Microorganism discrimination method and microorganism discrimination device

By comparing the mass-to-charge ratio list of samples with the genomic database in the microbial discrimination method, and weighted screening of protein mass-to-charge ratios in specific groups, the problem of insufficient microbial discrimination accuracy in the prior art is solved, and higher accuracy is achieved Sex and reliability.

CN119095976APending Publication Date: 2024-12-06SHIMADZU SEISAKUSHO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380034781.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-21
Filing Date
2023-04-03
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing microbial discrimination methods based on genomic databases have problems with insufficient accuracy in the construction and use of mass-to-charge ratio databases, especially misjudgment caused by false peaks.

Method used

By obtaining the mass-to-charge ratio list of samples, it is compared with the mass-to-charge ratio database predicted by genomic data, and weighted on the mass-to-charge ratio of proteins contained in a specific group, combining the expression, function and similarity of proteins in a specific group. Secondary screening is carried out to improve discrimination accuracy.

Benefits of technology

It improves the accuracy of microbial discrimination, reduces the misjudgment caused by false peaks, and enhances the accurate identification ability of samples.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119095976A_ABST
    Figure CN119095976A_ABST
Patent Text Reader

Abstract

The method for discriminating microorganisms comprises: a step (ST32) for acquiring a sample list, which is a list of mass-to-charge ratios of samples; a step (ST40-ST44, ST40A) for determining a sample by comparing a mass-to-charge ratio database, which is a database of the sample list and a mass-to-charge ratio list for each microorganism predicted from genomic data, on the basis of weighting a mass-to-charge ratio corresponding to a protein contained in a specific group; and steps (ST50, ST54) for outputting the result of the determination.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a microorganism identification method and a microorganism identification device. Background Art

[0002] Non-patent document 1 discloses two methods that can be adopted for identification of microorganisms using mass spectrometry.

[0003] The first method is the fingerprint method, which compares the mass spectrum measured for an unknown microorganism with a database of mass spectra measured for each known microorganism to identify the unknown microorganism. However, this method has the problem that the mass spectrum pattern of the microorganism is easily strongly affected by the culture medium conditions and the measurement reproducibility method.

[0004] In order to solve the problems of the fingerprint method, a second method is to focus on a method based on bioinformatics using a genome database. In this method, the mass spectrum measured for an unknown microorganism is compared with a database of mass-to-charge ratios of proteins predicted from the genome database, thereby distinguishing the unknown microorganism. In this method, the predicted mass-to-charge ratio is not affected by the culture medium conditions and the measurement reproducibility method, so the above-mentioned problems of the fingerprint method can be solved.

[0005] Prior art literature

[0006] Non-patent literature

[0007] Non-patent document 1: Hiroto Tamura et al., "Bacterial Identification Using MALDI-TOF MS Using Ribosomal Proteins as Biomarkers - S10 - Rapid Identification of Bacteria Using GERMS -", Shimadzu Review Supplement, Vol. 70, No. 3-4, 2013 Summary of the invention

[0008] Technical problem to be solved by the invention

[0009] In the second method, further improvement is desired in the step of comparing the mass spectrum of an unknown microorganism with a database of predicted mass-to-charge ratios to identify the microorganism.

[0010] The present disclosure is made to solve the above-mentioned technical problems, and its purpose is to improve the accuracy of distinguishing microorganisms in a microorganism distinguishing method using a mass-to-charge ratio database constructed based on a genome database.

[0011] Solutions for solving the above technical problems

[0012] The microorganism identification method of the first aspect of the present disclosure comprises: a step of obtaining a list of mass-to-charge ratios of samples, i.e., a sample list; a step of identifying samples by comparing the sample list and a database of mass-to-charge ratio lists of each microorganism predicted based on genomic data, i.e., a mass-to-charge ratio database, based on weighted mass-to-charge ratios corresponding to proteins included in a specific group; and a step of outputting the identification result.

[0013] The microorganism identification device of the second aspect of the present disclosure includes a storage unit, a processor, and an output unit. The storage unit stores a database of mass-to-charge ratio lists of each microorganism predicted based on genome data, namely a mass-to-charge ratio database. The processor obtains a list of mass-to-charge ratios of samples, namely a sample list, and for the sample list and the mass-to-charge ratio list included in the mass-to-charge ratio database, weights the mass-to-charge ratios corresponding to the proteins included in a specific group for comparison, thereby identifying the sample. The output unit outputs the result of the identification.

[0014] Effects of the Invention

[0015] According to the microorganism identification method of the present disclosure, when comparing the list of mass-to-charge ratios of the sample with the mass-to-charge ratio database of each microorganism predicted based on genomic data, weighting is performed on the mass-to-charge ratios corresponding to the proteins included in a specific group. Therefore, in the microorganism identification method using the genomic database and mass analysis method, the accuracy of identifying microorganisms can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram showing the configuration of a microorganism identification system according to an embodiment of the present invention.

[0017] Figure 2 This is a flowchart showing an overview of processing performed by the device.

[0018] Figure 3 This is a functional block diagram of an apparatus related to the construction of a mass-to-charge ratio database.

[0019] Figure 4 This is a functional block diagram of a device related to sample identification.

[0020] Figure 5 is a flowchart showing the process of constructing a mass-to-charge ratio database.

[0021] Figure 6 : is a flowchart showing a subroutine for determining genome data.

[0022] Figure 7 : is a flowchart showing the process of adding new genome data.

[0023] Figure 8: is a flowchart showing the processing related to the discrimination of the sample.

[0024] Fig. 9 This is a flowchart showing another example of processing related to sample discrimination.

[0025] Fig.10 This is a diagram showing the relationship between the total number of base sequences of gene sites in a genome and the estimated number of genes per genome. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In addition, the same reference numerals are given to the same or corresponding parts in the drawings, and the description thereof will not be repeated in principle.

[0027] [1. Structure of the microbial identification system]

[0028] Figure 1 1 is a schematic diagram showing the configuration of a microorganism identification system 1000 according to an embodiment of the present invention.

[0029] Reference Figure 1 The microorganism identification system 1000 includes a public genome database 70, a public classification database 80, a network 90, and an apparatus 100. In addition, in this specification, "database" is also described as "DB".

[0030] The public genome DB70 is a database containing genome data of organisms. Genome refers to the genetic information on nucleic acids (deoxyribonucleic acid (DNA), ribonucleic acid (RNA)) possessed by a certain organism, including the base sequence of nucleic acids. In this specification, genome data mainly refers to DNA sequence data.

[0031] The public genome DB 70 is typically a DB containing a large amount of genome data of organisms that are generally disclosed, such as NCBI (National Center for Biotechnology Information), DDBJ (DNA Data Bank of Japan), and EMBL (European Molecular Biology Laboratory). However, examples of the public genome DB 70 are not limited to these, and for example, genome DBs that are generally not disclosed may also be included.

[0032] The public classification DB80 is a database containing data related to the classification of organisms (hereinafter referred to as classification data). The classification of organisms generally refers to the classification based on the relationship between organisms shown by the hierarchy of family, genus, species, etc. In the classification of microorganisms, classification is traditionally based on both phenotype and genome, based on multiple indicators such as morphological observation, performance characteristics, chemical classification indicators, protein analysis and DNA analysis, but there are also classification systems based only on genome information, and there are multiple classification systems.

[0033] The public classification DB 80 is typically a DB containing generally disclosed biological classification data, such as GTDB (Genome Taxonomy Database), RDP (Ribosomal Database Project), Silva, etc. However, examples of the public classification DB 80 are not limited thereto, and may include, for example, generally undisclosed DBs.

[0034] The network 90 is a network for the device 100 to communicate with the public genome DB 70 and the public classification DB 80. The network 90 is, for example, the Internet that interconnects a large number of government, corporate, public, and private networks on the earth.

[0035] The device 100 is a device for constructing a mass-to-charge ratio (m / z) DB for discriminating microorganisms using mass spectrometry. In the present specification, discriminating microorganisms means identifying microorganisms taxonomically. That is, for example, identifying at least one of the genus, species, strain, and system of a microorganism. Therefore, the device 100 corresponds to an embodiment of a "device for constructing a database for discriminating microorganisms." In addition, the device 100 is a device for utilizing the m / zDB and discriminating microorganisms using mass spectrometry. Therefore, the device 100 also corresponds to an embodiment of a "device for discriminating microorganisms." In addition, in the present specification, the "type" of a microorganism or organism includes, for example, at least one of the "levels of systematic taxonomic groups such as genotypes, strains, or subspecies, species, genera, and families" of the microorganism or organism.

[0036] The device 100 includes a controller 101, a display 15, and an operation unit 14. The display 15 and the operation unit 14 are connected to the controller 101. The operation unit 14 is typically composed of a touch panel, a keyboard, a mouse, etc. The operation unit 14 receives the user's operation input to the processor 10. The display 15 is composed of, for example, a liquid crystal panel capable of displaying images. The display 15 displays images related to the acceptance of the user's operation input and displays the results of the processing of the processor 10.

