A novel method for typing cgcmlst-based bacillus cereus population

Abstract

This invention discloses a novel typing method for Bacillus cereus strains based on cgcMLST. The method includes constructing the original cgcMLST characteristic genome set, obtaining effective cgcMLST characteristic genes, obtaining effective cgcMLST characteristic gene combinations, obtaining whole genome information of bacterial samples, and evaluating the resolution of the novel cgcMLST-based typing method for Bacillus cereus strains. Compared with existing technologies, this invention has the following advantages: it enables the evaluation of cgcMLST resolution for Bacillus cereus strains, verifies that cgcMLST can effectively type Bacillus cereus strains, and solves the problem of poor typing of closely related Bacillus cereus strains by existing MLST methods. In the analysis of 198 actual samples, the efficiency is improved by approximately 99.17% compared to rMLST, enabling more effective identification of Bacillus cereus strains, which is crucial for food safety and public health.

Description

Technical Field

[0001] This invention belongs to the field of bacterial typing technology, specifically relating to a novel typing method for Bacillus cereus based on cgcMLST. Background Technology

[0002] The Bacillus cereus group, also known as Bacillus cereus sensulato (sl), is currently classified as a species complex within the genus Bacillus. This species complex consists of many closely related species within the genus. Some species within the group can cause anthrax and anthrax-like diseases, vomiting, and diarrhea, and in severe cases, gastrointestinal infections. Other species have important applications in agriculture and industry. Different species exhibit varying abilities to cause diseases in humans and animals, and misclassification of species can pose serious potential hazards. Therefore, converting isolated strains into species-level taxonomic units is crucial for food safety and public health. Taxonomical information indicates that this group consists of at least 11 closely related species, commonly including the foodborne pathogen *Bacillus cereus*, the entomopathogenic bacterium *Bacillus thuringiensis*, the zoonotic pathogen *Bacillus anthracis*, the fungal-like *Bacillus mycoides*, *Bacillus weihenstephanensis*, *Bacillus pseudodomycoides*, *Bacillus cytotoxicus*, and *Bacillus toyonensis*. Some species, while their names are recognized, are not considered validly published under the International Code of Nomenclature for Prokaryotes, such as *Bacillus gaemokensis*, *Bacillus manliponensis*, and *Bacillus bingmayongensis*. In recent years, new species have been continuously discovered and added to the list, including the cold-resistant and cytotoxic Bacillus wiedmannii, which has been isolated from dairy products and their environment.In 2017, Liu Yang et al. used phenotypic and phylogenetic data to discover nine new species of B. cereus sl: Bacillus paranthracis, Bacillus pacificus, Bacillus tropicalus, Bacillus paramycoides, Bacillus albus, Bacillus luti, Bacillus mobilis, Bacillus proteolyticus, and Bacillus nitratireducens. In 2020, Parks et al. discovered a species of Bacillus bombysepticus, first reported in 1931, in the Genome Taxonomy Database (GTDB), which is very similar to B. cereus and was ultimately classified under B. cereus sl by NCBI. In the same year, Liu Xin et al. isolated Bacillus fungorum from mushroom substrate, which was also classified under B. cereus sl by NCBI. In 2021, Tohya et al. discovered three new species of B. cereus sl from clinical samples in Japan: Bacillus sanguinis sp. nov., Bacillus paramobilis sp. nov., and Bacillus hominis sp. nov. (Note: Some of the species mentioned above are defined based on ANI values. In some cases, these new species have only one known isolate, and at most no more than 13. This contradicts the formal bacterial species classification rules and has low research reference value, so it is marked with quotation marks.)

[0003] The classification of Bacillus cereus is highly complex. While confusion stemming from taxonomic ambiguities existed before the widespread adoption of whole-genome sequencing, new insights from the omics era have often led to conflicting proposals for taxonomic changes to this group. Over the past 20 years, researchers have proposed numerous new ideas about Bacillus cereus, but the inherent taxonomic contradictions make it difficult to stay up-to-date. Bacterial taxonomy relies heavily on ANI-based methods, which have become the gold standard for species description in the era of high-throughput sequencing. However, ANI alone is insufficient for the classification and identification of Bacillus cereus; for example, the ANI values ​​of related species like Bacillus thuringiensis and Bacillus cereus exceed 0.96. Currently, the authoritative method for typing Bacillus cereus remains the traditional biochemical method, which is also the national standard method for food microbiological safety testing in my country and the US FDA.

