Rapid bacterial identification method based on thermally driven death-melting curve analysis

Abstract

The invention discloses a rapid bacterial identification method based on thermally driven death-melting curve analysis, and belongs to the technical field of microbiological detection. According to the method, a death-melting curve of bacterium / strain specificity is constructed by utilizing membrane thermal stability difference of bacteria and genome dsDNA sequence difference, and Pearson correlation analysis is performed on the death-melting curves of different bacteria by using an SPSSPRO data analysis platform, so that bacteria variety differentiation and rapid identification of the bacteria are realized. The identification method provided by the invention has the characteristics of rapid and efficient detection, simple operation, low cost, high identification accuracy, high specificity and strong universality.

Description

Technical Field

[0001] This invention relates to the field of microbial detection technology, and in particular to a rapid bacterial identification method based on thermally driven death-melting curve analysis. Background Technology

[0002] Bacterial identification plays a crucial role in biological research, food safety control, clinical diagnosis, and effective treatment. Currently, mainstream bacterial identification technologies mainly include the following: 16S rRNA sequencing technology: Identification is achieved by amplifying the conserved 16S rRNA gene of bacteria and performing sequencing comparison; it is a commonly used method for microbial classification. However, this technology requires steps such as DNA extraction, PCR amplification, sequencing, and data analysis, resulting in a long operation cycle (usually several hours to several days), and it relies on specialized sequencing equipment and data analysis capabilities, leading to high costs. Antibody immunoassay technology: Based on the principle of antigen-antibody specific binding, it achieves bacterial detection and has a certain degree of specificity. However, it requires the preparation of specific antibodies against different bacteria; the antibody screening and production process is complex and costly, and its applicability is limited to known target bacteria, making it unable to identify unknown bacteria.

[0003] Mass spectrometry (such as MALDI-TOF MS): Identification is achieved by analyzing the protein fingerprint of bacteria. It is faster than sequencing, but relies on high-precision mass spectrometry instruments (expensive equipment) and a complete protein database, making it difficult for grassroots laboratories to implement. PCR technology and its derivatives: Requires the design of specific primers, can only detect known bacteria, and carries the risk of cross-contamination. The process involves nucleic acid extraction and amplification, making it impossible to simultaneously identify bacteria and assess their activity. Biosensor technology: Its application has gradually increased in recent years, offering advantages such as speed and sensitivity. However, most sensors rely on antibodies or aptamers as recognition elements. The preparation and screening of these elements are complex, limiting their universal identification capability for a wide range of bacteria.

[0004] Existing technologies suffer from the aforementioned limitations. In scenarios such as soil microbiology surveys, it is difficult to quickly and cost-effectively distinguish and analyze the abundance of complex bacterial communities. In clinical diagnosis, long testing cycles may delay the treatment of infectious diseases. Furthermore, most existing technologies cannot utilize the inherent characteristics of bacteria (such as membrane thermal stability and differences in genomic DNA sequences) to identify bacteria without recognition elements, thus limiting the application of these technologies.

[0005] Based on this, a rapid bacterial identification method based on thermally driven death-melting curve analysis is provided. Summary of the Invention

[0006] To address the aforementioned technical problems, the present invention aims to provide a rapid bacterial identification method based on heat-driven death-melting curve analysis. This method utilizes the differences in bacterial membrane thermal stability and genomic dsDNA sequence to construct strain-specific "death-melting curves" for rapid identification.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A rapid bacterial identification method based on heat-driven death-melting curve analysis includes the following steps: (1) Add the bacterial suspension to be tested and SYBR Green fluorescent dye to the qPCR tube; (2) Place the qPCR tube into a temperature-controlled fluorescence detector to monitor the changes in bacterial fluorescence intensity and corresponding temperature data in real time, and to produce the death curve and melting curve. (3) Extract the key parameters of the death curve and melting curve and normalize them to ensure that the curve shapes of samples with different concentrations are comparable; (4) Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria to distinguish bacterial species.

[0008] Preferably, the method further includes the following steps: (5) Dimension reduction is performed on the curve dataset of multiple strains to form clear strain clusters and achieve rapid identification.

[0009] Preferably, the concentration of the bacterial suspension to be tested is 1×10⁻⁶. 7 -1×10 9 CFU / mL.