[0037] The controller 101 includes a processor 10, a memory 11, a communication interface (I / F) 12, and an input / output I / F 13 as main components. These components are connected to each other via a bus so as to be communicable with each other.

[0038] The processor 10 is typically a calculation processing unit such as a CPU (Central Processing Unit) or an MPU (MultiProcessing Unit) and controls the operation of the device 100 by reading and executing a program stored in the memory 11 .

[0039] The memory 11 is implemented by storage devices such as ROM (Read Only Memory), RAM (Random Access Memory) and HDD (Hard Disk Drive). ROM can store programs executed by the processor 10. RAM can temporarily store data used in the execution of programs in the processor 10, and can function as a temporary data storage device used as a work area. HDD is a non-volatile storage device. In addition to or instead of HDD, semiconductor storage devices such as flash memory can also be used. In addition, the above-mentioned programs and / or data can also be stored in an external storage device that can be accessed by the processor 10. The memory 11 corresponds to an embodiment of the "storage unit".

[0040] The communication I / F 12 is a communication interface for exchanging various data with external devices including the public genome DB 70 and the public classification DB 80, and is implemented by an adapter or a connector, etc. In addition, the communication method can be a wireless communication method based on a wireless LAN (Local Area Network) or the like, or a wired communication method using a USB (Universal Serial Bus) or the like.

[0041] The input / output I / F 13 is an interface for exchanging various data between the processor 10 and an external device connected to the input / output I / F 13. The external device includes an operation unit 14 and a display 15. A mass spectrometer (MS) 16 may also be connected to the input / output I / F 13. In this specification, the input / output I / F 13 also includes a device for exchanging data between a storage terminal such as a USB memory connected to the device 100 and the processor 10.

[0042] MS16 is a device for performing mass analysis of components contained in a sample, such as MALDI-TOF MS (Matrix-Assisted Laser Desorption / Ionization Time-of-Flight Mass Spectrometry), MALDI-IT-TOF (Matrix-Assisted Laser Desorption / Ionization Ion Trap Time-of-Flight Mass Spectrometry), or scanning IT-MS, but is not limited thereto. In the case where MS16 is MALDI-TOF MS, ions generated by laser irradiation are drawn into a flight tube to make them fly, and are separated according to the flight time and then detected. The flight time is related to the mass-to-charge ratio m / z of the component. As a result, a mass spectrum is obtained in which m / z is shown on the horizontal axis and the detected ion intensity is shown on the vertical axis.

[0043] In this specification, MS16 performs mass analysis of proteins in a sample. Therefore, in a mass spectrum, peaks are detected based on the m / z of proteins in a sample. Therefore, if a pattern of a mass spectrum is referred to, more specifically, a list of m / z of peaks having a height above a predetermined threshold value is obtained (also referred to as an m / z list in this specification), the proteins contained in the sample can be known. In this specification, the m / z contained in the m / z list indicates the m / z corresponding to the peak of the mass spectrum.

[0044] Different types of organisms contain different proteins, and therefore their mass spectrum patterns and m / z lists are also different. Therefore, organisms can be identified based on the mass spectrum patterns and m / z lists.

[0045] After performing mass analysis on unknown microorganisms as samples, the MS 16 transmits a sample list, which is a list of m / z of the samples, to the apparatus 100. The processor 10 identifies the samples based on the sample list.

[0046] Furthermore, the apparatus 100 does not need to be constituted by one computer, and may be constituted by a plurality of computers.

[0047] [2. Comparison with existing devices]

[0048] Conventionally, as a method for identifying microorganisms using such a mass spectrometer, a fingerprint method has been performed in which a database including mass spectra actually measured for each microorganism is constructed and the database is compared with mass spectra of unknown microorganisms.

[0049] However, in order to construct a practical database for fingerprinting, the mass spectrum data of a large number of types (e.g., hundreds of types) of microorganisms are required. In addition, even for the same type of microorganisms, the mass spectrum pattern may change due to genetic diversity, culture conditions, pre-treatment of mass spectrometry, and deviations associated with repeated measurements. Therefore, in view of these actual conditions, a large amount of mass spectrum data is required in a practical database. For example, dozens of mass spectrum data are required for each type of microorganism, and a total of tens of thousands of mass spectrum data are required for all types of microorganisms. That is, in order to construct a practical database, it is necessary to actually carry out a very large number of times (e.g., tens of thousands of times) of microbial cultivation and mass spectrum measurement, which consumes huge costs.

[0050] Therefore, as a method for identifying microorganisms using a new mass spectrometry method, the following method is focused on: using a public genome database to predict the expressed protein, constructing a database of m / z lists predicted from the protein, i.e., m / zDB, and using the m / zDB. In this method, the sample is identified by comparing the m / z list contained in the m / zDB with the m / z list corresponding to the peak of the mass spectrum of the unknown microorganism, i.e., the sample list. In this method, there is no need to actually culture the microorganism and measure the mass spectrum, and compared with the above-mentioned fingerprint method, it is possible to more easily construct a mass spectrum database.

[0051] However, this method also has room for improvement in terms of the quality of the predicted m / zDB and the accuracy of microorganism discrimination using the m / zDB.

[0052] For example, in this method, low-quality genome data (e.g., genome data containing a large number of undetermined bases) included in the published genome database is also reflected in the m / zDB. As a result, there is a concern that the quality of the m / zDB may be reduced, and the accuracy of microbial identification using the m / zDB may also be reduced.

[0053] Therefore, in the device 100 according to the present embodiment, the m / zDB is constructed based only on high-quality genome data that meets the prescribed criteria among the genome data obtained from the public genome DB 70. This can improve the quality of the m / zDB. In addition, the accuracy of microorganism identification using the m / zDB can be improved.

[0054] Furthermore, in the previous method of distinguishing microorganisms using the predicted m / zDB, other problems are also worried. For example, there is a worry that the predicted m / z list contains false peaks that do not appear in the mass spectrum when actually measured. This is because, even if it is a sequence predicted to express a protein based on genomic data, in fact, for some reason, the protein is not expressed, or even if it is expressed, it is not ionized, and it may not be detected as a peak in the measured mass spectrum. As a result, when the sample list is compared with the predicted m / z list, it is considered that the false peak becomes noise, and the sample list may match the m / z list of microorganisms of a type unrelated to the sample. Therefore, it is considered possible to distinguish the sample as an unrelated type of microorganism. That is, there is a worry that the accuracy of distinguishing microorganisms will be reduced.

[0055] Therefore, in the apparatus 100 according to this embodiment, proteins with a low probability of becoming false peaks, that is, "proteins with a high probability of being expressed in the microorganism and detected as peaks when measuring the mass spectrum" are weighted to distinguish microorganisms. Therefore, the possibility of erroneously distinguishing different microorganisms due to the influence of false peaks is reduced. As a result, the accuracy of distinguishing microorganisms can be improved.

[0056] [3. Overview of the Processing of the Device]

[0057] Figure 2 1 is a flowchart showing an overview of the processing performed by the device 100. In step (hereinafter also referred to as ST) 101, the processor 10 of the device 100 constructs an m / zDB based on the genome data of the public genome DB 70. In ST102, the processor 10 uses the m / zDB to identify a sample as an unknown microorganism.

[0058] (3-1. Functional blocks related to the construction of m / zDB)

[0059] Figure 3 is with Figure 2 Functional block diagram of the apparatus 100 related to the construction of m / zDB corresponding to ST101 of FIG. Figure 3 The device 100 includes a genome data collecting unit 21 , a genome data determining unit 22 , a protein predicting unit 23 , an m / zDB constructing unit 24 and a storage unit 25 .

[0060] The genome data collection unit 21 collects genome data from the public genome DB 70 .

[0061] The genome data determination unit 22 determines whether the collected genome data satisfies a predetermined criterion related to the quality of the genome data.

[0062] The protein prediction unit 23 predicts the expressed protein for the genome data that meets the specified criteria. Specifically, the inferred gene region is predicted from the DNA sequence, and the amino acid sequence is predicted based on the inferred gene region. Then, the expressed protein is predicted based on the amino acid sequence.

[0063] The m / zDB construction unit 24 predicts an m / z list based on the predicted proteins, constructs an m / zDB, and stores it in the storage unit 25. The m / zDB includes, for example, two types of m / zDBs. One m / zDB is a global m / zDB that includes m / z corresponding to all proteins predicted from the genome data. The other m / zDB is a specific m / zDB that includes only m / z corresponding to proteins included in a specific group among the proteins predicted from the genome data. The two m / zDBs are used to Figure 4 Discrimination of samples as unknown microorganisms as described in .

[0064] The genome data collection unit 21, the genome data determination unit 22, the protein prediction unit 23 and the m / zDB construction unit 24 are Figure 1 The storage unit 25 corresponds to Figure 1 Memory 11.