[0004] Several DNA sequence-based methods have been reported for resolving the phylogeny of Bacillus cereus strains, such as multilocus sequence typing (MLST). Core genome MLST (cgMLST) has been shown to offer significantly better resolution than MLST. Furthermore, ribosomal multilocus sequence typing (rMLST), based on 53 universally conserved ribosomal protein subunits, has become a universal bacterial typing scheme applicable to all bacteria. However, Bacillus cereus bacteria are ubiquitous in the environment, often allowing for the isolation of large numbers of samples. CgMLST and rMLST use too much genetic information, requiring significant computational resources and time for accurate identification, which hinders food and environmental safety efforts and makes it difficult to rapidly identify, track, and trace outbreaks, as well as for epidemiological monitoring and evolutionary research. Summary of the Invention

[0005] To address the shortcomings of existing MLST typing methods, such as poor typing performance for closely related Bacillus cereus groups, and the long typing time and high computational requirements of cgMLST and rMLST, this invention provides a novel typing method for Bacillus cereus groups based on cgcMLST.

[0006] The first objective of this invention is to provide a method for screening and differentiating cgcMLST characteristic gene combinations of Bacillus cereus, comprising the following steps:

[0007] S1. Construction of the original cgcMLST feature genome set: Obtain whole genome data of Bacillus cereus and perform screening and true species identification assessment;

[0008] S2. Obtaining effective cgcMLST characteristic genes: Grouping the Bacillus cereus strains with verified identity information, performing pan-genome analysis on each group, setting a threshold of 95%-99%, and converting the gene deletion and presence information in the pan-genome analysis into a matrix. Using the matrix, differential genes are mined and screened, and core genes that can be separated at the threshold are selected as potential candidate characteristic genes. The effect of candidate characteristic genes on the typing and identification of each group of strains is evaluated based on the strain information with verified true species identity determined in step S1. If the maximum likelihood tree clustering results of phylogenetic analysis are consistent with the true identity, then the characteristic genes are proven to be effective.

[0009] S3. Obtain effective cgcMLST characteristic gene combinations: Combine each group of effective characteristic genes, compare, align, and tandem the nucleotide sequence files of characteristic genes of different species of Bacillus cereus, determine the optimal partitioning scheme and the optimal evolutionary model, construct a phylogenetic maximum likelihood tree, evaluate the identification results of cgcMLST characteristic gene combinations for Bacillus cereus strains based on the strain information of the true species identity determined in step S1, and if the phylogenetic maximum likelihood tree clustering results are consistent with the true species identity, then the cgcMLST characteristic gene combination is effective.

[0010] Preferably, in step S2, the grouping is based on domestic and international standards and the similarity of the bacterial whole genome, dividing common Bacillus cereus bacteria into three groups:

[0011] Group BC1: Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, Bacillus megaterium;

[0012] Group BC2: emetic Bacillus cereus, Bacillus paranthracis, Bacilluspacificus, Bacillus tropicus;

[0013] Group BC3: Bacillus cytotoxicus, Bacilluspseudomycoides, Bacillustoyonensis, Bacillus wiedmannii.

[0014] Preferably, in Group BC1, differentially expressed genes are mined and screened using a matrix, and core genes that can be separated at a threshold are selected as potential candidate feature genes. The screening criteria include:

[0015] ① It is present in 100% of all Bacillus cereus, while Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, and Bacillus megaterium either do not contain this gene or have a threshold of less than 95%;

[0016] ② It is present in 100% of all Bacillus thuringiensis, while Bacillus cereus, Bacillus anthracis, Bacillus mycoides and Bacillus megaterium either do not contain this gene or have a threshold of less than 95%;

[0017] ③ It is present in 100% of all Bacillus anthracis, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus mycoides, and Bacillus megaterium.

[0018] ④ It is present in 100% of all Bacillus mycoides, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus megaterium.

[0019] ⑤ It is present in 100% of all Bacillus megaterium, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus mycoides.

[0020] ⑥ It is 100% present in all Bacillus cereus, Bacillus thuringiensis, Bacillus santhracis, Bacillus mycoides, and Bacillus megaterium, meaning it belongs to the core genome, but can be separated at the 95% homology threshold.

[0021] Preferably, the candidate characteristic genes of Group BC1 include: ispD, ytlP, est, mcsA, hemC, sigJ, and mleN; the candidate characteristic genes of Group BC2 include: ttr, yokD, arcB, racE, dhbB, mrdB, miscA, grpE, and trsA; and the candidate characteristic genes of Group BC3 include: yigZ, mtrB, ehaG, ppsC, and lgrD.