[0010] Preferably, the temperature-controlled fluorescence detector is set with the following program parameters: 37°C maintained for 5 seconds → temperature increased from 37°C to 98°C at a rate of 0.1°C / second → 98°C maintained for 5 seconds.

[0011] Preferably, the emission wavelength of the fluorescence detector is 530 nm.

[0012] Preferably, the temperature range of the death curve is 37-70℃; the temperature range of the melting curve is 70-98℃.

[0013] Preferably, the key parameters include the death temperature Td, the melting temperature Tm, the fluorescence rise rate, and the fluorescence fall rate.

[0014] Preferably, the normalization process refers to normalizing the fluorescence curve to eliminate the influence of bacterial concentration on fluorescence intensity.

[0015] Preferably, the bacterial species differentiation criteria are: curve similarity ≥98% for the same strain and <98% for different strains.

[0016] Preferably, the dimensionality reduction employs any one or a combination of two or more of principal component analysis, kernel principal component analysis, linear discriminant analysis, and t-distributed random neighborhood embedding.

[0017] The advantages of this invention compared to the prior art are as follows: This invention's identification method requires no recognition elements such as antibodies or aptamers. It is simple to operate, rapid (completed within 10 minutes), low-cost, and can simultaneously identify multiple bacteria. Furthermore, it eliminates the need for DNA extraction, PCR amplification, or prolonged culture, simplifying the process and reducing the risk of contamination. This invention utilizes the differences in bacterial membrane thermal stability and genomic dsDNA sequence to construct strain-specific "death-melting curves" for rapid identification.

[0018] The entire detection process of this invention (from sample loading to curve analysis) can be completed within 10 minutes, which is much faster than 16S rRNA sequencing (several hours to several days), mass spectrometry detection (more than 30 minutes) and traditional culture methods (several days). It is especially suitable for time-sensitive scenarios such as emergency clinical diagnosis and rapid food safety screening.

[0019] This invention eliminates the need for expensive sequencers and mass spectrometers, requiring only a conventional fluorescence PCR instrument; it eliminates the need for high-cost recognition elements such as antibodies and aptamers, with the core reagent being only SYBR Green dye; and it eliminates the need for DNA extraction, PCR amplification, or long-term culture, simplifying the operation steps, reducing the technical requirements for operators, and making it suitable for widespread use in grassroots laboratories.

[0020] This invention utilizes inherent bacterial characteristics (membrane thermal stability, dsDNA sequence) to construct species / strain-specific "fingerprint curves." The curve similarity of parallel experiments for the same strain is ≥98%, while the similarity between different strains is <98%. It is applicable to the identification and abundance analysis of bacterial communities in pure cultured bacteria (pathogenic bacteria, probiotics, etc.) and complex environmental samples (soil, water, etc.), and has strong universality.

[0021] In summary, the identification method of this invention has the characteristics of rapid and efficient detection, simple operation, low cost, high identification accuracy, high specificity, and strong universality. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the principle of thermally driven death-melting curve analysis (DMCA).

[0023] Figure 2The results are the death melting curves of nine different bacteria, including (a) Corynebacterium glutamicum, (b) Lactobacillus casei, (c) Staphylococcus aureus, (d) Staphylococcus albus, (e) Escherichia coli O157:H7, (f) Streptococcus mutans, (g) Pseudomonas aeruginosa, (h) Acinetobacter baumannii, (i) Escherichia coli TOP10, and (j) a mixture of death melting curves from different bacteria.

[0024] Figure 3 The analysis is a dimensionality reduction analysis of the death-melting curves; (a) is a mixture of normalized death-melting curves of different bacteria, and (bf) is a dimensionality reduction analysis of the above curves by KPCA (bc), LDA (d), PCA (e) and T-SNE (f).

[0025] Figure 4 Soil bacteria maps were constructed using death-melting curve analysis; (a) soil sampling and pretreatment, (b) single-cell derived soil bacterial death-melting curves, (c) T-SNE dimensionality reduction analysis of the above curves, and (d) sequencing results based on 16S rRNA and bacterial distribution in the dimensionality reduction map. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and preferred embodiments. However, it should be noted that many details listed in the specification are merely to provide the reader with a thorough understanding of one or more aspects of the present invention, and these aspects of the invention can be implemented even without these specific details.