[0065] (3-2. Functional blocks related to sample identification)

[0066] Figure 4 is with Figure 2 Functional block diagram of the device 100 related to sample identification corresponding to ST102 of FIG. Figure 4 The device 100 includes an acquisition unit 31, a sample identification unit 32, an annotation unit 33, an output unit 34 and a storage unit 25.

[0067] The acquisition unit 31 acquires a sample list. The sample list is acquired, for example, from the MS 16 connected to the device 100. The method of acquiring the sample list is not limited thereto, and for example, the sample list may be acquired from an external device that communicates with the device 100, or from a storage terminal connected to the device 100. The acquisition unit 31 also estimates the measurement error of the m / z included in the sample list as needed and performs correction. The acquisition unit 31 corresponds to Figure 1 Processor 10.

[0068] The sample identification unit 32 compares the sample list and the m / zDB stored in the storage unit 25 on the basis of weighting the m / z corresponding to the proteins included in the specific group, thereby identifying the sample. The sample identification unit 32 includes, for example, a primary screening unit 321 and a secondary screening unit 322. The primary screening unit 321 uses the m / z list included in the specific m / zDB to perform screening based on the m / z corresponding to the proteins included in the specific group. The secondary screening unit 322 performs screening based on the m / z corresponding to all proteins in the m / z list corresponding to the genome data locked in the primary screening in the entire m / zDB, thereby identifying the sample. The sample identification unit 32 corresponds to Figure 1 Processor 10.

[0069] The annotation unit 33 links information related to the predicted protein, i.e., annotations, to each m / z included in the sample list. In linking the annotations, for example, software is used to search for the name of the corresponding protein based on the mass of the protein. The annotation unit 33 corresponds to Figure 1 Processor 10.

[0070] The identification result and m / z annotation of the sample identification unit 32 are stored in the storage unit 25 and / or outputted by the output unit 34. The output unit 34 is equivalent to Figure 1 The processor 10, the display 15 or the communication I / F 12. That is, the identification result and the comment are displayed on the display 15, and / or transmitted to an external device via the communication I / F 12. Thus, the user can recognize the identification result and the comment.

[0071] [4. Flow of processing related to the construction of m / zDB]

[0072] (4-1. Construction of m / zDB)

[0073] Next, the flow of processing performed by the device 100 will be described in detail.

[0074] Figure 5 is a flowchart showing the process of building an m / zDB. Figure 5 The processing of ST02 to ST28 shown is equivalent to Figure 2 The processing in ST101.

[0075] Reference Figure 5 In ST02, the processor 10 acquires genome data of microorganisms from the public genome DB 70. At this time, by acquiring genome data from a plurality of public genome DBs 70, genome data of clinically or industrially important microorganism species can be comprehensively collected.

[0076] In ST04, the processor 10 integrates the acquired genome data and constructs a collected genome DB.

[0077] In ST06, the processor 10 determines whether the genome data in the collected genome DB meets a predetermined benchmark. The benchmark is set in such a way that only high-quality genome data meets the benchmark. Figure 6 The specific contents of the benchmark are described in.

[0078] In ST08 , the processor 10 constructs a high-quality genome DB including genome data determined to satisfy the criteria.

[0079] In ST10, the processor 10 predicts the genes contained in the genome data contained in the high-quality genome DB. A gene refers to a specific region that is translated into a protein on DNA or the information contained in the region. Gene prediction includes, for example, using the translation start codon (ATG sequence) and stop codon (TGA sequence) as clues to infer the putative gene region that is translated into a protein on the genome data.

[0080] In ST12, the processor 10 predicts the amino acid sequence after translation based on the predicted gene. The prediction of the amino acid sequence includes, for example, estimating the amino acids corresponding to each codon (three-base sequence) included in the predicted gene region and linking them.

[0081] In ST14, the processor 10 predicts post-translational modifications for the protein composed of the predicted amino acid sequence. Post-translational modifications are modifications made to proteins in order to make the proteins just translated into proteins that actually function in various places in the body. Post-translational modifications include, for example: decomposition of proteins including removal of methionine and removal of signal peptides; and specific chemical modifications including phosphorylation. Post-translational modifications are applied to almost all proteins to change their m / z. Therefore, in view of the post-translational modifications, a more accurate m / z of the protein can be calculated.

[0082] In ST16 , the processor 10 predicts proteins to which the predicted post-translational modification is applied.

[0083] In ST18, the processor 10 predicts the m / z list of each genome data based on the protein. Specifically, the m / z corresponding to the protein is calculated based on the mass of the atoms contained in the protein. In addition, as the mass of the atom, it is preferred to use the average mass of the element reflecting the isotopic distribution of the element in nature. In this way, a more accurate m / z can be calculated.

[0084] In ST20A, the processor 10 constructs a database including mass-to-charge ratios of the m / z list, namely, a global m / z DB. The global m / z DB includes all m / z predicted for each genome data.

[0085] On the other hand, in ST22, the processor 10 links the annotation to the data of the protein predicted in ST16. Annotation generally refers to information related to the protein, including the name and function of the protein. The annotation linking is performed, for example, using a common software for adding annotations according to m / z, but is not limited to this. For example, the device 100 may create a table showing the association between m / z and annotation based on the public genome DB 70 and the public classification DB 80, and use the table.

[0086] In this specification, annotation refers to information related to proteins, including information about groups containing proteins. Information related to groups of proteins includes at least one of the name, function, and family of the protein.

[0087] One of the advantages of linking annotations is that it is possible to select m / z corresponding to proteins included in a specific group based on the annotations and process them separately from m / z corresponding to other proteins. Therefore, for example, it is possible to selectively weight the m / z corresponding to "a group of proteins that are highly likely to be expressed in the microorganism and are also highly likely to be detected as peaks when measuring the mass spectrum" to distinguish microorganisms. Thus, in the m / z list predicted from genomic data, it is possible to weight "m / z corresponding to proteins that are highly likely to be expressed in the microorganism and are also highly likely to be detected as peaks when measuring the mass spectrum" over "m / z (false peaks) corresponding to proteins that are not actually expressed as proteins in the microorganism or do not appear in the mass spectrum even if expressed" to screen samples. Therefore, it is possible to prevent false peaks included in the predicted m / z list from becoming noise, thereby reducing the accuracy of sample screening.

[0088] In order to "have a high possibility of being expressed in vivo and detected as a peak when measuring a mass spectrum", it is preferred to select a group based on at least one of the following conditions: a condition that the expression amount is above a specified threshold value, a condition that it has a function that is indispensable for maintaining life, a condition that microorganisms whose amino acid sequence similarity (homology) is above a specified threshold value exist at a specified ratio or above in microorganisms classified as a specified type (for example, microorganisms belonging to a specified family), a condition that it is a basic protein, a condition that the mass-to-charge ratio can be analyzed within an error range of ±14 Da (more preferably within ±3 Da) when measured by MALD I-MS measurement, a condition that the mass of the protein is included in 4 to 30 kDa (more preferably 2 to 20 kDa), a condition that the number of types of proteins included in the group is above a specified number, and a condition that the number of microorganisms containing the genome data among microorganisms classified as a specified type (for example, microorganisms belonging to a specified family) is above a specified ratio.

[0089] Furthermore, the functions essential for maintaining life include at least one of the maintenance and proliferation of cells.

[0090] The group determined based on such conditions is, for example, ribosomal proteins. Other examples of the group are chaperone proteins and DNA binding proteins.

[0091] In addition, the group is not limited to the proteins significantly expressed in all microorganisms as exemplified above, and may also be proteins known to be significantly expressed in specific microorganisms. For example, if specific proteins known to be significantly expressed in each genus are weighted to discriminate a sample, the possibility of discriminating the sample as the correct genus can be increased. In addition, an example of a "significantly expressed protein" in this specification is a protein showing an expression amount above a predetermined threshold.

[0092] In ST24, the processor 10 selects proteins predicted to be included in a specific group based on information related to the group included in the annotation. In the next ST26, the processor 10 predicts a specific m / z list containing only m / z predicted based on the selected protein. In ST20C, the processor 10 constructs an m / z database, i.e., a specific m / z DB, containing the specific m / z list.

[0093] Another advantage of linking annotations is that it is easy for the user to understand whether the m / z included in the m / z list corresponds to which protein. From this point of view, in order to easily use the annotations for m / z, in ST20B, the processor 10 constructs an annotation DB that summarizes the annotations for m / z included in the entire m / z DB.

[0094] Another advantage of linking annotations is that the appropriateness of the comparison between the sample list and the m / z list contained in the m / zDB can be studied with reference to the annotations. For example, it is possible to refer to the annotations in the m / z list of the m / zDB that is judged to have a high degree of pattern matching between the sample list and the m / z. At this time, when the m / z list contains a large number of m / z corresponding to proteins that are presumed to be not expressed in the microorganism according to the annotations, the reliability of the m / z list itself is questionable, so the appropriateness of the comparison with the sample list is low, and the reliability of the discrimination of the sample is also low. In addition, it can be seen that in the sample list, when the m / z corresponding to the protein that is considered to be functionally important and preserved from an evolutionary perspective is consistent with the m / z of the noise in the m / z list, it can be seen that the appropriateness of the comparison is low, and the reliability of the discrimination of the sample is also low. In this way, when an m / z list with low appropriateness compared with the sample list is found, the user can remove the m / z list, etc., to improve the reliability of the discrimination.