[0022] Preferably, in step S1, the construction of the original cgcMLST characteristic genome set is as follows: collecting whole genome data of Bacillus cereus and filtering it to obtain candidate genomes. The filtering conditions include: genomes with contig or scaffold ≥ 200; filtering the candidate genomes: performing average nucleotide homology analysis with the reference genome to exclude strains with an ANI value less than 0.95; performing rMLST analysis on the candidate genomes to construct a maximum likelihood tree, and determining the true species identity of strains whose clustering results with the reference genome are consistent with the bacterial homology; in step S2, the matrix is ​​a matrix of 0 and 1, where 0 indicates that the gene does not exist and 1 indicates that the gene exists.

[0023] Preferably, in step S1, the construction of the original cgcMLST feature genome set is as follows:

[0024] 1) Download the whole genome data of Bacillus cereus from the NCBI public database, filter out genomes with contigs or scaffolds ≥ 200, and use Prokka software for gene annotation;

[0025] 2) Verification analysis of the authenticity of Bacillus cereus taxa genomic classification information: Using the internationally current standard strains of Bacillus cereus taxa and closely related species as reference strains, the whole genomes (reference genomes) of unknown strains and reference strains were analyzed by ANI. The unknown strains with an ANI value greater than 0.95 and the highest similarity to the above reference strains were initially considered to be of this species. Then, rMLST analysis was performed using PhyloSuite software. Strains that were consistent with the clustering results of the reference genome among bacteria of the same species were identified as their true species identity.

[0026] Preferably, the reference genome is selected as the following Bacillus cereus group standard strains: Bacilluscereus ATCC 14579, Bacillus thuringiensis ATCC 10792, Bacillus anthracis Ames, Bacillus mycoides DSM 2048, Bacillus megaterium ATCC 14581, emetic Bacilluscereus AH187, Bacillus paranthracis Mn5, Bacillus pacificus EB422, Bacillus tropicus N24, Bacillus cytotoxicus NVH 391-98, Bacillus pseudomycoides DSM12442, Bacillus toyonensis BCT-7112, and Bacillus wiedmannii FSLW8-0169.

[0027] The second objective of this invention is to provide a novel typing method for Bacillus cereus based on cgcMLST, which includes a combination of cgcMLST characteristic genes that distinguish Bacillus cereus strains obtained using the method described above.

[0028] Preferably, it includes the following steps:

[0029] S1. Extract DNA from the samples of the bacterial strains to be tested, complete library preparation, and obtain the whole genome and whole genome information of all bacterial strain samples;

[0030] S2. Perform cgcMLST analysis on the whole genome information of the strain sample and the reference genome of Bacillus cereus: extract the characteristic genes from the cgcMLST characteristic gene combination and string them together to construct a maximum likelihood tree to identify the species identity of the test strain. The cgcMLST characteristic gene combination is the cgcMLST characteristic gene combination that distinguishes Bacillus cereus bacteria obtained by the above method.

[0031] Preferably, the cgcMLST characteristic gene combination is ytlP, yokD, grpE, and yigZ.

[0032] ytlP, yokD, grpE, and yigZ encode RNA 2',3'-cyclic phosphodiesterase, SPBc2 prophage-derived aminoglycoside N(3')-acetyltransferase-like protein, Gro-P like protein E, and impact RWD Domain Protein family member, respectively.

[0033] Preferably, the specific steps are as follows:

[0034] S1. Extract DNA from bacterial samples using an extraction kit. After DNA fragmentation, follow the instructions of the library preparation kit to complete library preparation by end repair, A-tailing, sequencing adapter addition, purification, and PCR amplification of the DNA fragments to obtain the genome of all gene samples. Then, after sequencing and assembly, obtain the whole genome information of each bacterial sample.

[0035] S2. Add the reference genome of Bacillus cereus to the whole genome information of bacterial samples, perform cgcMLST analysis using PhyloSuite software, extract characteristic genes ytlP, yokD, grpE and yigZ and connect them in series to construct a maximum likelihood tree to identify the species identity of related strains.

[0036] Preferably, the bacterial strain sample is collected from pasteurized milk, Bacillus cereus is isolated and sequenced to obtain whole genome information.

[0037] A third objective of this invention is to provide the application of ytlP, yokD, grpE, and yigZ as characteristic gene combinations in the typing of Bacillus cereus taxa.

[0038] This invention selected 13 species of Bacillus cereus from the Bacillus group that are within the standard definition, recognized as new species and considered validly published in the International Code of Nomenclature for Prokaryotes, and have a certain sample size in the NCBI public database as research subjects. Through comparative genomics, it identified commonly conserved characteristic genes among the Bacillus cereus species. By screening and combining characteristic genes, a multi-site typing method for Bacillus cereus characteristic genes based on conserved gene combination MLST (cgcMLST) was established to reduce the time and cost of Bacillus cereus typing and improve identification efficiency. Furthermore, the effectiveness of the cgcMLST method in Bacillus cereus typing was verified by combining rMLST results and whole-genome data from 81 Bacillus strains isolated from pasteurized milk in our laboratory.

