A colorimetric LAMP method for simultaneous visual detection of vibrio vulnificus and vibrio alginolyticus

Abstract

This invention relates to the field of food testing technology, and more particularly to a colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus. The method employs a LAMP detection system composed of DNA-functionalized gold nanospheres and gold nanorods for detecting Vibrio vulnificus and Vibrio alginolyticus, respectively, and LAMP primer pairs. This allows for the detection and determination of the following states of Vibrio vulnificus and Vibrio alginolyticus: when neither pathogen is present, the solution is transparent; when both pathogens are present, the solution is purplish-red; when only Vibrio alginolyticus is present, the solution is gray; and when only Vibrio vulnificus is present, the solution is purple. This method can accurately determine the presence or absence of Vibrio vulnificus and Vibrio alginolyticus, and the specific type of bacteria, avoiding misjudgments, missed detections, and inaccurate traceability of these two types of pathogen contamination in food safety monitoring, clinical practice, and aquaculture protection.

Description

Technical Field

[0001] This invention relates to the field of food testing technology, and in particular to a colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus. Background Technology

[0002] Vibrio vulnificus ( V. vulnificus ) and Vibrio alginolyticus ( V. alginolyticus Both Vibrio vulnificus and Vibrio alginolyticus belong to the Vibrio genus of the Vibrioceae family. They are Gram-negative, halophilic, facultative anaerobic bacilli, sharing similar living environments and transmission routes, and exhibiting highly similar phenotypic characteristics. Infection with either pathogen is easily achieved through consuming raw seafood or through wounds coming into contact with seawater. Vibrio vulnificus is known as the "invisible killer of the ocean," posing a serious threat to human survival after infection. Vibrio alginolyticus, on the other hand, is the leading pathogen in aquaculture, easily causing large-scale mortality of fish, shrimp, and shellfish, resulting in severe economic losses; human infection with either pathogen usually presents with mild symptoms.

[0003] Given the differences in infection risk and impact on actual production between the two, accurate detection and differentiation are of great significance for food safety, aquaculture control, and clinical diagnosis. Currently, traditional culture methods, molecular biology methods, and whole-genome sequencing can be used for differentiation. However, traditional culture methods are only for preliminary differentiation, are time-consuming, and have limited resolution. Whole-genome sequencing is costly and more suitable for scientific research and source tracing analysis. Therefore, molecular biology methods have become the mainstream method for differentiating these pathogens. Among these, LAMP technology, as a powerful nucleic acid detection platform, has significant advantages, including isothermal operation without temperature-controlled instruments and rapid target amplification (usually completed within 1 hour). Its 4-6 primer design enables LAMP to achieve superior sequence specificity compared to traditional PCR. However, existing LAMP detection methods also face certain limitations. For example, fluorescence-based LAMP methods require expensive probes and specialized equipment to detect signals, while colorimetric detection cannot simultaneously detect and differentiate multiple targets.

[0004] To overcome the aforementioned problems, gold nanoparticles have demonstrated superior properties as colorimetric probes. Their localized surface plasmon resonance characteristics and high molar extinction coefficient enable them to exhibit significant color transitions, proving their ability to replace traditional fluorescent dyes. DNA-functionalized gold nanoparticles (DNA-AuNPs), based on their surface-specific DNA sequences, can target and trigger the aggregation and dispersion of nanoparticles, resulting in color changes. They have been used for rapid detection of pathogens, but most studies focus on single DNA-functionalized gold nanoparticles and generally only detect a single pathogen at a time. For example, patent application 2022114465920 discloses a LAMP reaction reagent and its application for detecting Vibrio parahaemolyticus and / or Vibrio vulnificus, using 16nm gold nanoparticles in a LAMP system for detection, with two color changes used to determine the presence or absence of Vibrio parahaemolyticus and Vibrio vulnificus. However, this detection result cannot further clarify the specific pathogen, and it cannot accurately identify the coexistence or non-coexistence of these pathogens in the sample. Furthermore, for LAMP detection of different combinations of pathogens, especially combinations of more closely related pathogens, the presence of more homologous sequences poses greater challenges to target gene selection, primer design, and amplification conditions. In addition, the potential cross-reactivity of different combinations of pathogens also introduces many interferences to the accurate detection and differentiation of targets.

[0005] Based on this, this application proposes a novel colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus. The research and development goal is to achieve a more accurate distinction between the different states of these two specific combinations of pathogens, and to achieve the purpose of precise typing detection. Summary of the Invention

[0006] This invention provides a colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus. It can accurately determine the presence or absence of Vibrio vulnificus and Vibrio alginolyticus, as well as the specific type of bacteria, avoiding misjudgment, missed detection, and inaccurate traceability in food safety monitoring, clinical practice, and aquaculture protection. This solves the problems existing in the prior art.