[0095] The annotations of the annotation DB are linked to the m / z included in the m / zDB. As an example, the m / zDB is associated with the annotation DB so that when the m / z in the m / z list included in the m / zDB is referred to, the corresponding annotations included in the annotation DB can be referred to together. In addition, as another example, the annotation DB can be constituted as a part of the m / zDB by adding corresponding annotations to the m / z included in the m / zDB.

[0096] In ST28, the processor 10 acquires the classification data of the public classification DB 80. In ST20D, the processor 10 constructs a collection classification DB obtained by integrating the collected classification data. At this time, if the collection classification DB is constructed based on the classification data of multiple public classification DBs 80, a wide range of taxonomic systems can be introduced. Therefore, if the collection classification DB is used, various taxonomic systems can be reflected in the discrimination results of microorganisms.

[0097] In addition, the collection classification DB may include a genome ID as an ID for each genome. The genome ID is generated based on the collected classification data, for example.

[0098] The classification data of the collection classification DB is associated with the data contained in the entire m / zDB, the specific m / zDB, and the annotation DB. Therefore, a genome ID can be added to each genome data of the entire m / zDB and the specific m / zDB. In addition, the content of the collection classification DB can also be used for the arrangement of the entire m / zDB and the specific m / zDB, or reflected in the content. In addition, the collection classification DB can also be used for other purposes in the device 100, such as determining the above-mentioned "specific protein known to be significantly expressed only in a specific species".

[0099] The four corresponding DBs are collectively referred to as microorganism DBs. After the processor 10 constructs the microorganism DB in ST20A to ST20D, the processing is temporarily terminated. Thus, the device 100 can use the microorganism DB to perform Figure 8 , Fig. 9 Identification of samples using mass spectrometry as detailed in .

[0100] Figure 5 The processing shown is performed, for example, once a year in accordance with the update of the public genome DB 70. As a result, the content updated in the public genome DB 70 can be appropriately reflected in the microorganism DB, so the content of the microorganism DB is further improved.

[0101] (4-2. Determination of genome data)

[0102] Figure 6 It is a diagram showing the determination process of genome data. Figure 6 The ST060~ST069 shown are Figure 2 The processing corresponding to ST06. Figure 6 The processing shown is performed to remove low-quality genome data included in the collected genome DB.

[0103] In ST060, the processor 10 determines the quality of the genome data based on the integrity of the genome data. The integrity of the genome is, for example, based on a group of single copy marker genes (single copy marker gene) known to exist in each of the genomes of microorganisms as an indicator. When the genome data is complete, all single copy marker genes should be present in the sample. However, for example, if a part of the genome data is missing or read incorrectly, in the case of incomplete genome data, the single copy marker gene contained in the missing portion is lost. Therefore, the larger the part of the genome data that is missing or read incorrectly, the fewer the number of single copy marker genes on the genome data. Therefore, the number of single copy marker genes can be used as an indicator of the integrity of the genome data. Specifically, the situation where all single copy marker genes exist on the genome data is set to 100%, and the integrity is calculated as a ratio in direct proportion to the number of single copy marker genes present.

[0104] Specifically, in ST060, the processor 10 determines whether the integrity of the genome data is greater than a reference value T1. The reference value T1 is, for example, 50%. If the integrity is less than the reference value T1 (no in ST060), the processor 10 removes the genome data in ST061. If the integrity is greater than the reference value T1 (yes in ST060), the processor 10 causes the process to enter ST062.

[0105] In ST062, the processor 10 determines the quality of the genomic data based on the proportion of contamination of the genome. Contamination refers to the phenomenon that, for some reason, the DNA sequence of other genomic data is mixed into the DNA sequence of one genomic data. That is, contamination typically occurs when the DNA sequences of multiple microorganisms are mixed. If the proportion of single-copy marker genes found is set to 100% when the genomic data is not contaminated, then in the case of contamination, the proportion will be greater than 100%. Therefore, for example, the situation where no contamination occurs and all single-copy marker genes exist in the genomic data is set to 100%, and the proportion of contamination is calculated based on the number of single-copy marker genes found. In the case where the number of single-copy marker genes found is equivalent to (100+n)%, the proportion of contamination is n%. n is a real number satisfying n>0. If the proportion of contamination is high, it can be considered that the possibility of the DNA sequences of multiple microorganisms being mixed is high.

[0106] Specifically, in ST062, the processor 10 determines whether the contamination ratio is less than a reference value T2. The reference value T2 is, for example, 20%. If the contamination ratio is greater than the reference value T2 (no in ST062), the processor 10 removes the genome data in ST063. If the contamination ratio is less than the reference value T2 (yes in ST062), the processor 10 causes the process to proceed to ST064.

[0107] In ST064, processor 10 determines the quality of genomic data based on the number of contigs. Contigs refer to sequences obtained by the segmentation when one DNA sequence is originally segmented into multiple DNA sequences. Therefore, the more the number of contigs, the finer the DNA sequence is segmented. If the number of contigs is too much, the gene region expressing the protein is also segmented, and it may not be possible to read accurately. If the DNA sequence contained in the genomic data is segmented into how many counts, the number of contigs can be known.

[0108] Specifically, in ST064, the processor 10 determines whether the number of overlapping groups is less than the reference value T3. The reference value T3 is, for example, 1000. When the number of overlapping groups is greater than the reference value T3 (no in ST064), the processor 10 removes the genome data in ST065. When the number of overlapping groups is less than the reference value T3 (yes in ST064), the processor 10 causes the processing to enter ST066.

[0109] In ST066, the processor 10 determines the quality of the genome data based on the number of undetermined bases. Undetermined bases are bases that cannot be determined as any of AGCT when the DNA base sequence is interpreted. There is a high possibility that genes cannot be properly found from a DNA sequence containing a large number of undetermined bases.

[0110] Specifically, in ST066, the processor 10 determines whether the number of undetermined bases is less than the reference value T4. The reference value T4 is, for example, 100,000. When the number of undetermined bases is greater than the reference value T4 (no in ST066), the processor 10 removes the genome data in ST067. When the number of overlapping groups is less than the reference value T4 (yes in ST067), the processor 10 causes the process to enter ST068.

[0111] In ST068, the processor 10 determines the quality of the genomic data based on whether the number of genes meets the benchmark value. The benchmark is used to determine whether the number of genes inferred from the genomic data is included in an appropriate range. For example, in the case where the number of genes inferred from the genomic data is abnormally large, it can be considered that for some reason, the part that was not originally a gene is inferred to be a gene. Some reasons refer to, for example, when interpreting the DNA base sequence, an error in the interpretation of the base sequence occurs, and a sequence that is originally unrelated to the start or end of transcription and translation is interpreted as a sequence related to the start or end of transcription and translation. In this case, a sequence that does not express protein is also misunderstood as a sequence that expresses protein, and a large number of erroneous peaks may be included in the predicted m / z list. If such an m / z list is included in the m / zDB, the quality of the m / zDB is reduced and the accuracy of the sample discrimination is also reduced.

[0112] Specifically, in ST068, the processor 10 determines whether the number obtained by dividing the number of genes in the genome data by the number of bases encoding the genes is less than the reference value T5. The bases encoding the genes generally refer to the bases contained in the region on the DNA sequence that is related to the expression of the protein. The reference value T5 is, for example, 0.00180. When the number obtained by the division operation is greater than the reference value T5 (no in ST068), in ST069, the processor 10 removes the genome data. When the number obtained by the division operation is less than the reference value T5 (yes in ST068), the processor 10 adds the genome data to the high-quality genome DB.

[0113] The processor 10 performs ST060 to ST069 on all genome data included in the collected genome DB.

[0114] In addition, the calculation method of each benchmark for completeness, contamination ratio, number of overlapping groups, number of undetermined bases, and appropriateness of the number of genes is not limited to the above examples. For example, the appropriateness of the number of genes can be determined by whether the number of genes included in one genome data is less than a predetermined benchmark value.

[0115] pass Figure 6 The processing shown removes genome data that does not meet the benchmark contained in the collected genome DB. That is, among the genome data contained in the public genome DB 70, low-quality data is removed, and only high-quality data is used to construct the m / zDB. Therefore, the quality of the m / zDB in the device 100 is improved.

[0116] (4-3. Addition of new genome data)

[0117] The device 100 is configured to be able to add new genome data to the m / zDB. This addition is performed, for example, when a user using the device 100 discovers a new microorganism and desires to add the genome data of the microorganism.

[0118] Figure 7 is a diagram showing the process of adding new genome data. Figure 7 In the flowchart of Figure 5 The ST02 of the flowchart was changed to ST02A and deleted Figure 5 Steps ST04 and ST08. Figure 7 The processing after ST12 of the flowchart corresponds to Figure 5 The processing after ST12 of the flowchart.