[0039] This invention verifies the authenticity of existing strain information by collecting genomes of Bacillus cereus strains from public databases and previous studies. Comparative genomics is used to mine characteristic gene combinations for Bacillus cereus typing. Using strains isolated in our laboratory to validate the cgcMLST method, in analyzing 198 actual samples, it showed an efficiency improvement of approximately 99.17% compared to rMLST, enabling more effective identification of Bacillus cereus strains. Compared to existing technologies, this invention has the following advantages: By rationally setting the screening conditions for cgcMLST characteristic genes, it obtains effective characteristic gene combinations for cgcMLST, solving the problem of poor typing results of traditional MLST for closely related Bacillus cereus strains. It significantly shortens the typing time compared to rMLST, achieving efficient and accurate typing. It can be applied to the transmission analysis of Bacillus cereus strains, facilitating food and environmental safety work and enabling rapid identification, tracking, and tracing of outbreaks. Attached Figure Description

[0040] Figure 1 Phylogenetic analysis of Bacillus cereus Group BC1 based on the ytlP characteristic gene.

[0041] Figure 2 Phylogenetic analysis of Group BC2 of Bacillus cereus based on the yokD and grpE characteristic genes.

[0042] Figure 3 Phylogenetic analysis of Group BC3 of Bacillus cereus based on the yigZ characteristic gene.

[0043] Figure 4 This study aims to perform a phylogenetic analysis of Bacillus cereus taxa based on rMLST.

[0044] Figure 5 This study aims to perform a phylogenetic analysis of Bacillus cereus taxa based on cgcMLST.

[0045] Figure 6 Phylogenetic analysis of 81 Bacillus cereus strains isolated from pasteurized milk and 56 Bacillus cereus strains downloaded from the NCBI public database based on rMLST.

[0046] Figure 7 Phylogenetic analysis of 81 Bacillus cereus strains isolated from pasteurized milk and 56 Bacillus cereus strains downloaded from the NCBI public database, based on cgcMLST. Detailed Implementation

[0047] The technical approach of this invention is as follows:

[0048] (1) Construction of the original cgcMLST feature genome set

[0049] Based on domestic and international standards and the similarity of the bacterial whole genome, common Bacillus cereus bacteria are divided into three groups:

[0050] Group BC1: Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, Bacillus megaterium;

[0051] Group BC2: emetic Bacillus cereus, Bacillus paranthracis, Bacilluspacificus, Bacillus tropicus;

[0052] Group BC3: Bacillus cytotoxicus, Bacilluspseudomycoides, Bacillustoyonensis, Bacillus wiedmannii;

[0053] The following Bacillus cereus strains were selected as standard strains: Bacillus cereus ATCC 14579, Bacillus thuringiensis ATCC 10792, Bacillus anthracis Ames, Bacillus mycoides DSM 2048, Bacillus megaterium ATCC 14581, Bacillus cereus AH187, Bacillus paranthracis Mn5, Bacillus pacificus EB422, Bacillus tropicus N24, Bacillus cytotoxicus NVH 391-98, Bacillus pseudomycoides DSM 12442, Bacillus stoyonensis BCT-7112, and Bacillus wiedmannii. FSLW8-0169 was used as a reference genome to filter and screen the whole genome information of Bacillus cereus group bacteria, and a preliminary original cgcMLST characteristic genome set was obtained.

[0054] (1.1) Obtain whole genome data of Bacillus cereus group

[0055] Download the whole genome data of Bacillus cereus taxa from NCBI public database and filter and screen them to obtain candidate genomes;

[0056] The filtering criteria include: genomes with contigs or scaffolds ≥ 200;

[0057] (1.2) Screening of the whole genome library of Bacillus cereus group

[0058] Based on the gold standard of bacterial classification, the genomes of candidate Bacillus cereus groups were filtered and screened.

[0059] The filtering criteria included: performing average nucleotide identity (ANI) analysis among the same species as the reference genome, and excluding strains with an ANI value less than 0.95;

[0060] (1.3) Assessment of the true species identity of the Bacillus cereus group whole genome database

[0061] Based on the bacterial universal typing scheme ribosomal multilocussequence typing (rMLST), candidate genomes were typed to obtain typing results. Phylogeny was used to determine the optimal partitioning scheme and optimal evolutionary model for 53 ribosomal protein subunit (rps) gene data, and the data were integrated to obtain the maximum likelihood tree of phylogeny. Strains that were consistent with the clustering results of the reference genome among bacterial species were identified as their true species identity.