[0007] The technical solution adopted in this invention is:

[0008] A colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus is provided. The LAMP detection system consists of DNA-functionalized gold nanospheres and gold nanorods, used to identify Vibrio vulnificus and Vibrio alginolyticus respectively, and LAMP primer pairs. This allows for the detection and determination of the following states of Vibrio vulnificus and Vibrio alginolyticus: when neither pathogen is present, the solution is transparent; when both pathogens are present, the solution is purplish-red; when only Vibrio alginolyticus is present, the solution is gray; and when only Vibrio vulnificus is present, the solution is purple.

[0009] Furthermore, the DNA-functionalized gold nanospheres and gold nanorods are respectively modified with Vibrio vulnificus probes. VV and Vibrio alginolyticus probe VA The Probe VV The sequence is shown in SEQ ID NO.1, the Probe VA The sequence is shown in SEQ ID NO.2;

[0010] The LAMP primer pairs include primer set 1 for amplifying Vibrio vulnificus and primer set 2 for amplifying Vibrio alginolyticus; primer set 1 includes the outer primer pairs F3 and B3 as shown in SEQ ID NO.3 and SEQ ID NO.4, the inner primer pairs FIP and BIP as shown in SEQ ID NO.5 and SEQ ID NO.6, and the loop primer pairs LF and LB as shown in SEQ ID NO.7 and SEQ ID NO.8; primer set 2 includes the outer primer pairs F3 and B3 as shown in SEQ ID NO.9 and SEQ ID NO.10, the inner primer pairs FIP and BIP as shown in SEQ ID NO.11 and SEQ ID NO.12, and the loop primer pairs LF and LB as shown in SEQ ID NO.13 and SEQ ID NO.14.

[0011] Furthermore, the molar ratio of probe-modified gold nanorods to probe-modified gold nanospheres is 0.5-1.5:3.

[0012] Furthermore, Probe is modified on the gold nanorods. VA DNA-functionalized gold nanorod probes were obtained, and Probe was modified on the gold nanospheres. VV DNA-functionalized gold nanosphere probes were obtained. The probes... VA -AuNRs probes and probes VV The preferred molar concentration ratio of the AuNSs probe is 1:3.

[0013] Furthermore, the LAMP detection system also includes buffer, dNTP, MgSO4, and BstDNA polymerase, with amplification conditions of 65°C for 30 minutes.

[0014] Furthermore, the above-mentioned colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus includes the following steps:

[0015] DNA-functionalized gold nanospheres were prepared separately. VV -AuNSs and DNA-functionalized gold nanorods Probe VA-AuNRs; LAMP primer set for Vibrio vulnificus and Vibrio alginolyticus amplification: outer primer pair F3 and B3, inner primer pair FIP and BIP, loop primer pair LF and LB. Prepare the following detection system and amplify at 65℃ for 30 minutes. Add MgCl2 to induce color development, and judge based on the color change of the solution; Detection system: containing 1 μM FIP, 1 μM BIP, 0.2 μM F3, 0.2 μM B3, 0.4 μM LF and 0.4 μM LB primers, 1.2 mM each of dNTP mixture, Isothermal Amplification Buffer (1×), 4 mM MgSO4, 0.24 U Bst 2.0 DNA polymerase, Probe VV -AuNSs 3 nM、Probe VA -AuNRs 1 nM and 1 μL target DNA, total volume 10 μL.

[0016] Furthermore, the concentration of MgCl2 is 90 mM.

[0017] Furthermore, the above MgCl2 addition adopts the following operation steps: a certain concentration of MgCl2 solution is first added to the cap of the centrifuge tube used for amplification reaction. After LAMP amplification is completed, further centrifugation can ensure that it falls from the cap into the centrifuge tube, achieving 90mM and realizing the purpose of inducing color change.

[0018] Furthermore, the above-mentioned MgCl2 solution of a certain concentration is 1M, and the amount added ensures that its concentration in the system after entering the amplification is 90mM.

[0019] Furthermore, Probe VV -AuNSs and Probe VA The preparation steps for AuNRs are as follows:

[0020] (1) Preparation of DNA-functionalized gold nanospheres

[0021] S1. In the presence of 20 mM NaNO3, Probe VV Added to the gold nanosphere solution, the total volume of the mixture was 100 μL; Probe VV The final concentration was 10 μM, and the final concentration of gold nanospheres was 10 nM.

[0022] S2. Add 700 μL of n-butanol to the above mixture solution and vortex mix rapidly;

[0023] S3. Heat the mixture after S2 treatment at 80℃ for 30 min, and then add 200 μL of 0.5×TBE buffer.

[0024] S4. Heat the mixture obtained in S3 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VV - After adding ultrapure water to the gold nanosphere solution and centrifuging, the supernatant was removed to obtain DNA-functionalized gold nanospheres. VV -AuNSs, for backup;

[0025] (2) Preparation of DNA-functionalized gold nanorods

[0026] S5. In the presence of 15 mM NaNO3, Probe VA Added to the gold nanorod solution, the total volume of the mixture was 100 μL; Probe VA The final concentration was 15 μM, and the final concentration of gold nanorods was 2 nM.