[0119] In ST02A, the processor 10 acquires new genome data. Specifically, for example, the processor 10 acquires the genome data from an external device such as a DNA sequencer or a storage device, or from a storage terminal such as a USB memory, via the input / output I / F 13 or the communication I / F 12.

[0120] In ST06, the processor 10 determines whether the genome data satisfies the benchmark. The benchmark is set in such a way that only high-quality genome data satisfies the benchmark. If the new genome data satisfies the benchmark, the processor 10 causes the process to enter ST10. If the new genome data does not satisfy the benchmark, the processor 10 removes the new genome data.

[0121] In ST10, the processor 10 predicts the genes included in the genome data and proceeds to ST12. Figure 5 Therefore, the processor 10 can also add proteins predicted to be expressed based on new genome data to the m / zDB if their quality satisfies a predetermined standard.

[0122] With this structure, the m / z list predicted from the newly acquired genomic data can be added to the m / zDB, which can enrich the content of the m / zDB. As a result, the quality of the m / zDB is further improved, and the accuracy of distinguishing samples using the m / zDB is also further improved.

[0123] [5. Flow of processing related to sample identification]

[0124] (5-1.2 Stage Screening)

[0125] The apparatus 100 uses the m / zDB constructed as described above to discriminate samples.

[0126] Figure 8 : is a flowchart showing the processing related to the discrimination of the sample. Figure 8 The processing of ST32 to ST54 shown is equivalent to Figure 2 Processing of ST102.

[0127] Reference Figure 8 In ST32, the processor 10 obtains a sample list. For example, the sample list is obtained from the MS 16. In ST34, the processor 10 determines whether to calibrate the m / z of the sample list. For example, the user pre-sets whether to calibrate the sample list.

[0128] When performing analysis in a mass spectrometer such as MALDI-TOF MS, an m / z greater or smaller than the actual value may be detected depending on the mass of the protein contained in the sample, the device used, etc. That is, the sample list may contain a slight shift in m / z as a measurement error. On the other hand, the m / zDB contained in the device 100 is a theoretical value and does not contain a measurement error. Therefore, after offsetting the m / z of the sample list to eliminate the measurement error, the sample list can be compared with the m / zDB contained in the device 100 to more accurately identify the sample.

[0129] The estimation of the measurement error is performed according to the following steps. First, the sample list containing the measurement error is directly compared with the "m / z list assumed to contain (roughly) no measurement error". Next, when the sample list is offset by a specified value, a specified value is searched for a higher matching rate with the "m / z list assumed to contain no measurement error". The specified value corresponds to the measurement error. In addition, the specified value is searched within the range of values ​​that can be taken as the measurement error.

[0130] In addition, the "m / z list assumed to contain no measurement errors" is, for example, an m / z list contained in a specific m / zDB that is considered unlikely to contain false peaks, but is not limited to this. For example, it may be an m / z list contained in the entire m / zDB, or it may be another m / z list prepared for correcting the measurement errors of the sample list.

[0131] When the m / z of the sample list is corrected (YES in ST34), the processor 10 estimates the measurement error included in the sample list based on the specific m / z DB in ST36. In ST38, the processor 10 performs correction to shift the m / z of the sample list by the estimated measurement error.

[0132] Without correcting the m / z of the sample list (NO in ST34), or following ST38, in ST40 to ST44, the processor 10 compares the sample list and the m / zDB based on weighting the m / z corresponding to the proteins included in the specific group to identify the sample.

[0133] In ST40, as a primary screening, the processor 10 selects from a specific m / zDB an m / z list with a matching rate above a specified rank with respect to the sample list. More specifically, the "m / z list with a matching rate above a specified rank" refers to an m / z list in the m / z list of the m / zDB used for screening that has a matching rate above a specified rank with respect to the sample list. For example, the m / z list with the top N1 matching rates is selected as the m / z list with a matching rate above a specified rank. N1 is, for example, an integer between 500 and 5000. Another example of the "m / z list with a matching rate above a specified rank" is an m / z list with a matching rate above a specified value. The "m / z list with a matching rate above a specified value" can be considered as an "m / z list with a rank above that corresponding to the number of m / z lists with a matching rate above a specified value".

[0134] In ST42, the processor 10 selects the m / z lists in the entire m / zDB corresponding to the top N1 m / z lists. In other words, it selects the m / z lists in the entire m / zDB corresponding to the genomes of the top N1 m / z lists.

[0135] In ST44, as a secondary screening, the processor 10 discriminates the sample by selecting, from the m / z lists in the entire m / zDB that have been selected, the m / z lists with a high matching rate with respect to the sample list. For example, the top N2 m / z lists among the m / z lists in the entire m / zDB that have been selected are selected as the m / z lists with a high matching rate. Additionally, N1 is an integer where N2 < N1, for example, an integer between 1 and 100.

[0136] If the discrimination of the sample is completed, in ST46, the processor 10 reflects the classification data in the discrimination result. For example, the processor 10 attaches the corresponding taxonomic information (family, genus, species, system, etc.) of the microorganism to each of the N1 m / z lists that have been screened.

[0137] In addition, N1 m / z lists can also be sorted based on classification data. For example, a table in which N1 m / z lists are rearranged in the order of classification information can also be generated. In addition, for example, a diagram in which the microorganisms corresponding to N1 m / z lists are recorded on a phylogenetic tree can also be generated. In addition, for example, the number of microorganisms corresponding to specific families, genera, species, and systems contained in the microorganisms corresponding to N1 m / z lists can also be digitized. Specifically, a table recording the families, genera, species, and systems with the largest number of microorganisms corresponding to N1 m / z lists can be generated. In addition, for example, the discrimination results can also be further locked by reflecting the classification data in N1 m / z lists. Specifically, processing such as removing m / z lists that are obvious outliers in taxonomy can be added. Through the processing illustrated above, a discrimination result reflecting a taxonomic viewpoint can be output. In addition, the processing illustrated above can also be performed based on more than two taxonomic systems. Thus, a discrimination result reflecting multiple taxonomic viewpoints can be generated.

[0138] In ST48, the processor 10 determines whether to predict a protein corresponding to the m / z included in the sample list, that is, a protein considered to be expressed in the sample. For example, whether to predict a protein is set in advance by the user.

[0139] When a protein is not predicted (NO in ST48), processor 10 outputs the determination result in ST50 and ends the process. The determination result is output by displaying it on display 15, for example.

[0140] In the case of predicting a protein (YES in ST48), in ST52, the processor 10 links the annotation of the protein corresponding to the m / z included in the sample list. As described above, the annotation is information related to the protein, including information related to the group containing the protein. Specifically, for example, the processor 10 adds the item of the name and function of the protein corresponding to the m / z in the sample list. In addition, for example, the processor 10 can also generate a table of the name and function of the protein corresponding to the m / z included in the sample list, which is independent of the sample list.

[0141] If the annotation of the protein is linked, then in ST54, the processor 10 outputs the discrimination result generated in ST44 and ST46, and the annotation corresponding to the m / z included in the sample list in ST52, and ends the processing. The discrimination result and the annotation are output, for example, by being displayed on the display 15. In this way, when outputting information related to the expressed protein predicted according to the sample list, the user can easily identify the information related to the protein predicted to be expressed in the sample, thereby deepening the understanding of the sample. In addition, the information related to the protein can also be referenced when studying the discrimination result of the sample, and can also be referenced when performing other analyses on the sample, thereby improving the convenience of the user.

[0142] exist Figure 8 In the processing, the primary screening based on the specific m / z DB and the secondary screening based on the entire m / z DB are performed, which has the following advantages. First, by narrowing the primary screening to the m / z of proteins that are functionally important and have a high expression level, it is possible to lock in an m / z list with a high matching rate with less influence of false peaks. Then, by performing a secondary screening on all m / z, it is possible to reflect the similarity of proteins other than the proteins included in the specific group obtained by the primary screening.

[0143] Alternatively, secondary screening may be performed by narrowing down the m / z of proteins included in a specific group different from that in the primary screening. In this case, samples can be discriminated by focusing on two important proteins.

[0144] Furthermore, such different screenings may be combined three or more times.

[0145] In summary, the device 100 can classify samples from a taxonomic perspective by screening in two or more stages, including screening based on m / z corresponding to proteins included in a specific group. By performing multiple different screenings in this way, the characteristics of each screening can be utilized, thereby improving the accuracy of sample discrimination as a whole.

[0146] Furthermore, the device 100 can also be configured to discriminate a sample based on classification data related to classification of microorganisms.

