination incidents have occurred frequently. The recalls of a wide range of food types, from various dairy products, meats to vegetables, indicate the severity of this problem. Since many tests recognized and regulated by government agencies such as the US Food and Drug Administration cannot distinguish Listeria species, and the emergence of non-pathogenic Listeria is likely to mean the potential risk of pathogenic Listeria, therefore, achieving accurate identification of Listeria at the "species" level is of great significance for improving the guarantee level of food quality and safety.

[0003] The identification methods of Listeria include 16S rRNA gene sequencing analysis, matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF), etc. Although these methods have the advantages of being fast and efficient, they are difficult to be popularized on a large scale due to the need for professional personnel and expensive instrument equipment. Traditional biochemical identification is the "gold standard" for Listeria detection, which identifies bacteria according to the different abilities of bacteria to utilize nutrients and the differences in their metabolites. National Standard GB 4789.30-2016 "National Food Safety Standard Food Microbiology Examination - Examination of Listeria monocytogenes" stipulates that the test sample first needs to go through steps such as enrichment and isolation and purification, and the final identification is carried out through biochemical analysis methods, but this national standard can only identify six species. The numerical identification method is a rapid detection method that combines traditional biochemical reaction detection technology and modern computer technology. API is an internationally recognized standard numerical identification system and is currently the numerical identification system with the widest identification range in the world, but its existing related products can only identify six common species within the genus Listeria. There are many difficulties in the numerical identification of Listeria. On the one hand, it is difficult to determine an accurate and effective combination of biochemical tests. The physiological characteristics of different species within the genus Listeria are relatively similar. Screening out a combination that can accurately distinguish each species from numerous biochemical experiments requires a large amount of experimental research and data analysis, and strains from different regions and different sources may have differences, making the screening work more complicated. On the other hand, establishing a reliable probability database is a key problem. The positive probability data in the database needs to be statistically based on a large number of representative strain experiments, but it is difficult to obtain strain samples covering all possible situations in actual operation, and data deviation may lead to inaccurate identification results. In addition, the calculation process is complex, involving the calculation of multiple parameters such as mode frequency and total occurrence frequency. Manual calculation is inefficient and prone to errors. Developing a supporting calculation software not only requires professional programming technology but also needs to ensure the accuracy and stability of the software, which also increases the implementation difficulty of the numerical identification method. So far, there have been no relevant reports on numerical identification methods for distinguishing more species within the genus Listeria at home and abroad. Therefore, it is urgent to develop a Listeria identification system with independent intellectual property rights and a supporting biochemical kit to identify more species.

[0004] Based on this, a numerical biochemical kit for Listeria bacteria with independent intellectual property rights in China has been developed, which can provide high-quality and low-cost Listeria species identification products for domestic users, and has important significance for the development of China's food hygiene inspection industry and ensuring food safety. Summary of the Invention

[0005] In order to achieve accurate identification of Listeria at the "species" level, the purpose of the present invention is to overcome at least one deficiency of the prior art and provide a numerical identification method for Listeria.

[0006] The technical solution adopted by the present invention is as follows:

[0007] A numerical identification method for Listeria, comprising the following steps:

[0008] Performing biochemical reaction experiments on known Listeria of different species. The biochemical reaction experiments include hemolysis test, xylose, ribose, tagatose, mannitol, L - arabinose, D - arabinitol, mannosidase, maltose, glucose - 1 - phosphate, rhamnose and arylamidase experiments, a total of 12 kinds of biochemical experiments. Obtaining the positive rates of biochemical reactions of different species of Listeria, taking different species of known Listeria as different taxonomic units, and determining the probability matrix of each taxonomic unit;

[0009] Calculating the mode frequency of the taxonomic unit Total occurrence frequency P (u / J) , multiple total occurrence frequency S, identification percentage ID and mode likelihood ratio T, where:

[0010] Mode frequency Is the product of the maximum total occurrence probabilities of the biochemical experiment combinations under this taxonomic unit,