[0062] (2) Obtain effective cgcMLST characteristic genes

[0063] (2.1) Screening of effective cgcMLST characteristic genes

[0064] Based on the whole genome library of Bacillus cereus, the gene groups were grouped according to step (1), and a pan-genome analysis with a 95% threshold was performed using Roary. The gene deletion and presence in the above pan-genome analysis were converted into a matrix of 0 and 1. The conversion was achieved using a local script (0 indicates that the gene does not exist, and 1 indicates that the gene exists). Taking Group BC1 as an example, the matrix was used to mine and screen differential genes, and core genes that can be separated at the threshold were selected as potential candidate feature genes.

[0065] The filtering criteria include:

[0066] ① It is present in 100% of all Bacillus cereus, while it is absent or present in less than 95% of Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, and Bacillus megaterium;

[0067] ② It is present in 100% of all Bacillus thuringiensis, while it is absent or present in less than 95% of Bacillus cereus, Bacillus anthracis, Bacillus mycoides, and Bacillus megaterium;

[0068] ③ It is present in 100% of all Bacillus anthracis, while it is absent or present in less than 95% of Bacillus cereus, Bacillus thuringiensis, Bacillus mycoides, and Bacillus megaterium;

[0069] ④ It is present in 100% of all Bacillus mycoides, while it is absent or present in less than 95% of Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus megaterium.

[0070] ⑤ It is present in 100% of all Bacillus megaterium, while it is absent or present in less than 95% of Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus mycoides;

[0071] ⑥ It is 100% present in all Bacillus cereus, Bacillus thuringiensis, Bacillus santhracis, Bacillus mycoides, and Bacillus megaterium, that is, it belongs to the core genome, but can be separated at the 95% homology threshold;

[0072] (2.2) Evaluation of the representativeness of cgcMLST characteristic genes

[0073] The effectiveness of candidate feature genes in identifying species in Group BC is evaluated based on the strain information with the true identity determined in step (1). If the maximum likelihood tree clustering result of phylogenetic analysis is consistent with the true identity, then the feature genes are proven to be effective.

[0074] (3) Obtain the cgcMLST characteristic gene combination

[0075] (3.1) Screening of cgcMLST characteristic gene combinations

[0076] Each group of effective characteristic genes was combined, and the nucleotide sequence files of characteristic genes of different species of Bacillus cereus were compared, aligned and tandemd using PhyloSuite software to determine the optimal partitioning scheme and the optimal evolutionary model. The results were then integrated to obtain the maximum likelihood tree of phylogeny.

[0077] (3.2) Evaluation of the representativeness of effective cgcMLST characteristic gene combinations

[0078] Based on the strain information that determines the true species identity in step (1), evaluate the cgcMLST characteristic gene combination for the identification of Bacillus cereus strains. If the maximum likelihood tree clustering result of phylogeny is consistent with the true species identity, it proves that the cgcMLST characteristic gene combination is effective.

[0079] (4) Assess the resolution of cgcMLST for Bacillus cereus taxa.

[0080] Prokka was used to annotate the whole genome of the unknown bacteria; the whole genome of the unknown strain was compared with that of the reference strain using ANI, and the strain with the highest similarity was selected as the species; cgcMLST typing was performed, characteristic genomes were extracted and merged, and PhyloSuite software was used to align and tandem to construct a maximum likelihood tree to confirm the species identity of the strain.

[0081] The present invention will now be described in detail with reference to specific embodiments and accompanying drawings. These embodiments are further illustrative of the invention, but not intended to limit it.

[0082] Example 1:

[0083] 1. Thirteen species of the Bacillus cereus group, defined by standards, recognized as new species and validly published according to the International Code of Nomenclature for Prokaryotes, and with a certain sample size in the NCBI public database, were selected as research subjects. Based on the genomes of Bacillus cereus groups or closely related species in the NCBI public database and existing laboratory strains, genomes were downloaded and data that did not meet the analysis requirements (contigs or scaffolds ≥ 200) were removed for preliminary screening. Genome names were revised, and 13 standard strains of the Bacillus cereus group or closely related species currently recognized internationally (Table 1) were used as reference strains for the Bacillus cereus group. The processed genome data was compared with those of the Bacillus cereus group. Average nucleotide homology analysis was performed on the reference genome to exclude strains with an ANI value less than 0.95. Based on the highest similarity to the reference strain, the unknown strain was initially identified as belonging to this species. Genome annotation was then performed using Prokka software. rMLST analysis was performed using PhyloSuite software to extract 53 ribosomal protein genes (Table 2) and tandem them to construct a maximum likelihood tree. Strains with consistent clustering results with the reference genome among bacterial species were identified as having the true species identity, as shown in Table 3. A total of 425 strains were identified as true species in the Bacillus cereus group (details are shown in Table 5; the bolded parts are standard strains, those with assembly numbers are strains from the NCBI public database, and those without assembly numbers are strains currently existing in the laboratory).