[0147] For example, a specific m / zDB is constructed to include only m / z corresponding to a group of proteins co-expressed in a higher taxonomic group (e.g., a genus that is a taxonomic group higher than species / strains). Conceptually, a specific m / zDB includes a group PA of proteins co-expressed in a certain genus A. In this case, by screening using the specific m / zDB, it is possible to accurately determine whether a sample is included in genus A. Similarly, if a specific m / zDB is constructed to include each group of proteins co-expressed in each genus, the genus of the sample can be accurately determined by screening using the specific m / zDB. In addition, it is possible to subsequently perform the determination of species / strains within the determined genus as a secondary screening. Therefore, in the primary screening, the possibility of erroneously determining the genus of the sample can be reduced, and in the secondary screening, the screening can be performed in a state suitable for the determination of species / strains. Therefore, the accuracy of sample determination can be improved.

[0148] (5-2. Weighting based on score)

[0149] In the discrimination of samples, the method of weighting the m / z corresponding to the proteins included in the specific group is not limited to the method using the specific m / z DB described above. For example, the following method may be used: when comparing the sample list with the m / z list included in the entire m / z DB, when the m / z corresponding to the proteins included in the specific group is consistent, the matching rate is calculated in a manner that is higher than when the m / z corresponding to other proteins is consistent.

[0150] Fig. 9 This is a flowchart showing another example of processing related to sample discrimination. Fig. 9 The flowchart will Figure 8 The flowchart of ST40 to ST42 is changed to ST40A. Fig. 9 The other steps are Figure 8 same.

[0151] exist Fig. 9 In ST40A, the processor 10 selects an m / z list with a high matching rate with the sample list from the m / z lists in the entire m / zDB. For example, the first N3 m / z lists in the entire m / zDB are selected as the m / z lists with a high matching rate. Specifically, the processor 10 first calculates the score for each of the m / z lists contained in the entire m / zDB by multiplying the number of m / z contained in the m / z list that is consistent with the m / z contained in the sample list by a specified coefficient. Then, in the entire m / zDB, an m / z list with a score above a specified rank is selected. In more detail, "an m / z list with a score above a specified rank" refers to an m / z list with a score above a specified rank in the m / z list in the m / zDB used for screening. For example, an m / z list with a score of the first N3 is selected as an m / z list with a score above a specified rank. Another example of "an m / z list with a score above a specified rank" is an m / z list with a score above a specified numerical value. The “m / z lists having scores equal to or greater than a predetermined value” can be considered to be “m / z lists having a rank equal to or greater than the number of m / z lists having scores equal to or greater than a predetermined value”.

[0152] At this time, the coefficient used in calculating the score is set in the following manner: when the m / z corresponding to the protein included in the specific group is consistent, it is greater than the case where the m / z corresponding to the protein not included in the specific group is consistent. For example, the coefficient is set to 10 times. That is, the score for the case where the m / z corresponding to the protein included in the specific group is consistent is likely to be greater than the case where the m / z corresponding to the protein not being the case is consistent, resulting in a high matching rate that is easy to calculate. Therefore, the sample can be discriminated in a state where the proteins included in the specific group are weighted. Therefore, for example, it is possible to weight the protein that is important in function and is difficult to contain false peaks to discriminate the sample, so the accuracy of sample discrimination is improved.

[0153] In ST40A, the m / z corresponding to the protein included in the specific group may be selected as the m / z included in the specific m / z DB, but may also be selected by referring to the annotation DB. In addition, as described above, the specific m / z DB is not necessarily used in the estimation of the measurement error shown in ST36. Therefore, in the method for distinguishing microorganisms weighted by the coefficient of the score, the construction of the specific m / z DB is not necessarily required.

[0154] In addition, as a weighting method, a method of performing multi-stage screening using a specific m / z DB and a method of changing the coefficient of classification can also be combined. For example, a method of using a specific m / z DB of a certain group of proteins for screening as a primary screening and increasing the coefficient when the m / z corresponding to the protein of another group matches as a secondary screening can be used to discriminate samples.

[0155] [6. Experimental example]

[0156] An example of an experiment performed using the microorganism identification system 1000 will be described.

[0157] (6-1. Database construction)

[0158] The genome sequences of bacteria and archaea were obtained from the National Center for Biotechnology Information (NCBI) of the United States via an FTP server (using RefSeq v95, more than 270,000). For all genome sequences, gene inference (using Prodigal) was performed, and loci and their products were predicted. The integrity and contamination of the results were inferred by checkM. In addition, the number of overlapping groups in the genome sequence, the number of undetermined bases (N), N50, and the number of genes relative to the genome base length were measured on the computer. In addition, N50 is one of the indicators that represents the quality of genome information (assembly: assembly), which represents the weighted average of the sequence length of overlapping groups in the assembly of the genome sequence. N50 is the length (base length) of the sequence when the overlapping groups are arranged in order from long to short and added sequentially from the top to reach half of the overall length. For the product (protein) predicted by the inferred gene, according to its amino acid composition, methionine removal, signal protein prediction and cleavage fragment prediction based on it (using SignalP) are implemented to calculate the quality of the predicted final protein. The theoretical protein mass information was digitized (all m / zDB), and the systematic classification information of each genome (existing taxonomic information such as GTDB, Silva, GreenGenes, etc.) was collected to generate a database (classification DB) that connected them with the same ID. In addition, for the predicted gene products (proteins), the functions of each protein were inferred using the similarity with the registration information of existing protein databases such as Uni ProKB and PFAM, and a database (annotation DB) was generated that associated all theoretical protein masses with the names of proteins.

[0159] For each genome sequence, a genome sequence with any of the benchmark values ​​of estimated completeness of less than 50%, contamination of more than 10%, number of overlapping groups of more than 1,000, N50 of less than 5 kbp, and undetermined bases (N) of more than 100,000 was judged to have low quality of genome sequence and excluded from the data. In addition, genome sequences with (number of genes / total length of bases in loci in the genome) of more than 0.00180 were deleted.

[0160] When a database was constructed based on all genome entries (ENTRY) without considering these benchmarks, and microorganism identification was performed based on a protein measurement peak list of known microbial strains obtained by MALDI-MS (AXIMA (registered trademark) manufactured by Shimadzu Corporation), it was observed that genome entries with extremely high numbers of genes per genome were consistent with a high probability, and correct results could not be obtained.

[0161] Specific experimental procedures for known microbial strains are described below. Microorganisms obtained from the National Institute of Technology and Evaluation Bioresource Center (NBRC) of Japan, such as Escherichia coli NBRC 3301, Bacillus subtilis subsp. Subtilis NBRC 13719, Microlunatus phosphovorus NBRC 101784, Bifidobacterium longum ATCC15707, Clostridium acetobutylicum NBRC 13948, Arthrobacter globiformis NBRC 12137, Brachybacterium conglomeratum NBRC 15472, Streptomyces griseus subsp. Griseus NBRC 12875, Tetrasphaera Bacteria and archaea such as Bacteroides duodecadis NBRC12959, Bacteroides fragilis ATCC 25285, Sphingomonas yanoikuyae NBRC 15102, Xanthobacter autotrophicus NBRC 102463, Rhodobacter azotoformans NBRC 16436, Methanosarcina thermophila MST-A1, and Thaueralinaloolentis NBRC 102519 were cultured in a designated medium, the culture solution was centrifuged (10,000 g, 2 minutes) to remove the medium components, an equal amount of pure water was added to disperse the cells, and centrifuged under the same conditions to remove the supernatant. 500 μL of pure water was added to the obtained bacterial cell precipitate to disperse the bacterial cells, thereby obtaining a bacterial cell dispersion. 500 μL of zirconium oxide beads (φ0.5 mm) were added to a 1.5 mL tube with a screw cap, and 500 μL of the above bacterial cell dispersion was added.The beads were crushed at 4000 rpm for 3 minutes in total, and the crushed liquid was centrifuged (15000 g, 5 minutes). 1 μL of the obtained supernatant was mixed with 9 μL of 10 mg / ml CHCA (α-cyano-4-hydroxycinnamic acid) solution (50% acetonitrile aqueous solution containing 1% TFA (trifluoroacetic acid)), and 1 μL was added to the sample plate for MALDI-MS, and air-dried to prepare sample / matrix mixed crystals. These bacterial samples were measured in the range of m / z 2000-20000 in MALDI-MS linear mode to obtain a MALDI mass spectrum. Peak detection (peak pick) was performed to generate a peak list consisting of the m / z value and peak intensity (mV) of the detected peak.

[0162] Then, using the database generated above, the consistency was confirmed by comparing the peaks in the peak list with the theoretical m / z values ​​in the database. As a result, many measured peaks were consistent with the theoretical peaks inferred from the genome information of the strain within a certain range, but there was also genome information that showed high consistency with the theoretical peaks inferred from the genome information of other strains.

[0163] Fig.10 This is a graph showing the relationship between the total number of base sequences of gene sites in a genome and the number of estimated genes per genome. Fig.10 The relationship between the base length of the locus and the estimated number of genes in each genome estimated from the bacterial and archaeal genome information in RefSeq95 (more than 270,000). Fig.10 In the case where the relationship between the total number of base sequences and the putative genes is shown by all genome entries (more than 270,000), Fig.10 A large number of genomes were detected above the single-dotted dash line in the figure (the line with the total length of 0.00180 for the number of genes / the number of bases in the loci in the genome). This is presumed to be because proteins that do not actually exist were predicted due to errors in reading the base sequence during genome interpretation, and more theoretical peaks than actually exist were predicted, resulting in the observation of peak coincidence as false positives.