[0011] In the formula: LJ represents the probability of the taxonomic unit J for all biochemical tests, P(X i / J) represents the probability of a certain biochemical test for the taxonomic unit J, and X i Is the i - th biochemical test;

[0012] Total occurrence frequency P (u / J) Is the product of the actual occurrence probabilities of the biochemical experiment combinations of a certain unknown bacterium to be identified under this taxonomic unit, P (u / J) = LJ = Π I P(X i / J);

[0013] Multiple total occurrence frequency S is the sum of the total occurrence frequencies of each taxonomic unit, S = ∑P (u / J) = ∑(Π I P(X i / J));

[0014] Identification percentage ID is the total occurrence frequency of each taxonomic unit divided by the multiple total occurrence frequency and then multiplied by 100%, and the sum of all identification percentages should be equal to 100%,

[0015] Mode likelihood ratio T is the total occurrence frequency of each taxonomic unit divided by the mode frequency of this taxonomic unit and then multiplied by 100%,

[0016] After the strain to be identified is pure cultured on a selective medium for Listeria, a biochemical reaction experiment is carried out. According to the positive probability data of the biochemical reaction of Listeria, the negative / positive experimental results are converted into probability values, and the mode frequency of the strain to be identified is calculated. Total occurrence frequency P (u / J) , multiple total occurrence frequency S, identification percentage ID, and mode likelihood ratio T;

[0017] According to the identification percentage ID and mode likelihood ratio T of the strain to be identified, the reliability of the identification result of the strain to be tested is evaluated, and the taxonomic unit with the highest score is the identification result of the strain to be tested.

[0018] In some examples, when converting the negative / positive experimental results into probability values, the probability value of the positive experimental result is the known positive probability value, and the probability value of the negative experimental result is 100% - the known positive probability value.

[0019] In some examples, to avoid extreme values, if the probability value is 0% or 100%, a close value greater than 0 or a close value less than 100% is used for substitution respectively.

[0020] In some examples, the close value greater than 0 is 1%, and the close value less than 100% is 99%.

[0021] In some instances, the genus Listeria includes Listeria monocytogenes (Lmo), Listeria innocua (Lin), Listeria seeligeri (Lse), Listeria ivanovii (Liv), Listeria welshimeri (Lws), Listeria grayi (Lgy), Listeria marthii (Lma), Listeria rocourtiae (Lro), Listeria weihenstephanensis (Lwp), Listeria cornellensis (Lcn), Listeria riparia (Lri), Listeria grandensis (Lgd), Listeria fleischmannii (Lfc), Listeria aquatica (Laq), Listeria floridensis (Lfl), Listeria newyorkensis (Lny), and Listeria booriae (Lbo).

[0022] In some instances, the positive probability data of the biochemical reactions of the standard strains and isolated strains of the genus Listeria with known biochemical reaction positive probabilities are obtained through statistical analysis of the biochemical reaction experiments, or are further determined by combining public data.

[0023] In some instances, the positive probabilities of different biochemical reaction experiments of the Listeria are shown in the following table:

[0024]

[0025] The data in the table are the percentage positive rates of each biochemical reaction. HEM refers to the hemolysis test reaction; DXYL refers to the xylose reaction; RIB refers to the ribose reaction; TAG refers to the tagatose reaction; MAN refers to the mannitol reaction; LARA refers to the L - arabinose reaction; DARL refers to the D - arabinitol reaction; αMAN refers to the mannosidase reaction; DMAL refers to the maltose reaction; G1P refers to the glucose - 1 - phosphate reaction; LRHA refers to the rhamnose reaction; DIM refers to the arylamidase reaction.

[0026] In some examples, the criteria for judging the credibility of the identification results are as follows:

[0027]

[0028] In some examples, the strains to be tested are from food samples and environmental samples.

[0029] In some examples, the inconsistent biochemical experiments of the strains to be tested with each taxonomic unit are listed separately, and other known methods are used to identify and determine the strains to be tested.

[0030] The above features can be combined arbitrarily under the condition of non-conflict.