[0084] Table 1. Thirteen reference strains of the currently existing Bacillus cereus group or closely related species in the international community.

[0085]

[0086] Table 253 ribosomal protein genes

[0087]

[0088]

[0089] 2. Based on domestic and international standards and the similarity of the bacterial whole genome, common Bacillus cereus bacteria are divided into three groups:

[0090] Group BC1: Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, Bacillus megaterium;

[0091] Group BC2: emetic Bacillus cereus, Bacillus paranthracis, Bacilluspacificus, Bacillus tropicus;

[0092] Group BC3: Bacillus cytotoxicus, Bacilluspseudomycoides, Bacillustoyonensis, Bacillus wiedmannii.

[0093] For each group, Roary was used to perform pan-genome analysis, with a threshold (Identity) set at 95%. The gene deletion and presence status in the above pan-genome analysis were converted into a matrix of 0 and 1 using a local script (0 indicates that the gene does not exist, and 1 indicates that the gene exists). Differential genes were mined and screened through the matrix, and core genes that could be separated at the threshold were selected as potential candidate feature genes.

[0094] The filtering criteria include:

[0095] ① It is present in 100% of all Bacillus cereus, while Bacillus thuringiensis, Bacillus anthracis, Bacillus mycoides, and Bacillus megaterium either do not contain this gene or have a threshold of less than 95%;

[0096] ② It is present in 100% of all Bacillus thuringiensis, while Bacillus cereus, Bacillus anthracis, Bacillus mycoides and Bacillus megaterium either do not contain this gene or have a threshold of less than 95%;

[0097] ③ It is present in 100% of all Bacillus anthracis, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus mycoides, and Bacillus megaterium.

[0098] ④ It is present in 100% of all Bacillus mycoides, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus megaterium.

[0099] ⑤ It is present in 100% of all Bacillus megaterium, while it is absent or present in Bacillus cereus, Bacillus thuringiensis, Bacillus anthracis, and Bacillus mycoides.

[0100] ⑥ It is 100% present in all Bacillus cereus, Bacillus thuringiensis, Bacillus santhracis, Bacillus mycoides, and Bacillus megaterium, that is, it belongs to the core genome, but can be separated at the 95% homology threshold;

[0101] Based on the screening criteria, candidate characteristic genes and related annotation information were obtained in Group BC1. Similarly, candidate characteristic genes for each group were screened in turn. The candidate characteristic gene information is shown in Table 4.

[0102] Table 4. Annotation Functional Information of Candidate Feature Genes

[0103]

[0104] 3. Evaluate the representativeness of each group of candidate feature genes: For each group, sequentially attempt to construct a phylogenetic maximum likelihood tree using a single candidate feature gene, and then proceed to constructing a phylogenetic maximum likelihood tree using multiple candidate feature genes. Based on the strain information determined in step 1, evaluate the identification effect of the candidate feature genes for each group of Bacillus cereus species. Continue until the phylogenetic maximum likelihood tree clustering result matches the true identity, thereby selecting high-quality feature genes from each group. According to the phylogenetic tree clustering results, different gray levels represent different Bacillus species, such as... Figure 1As shown, the ytlP characteristic gene can effectively distinguish the species in Group BC1; however, during the screening of characteristic genes for Group BC2, no single gene that can be directly used for this subtyping was obtained, and at least two characteristic genes are required, such as... Figure 2 As shown, the yokD and grpE characteristic genes can effectively distinguish the species in Group BC2; for example... Figure 3 As shown, the yigZ characteristic gene can effectively distinguish the bacterial species in Group BC3.