[0164] On the other hand, by deleting from the database the genome entries with the used data (number of genes / total length of bases in the loci in the genome) of about 0.00180 or more, the peak list from the above microbial group and the estimated theoretical peak obtained from the corresponding genome information were found to be at a higher level of consistency. In other words, by deleting from the database the peaks at Fig.10These results indicate that it is essential to construct a database for microorganism identification based on genome information registered in public databases and to construct a database appropriately screened by the above method.

[0165] As a result of these screenings, a total m / zDB with 193,197 entries was generated. In addition, a species-level representative database was generated using GTDB as a reference, and a species-level representative database consisting of 31,760 entries was constructed.

[0166] (6-2. Algorithm construction)

[0167] Microorganisms obtained from NBRC, such as Escherichia coli NBRC 3301, Bacillus subtilis subsp. subtilis NBRC 13719, Microlunatus phosphovorus NBRC 101784, Bifidobacterium longum ATCC15707, Clostridium acetobutylicum NBRC 13948, Arthrobacter globiformis NBRC 12137, Brachybacterium conglomeratum NBRC 15472, Streptomyces griseus subsp. griseus NBRC12875, Tetrasphaera Bacteria and archaea such as Bacteroides duodecadis NBRC12959, Bacteroides fragilis ATCC 25285, Sphingomonas yanoikuyae NBRC 15102, Xanthobacter autotrophicus NBRC 102463, Rhodobacter azotoformans NBRC 16436, Methanosarcina thermophila MST-A1, and Thaueralinaloolentis NBRC 102519 were cultured in a designated medium, the culture solution was centrifuged (10,000 g, 2 minutes) to remove the medium components, an equal amount of pure water was added to disperse the cells, and centrifuged under the same conditions to remove the supernatant. These microbial communities include aerobic and anaerobic bacteria, methanogenic archaea and other phyla, with various cell wall structures such as Gram-positive and Gram-negative, and also include actinomycetes and the like. 500 μL of pure water was added to the obtained bacterial precipitate to disperse the bacterial cells to obtain a bacterial dispersion. 500 μL of zirconium oxide beads (φ0.5 mm) were added to a 1.5 mL tube with a screw cap, and 500 μL of the above bacterial dispersion was added.The bead crusher (MS-100 manufactured by TOMY Seiko Co., Ltd.) was used for a total of 3 minutes at 4000 rpm, and the crushed liquid was centrifuged (15000 g, 5 minutes). 1 μL of the obtained supernatant was mixed with 9 μL of a 10 mg / ml CHCA solution (50% acetonitrile aqueous solution containing 1% TFA), and 1 μL was added dropwise to a MALDI-MS sample plate, and air-dried to prepare a sample / matrix mixed crystal. Then, the mass spectrum of each sample strain was obtained by measuring with MALDI-MS (AXIMA (registered trademark) manufactured by Shimadzu Corporation).

[0168] The theoretical peak list in the database generated above is compared with the measured peaks actually obtained from the cultured microbial community. The peaks in the case where there are measured peaks within a certain range from the theoretical peak are set to be consistent and the range is set to 200ppm. When the number of consistent peaks is calculated in all entries in the entire m / zDB, it is found that the genome entry with the highest consistency is not necessarily the genome entry corresponding to the measured strain. Next, a database (specific m / zDB) is constructed to screen out proteins that are easily detected at a high frequency in MALDI measurements. Here, as a reference, ribosomal proteins that are easily detected at a high frequency are extracted from the database (annotation DB), and genome entries with high consistency with the measured peak list are screened in the database (for example, 500 to 5,000 entries are extracted), and a two-stage search algorithm for calculating consistency using all theoretical protein peak lists is implemented for these entries. The results are shown in Table 1 below. In all of the above 15 strains, an algorithm that can select closely related genome entries from the measured peak list and can correctly infer systematic taxonomic groups (genus, species) is constructed. Table 1 is a diagram showing the results of discrimination of bacteria and archaea having various systems, physiological characteristics, and cell wall characteristics using this algorithm.

[0169] [Table 1]

[0170]

[0171] [plan]

[0172] Those skilled in the art will appreciate that the above-mentioned multiple exemplary embodiments are specific examples of the following schemes.

[0173] (Item 1) A method for identifying microorganisms of one scheme may include: a step of obtaining a list of mass-to-charge ratios of samples, i.e., a sample list; a step of identifying samples by comparing the sample list and a database of mass-to-charge ratio lists of each microorganism predicted based on genomic data, i.e., a mass-to-charge ratio database, based on weighted mass-to-charge ratios corresponding to proteins included in a specific group; and a step of outputting the identification result.

[0174] According to the microorganism identification method described in Item 1, when comparing the list of mass-to-charge ratios of the sample with the mass-to-charge ratio database of each microorganism predicted based on genomic data, weighting is performed on the mass-to-charge ratios corresponding to the proteins included in a specific group. Therefore, in the microorganism identification method using the genomic database and mass spectrometry, the accuracy of identifying microorganisms can be improved.

[0175] (Item 2) In the microorganism identification method described in Item 1, the group can be defined by at least one of the name of the protein, the function of the protein, and the family.

[0176] According to the microorganism identification method described in Item 2, the mass-to-charge ratios corresponding to the same protein, proteins having the same function, or proteins of the same family can be weighted to identify samples. In particular, based on the name, function, or family of the protein, the mass-to-charge ratios corresponding to "a group of proteins that are highly likely to be expressed in the organism of a microorganism and are also highly likely to be detected as peaks when measuring a mass spectrum" can be weighted to identify samples.

[0177] (Item 3) In the method for identifying microorganisms described in Item 1 or Item 2, the group may be selected based on at least one of the following conditions: a condition that the expression level is above a prescribed threshold value, a condition that the group has a function that is essential for maintaining life, a condition that microorganisms having an amino acid sequence similarity above a prescribed threshold value are present at a prescribed ratio or above, a condition that the group is an alkaline protein, a condition that the mass-to-charge ratio can be analyzed within an error range of ±14 Da when measured by MALD I-MS measurement, a condition that the mass of the protein is within the range of 4 to 30 kDa, a condition that the number of types of proteins included in the group is above a prescribed number, and the function that is essential for maintaining life includes at least one of the essential functions of cell maintenance and proliferation.

[0178] According to the microorganism identification method described in item 3, the sample can be identified by weighting the mass-to-charge ratio corresponding to the "protein that is highly likely to be expressed in vivo and detected as a peak when the mass spectrum is measured" that satisfies the above condition.

[0179] (Item 4) In the microorganism identification method described in any one of Items 1 to 3, the group may include at least one of a ribosomal protein, a chaperone protein, and a DNA binding protein.

[0180] According to the microorganism identification method described in Item 4, samples can be identified by weighting the mass-to-charge ratio corresponding to "proteins that are highly likely to be expressed in vivo and are detected as peaks when measuring mass spectra", such as ribosomal proteins, chaperone proteins, and DNA-binding proteins.

[0181] (Item 5) In the microorganism identification method described in any one of Items 1 to 4, the step of identifying the sample may include: a step of taxonomically identifying the sample through two or more stages of screening including screening based on a mass-to-charge ratio corresponding to a protein included in a specific group.

[0182] According to the microorganism identification method described in Item 5, by screening based on the mass-to-charge ratio corresponding to the proteins contained in a specific group, for example, the mass-to-charge ratio of the proteins that are functionally important and expressed in large amounts can be narrowed down for screening. Then, by other screening, screening can be performed based on the mass-to-charge ratio of the proteins contained in other groups or all mass-to-charge ratios. By performing multiple different screenings in this way, the characteristics of each screening can be brought into play, and the discrimination accuracy of the sample can be improved as a whole.

[0183] (Item 6) In the microorganism identification method described in any one of Items 1 to 5, the mass-to-charge ratio database includes: a specific mass-to-charge ratio database including a specific mass-to-charge ratio list having only mass-to-charge ratios predicted to be included in a specific group; and a total mass-to-charge ratio database including a mass-to-charge ratio list having all mass-to-charge ratios. The step of identifying the sample includes: the step of selecting a specific mass-to-charge ratio list having a matching rate of a specified rank or higher with the sample list from the specific mass-to-charge ratio database; and the step of selecting a mass-to-charge ratio list having a matching rate of a specified rank or higher with the sample list from the mass-to-charge ratio list corresponding to the selected specific mass-to-charge ratio list in the total mass-to-charge ratio database.

[0184] According to the microorganism identification method described in item 6, first, a specific mass-to-charge ratio database can be used to narrow down the mass-to-charge ratio of proteins contained in a specific group (e.g., functionally important and highly expressed proteins) for primary screening. Therefore, it is possible to lock in an m / z list with a high matching rate under a condition where the influence of false peaks is small. Next, a secondary screening can be performed using the entire mass-to-charge ratio database, thereby reflecting the similarity of proteins other than the proteins contained in the specific group obtained by the primary screening to identify the sample.