[0027] A rapid bacterial identification method based on heat-driven death-melting curve analysis includes the following steps: (1) Add the bacterial suspension to be tested and SYBR Green fluorescent dye to the qPCR tube; (2) Place the qPCR tube into a temperature-controlled fluorescence detector to monitor the changes in bacterial fluorescence intensity and corresponding temperature data in real time, and to produce the death curve and melting curve. (3) Extract the key parameters of the death curve and melting curve and normalize them to ensure that the curve shapes of samples with different concentrations are comparable; (4) Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria to distinguish bacterial species.

[0028] In one embodiment, the following steps are also included: (5) Dimension reduction is performed on the curve dataset of multiple strains to form clear strain clusters and achieve rapid identification.

[0029] In one embodiment, the concentration of the bacterial suspension to be tested was 1 × 10⁻⁶. 7 -1×10 9 CFU / mL.

[0030] In one embodiment, the temperature-controlled fluorescence detector is programmed to maintain a temperature of 37°C for 5 seconds, then increase the temperature from 37°C to 98°C at a rate of 0.1°C / second, and maintain the temperature at 98°C for 5 seconds.

[0031] In one embodiment, the emission wavelength of the fluorescence detector is 530 nm.

[0032] In one embodiment, the temperature range of the death curve is 37-70°C; the temperature range of the melting curve is 70-98°C.

[0033] In one embodiment, the key parameters include the death temperature Td, the melting temperature Tm, the fluorescence rise rate, and the fluorescence fall rate.

[0034] In one embodiment, the normalization process refers to normalizing the fluorescence curve to eliminate the influence of bacterial concentration on fluorescence intensity.

[0035] In one embodiment, the bacterial species differentiation criteria are: curve similarity ≥98% for the same strain and <98% for different strains.

[0036] In one embodiment, the dimensionality reduction employs any one or a combination of two or more of principal component analysis, kernel principal component analysis, linear discriminant analysis, and t-distributed random neighborhood embedding.

[0037] SYBR Green is a fluorescence-enhancing dye targeting double-stranded DNA (dsDNA). It exhibits weak fluorescence in its free state and becomes strongly fluorescent upon binding to dsDNA. When heated from 37°C to 70°C, the damaged membranes of bacteria gradually die, increasing the likelihood of SYBR Green penetrating from the outside to the inside of the bacteria, thus enhancing the fluorescence signal. Conversely, as the temperature continues to rise, the bacterial genomic double-stranded DNA gradually melts, leading to a decrease in the fluorescence signal. Differences in bacterial thermal stability and double-stranded DNA composition allow for the identification of these bacteria using multifactorial dimensionality-reduced death-melting curves.

[0038] The instruments, reagents, and bacteria used in this invention are as follows: reagents Core reagents: SYBR Green dye (purchased from Thermo Fisher Scientific), prepared at 2× or 4× concentration; Culture media: LB medium (containing 0.5g NaCl, 0.5g yeast extract, and 1.0g peptone, dissolved in 100mL ultrapure water, self-prepared), MRS medium (purchased from Best Biotechnology Co., Ltd.), M9 medium (for soil bacterial amplification); PBS buffer: containing 8g NaCl, 0.2g KCl, 1.44g Na₂HPO₄, and 0.24g KH₂PO₄, dissolved in 800mL ultrapure water, pH adjusted to 7.4, and filtered through a 0.22μm filter membrane; Ultrapure water: prepared using a Milli-Q RG system for reagent preparation.

[0039] strain Corynebacterium glutamicum, Lactobacillus casei, Staphylococcus aureus, Staphylococcus albus, Streptococcus mutans, Pseudomonas aeruginosa, Acinetobacter baumannii, Escherichia coli TOP10, Escherichia coli O157:H7, and unknown bacterial groups in the soil environment.

[0040] Bacterial sample preparation The bacteria to be identified (such as C. glutamicum, S. aureus, etc.) were inoculated into their respective culture media: Except for *L. casei*, which was inoculated on MRS medium, all other strains were inoculated on LB medium; cultured until the bacterial concentration reached OD500. 600 =1.0 (logarithmic growth phase), collect bacterial cells by centrifugation at 5000 rpm for 5 minutes; Wash bacterial cells three times with PBS buffer to remove residual culture medium; resuspend bacteria in PBS buffer and adjust concentration to OD. 600 =0.5 (100% viability), for later use.