[0027] S6. Add 700 μL of n-butanol to the mixture solution in S5 and vortex mix rapidly;

[0028] S7. Heat the mixture treated in S6 at 80°C for 30 min, then add 200 μL of 0.5×TBE buffer.

[0029] S8. Heat the mixture obtained in S7 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VA -AuNRs solution, after adding ultrapure water and centrifugation, the supernatant was removed to obtain DNA-functionalized gold nanorods Probe. VA -AuNRs, for backup.

[0030] Furthermore, the detection and judgment of color changes in the sample solution can be carried out by direct visual interpretation or by inputting the result photos into a computer for automatic interpretation after training and evaluation using a machine learning model.

[0031] The above-described computer-based machine learning-based automatic interpretation method, when used in the field of molecular diagnostics, can eliminate human visual interpretation errors and reduce workload. Furthermore, the machine learning model training and evaluation operations of this invention are as follows:

[0032] The resulting photos were cropped using Photoshop to a consistent size of 170 pixels wide, 240 pixels long, with a horizontal resolution of 96 dpi, a vertical resolution of 96 dpi, and a bit depth of 32. The model was trained using 680 sample data points, divided into four groups: a negative sample group (170 samples). V. vulnificus Group (170 samples) V. alginolyticus Group (170 samples) V. vulnificus and V. alginolyticusThere are groups (170 samples). 34 samples are randomly selected from each group as the test set (136 samples in total), and the remaining 136 samples form the training set (544 samples in total).

[0033] A Random Forest (RF) model based on ResNet-50 is used. Pre-defined training and validation sets of images are loaded into a dataset, and image preprocessing is performed, including image scaling, cropping, tensorization, and ImageNet normalization. Then, a ResNet-50 network pre-trained on ImageNet is used to remove the final classification layer, serving as a feature extractor to extract a 2048-dimensional depth feature vector for each image. This feature vector is used as input to train the RF classifier, which is then used for prediction and performance evaluation on the validation set. Classification accuracy and a classification report are output, and the trained RF model is saved locally. Based on this, a learning curve for the RF model is plotted using cross-validation based on the training set features. This is used to analyze the changes in model training performance and generalization ability under different training sample sizes, and the learning curve graph and corresponding statistical results are saved. In the inference phase, the trained and saved RF model is first loaded, and a model is constructed to support the training phase. A completely identical ResNet-50 feature extractor is used to perform the same preprocessing procedure on the images to be predicted, extracting 2048-dimensional depth features for each image. The features are then input into the RF model for class prediction, obtaining the predicted class, confidence score, and predicted probability for each class for each sample. All results are then compiled and saved as a CSV file. Finally, in the evaluation phase, the CSV file generated by inference is read, and true labels are constructed based on the known distribution of sample numbers for each class. The overall classification accuracy is calculated, and the accuracy and macro-average F1 score are further evaluated to generate and save a complete classification report. At the same time, the original and normalized confusion matrices are plotted and saved. When the predicted probabilities are available, the multi-class receiver operating characteristic (ROC) curve is calculated and plotted, and the corresponding area under the curve (AUC) index and its original data file are output to achieve a systematic analysis and visualization of the model's discriminative performance and confidence characteristics.

[0034] This invention also provides the application of the method in visually detecting and distinguishing between any of the following infection states: Vibrio vulnificus and Vibrio alginolyticus in food or clinical samples, with a detection limit of 10. 2 CFU / mL:

[0035] a) Vibrio vulnificus infection, the sample solution turns purple;

[0036] b. Infection with Vibrio alginolyticus, the sample solution appears gray;

[0037] c. Infection with Vibrio vulnificus and Vibrio alginolyticus will cause the sample solution to appear purple-red.

[0038] d. No Vibrio vulnificus or Vibrio alginolyticus infection; the sample solution is transparent.

[0039] The beneficial effects of this invention are:

[0040] 1. This invention uses a combination of DNA-functionalized gold nanospheres and gold nanorods, along with a LAMP primer set, to form a LAMP detection system. This system enables accurate detection and differentiation of Vibrio vulnificus and Vibrio alginolyticus in various states, including presence, coexistence, and specific identification. Color changes can accurately indicate whether the infection is caused by Vibrio vulnificus, Vibrio alginolyticus, co-infection, or no infection with either bacterium. This avoids misjudgment or missed detection of infection by either bacterium in food safety monitoring, clinical practice, or aquaculture protection in actual production.