[0105] 4. Evaluation of the representativeness of effective cgcMLST characteristic gene combinations: Each group of effective characteristic genes was combined, and the nucleotide sequence files of characteristic genes from different species within the *Bacillus cereus* group were compared, aligned, and tandemd using PhyloSuite software to determine the optimal partitioning scheme and optimal evolutionary model. This was then integrated to obtain a phylogenetic maximum likelihood tree. Ten strains from each of the identified *Bacillus cereus* group species, plus one *Bacillus subtilis* strain, were selected as outgroups. PhyloSuite software was used to perform rMLST and cgcMLST analyses respectively, and a phylogenetic maximum likelihood tree was constructed. For example... Figure 4 As shown, the rMLST method can effectively cluster and genotype Bacillus cereus; for example... Figure 5 As shown, the cgcMLST method based on four characteristic gene combinations (ytlP, yigZ, yokD, and grpE) can obtain clustering results consistent with rMLST, proving that the cgcMLST method can effectively identify and classify Bacillus cereus groups.

[0106] 5. Bacterial samples were collected from pasteurized milk, and 81 strains of Bacillus were isolated. DNA was extracted from the bacterial samples using the HiPureBacterial DNA Kit D3146-03 (Guangzhou Meiji Biotechnology Co., Ltd.). After DNA fragmentation, the DNA fragments were prepared according to the instructions of the Ion Plus Fragment Library Kit (catalog number 4471252, Thermo Fisher Scientific). The DNA fragments underwent end repair, A-tailing, sequencing head addition, purification, and PCR amplification to complete library preparation, obtaining the whole genome of all bacterial samples. The whole genome information of the 81 bacterial samples was then obtained by sequencing using the Illumina NextSeq sequencing platform and assembled using SPAdes v3.6.2 software.

[0107] 6. Prokka was used to annotate the whole genomes of 81 unknown bacteria; the whole genomes of unknown strains were compared with those of reference strains using ANI, and the strains with the highest similarity to the reference strains were preliminarily identified as belonging to this species; cgcMLST typing was performed, characteristic gene combinations (ytlP, yigZ, yokD, and grpE) were extracted, and tandem alignment was performed using PhyloSuite software to construct a maximum likelihood tree for species identification. From the 81 Bacillus strains isolated in the laboratory and the 56 Bacillus cereus strains downloaded from the NCBI public database, 5 strains from each of the 12 confirmed Bacillus cereus taxa were selected, along with one Bacillus subtilis standard strain 6051-HGW as an outgroup. rMLST and cgcMLST typing were performed, and a phylogenetic maximum likelihood tree was constructed. Figure 6 and Figure 7 ).like Figure 6 and Figure 7 As shown, the typing results of cgcMLST are consistent with the clustering results of rMLST, and the bacterial species classification is consistent with their true identities. During typing, cgcMLST took 8 minutes for 198 bacterial strains, while rMLST took 16 hours. cgcMLST is approximately 99.17% more efficient than rMLST. cgcMLST can achieve efficient and accurate typing of Bacillus cereus, solving the problem of poor typing results of traditional MLST for closely related Bacillus cereus groups, and significantly shortening the time required for identification and typing compared to rMLST while maintaining accuracy.

[0108] Table 5. Bacillus cereus group information for verifying authenticity.

[0109]

[0110]

[0111]

[0112]

[0113]

[0114]

[0115]

[0116] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for screening and distinguishing cgcMLST characteristic gene combinations of Bacillus cereus group bacteria, characterized in that, Includes the following steps: S1. Construction of the original cgcMLST feature genome set: Obtain whole genome data of Bacillus cereus and perform screening and true species identification assessment; S2. Obtain effective cgcMLST feature genes: Group the Bacillus cereus groups whose identity information has been verified, perform pan-genome analysis on each group, set a threshold of 95%-99%, and convert the gene deletion and presence in the pan-genome analysis into a matrix. Use the matrix to mine and screen differential genes, and select the core genes that can be separated at the threshold as potential candidate feature genes. The effectiveness of candidate feature genes in the typing and identification of each group of species is evaluated based on the strain information that determines the true species identity in step S1. If the maximum likelihood tree clustering result of phylogeny is consistent with the true identity, then the feature genes are proven to be effective. S3. Obtain effective cgcMLST characteristic gene combinations: Combine each group of effective characteristic genes, compare, align, and tandem the nucleotide sequence files of characteristic genes of different species of Bacillus cereus, determine the optimal partitioning scheme and the optimal evolutionary model, construct a phylogenetic maximum likelihood tree, evaluate the identification results of cgcMLST characteristic gene combinations for Bacillus cereus strains based on the strain information of the true species identity determined in step S1, and prove that the cgcMLST characteristic gene combination is effective if the phylogenetic maximum likelihood tree clustering results are consistent with the true species identity.