[0031] The beneficial effects of the present invention are:

[0032] In the numerical identification method of Listeria in some examples of the present invention, the negative or positive identification results presented by the biochemical reactions are distinct and easy to distinguish.

[0033] In the numerical identification method of Listeria in some examples of the present invention, effective differentiation of 17 species of Listeria genus is achieved, and the identification accuracy rate is as high as 99.23%. However, the existing commercially available numerical methods at home and abroad can usually only identify 6 species.

[0034] In the numerical identification method of Listeria in some examples of the present invention, the positive rate data of different species among the Listeria genus are obtained by counting after biochemical experiments on a large number of isolates, ensuring the accuracy and timeliness of the basic data. Description of the Drawings

[0035] Figure 1 It is the biochemical test result of Listeria ATCC 19115. Detailed Embodiments

[0036] A numerical identification method of Listeria includes the following steps:

[0037] Carry out biochemical reaction experiments on known Listeria of different species. The biochemical reaction experiments include a total of 12 biochemical experiments such as hemolysis test, xylose, ribose, tagatose, mannitol, L-arabinose, D-arabitol, mannosidase, maltose, glucose-1-phosphate, rhamnose and arylamidase experiments, obtain the biochemical reaction positive rates of different species of Listeria, and use the known Listeria of different species as different taxonomic units to determine the probability matrix of each taxonomic unit;

[0038] Calculate the mode frequency of the taxonomic unit Total occurrence frequency P (u / J) Multiple total occurrence frequency S, identification percentage ID and mode likelihood ratio T, where:

[0039] Mode frequency It is the product of the maximum total occurrence probabilities of the biochemical experiment combinations under this taxon,

[0040] where: LJ represents the probability of the occurrence of all biochemical tests for taxon J, and P(X i / J) represents the probability of the occurrence of a certain biochemical test for taxon J, and X i is the i-th biochemical test;

[0041] The total occurrence frequency P (u / J) is the product of the actual occurrence probabilities of the biochemical experiment combinations of a certain unknown bacterium to be identified under this taxon, P (u / J) = LJ = Π I P(X i / J);

[0042] The multinomial total occurrence frequency S is the sum of the total occurrence frequencies of each taxon, S = ∑P (u / J) = ∑(Π I P(X i / J));

[0043] The identification percentage ID is the total occurrence frequency of each taxon divided by the multinomial total occurrence frequency and then multiplied by 100%, and the sum of all identification percentages should be equal to 100%,

[0044] The mode likelihood ratio T is the total occurrence frequency of each taxon divided by the mode frequency of this taxon and then multiplied by 100%,

[0045] After the strain to be identified is pure-cultured on the Listeria selective medium, a biochemical reaction experiment is carried out. According to the positive probability data of the Listeria biochemical reaction, the negative / positive experimental results are converted into probability values, and the mode frequency of the strain to be identified is calculated The total occurrence frequency P (u / J) 、the multinomial total occurrence frequency S, the identification percentage ID, and the mode likelihood ratio T;

[0046] According to the identification percentage ID and the mode likelihood ratio T of the strain to be identified, the reliability of the identification result of the strain to be tested is evaluated, and the taxon with the highest score is the identification result of the strain to be tested.

[0047] In some examples, when converting the negative / positive experimental results into probability values, the probability value of the positive experimental result is the known positive probability value, and the probability value of the negative experimental result is 100% - the known positive probability value.

[0048] In some examples, to avoid extreme values, if the probability value is 0% or 100%, a close value greater than 0 or a close value less than 100% is used for substitution respectively.

[0049] In some instances, the proximity value >0 is 1%, and the proximity value <100% is 99%. Using such proximity values can better distinguish different Listeria species.

[0050] In some instances, the genus Listeria includes Listeria monocytogenes (Lmo), Listeria innocua (Lin), Listeria seeligeri (Lse), Listeria ivanovii (Liv), Listeria welshimeri (Lws), Listeria grayi (Lgy), Listeria marthii (Lma), Listeria rocourtiae (Lro), Listeria weihenstephanensis (Lwp), Listeria cornellensis (Lcn), Listeria riparia (Lri), Listeria grandensis (Lgd), Listeria fleischmannii (Lfc), Listeria aquatica (Laq), Listeria floridensis (Lfl), Listeria newyorkensis (Lny), and Listeria booriae (Lbo).