[0185] (Item 7) In the microorganism identification method described in any one of Items 1 to 5, the step of identifying the sample includes: for each mass-to-charge ratio list included in the mass-to-charge ratio database, the step of multiplying the number of mass-to-charge ratios included in the mass-to-charge ratio list that match the mass-to-charge ratios included in the sample list by a predetermined coefficient to calculate a score; and the step of selecting a mass-to-charge ratio list having a score above a predetermined rank. The coefficient is set in such a way that, in the case where the mass-to-charge ratio corresponding to the protein included in the specific group is consistent, the coefficient is made greater than in the case where the mass-to-charge ratio is consistent with the protein not included in the specific group.

[0186] According to the microorganism discrimination method described in Item 7, the score tends to be larger when the mass-to-charge ratio corresponding to the protein included in the specific group is consistent than when the mass-to-charge ratio corresponding to the protein not included in the specific group is consistent, and as a result, a higher matching rate is easily calculated. Therefore, the sample can be discriminated in a state where the proteins included in the specific group are weighted. Therefore, for example, the protein that is functionally important and is difficult to contain false peaks can be weighted to discriminate the sample, so the accuracy of sample discrimination is improved.

[0187] (Item 8) In the microorganism identification method described in any one of Items 1 to 7, the step of identifying the sample may include a step of identifying the sample based on classification data related to classification of the microorganism.

[0188] According to the microorganism identification method described in Item 8, it is possible to output identification results reflecting a taxonomic viewpoint.

[0189] (Item 9) In the microorganism identification method described in any one of Items 1 to 8, the step of acquiring may further include: a step of estimating a measurement error of the mass-to-charge ratio of the sample list; and a step of correcting the estimated measurement error.

[0190] According to the microorganism identification method described in Item 9, the identification accuracy of the sample is improved compared to the case where the measurement error of the mass-to-charge ratio of the sample list is not corrected.

[0191] (Item 10) In the microorganism identification method described in any one of Items 1 to 9, the output step may include a step of outputting a result obtained by identifying the sample based on two or more taxonomic systems.

[0192] According to the microorganism identification method described in Item 10, it is possible to output identification results reflecting a plurality of taxonomic viewpoints.

[0193] (Item 11) In the microorganism identification method described in any one of Items 1 to 9, the step of outputting may include the step of outputting information related to the expressed protein predicted based on the sample list.

[0194] According to the microorganism identification method described in Item 11, the user can easily identify information related to proteins predicted to be expressed in the sample, thereby deepening the understanding of the sample. In addition, the information related to the protein can also be referenced when studying the identification results of the sample, and can also be referenced when performing other analyses on the sample.

[0195] (Item 12) A microorganism identification device of one scheme comprises a storage unit, a processor, and an output unit. The storage unit stores a mass-to-charge ratio database, which is a database of a list of mass-to-charge ratios of each microorganism predicted based on genome data. The processor obtains a list of mass-to-charge ratios of samples, i.e., a sample list, and compares the mass-to-charge ratios corresponding to the proteins included in the sample list and the mass-to-charge ratio list included in the mass-to-charge ratio database by weighting them, thereby identifying the sample. The output unit outputs the result of the identification.

[0196] According to the microorganism identification method described in Item 12, when comparing the list of mass-to-charge ratios of the sample with the mass-to-charge ratio database of each microorganism predicted based on genomic data, weighting is performed on the mass-to-charge ratios corresponding to the proteins included in a specific group. Therefore, in the microorganism identification method using the genomic database and mass spectrometry, the accuracy of identifying microorganisms can be improved.

[0197] The embodiments disclosed this time should be considered as illustrative in all aspects and not restrictive. The scope of the present invention is indicated by the claims rather than the above description, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0198] Description of Reference Numerals

[0199] 1 Purification device

[0200] 10 Processor

[0201] 11 Memory

[0202] 12 Communication I / F

[0203] 13 Input and output I / F

[0204] 14 Operation section

[0205] 15 Display

[0206] 16MS

[0207] 21 Genome Data Collection Department

[0208] 22 Genome Data Evaluation Department

[0209] 23 Protein Prediction Department

[0210] 24 Construction Department

[0211] 25 Storage

[0212] 31 Acquisition Department

[0213] 32 Sample identification department

[0214] 33 Notes

[0215] 34 Output

[0216] 70 Public genome databases

[0217] 80 Public classification databases

[0218] 90 Network

[0219] 100 Installations

[0220] 101 Controller

[0221] 321 Primary Screening Department

[0222] 322 Secondary Screening Department

[0223] 1000 microorganism identification system.

Claims

1. A method for identifying microorganisms, characterized in that: have: The step of obtaining a list of mass-to-charge ratios of samples, i.e., a sample list; A step of comparing the sample list and a database of mass-to-charge ratio lists of each microorganism predicted based on genome data, i.e., a mass-to-charge ratio database, based on weighted mass-to-charge ratios corresponding to proteins included in a specific group to discriminate the sample; as well as The step of outputting the result of the discrimination.

2. The microorganism identification method according to claim 1, characterized in that: The group is defined by at least one of the name of the protein, the function of the protein, and the family.

3. The microorganism identification method according to claim 1 or 2, characterized in that: The group is selected based on at least one of the following conditions: a condition that the expression amount is above a predetermined threshold, a condition that the group has a function that is essential for life maintenance, a condition that the microorganisms with an amino acid sequence similarity of above a predetermined threshold are present at a predetermined ratio or above, a condition that the group is an alkaline protein, a condition that the mass-to-charge ratio can be analyzed within an error range of ±14 Da when measured by MALDI-MS measurement, a condition that the mass of the protein is within 4 to 30 kDa, and a condition that the number of types of proteins included in the group is above a predetermined number, The functions essential for maintaining life include at least one of the maintenance and proliferation of cells.

4. The microorganism identification method according to claim 1 or 2, characterized in that: The group comprises at least one of a ribosomal protein, a chaperone protein, and a DNA binding protein.

5. The microorganism identification method according to claim 1 or 2, characterized in that: The step of distinguishing the samples includes the step of distinguishing the samples taxonomically through two or more stages of screening including screening based on the mass-to-charge ratio corresponding to the specific proteins included in the group.

6. The microorganism identification method according to claim 1 or 2, characterized in that: The mass-to-charge ratio database comprises: a specific mass-to-charge ratio database comprising a specific mass-to-charge ratio list having only mass-to-charge ratios predicted to be included in a specific said group; as well as The global mass-to-charge ratio database contains a mass-to-charge ratio list with all mass-to-charge ratios. The step of identifying the sample comprises: The step of selecting a specific mass-to-charge ratio list having a matching rate with the sample list of a specified rank or higher from the specific mass-to-charge ratio database; as well as A step of selecting, from the mass-to-charge ratio lists corresponding to the selected specific mass-to-charge ratio list, a mass-to-charge ratio list having a matching rate with the sample list of a predetermined rank or higher in the entire mass-to-charge ratio database.

7. The microorganism identification method according to claim 1 or 2, characterized in that: The step of identifying the sample comprises: For each of the mass-to-charge ratio lists contained in the mass-to-charge ratio database, a step of calculating a score by multiplying the number of mass-to-charge ratios contained in the mass-to-charge ratio list that are consistent with the mass-to-charge ratios contained in the sample list by a specified coefficient; and The step of selecting the mass-to-charge ratio list having a score above a specified rank, The coefficient is set so that, when the mass-to-charge ratio corresponding to a protein included in the specific group matches, the coefficient is larger than when the mass-to-charge ratio corresponding to a protein not included in the specific group matches.

8. The microorganism identification method according to claim 1 or 2, characterized in that: The step of discriminating the sample includes the step of discriminating the sample based on classification data related to classification of microorganisms.

9. The microorganism identification method according to claim 1 or 2, characterized in that: The step of obtaining further comprises: The step of estimating the measurement error of the mass-to-charge ratio of the sample list; and Procedure for correcting estimated measurement errors.

10. The microorganism identification method according to claim 1 or 2, characterized in that: The step of outputting includes the step of outputting a result of discriminating the sample based on two or more taxonomic systems.

11. The microorganism identification method according to claim 1 or 2, characterized in that: The step of outputting includes the step of outputting information related to expressed proteins predicted based on the sample list.

12. A microorganism identification device, characterized in that: have: A storage unit storing a database including a mass-to-charge ratio list of each microorganism predicted based on genome data, namely, a mass-to-charge ratio database; A processor obtains a list of mass-to-charge ratios of samples, i.e., a sample list, and compares the sample list and the mass-to-charge ratio list contained in the mass-to-charge ratio database by weighting the mass-to-charge ratios corresponding to the proteins contained in the specific group, thereby discriminating the sample; as well as The output unit outputs the result of the discrimination.