[0041] 2. Specifically, this invention obtains a gold nanomaterial probe combination by modifying gold nanospheres and gold nanorods with probe sequences for the precise identification of Vibrio vulnificus and Vibrio alginolyticus. This overcomes the existing established mode of using only single DNA-functionalized gold nanoparticles for target bacteria identification. It forms a special detection system with LAMP primers for the specific amplification of the two pathogens. At the same time, by optimizing the ratio of the two DNA-functionalized gold nanomaterials, the purpose of accurately detecting and determining the aforementioned multiple states of Vibrio vulnificus and Vibrio alginolyticus is achieved.

[0042] 3. By comparing the results with existing LAMP fluorescence methods, the detection limits of this invention for Vibrio vulnificus and Vibrio alginolyticus are comparable to existing methods, and it achieves colorimetric detection in more states. Actual seafood sample testing comparisons show that the detection method of this invention can achieve accurate, rapid, and repeatable detection of these two pathogens in multiple states. Attached Figure Description

[0043] Figure 1 Vibrio vulnificus of the present invention V. vulnificus and Vibrio alginolyticus V. alginolyticus Simultaneous detection principle diagram;

[0044] Figure 2 For different Probe VA -AuNRs and Probe VV - An investigation into the impact of AuNSs probe ratio on colorimetric detection performance;

[0045] Figure 3 For the present invention to V. vulnificus and V. alginolyticus Simultaneous colorimetric detection results image;

[0046] Figure 4 For the present invention to V. vulnificus and V. alginolyticus Image of colorimetric test results for oyster samples;

[0047] Figure 5 For Probe VA-AuNRs and Probe VV -Investigation of the colorimetric effect of AuNSs probes at different MgCl2 concentrations;

[0048] Figure 6 for V. vulnificus and V. alginolyticus LAMP fluorescence detection results;

[0049] Figure 7 To V. vulnificus and V. alginolyticus LAMP fluorescence detection results of oyster samples;

[0050] Figure 8 The diagram shows the learning curve (a), ROC curve (b), normalized confusion matrix (c), and confusion matrix (d) of the RF model.

[0051] in: Figure 2 In the middle: 'a' represents Probe in different proportions. VA -AuNRs and Probe VV - UV-Vis spectrum of the AuNSs probe, inset is actual photograph; b represents probes at different scales. VA -AuNRs and Probe VV -AuNSs probe colorimetric detection V. vulnificus and V. alginolyticus The result;

[0052] Figure 3 In Chinese: a is V. vulnificus The ultraviolet-visible spectrum of colorimetric detection; b is V. alginolyticus The ultraviolet-visible spectrum of colorimetric detection; c is V. vulnificus and V. alginolyticus Simultaneous presence of UV-Vis spectra for colorimetric detection; insets are actual photographs;

[0053] Figure 4 In Chinese: a is a V. vulnificus Results of the detection limit for oyster samples; b is for V. alginolyticus Detection limit results for oyster samples; c indicates simultaneous presence. V. vulnificus and V. alginolyticus The detection limit results for oyster samples;

[0054] Figure 6 In Chinese: a is a V. vulnificus The fluorescence detection range; b is for V. alginolyticus The fluorescence detection range;

[0055] Figure 7 In Chinese: a is a V. vulnificus The fluorescence detection range of oyster samples; b is... V. alginolyticus Fluorescence detection range of oyster samples. Detailed Implementation

[0056] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific implementation methods.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application.

[0058] The following embodiments are for illustrative purposes only and are not intended to limit the scope of this application. Experimental methods in the following embodiments that do not specify specific conditions can be performed according to experimental manuals or conventional conditions in the art, or by referring to known experimental methods in the art. Unless otherwise specified, all methods are conventional methods in the art.

[0059] Example 1.

[0060] This embodiment is used for Vibrio vulnificus ( V. vulnificus ) and Vibrio alginolyticus ( V. alginolyticus The colorimetric LAMP method for simultaneous visualization detection includes the following steps:

[0061] (1) DNA-functionalized gold nanospheres Probe VV -AuNSs preparation

[0062] S1. Preparation of gold nanosphere solution: First, add 1 mL of HAuCl4 solution (25 mM) to 95 mL of ultrapure water, stir well and heat to boiling. Then, quickly add 4 mL of sodium citrate solution (34 mM); after the solution turns wine red, continue stirring and heating, maintaining a gentle boil for 15 min; store the prepared AuNSs at 4℃ in the dark for later use.

[0063] In the presence of 20 mM NaNO3, Probe VV The mixture was added to the aforementioned gold nanosphere solution, and the total volume of the mixture was 100 μL; Probe VV The final concentration was 10 μM, and the final concentration of gold nanospheres was 10 nM.

[0064] S2. Add 700 μL of n-butanol to the above mixture solution and vortex mix rapidly;

[0065] S3. Heat the mixture after S2 treatment at 80℃ for 30 min, and then add 200 μL of 0.5×TBE buffer.