[0051] In some instances, the positive probability data of the biochemical reactions of Listeria are obtained by statistically analyzing the biochemical reaction experiments on standard strains and isolated strains of the genus Listeria whose positive probability data of biochemical reactions are known, or are determined by further combining public data.

[0052] In some instances, the positive probabilities of different biochemical reaction experiments of Listeria are shown in the following table:

[0053]

[0054] The data in the table are the percentage positive rates of various biochemical reactions. HEM refers to the hemolysis test reaction; DXYL refers to the xylose reaction; RIB refers to the ribose reaction; TAG refers to the tagatose reaction; MAN refers to the mannitol reaction; LARA refers to the L-arabinose reaction; DARL refers to the D-arabitol reaction; αMAN refers to the mannosidase reaction; DMAL refers to the maltose reaction; G1P refers to the glucose-1-phosphate reaction; LRHA refers to the rhamnose reaction; DIM refers to the arylamidase reaction.

[0055] The data in the table were statistically obtained by the inventors through long-term biochemical experiments on a large number of isolates, ensuring the accuracy and timeliness of the basic data.

[0056] In some examples, the criteria for determining the credibility of the identification results are as follows:

[0057]

[0058] In some examples, for suspicious or unacceptable identification results, other known methods are used for further identification.

[0059] In some examples, the strains to be tested are from food samples and environmental samples.

[0060] In some examples, the inconsistent biochemical experiments of the strains to be tested with each taxonomic unit are listed separately, and other known methods are used to identify and determine the strains to be tested.

[0061] The above features can be combined arbitrarily under the condition of non-conflict.

[0062] The following describes a specific embodiment of the present invention in detail in combination with the embodiments and the accompanying drawings, but it is not a limitation of the present invention. Unless otherwise defined, the technical and scientific terms used in the following embodiments have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0063] Example 1: Determining the species of the genus Listeria

[0064] Listeria monocytogenes (Lmo), Listeria innocua (Lin), Listeria seeligeri (Lse), Listeria ivanovii (Liv), Listeria welshimeri (Lws), Listeria grayi (Lgy), Listeria marthii (Lma), Listeria rocourtiae (Lro), Listeria weihenstephanensis (Lwp), Listeria cornellensis (Lcn), Listeria riparia (Lri), Listeria grandensis (Lgd), Listeria fleischmannii (Lfc), Listeria aquatica (Laq), Listeria floridensis (Lfl), Listeria newyorkensis (Lny), and Listeria booriae (Lbo)

[0065] Example 2: Select the biochemical experiments required in the numerical method

[0066] Through a non-probabilistic data matrix cladogram, 12 biochemical tests were selected: namely, hemolysin test (HEM), D-xylose (DXYL), D-ribose (RIB), D-tagatose (TAG), D-mannitol (MAN), L-arabinose (LARA), D-arabitol (DARL), α-mannosidase (αMAN), maltose (DMAL), glucose-1-phosphate (G1P), L-rhamnose (LRHA), and D-arylamidase (DIM).

[0067] Example 3: Establish a mathematical model in the numerical method

[0068] (1) Establish a positive probability database

[0069] Based on the existing standard strains and isolated strains of Listeria in this laboratory, conduct the 12 biochemical experiments in Example 2. According to the results, statistically calculate the positive rates of various biochemical reactions of different bacteria, and combine with materials such as Bergey's Manual of Determinative Bacteriology to establish a positive probability database for 17 Listeria species in Example 1, as shown in Table 1 specifically.