2. The method according to claim 1, characterized in that, In step S2, the grouping is based on domestic and international standards and the similarity of the bacterial whole genome, dividing common Bacillus cereus bacteria into three groups: Group BC1: Bacillus cereus 、 Bacillus thuringiensis 、 Bacillus anthracis 、 Bacillus mycoides, Bacillus megaterium ; Group BC2: emetic Bacillus cereus, Bacillus paranthracis, Bacillus pacificus, Bacillus tropicus ; Group BC3: Bacillus cytotoxicus, Bacillus pseudomycoides, Bacillus toyonensis, Bacillus wiedmannii 。 3. The method according to claim 2, characterized in that, In Group BC1, differentially expressed genes are mined and screened using a matrix. Core genes that can be separated at a threshold are selected as potential candidate feature genes. The screening criteria include: ① 100% exists in all Bacillus cereus In the middle, and Bacillus thuringiensis , Bacillus anthracis , Bacillus mycoides and Bacillus megaterium The gene is absent or the threshold is below 95%; ② 100% exists in all Bacillus thuringiensis In the middle, and Bacillus cereus , Bacillus anthracis , Bacillus mycoides and Bacillus megaterium The gene is absent or the threshold is below 95%; ③ 100% exists in all Bacillus anthracis In the middle, and Bacillus cereus , Bacillus thuringiensis , Bacillus mycoides and Bacillus megaterium The gene is absent or the threshold is below 95%; ④ 100% exists in all Bacillus mycoides In the middle, and Bacillus cereus , Bacillus thuringiensis , Bacillus anthracis and Bacillus megaterium The gene is absent or the threshold is below 95%; ⑤ 100% exists in all Bacillus megaterium In the middle, and Bacillus cereus , Bacillus thuringiensis , Bacillus anthracis and Bacillus mycoides The gene is absent or the threshold is below 95%; ⑥ 100% exists in all Bacillus cereus , Bacillus thuringiensis , Bacillus anthracis , Bacillus mycoides and Bacillus megaterium In the middle, that is, belonging to the core genome, but can be separated at the 95% homology threshold.

4. The method according to claim 2, characterized in that, The candidate characteristic genes of Group BC1 include: ispD , ytlP , est , mcsA , hemC , sigJ , mleN The candidate characteristic genes of Group BC2 include: ttr , yokD , arcB , racE , dhbB , mrdB , miscA , grpE , trsA The candidate characteristic genes of Group BC3 include: yigZ , mtrB , ehaG , ppsC , lgrD .

5. The method according to claim 1, characterized in that, Step S1, the construction of the original cgcMLST feature genome set is as follows: collect whole genome data of Bacillus cereus and filter them to obtain candidate genomes. The filtering conditions include: genomes with contig or scaffold ≥ 200; filter the candidate genomes: perform average nucleotide homology analysis with the reference genome to exclude strains with ANI values ​​less than 0.95; perform rMLST analysis on the candidate genomes, construct a maximum likelihood tree, and determine the true species identity of strains whose clustering results with the reference genome are consistent with the bacterial homology; Step S2, the matrix is ​​a matrix of 0 and 1, where 0 indicates that the gene does not exist and 1 indicates that the gene exists.

6. The method according to claim 5, characterized in that, The reference genome is: a standard strain of the Bacillus cereus group. Bacillus cereus ATCC 14579 Bacillus thuringiensis ATCC 10792 Bacillus anthracis Ames Bacillus mycoides DSM 2048 Bacillus megaterium ATCC14581 emetic Bacillus cereus AH187 Bacillus paranthracis Mn5, Bacillus pacificus EB422, Bacillus tropicus N24, Bacillus cytotoxicus NVH 391-98 Bacillus pseudomycoides DSM 12442 Bacillus toyonensis BCT-7112 Bacillus wiedmannii FSL W8-0169 was used as the reference genome.

7. A novel typing method for Bacillus cereus groups based on cgcMLST, characterized in that, Includes the cgcMLST characteristic gene combination that distinguishes Bacillus cereus taxa, obtained by using the method described in any one of claims 1-6.

8. The classification method according to claim 7, characterized in that, Includes the following steps: S1. Extract DNA from the samples of the bacterial strains to be tested, complete library preparation, and obtain the whole genome and whole genome information of all bacterial strain samples; S2. Perform cgcMLST analysis on the whole genome information of the strain sample and the reference genome of Bacillus cereus: extract the characteristic genes from the cgcMLST characteristic gene combination and string them together to construct a maximum likelihood tree to identify the species identity of the strain to be tested. The cgcMLST characteristic gene combination is the cgcMLST characteristic gene combination that distinguishes Bacillus cereus bacteria obtained by the method described in any one of claims 1-6.

9. The classification method according to claim 7, characterized in that, The cgcMLST characteristic gene combination is as follows: ytlP , yokD , grpE and yigZ .

Images