[0066] S4. Heat the mixture obtained in S3 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VV - After adding ultrapure water to the gold nanosphere solution and centrifuging, the supernatant was removed to obtain DNA-functionalized gold nanospheres. VV -AuNSs, for backup;

[0067] (2) DNA-functionalized gold nanorods Probe VA -AuNRs preparation

[0068] S5. Preparation of gold nanorod solution:

[0069] 1) Preparation of seed solution: Mix 5.0 mL of cetyltrimethylammonium bromide (CTAB) solution (0.2 M) with 5.0 mL of HAuCl4 solution (0.5 mM) and stir until homogeneous. The solution is bright yellow. Under rapid stirring, add 0.6 mL of freshly prepared sodium borohydride solution (10 mM) all at once. The solution immediately turns brownish-yellow. Continue stirring vigorously for 2 min, then stop stirring and let the seed solution stand at 25°C for 5 min.

[0070] 2) Preparation of growth solution: Add 1 mL of silver nitrate solution (4 mM) to 50 mL of CTAB solution (0.2 M) and stir until well mixed; then add 50 mL of HAuCl4 solution (1 mM) and stir until well mixed, the solution turns dark yellow; then, add 700 μL of ascorbic acid solution (78.8 mM) dropwise, the solution becomes colorless and transparent;

[0071] 3) Growth of AuNRs: Add 120 μL of seed solution to the growth solution while stirring, mix well and stop stirring. Let it grow in a 30℃ water bath for 12 h. Store the obtained AuNRs at 4℃ for later use.

[0072] In the presence of 15 mM NaNO3, Probe VA Add to AuNRs solution, the total volume of the mixture is 100 μL; Probe VA The final concentration was 15 μM, and the final concentration of gold nanorods was 2 nM.

[0073] S6. Add 700 μL of n-butanol to the mixture solution in S5 and vortex mix rapidly;

[0074] S7. Heat the mixture treated in S6 at 80°C for 30 min, then add 200 μL of 0.5×TBE buffer.

[0075] S8. Heat the mixture obtained in S7 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VA -AuNRs solution, after adding ultrapure water and centrifugation, the supernatant was removed to obtain DNA-functionalized gold nanorods Probe. VA -AuNRs, for backup;

[0076] (3) Construction of detection system and colorimetric judgment

[0077] LAMP primer sets for the amplification of Vibrio vulnificus and Vibrio alginolyticus were designed using the NEB LAMP primer design tool: two sets of outer primer pairs F3 and B3, inner primer pairs FIP and BIP, and loop primer pairs LF and LB. The following detection system was prepared and amplified at 65℃ for 30 minutes. The results were then analyzed based on the color change of the solution.

[0078] Detection system: Contains 1 μM FIP, 1 μM BIP, 0.2 μM F3, 0.2 μM B3, 0.4 μM LF and 0.4 μM LB primers, 1.2 mM each of dNTP mixture, Isothermal Amplification Buffer (1×), 4 mM MgSO4, 0.24 U Bst 2.0 DNA polymerase, and Probe. VV -AuNSs 3 nM、Probe VA -AuNRs 1 nM and 1 μL target DNA, total volume 10 μL;

[0079] After amplification, MgCl2 solution was added to induce color development, ensuring a final concentration of 90 mM for colorimetric determination.

[0080] If the solution is transparent, it indicates that there are no Vibrio vulnificus or Vibrio alginolyticus in the sample; if the solution is purplish-red, it indicates that both Vibrio vulnificus and Vibrio alginolyticus are present in the sample; if the solution is gray, it indicates that only Vibrio alginolyticus is present in the sample; if the solution is purple, it indicates that only Vibrio vulnificus is present in the sample.

[0081] For Vibrio vulnificus ( V. vulnificus Primer set for amplification detection of Vibrio alginolyticus ( ) V. alginolyticus Primer set for amplification detection, and Probe VV Probe VA The probe sequence information is shown in Table 1 below.

[0082] Table 1. DNA sequence information used

[0083]

[0084] Example 2: Examining different probesVA -AuNRs and Probe VV - The impact of AuNSs ratio on colorimetric detection performance.

[0085] Based on the method of Example 1, this example uses the following different probes. VA -AuNRs and Probe VV - LAMP colorimetric detection using the following AuNSs ratios: DNA-AuNRs: DNA-AuNSs = 0.5nM: 3nM, 1nM: 3nM, 1.5nM: 3nM, 1nM: 1nM, 3nM: 1nM.

[0086] See results Figure 2 ,Depend on Figure 2 As shown in section a, with Probe VA -Increased AuNRs concentration, Probe VA The intensity of the UV-Vis characteristic peaks of -AuNRs gradually increases, while the intensity of the probe mixture in the probe solution also increases. VA -The color of AuNRs gradually becomes more distinct. Figure 2 As shown in b, with Probe VA The increased concentration of AuNRs mainly indicates that the Probe VA The color of AuNRs negatively impacts detection results. Therefore, Probe should be selected. VA -AuNRs and Probe VV -AuNSs ratio of 1 nM:3 nM yields the best colorimetric detection results.