[0070] Table 1. Positive probability database of Listeria

[0071]

[0072] (2) Determine the calculation method in the numerical method

[0073] Calculate and determine the pattern frequency of each taxonomic unit according to the following function Total occurrence frequency P (u / J) , multiple total occurrence frequency S, identification percentage ID, and pattern likelihood ratio T:

[0074] Pattern frequency Is the product of the maximum total occurrence probability of the biochemical experiment combination under this taxonomic unit,

[0075] In the formula: LJ represents the probability of taxonomic unit J for all biochemical tests, P(X i / J) represents the probability of a certain biochemical test for taxonomic unit J, and X i Is the i-th biochemical test;

[0076] Total occurrence frequency P (u / J) Is the product of the actual occurrence probability of the biochemical experiment combination of a certain unknown bacterium to be identified under this taxonomic unit, P (u / J) = LJ = Π I P(X i / J);

[0077] Multiple total occurrence frequency S is the sum of the total occurrence frequencies of each taxonomic unit, S = ∑P (u / J) = ∑(Π I P(X i / J));

[0078] Identification percentage ID is the total occurrence frequency of each taxonomic unit divided by the multiple total occurrence frequency and then multiplied by 100%, and the sum of all identification percentages should be equal to 100%,

[0079] Pattern likelihood ratio T is the total occurrence frequency of each taxonomic unit divided by the pattern frequency of this taxonomic unit and then multiplied by 100%,

[0080] (3) Determine the result evaluation criteria

[0081] According to the magnitudes of the identification percentage and the T value, judge the credibility of the identification result according to the evaluation criteria in Table 2, and display the inconsistent biochemical experiments. In the result display, provide Chinese and Latin name contrasts for the taxonomic units.

[0082] Table 2. Evaluation of Identification Results

[0083]

[0084] Example 4: Establish an identification process for the strain to be tested

[0085] After the strain to be tested is pure cultured on the Listeria selective medium, perform the 12 biochemical experiments described in Example 2, namely hemolysis test, xylose, ribose, tagatose, mannitol, L - arabinose, D - arabinitol, mannosidase, maltose, glucose - 1 - phosphate, rhamnose, and arylamidase. A commercial identification strip can be used, or biochemical tubes can be purchased or the culture medium can be prepared by oneself for the biochemical experiments, and record the experimental results.

[0086] After obtaining the 12 biochemical experiment results of the strain to be tested, convert the negative / positive values of the results into probability values: if a certain test result is positive, use the corresponding positive probability value of this test in the database; if it is negative, use the value obtained by (100% - positive probability value) to represent the negative probability. To avoid extreme values, if the positive probability of any test is exactly 0% or 100%, we use the approximate values 1% or 99% for substitution.

[0087] Calculate the mode frequency, total occurrence frequency, multiple total occurrence frequency, identification percentage, and mode likelihood ratio T value of the strain to be tested and 17 taxonomic units according to the calculation method in Example 3. Due to limited space, it is impossible to list all the 12 reaction positive rates of the 17 taxonomic units, so known bacteria A, known bacteria B, known bacteria C and their biochemical experiment 1, biochemical experiment 2, biochemical experiment 3, biochemical experiment 4, and biochemical test 5 are used for illustration. The biochemical experiment results of the strain to be tested and the reaction positive rates of known bacteria A, known bacteria B, and known bacteria C are shown in Table 3.

[0088] Table 3. Biochemical Experiment Results of the Bacteria to be Tested and Reaction Positive Rates of Known Bacteria

[0089]

[0090] Known bacteria A, known bacteria B, and known bacteria C belong to any three of the 17 species of the genus Listeria, and biochemical experiment 1, biochemical experiment 2, biochemical experiment 3, biochemical experiment 4, and biochemical test 5 are any 5 of the 12 biochemical experiments.