[0041] Positive samples after identification can be verified by 16S rRNA sequencing. Bacterial DNA was extracted from qPCR tubes and amplified by PCR using primers (forward: 5'-AGAGTTTGATCCTGGCTCAG-3'; reverse: 5'-GGTTACCTTGTTACGACTT-3'). The amplified products were sequenced, and the bacterial species were determined by NCBI BLAST comparison to verify the identification accuracy of the DMCA method.

[0042] Example 1

[0043] Identification principle as follows Figure 1 As shown, to culture bacteria: Take 15 μL of the prepared bacterial suspension (concentration 1×10⁻⁶). 7 -1×10 9 Add CFU / mL to the qPCR tube, then add 15 μL of 2×SYBR Green solution and mix well; Temperature-controlled fluorescence detection: Place the qPCR tube into the fluorescence detector (StepOnePlus™ real-time PCR system, Thermo Fisher Scientific); set the programmed temperature parameters: 37℃ for 5 seconds → increase from 37℃ to 98℃ at a rate of 0.1℃ / second → hold at 98℃ for 5 seconds; during the detection process, record the fluorescence intensity (FL, emission wavelength of 530nm) and corresponding temperature data in real time, and automatically generate the death curve (37-70℃, reflecting the bacterial death process) and melting curve (70-98℃, reflecting the dsDNA melting process).

[0044] Curve feature extraction: Key parameters were extracted from the original fluorescence curves: death temperature (Td, the temperature at which fluorescence intensity reaches its maximum value, corresponding to the complete death of bacteria), melting temperature (Tm, the inflection point temperature at which fluorescence intensity drops sharply, corresponding to the extensive denaturation of dsDNA), fluorescence rise rate (37-Td stage), and fluorescence fall rate (Td-98℃ stage); the fluorescence curves were normalized (to eliminate the influence of bacterial concentration on fluorescence intensity) to ensure that the curve shapes of samples with different concentrations were comparable.

[0045] Similarity analysis: Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria: the similarity of parallel experimental curves of the same strain was ≥98%, and the similarity of curves of different strains was <98%, which was used to distinguish species / strains; Dimensionality reduction analysis: Dimensionality reduction is performed on the curve dataset of multiple bacterial strains (≥100 curves per strain). Optional algorithms include principal component analysis (PCA), nuclear principal component analysis (KPCA, Cosin / Poly model), linear discriminant analysis (LDA), and t-distributed random neighborhood embedding (T-SNE). Among them, the T-SNE algorithm has the best discrimination effect, which can completely separate the curves of 9 different bacteria in a two-dimensional scatter plot, forming clear strain clusters and achieving rapid identification.

[0046] Example 2

[0047] The above nine bacteria were used as model organisms, with five parallel experiments for each strain. The specific steps are as follows: Pure bacterial culture: Take 15 μL of the prepared bacterial suspension (concentration 1×10⁻⁶). 7 -1×10 9 Add CFU / mL to the qPCR tube, then add 15 μL of 2×SYBR Green solution and mix well; Temperature-controlled fluorescence detection: Place the qPCR tube into the fluorescence detector (StepOnePlus™ real-time PCR system, Thermo Fisher Scientific); set the programmed temperature parameters: 37℃ for 5 seconds → increase from 37℃ to 98℃ at a rate of 0.1℃ / second → hold at 98℃ for 5 seconds; during the detection process, record the fluorescence intensity (FL, emission wavelength of 530nm) and corresponding temperature data in real time, and automatically generate the death curve (37-70℃, reflecting the bacterial death process) and melting curve (70-98℃, reflecting the dsDNA melting process).

[0048] Curve feature extraction: Key parameters were extracted from the original fluorescence curves: death temperature (Td, the temperature at which fluorescence intensity reaches its maximum value, corresponding to the complete death of bacteria), melting temperature (Tm, the inflection point temperature at which fluorescence intensity drops sharply, corresponding to the extensive denaturation of dsDNA), fluorescence rise rate (37-Td stage), and fluorescence fall rate (Td-98℃ stage); the fluorescence curves were normalized (to eliminate the influence of bacterial concentration on fluorescence intensity) to ensure that the curve shapes of samples with different concentrations were comparable.