[0087] Example 3: Investigation of the lowest detection limit for two pathogens.

[0088] Experimental Procedure and Methods: Components listed in Table 2 were added sequentially to 200 μL microcentrifuge tubes. After addition, 10 μL of PCR-grade paraffin oil was added to each reaction tube, and 1 μL of MgCl2 (1 M) was carefully dripped onto the center of the inner wall of the centrifuge tube cap. The reaction tubes were placed in a constant-temperature metal bath and amplified at 65℃ for 30 min. After amplification, the reaction tubes were removed, and the MgCl2 solution dripped from the inner wall of the cap was mixed using a handheld centrifuge.

[0089] Table 2 Colorimetric LAMP reaction systems for simultaneous visualization detection.

[0090]

[0091] See results Figure 3 ,like Figure 3 As shown in Figure a, the colorimetric LAMP system in this embodiment can detect a minimum of 10. 2 CFU / mLV. vulnificus ;like Figure 3 As shown in Figure b, the colorimetric LAMP system in this embodiment can detect a minimum of 10. 2 CFU / mL V. alginolyticus ;like Figure 3 As shown in Figure c, the colorimetric LAMP system in this embodiment can detect the simultaneous presence of at least 10 2 CFU / mL V. vulnificus and 10 2 CFU / mL V. alginolyticus The sample.

[0092] Example 4: Investigation of the lowest detection limit for colorimetric detection of oyster samples.

[0093] The experimental procedure and method are the same as those in Example 3.

[0094] See results Figure 4 ,Depend on Figure 4 As shown in Figure a, the colorimetric LAMP system in this embodiment can detect a minimum of 10. 2 CFU / mL V. vulnificus Oyster samples; Figure 4 As shown in b, the colorimetric LAMP system can detect a minimum of 10. 2 CFU / mL V. alginolyticus Oyster samples; [[ID=6 The data in section c shows that the colorimetric LAMP system can detect the simultaneous presence of at least 10 2 CFU / mL ​ and 10 2 CFU / mL ​ Oyster samples.

[0095] Example 5: The detection performance of gold nanospheres and gold nanorods modified with probe sequences at different MgCl2 concentrations was investigated.

[0096] Based on the method in Example 1, the performance of Probe was investigated under different MgCl2 concentrations. VA -AuNRs、Probe VV -The colorimetric effect of the AuNSs probe. Regarding the addition of MgCl2, a 1M magnesium chloride solution can be added to the cap of the centrifuge tube used in the amplification reaction system beforehand. After amplification, the tube is slightly centrifuged to allow the MgCl2 to fall into the post-amplification reaction solution and reach the desired concentration for colorimetric induction.

[0097] Experimental Procedure and Methods: Components listed in Table 3 were added sequentially to 200 μL microcentrifuge tubes. After addition, 10 μL of PCR-grade paraffin oil was added to each reaction tube. 1 μL of MgCl2 (250 mM, 500 mM, or 1 M) was carefully dripped onto the center of the inner wall of the centrifuge tube cap. The reaction tubes were placed in a constant-temperature metal bath and amplified at 65℃ for 30 min. After amplification, the reaction tubes were removed, and the MgCl2 solution dripped from the inner wall of the cap was mixed using a handheld centrifuge.

[0098] Table 3. Reaction systems for the two probe colorimetric salt concentration experiments.

[0099]

[0100] See results ​ ,Depend on ​ The colorimetric images corresponding to different MgCl2 concentrations: 23 mM, 45 mM, and 90 mM show that the colorimetric effect is most obvious at a MgCl2 concentration of 90 mM.

[0101] Comparative examples: The LAMP primers designed in Example 1 were applied to... ​ and V. ​ Fluorescence detection, and ​ and ​ Fluorescence detection of oyster samples.

[0102] ​ and ​ Fluorescence detection experimental procedure and method: Add the components listed in Table 4 sequentially to 200 μL microcentrifuge tubes. After sample addition, add 10 μL of PCR-grade paraffin oil to each reaction tube, and place the reaction tubes into a qPCR device for amplification reaction.

[0103] Table 4. Fluorescent LAMP reaction system for simultaneous visualization detection

[0104]

[0105] like ​ In the middle a, the fluorescent LAMP system can detect as low as 10 2 CFU / mL ​ Similarly, such as ​ As shown in Figure b, the fluorescent LAMP system can detect a minimum of 10 2 CFU / mL ​ .

[0106] Therefore, it can be seen that, compared with fluorescent LAMP, the detection system of this invention can achieve a detection accuracy comparable to that of fluorescent LAMP.