[0091] Mode frequency of known bacterium A: 0.99×0.96×0.98×0.99×0.95 = 0.8760;

[0092] Mode frequency of known bacterium B: 0.95×0.98×0.99×0.99×0.99 = 0.9033;

[0093] Mode frequency of known bacterium C: 0.80×0.98×0.99×0.99×0.98 = 0.7530;

[0094] Total occurrence frequency of known bacterium A: 0.99×0.04×0.98×0.99×0.95 = 0.0365;

[0095] Total occurrence frequency of known bacterium B: 0.95×0.98×0.99×0.99×0.99 = 0.9033;

[0096] Total occurrence frequency of known bacterium C: 0.80×0.02×0.01×0.99×0.98 = 0.0002;

[0097] Total occurrence frequency of multiple items: 0.0365 + 0.9033 + 0.0002 = 0.9400;

[0098] Identification percentage of known bacterium A: 0.0365 / 0.9400 × 100% = 3.883%; Identification percentage of known bacterium B: 0.9033 / 0.9400 × 100% = 96.096%; Identification percentage of known bacterium C: 0.0002 / 0.9400 × 100% = 0.021%;

[0099] T value of known bacterium A: 0.0365 / 0.8760 × 100% = 4.17%; T value of known bacterium B: 0.9033 / 0.9033 × 100% = 100%; T value of known bacterium C: 0.0002 / 0.7530 × 100% = 0.03%;

[0100] Since the present invention includes 12 biochemical reactions of 17 taxonomic units, the identification percentage of 17 taxonomic units and the mode likelihood ratio T value need to be calculated for the identification of each test strain, and the manual calculation has a large amount of calculation. Therefore, the calculation process is carried out in the corresponding supporting software.

[0101] Sort according to the size of the identification percentage and the mode likelihood ratio T value, and judge according to the identification result evaluation criteria described in Example 3. In this example, the identification result of the test bacterium is: known bacterium B, an excellent identification result. List the inconsistent biochemical experiments of the test strain and each taxonomic unit in the identification result separately. And in the result display, give the Chinese and Latin names in contrast for the taxonomic unit.

[0102] Example 5: Verification of the Accuracy of the Numerical Identification Biochemical Kit for Listeria

[0103] To verify the accuracy of the numerical identification biochemical kit for Listeria, 11 standard strains of Listeria and 260 Listeria isolates isolated by this laboratory from six major categories of foods, namely vegetables, frozen foods, milk powder, meat products, aquatic products, and cooked foods, across the country were selected as test subjects, and the known strains were identified according to the procedure described in Example 4.

[0104] The results of the numerical identification biochemical kit for 11 standard strains of Listeria are shown in Table 4. The results show that good identification results were obtained for all 11 standard strains. Among them, the biochemical test results of ATCC 19115 are as Figure 1 .

[0105] Table 4. Identification Results of the Standard Strains of Listeria by the Present Invention

[0106]

[0107] The identification results of the 260 known isolates by MALDI-TOF MS were compared with the identification results of the numerical identification biochemical kit for Listeria of the present invention. The results show that the identification accuracy of the present invention for 260 Listeria isolates is 99.23% (Table 5), which is relatively consistent with the results of MALDI-TOF MS. Generally speaking, the numerical identification biochemical kit for Listeria established by the present invention has a high identification level.

[0108] Table 5. Comparison of the Identification Results of Listeria Isolates by MALDI-TOF MS and the Numerical Identification Biochemical Kit

[0109]