[0049] Similarity analysis: Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria: the similarity of parallel experimental curves of the same strain was ≥98%, and the similarity of curves of different strains was <98%, which was used to distinguish species / strains; Dimensionality reduction analysis: Dimensionality reduction is performed on the curve dataset of multiple bacterial strains (≥100 curves per strain). Optional algorithms include principal component analysis (PCA), nuclear principal component analysis (KPCA, Cosin / Poly model), linear discriminant analysis (LDA), and t-distributed random neighborhood embedding (T-SNE). Among them, the T-SNE algorithm has the best discrimination effect, which can completely separate the curves of 9 different bacteria in a two-dimensional scatter plot, forming clear strain clusters and achieving rapid identification.

[0050] The results are as follows Figure 2As shown, the results indicate that all strains exhibited inverted V-shaped death-melting curves, suggesting that these two biological processes are prevalent in different bacteria. The differences in the rise / fall rates of FL and the peaks showing the strongest FL on the curves are significant. These differences are attributed to the thermostability of the bacteria and the variations in their genomic dsDNA, which hold great potential for bacterial identification.

[0051] Example 3

[0052] To extract and visualize the characteristics of different bacteria in a two-dimensional scatter plot, dimensionality reduction analysis was performed on the mortality-melting curves. Algorithms such as nuclear principal component analysis (KPCA, Cosin, and Poly models), linear discriminant analysis (LDA), principal component analysis (PCA), and t-distributed random neighborhood embedding (T-SNE) were applied during the dimensionality reduction process. More than 100 curves were used for each strain, such as... Figure 3 As shown. First, the curve is normalized to eliminate fluctuations in FL reporting caused by bacterial concentration, such as... Figure 3 As shown in figure a. After dimensionality reduction, the scatter plots representing bacterial cells of different strains are distributed in the figure, as shown in figure a. Figure 3 As shown in bf. Among these algorithms, T-SNE showed the best differentiation, being able to completely isolate nine strains.

[0053] Example 4

[0054] Soil bacterial sample preparation Soil sampling: Surface soil was collected from Nanhu Park in Nanning. 10g of soil sample was dissolved in 50mL of PBS buffer. Dispersion and separation: The soil-PBS mixture was sonicated at 25°C and 100W for 10 minutes to fully disperse the bacteria in the soil; Centrifugation: Centrifuge at 500 rpm for 10 minutes to remove soil particle precipitate and retain the supernatant containing bacteria; Serial dilution: Determine the total bacterial concentration in the supernatant by plate counting (M9 medium), and serially dilute it to 10² CFU / mL with PBS buffer to ensure that each tube contains no more than 1 bacterial cell in subsequent sampling; Single-bacterial amplification: Take 5 μL of diluted soil bacterial suspension and add it to a qPCR tube containing 5 μL of 4×SYBR Green solution and 10 μL of 2×M9 medium. Incubate at 37℃ and 200 rpm for 6 hours to achieve single-bacterial amplification.

[0055] Cultivate bacteria: Take 15 μL of the prepared bacterial suspension (concentration 1×10⁻⁶). 7 -1×10 9Add CFU / mL to the qPCR tube, then add 15 μL of 2×SYBR Green solution and mix well; Temperature-controlled fluorescence detection: Place the qPCR tube into the fluorescence detector (StepOnePlus™ real-time PCR system, Thermo Fisher Scientific); set the programmed temperature parameters: 37℃ for 5 seconds → increase from 37℃ to 98℃ at a rate of 0.1℃ / second → hold at 98℃ for 5 seconds; during the detection process, record the fluorescence intensity (FL, emission wavelength of 530nm) and corresponding temperature data in real time, and automatically generate the death curve (37-70℃, reflecting the bacterial death process) and melting curve (70-98℃, reflecting the dsDNA melting process).

[0056] Curve analysis and bacterial identification Curve feature extraction: Key parameters were extracted from the original fluorescence curves: death temperature (Td, the temperature at which fluorescence intensity reaches its maximum value, corresponding to the complete death of bacteria), melting temperature (Tm, the inflection point temperature at which fluorescence intensity drops sharply, corresponding to the extensive denaturation of dsDNA), fluorescence rise rate (37-Td stage), and fluorescence fall rate (Td-98℃ stage); the fluorescence curves were normalized (to eliminate the influence of bacterial concentration on fluorescence intensity) to ensure that the curve shapes of samples with different concentrations were comparable.