[0107] Includes​ and ​ Experimental procedure and method for fluorescence detection of oyster samples: same as above. V. ​ and ​ Fluorescence detection experimental procedures and methods.

[0108] The results are as follows ​ , ​ As shown in Figure a, the fluorescent LAMP system can detect a minimum of 10 2 CFU / mL V. ​ Oyster samples; similarly, such as ​ As shown in b, the fluorescent LAMP system can detect a minimum of 10 2 CFU / mL V. ​ Oyster sample.

[0109] Example 6: Using computer + machine learning to quickly interpret the results of color photos.

[0110] This embodiment acquires and trains the image of the detection result to enable the computer to recognize the result more quickly. Specifically, the machine learning model training and evaluation operations are as follows:

[0111] The resulting photos were cropped using Photoshop to a consistent size of 170 pixels wide, 240 pixels long, with a horizontal resolution of 96 dpi, a vertical resolution of 96 dpi, and a bit depth of 32. The model was trained using 680 sample data points, divided into four groups: a negative sample group (170 samples). ​ Group (170 samples) ​ Group (170 samples) ​ and ​ There are groups (170 samples). 34 samples are randomly selected from each group as the test set (136 samples in total), and the remaining 136 samples form the training set (544 samples in total).

[0112] A Random Forest (RF) model based on ResNet-50 is used. Pre-defined training and validation sets of images are loaded into a dataset, and image preprocessing is performed, including image scaling, cropping, tensorization, and ImageNet normalization. Then, a ResNet-50 network pre-trained on ImageNet is used to remove the final classification layer, serving as a feature extractor to extract a 2048-dimensional depth feature vector for each image. This feature vector is used as input to train the RF classifier, which is then used for prediction and performance evaluation on the validation set. Classification accuracy and a classification report are output, and the trained RF model is saved locally. Based on this, a learning curve for the RF model is plotted using cross-validation based on the training set features. This is used to analyze the changes in model training performance and generalization ability under different training sample sizes, and the learning curve graph and corresponding statistical results are saved. In the inference phase, the trained and saved RF model is first loaded, and a model is constructed to support the training phase. A completely identical ResNet-50 feature extractor is used to perform the same preprocessing procedure on the images to be predicted, extracting 2048-dimensional depth features for each image. The features are then input into the RF model for class prediction, obtaining the predicted class, confidence score, and predicted probability for each class for each sample. All results are then compiled and saved as a CSV file. Finally, in the evaluation phase, the CSV file generated by inference is read, and true labels are constructed based on the known distribution of sample numbers for each class. The overall classification accuracy is calculated, and the accuracy and macro-average F1 score are further evaluated to generate and save a complete classification report. At the same time, the original and normalized confusion matrices are plotted and saved. When the predicted probabilities are available, the multi-class receiver operating characteristic (ROC) curve is calculated and plotted, and the corresponding area under the curve (AUC) index and its original data file are output to achieve a systematic analysis and visualization of the model's discriminative performance and confidence characteristics.

[0113] See ​ Figure 'a' shows a 4-class classification model trained using RF based on the experimental data. The validation score initially rises steadily, while both the training and validation scores stabilize towards the end. Both scores are high and the difference is small. This indicates a good model fit without overfitting, demonstrating effective model training. In the RF test, the AUC for all four types of samples is above 0.91, indicating strong sample discrimination ability of the model. (See [link to relevant documentation]). ​ b. From ​ As shown in Figure C, the confusion matrix reveals the confusion between different categories of samples. Comparative analysis shows that the RF model achieves an F1 score of 0.914. The trained model achieves an accuracy of 0.917 in predicting the four types of samples, indicating that the ML model has good recognition performance. ​ The result of d.

[0114] The present invention has been described in detail above. The specific embodiments described above should not be construed as limiting the scope of protection of the present invention. Any alternative modifications or variations made to the embodiments of the present invention by those skilled in the art will fall within the scope of protection of the present invention. Any aspects of the present invention not described in detail are well-known techniques to those skilled in the art.

Claims

1. A colorimetric LAMP method for the simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus, characterized in that, A LAMP detection system was constructed using DNA-functionalized gold nanospheres and gold nanorods, along with LAMP primer pairs, specifically designed to identify Vibrio vulnificus and Vibrio alginolyticus, to amplify and develop colorimetric results. This system enabled the detection and identification of Vibrio vulnificus and Vibrio alginolyticus in the following states: when neither pathogen is present, the solution is transparent; when both pathogens are present, the solution is purplish-red; when only Vibrio alginolyticus is present, the solution is gray; and when only Vibrio vulnificus is present, the solution is purple.

2. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 1, characterized in that, The DNA-functionalized gold nanospheres and gold nanorods were respectively modified with the Vibrio vulnificus probe Probe. VV and Vibrio alginolyticus probe VA The Probe VV The sequence is shown in SEQ ID NO.1, the Probe VA The sequence is shown in SEQ ID NO.2; The LAMP primer pairs include primer set 1 for amplifying Vibrio vulnificus and primer set 2 for amplifying Vibrio alginolyticus; primer set 1 includes the outer primer pairs F3 and B3 as shown in SEQ ID NO.3 and SEQ ID NO.4, the inner primer pairs FIP and BIP as shown in SEQ ID NO.5 and SEQ ID NO.6, and the loop primer pairs LF and LB as shown in SEQ ID NO.7 and SEQ ID NO.8; primer set 2 includes the outer primer pairs F3 and B3 as shown in SEQ ID NO.9 and SEQ ID NO.10, the inner primer pairs FIP and BIP as shown in SEQ ID NO.11 and SEQ ID NO.12, and the loop primer pairs LF and LB as shown in SEQ ID NO.13 and SEQ ID NO.

14.

3. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 2, characterized in that, The molar ratio of probe-modified gold nanorods to probe-modified gold nanospheres is 0.5-1.5:

3.

4. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 1, characterized in that, The LAMP detection system also includes buffer, dNTP, MgSO4, and BstDNA polymerase, and the amplification conditions are: amplification at 65℃ for 30 minutes.

5. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to any one of claims 1-4, characterized in that, The operating steps include: DNA-functionalized gold nanospheres were prepared separately. VV -AuNSs and DNA-functionalized gold nanorods Probe VA -AuNRs; LAMP primer set for Vibrio vulnificus and Vibrio alginolyticus amplification: outer primer pair F3 and B3, inner primer pair FIP and BIP, loop primer pair LF and LB. Prepare the following detection system and amplify at 65℃ for 30 minutes. Add MgCl2 to induce color development, and judge based on the color change of the solution; Detection system: containing 1 μM FIP, 1 μM BIP, 0.2 μM F3, 0.2 μM B3, 0.4 μM LF and 0.4 μM LB primers, 1.2 mM each of dNTP mixture, Isothermal Amplification Buffer (1×), 4 mM MgSO4, 0.24 U Bst 2.0 DNA polymerase, Probe VV -AuNSs 3 nM、Probe VA -AuNRs 1 nM and 1 μL target DNA, total volume 10 μL.

6. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 5, characterized in that, The concentration of MgCl2 was 90 mM.

7. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 5, characterized in that, Probe VV -AuNSs and Probe VA The preparation steps for AuNRs are as follows: (1) Preparation of DNA-functionalized gold nanospheres S1. In the presence of 20 mM NaNO3, Probe VV Added to the gold nanosphere solution, the total volume of the mixture was 100 μL; Probe VV The final concentration was 10 μM, and the final concentration of gold nanospheres was 10 nM. S2. Add 700 μL of n-butanol to the above mixture solution and vortex mix rapidly; S3. Heat the mixture after S2 treatment at 80℃ for 30 min, then add 200 μL of 0.5×TBE buffer. S4. Heat the mixture obtained in S3 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VV - After adding ultrapure water to the gold nanosphere solution and centrifuging, the supernatant was removed to obtain DNA-functionalized gold nanospheres. VV -AuNSs, for backup; (2) Preparation of DNA-functionalized gold nanorods S5. In the presence of 15 mM NaNO3, Probe VA Added to the gold nanorod solution, the total volume of the mixture was 100 μL; Probe VA The final concentration was 15 μM, and the final concentration of gold nanorods was 2 nM. S6. Add 700 μL of n-butanol to the mixture solution in S5 and vortex mix rapidly; S7. Heat the mixture treated in S6 at 80°C for 30 min, then add 200 μL of 0.5×TBE buffer. S8. Heat the mixture obtained in S7 at 80°C for 30 min, centrifuge to separate the two phases, and take out the lower layer. VA -AuNRs solution, after adding ultrapure water and centrifugation, the supernatant was removed to obtain DNA-functionalized gold nanorods Probe. VA -AuNRs, for backup.

8. The colorimetric LAMP method for simultaneous visual detection of Vibrio vulnificus and Vibrio alginolyticus according to claim 5, characterized in that, The detection and judgment of color changes in sample solutions can be carried out by direct visual interpretation or by inputting the result photos into a computer for automatic interpretation after training and evaluation using a machine learning model.

9. The application of the method according to any one of claims 1-4 in the visual detection and differentiation of any of the following infection states of Vibrio vulnificus and Vibrio alginolyticus in food or clinical samples, with a detection limit of 10. 2 CFU / mL: a) Vibrio vulnificus infection, the sample solution turns purple; b. Infection with Vibrio alginolyticus, the sample solution appears gray; c. Infection with Vibrio vulnificus and Vibrio alginolyticus will cause the sample solution to appear purple-red. d. No Vibrio vulnificus or Vibrio alginolyticus infection; the sample solution is transparent.

Images