[0110] The above is a further detailed description of the present invention, and it should not be regarded as a limitation on the specific implementation of the present invention. For those of ordinary skill in the technical field to which the present invention pertains, simple deductions or substitutions without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A numerical identification method for Listeria, comprising the following steps: Carry out biochemical reaction experiments on known Listeria of different species. The biochemical reaction experiments include a total of 12 biochemical experiments: hemolysis test, xylose, ribose, tagatose, mannitol, L - arabinose, D - arabitol, mannosidase, maltose, glucose - 1 - phosphate, rhamnose, and arylamidase experiments. Obtain the positive rates of biochemical reactions of different species of Listeria. Take different known Listeria as different taxonomic units and determine the probability matrix of each taxonomic unit; Calculating the mode frequency of taxa Total occurrence frequency P (u / J) , multinomial total occurrence frequency S, identification percentage ID, and mode likelihood ratio T, where: Mode frequency is the product of the maximum total occurrence probabilities of the biochemical experiment combinations under this taxon, where: LJ represents the probability of the occurrence of all biochemical tests for taxonomic unit J, P(X i / J) represents the probability of occurrence of a certain biochemical test for taxon J, where X i is the i-th biochemical test; Total occurrence frequency P (u / J) is the product of the actual occurrence probabilities of a biochemical experiment combination of an unknown bacterium to be identified under this taxonomic unit, P (u / J) = LJ = Π I P(X i / J); The total frequency of occurrence S of multiple items is the sum of the total occurrence frequencies of each taxonomic unit, S = ∑P (u / J) = ∑(Π I P(X i / J)); The identification percentage ID is calculated by dividing the total occurrence frequency of each taxonomic unit by the total occurrence frequency of multiple items and then multiplying by 100%. The sum of all identification percentages should be equal to 100%. The pattern likelihood ratio T is the total occurrence frequency of each taxonomic unit divided by the pattern frequency of that taxonomic unit and then multiplied by 100%, After the strain to be identified is pure cultured on Listeria selective medium, a biochemical reaction experiment is carried out. According to the positive probability data of the Listeria biochemical reaction, the negative / positive experimental results are converted into probability values, and the pattern frequency of the strain to be identified is calculated. Total occurrence frequency P (u / J) , multiple total occurrence frequency S, identification percentage ID, and pattern likelihood ratio T; According to the identification percentage ID of the strain to be identified and the pattern likelihood ratio T, evaluate the credibility of the identification result of the strain to be tested. The taxonomic unit with the highest score is the identification result of the strain to be tested.

2. The numerical identification method according to claim 1, wherein The positive probability data of the biochemical reactions of Listeria is obtained by statistical analysis of the standard strains and isolated strains of the genus Listeria for which the data is known, or is determined by further combining public data.

3. The numerical identification method according to claim 1, characterized in that, When converting the negative / positive experimental results into probability values, the probability value of a positive experimental result is the known positive probability value, and the probability value of a negative experimental result is 100% - the known positive probability value.

4. The numerical identification method according to claim 2, characterized in that, To avoid extreme values, if the probability value is 0% or 100%, substitute it with a value close to >0 or <100% respectively.

5. The numerical identification method according to claim 4, characterized in that The value close to >0 is 1%, and the value close to <100% is 99%.

6. The numerical identification method according to any one of claims 1 to 5, characterized in that, The genus Listeria includes Listeria monocytogenes (Lmo), Listeria innocua (Lin), Listeria seeligeri (Lse), Listeria ivanovii (Liv), Listeria welshimeri (Lws), Listeria grayi (Lgy), Listeria marthii (Lma), Listeria rocourtiae (Lro), Listeria weihenstephanensis (Lwp), Listeria cornellensis (Lcn), Listeria riparia (Lri), Listeria grandensis (Lgd), Listeria fleischmannii (Lfc), Listeria aquatica (Laq), Listeria floridensis (Lfl), Listeria newyorkensis (Lny), and Listeria booriae (Lbo).

7. The numerical identification method according to claim 6, characterized in that, The positive probabilities of different biochemical reaction experiments of the Listeria are shown in the following table: The data in the table are the percentage positive rates of each biochemical reaction. HEM refers to the hemolysis test reaction; DXYL refers to the xylose reaction; RIB refers to the ribose reaction; TAG refers to the tagatose reaction; MAN refers to the mannitol reaction; LARA refers to the L - arabinose reaction; DARL refers to the D - arabinitol reaction; αMAN refers to the mannosidase reaction; DMAL refers to the maltose reaction; G1P refers to the glucose - 1 - phosphate reaction; LRHA refers to the rhamnose reaction; DIM refers to the arylamidase reaction.

8. The numerical identification method according to any one of claims 1 to 5, characterized in that, The credibility determination criteria for the identification results are as follows:

9. The numerical identification method according to any one of claims 1 to 5, characterized in that The test strains are from food samples and environmental samples.

10. The numerical identification method according to any one of claims 1 to 5, characterized in that List the inconsistent biochemical experiments of the test strains with each taxonomic unit separately, and use other known methods to identify and determine the test strains.

Images