[0057] Similarity analysis: Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria: the similarity of parallel experimental curves of the same strain was ≥98%, and the similarity of curves of different strains was <98%, which was used to distinguish species / strains; Dimensionality reduction analysis: Dimensionality reduction is performed on the curve dataset of multiple bacterial strains (≥100 curves per strain). Optional algorithms include principal component analysis (PCA), nuclear principal component analysis (KPCA, Cosin / Poly model), linear discriminant analysis (LDA), and t-distributed random neighborhood embedding (T-SNE). Among them, the T-SNE algorithm has the best discrimination effect, which can completely separate the curves of 9 different bacteria in a two-dimensional scatter plot, forming clear strain clusters and achieving rapid identification.

[0058] For complex samples such as soil, 16S rRNA sequencing can be performed to verify the positive samples after identification. Bacterial DNA was extracted from qPCR tubes and amplified by PCR using primers (forward: 5'-AGAGTTTGATCCTGGCTCAG-3'; reverse: 5'-GGTTACCTTGTTACGACTT-3'). The amplified products were sequenced, and the bacterial species were determined by NCBI BLAST comparison to verify the identification accuracy of the DMCA method.

[0059] In summary, bacterial composition, in turn, can be used to infer soil properties and functions. Here, soil samples were collected and tested from Nanhu Park (Nanning, China) to further investigate the practicality of the developed Thermally Driven Death Melting Curve Analysis (DMCA) for bacterial identification. First, the soil was melted in PBS and sonicated to disperse bacterial cells, as... Figure 4 Figure 4a shows the bacterial and soil separation process. The turbid solution was centrifuged at low speed to separate the bacteria from the soil, and then serially diluted to a final concentration of ~10² cfu / mL. Finally, 5 μL aliquots of the bacterial suspension were inoculated into qPCR tubes containing M9 medium and SYBR Green for bacterial proliferation and cell death / melting curve analysis. This protocol ensures that each tube contains only one or more original bacterial cells. Figure 4b shows the cell death / melting curves, each curve representing a single bacterial cell. Dimensionality reduction was performed on 219 positive samples out of 384 samples (4 × 96 tubes / plate). Figure 4 As shown in c, the bacteria differentiated into more than 14 populations, indicating that more than 14 strains were present in the soil.

[0060] Since 16S rRNA sequences are widely used in taxonomic classification in microbial research, this invention also performs 16S rRNA sequencing on all samples to confirm the accuracy of DMCA-based differentiation.

[0061] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A rapid bacterial identification method based on heat-driven death-melting curve analysis, characterized in that, Includes the following steps: (1) Add the bacterial suspension to be tested and SYBR Green fluorescent dye to the qPCR tube; (2) Place the qPCR tube into a temperature-controlled fluorescence detector to monitor the changes in bacterial fluorescence intensity and corresponding temperature data in real time, and to produce the death curve and melting curve. (3) Extract the key parameters of the death curve and melting curve and normalize them to ensure that the curve shapes of samples with different concentrations are comparable; (4) Using the SPSSPRO data analysis platform, Pearson correlation analysis was performed on the death-melting curves of different bacteria to distinguish bacterial species.

2. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, It also includes the following steps: (5) Dimension reduction is performed on the curve dataset of multiple strains to form clear strain clusters and achieve rapid identification.

3. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The concentration of the bacterial suspension to be tested was 1×10⁻⁶. 7 -1×10 9 CFU / mL.

4. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The temperature-controlled fluorescence detector is set to the following parameters: maintain 37℃ for 5 seconds → increase the temperature from 37℃ to 98℃ at a rate of 0.1℃ / second → maintain 98℃ for 5 seconds.

5. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The emission wavelength of the fluorescence detector is 530nm.

6. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The temperature range of the death curve is 37-70℃; the temperature range of the melting curve is 70-98℃.

7. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The key parameters include death temperature Td, melting temperature Tm, fluorescence rise rate, and fluorescence fall rate.

8. The rapid bacterial identification method based on thermally driven death-melting curve analysis according to claim 1, characterized in that, The normalization process refers to normalizing the fluorescence curve to eliminate the influence of bacterial concentration on fluorescence intensity.

9. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The criteria for distinguishing bacterial species are: curve similarity ≥98% for the same strain and <98% for different strains.

10. The rapid bacterial identification method based on heat-driven death-melting curve analysis according to claim 1, characterized in that, The dimensionality reduction employs one or more of the following methods: principal component analysis, kernel principal component analysis, linear discriminant analysis, and t-distributed random neighborhood embedding.

Images