A virus infection detection method and related device based on machine learning

Abstract

The present application discloses a virus infection detection method and related devices based on machine learning, which relate to the technical field of virus infection detection. The method includes: obtaining two-color labeled fluorescence trajectory data containing the position information of the fluorescent spots of the virus envelope and the virus genome changing over time, and extracting the virus envelope trajectory data and the virus genome trajectory data respectively; matching the virus envelope trajectory data and the virus genome trajectory data according to preset matching rules to determine the paired trajectory data; finally, according to the paired trajectory data, using a virus release detection model constructed and trained based on a machine learning algorithm, the virus release detection result is obtained. The above-mentioned scheme of the present application greatly saves manpower, material resources and time consumption through a fully automatic trajectory extraction, pairing and detection scheme, and can quickly and batch detect viral nucleic acid release events based on fluorescent trajectory data, thereby quickly and large-scale calculating and evaluating the virus infection rate.

Description

Technical Field

[0001] The present application relates to the technical field of virus infection detection, and in particular to a virus infection detection method and related devices based on machine learning. Background Art

[0002] The release of nucleic acids is a critical moment in the virus infection process and the first step for the virus to successfully infect host cells. By monitoring and analyzing the release of nucleic acids, the spread of the virus can be mastered in real time. In the current research context, in order to conduct large-scale virus infection rate analysis, an efficient quantitative single virus tracing method with dual-color labeled viruses is usually used. This method uses streptavidin-modified quantum dots to mark the viral envelope and uses a hybridization chain reaction to mark the new coronavirus genome. When photographed using a TIRF microscope, the viral envelope appears green, while the viral genome marker appears red, achieving multi-color tracking and recording trajectory data. Over time, if the viral genome enters the cell, it will separate from the viral envelope. However, it is time-consuming and laborious to distinguish these changes only by the naked eye, especially when large-scale data needs to be processed, this defect of the existing technology is more obvious. Summary of the invention

[0003] The purpose of this application is to provide a method and related device for detecting viral infection based on machine learning, which can quickly and batch detect viral nucleic acid release events based on fluorescence trajectory data.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a method for detecting virus infection based on machine learning, comprising the following steps:

[0006] The dual-color labeled fluorescence trajectory data of the virus particles are obtained; the dual-color labeled fluorescence trajectory data includes the position information of the fluorescent spots of the virus envelope changing with time and the position information of the fluorescent spots of the virus genome changing with time.

[0007] According to the dual-color labeled fluorescence trajectory data, virus envelope trajectory data and virus genome trajectory data are extracted respectively; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames.

[0008] The virus envelope trajectory data and the virus genome trajectory data are matched according to preset matching rules to determine paired trajectory data; the paired trajectory data includes the position information of the virus envelope and the position information of the virus genome that meet the preset matching rules.

[0009] According to the paired trajectory data, the virus release detection result is output using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

[0010] Optionally, obtaining dual-color labeled fluorescence trajectory data of virus particles specifically includes:

[0011] Biotinylated lipid molecules are inserted into the lipid layer of the viral envelope, and then the viral envelope of the virus particles is fluorescently labeled by utilizing the specific binding of streptavidin-modified quantum dots to the lipid molecules.

[0012] The viral genome of the virus particles was fluorescently labeled using hybridization chain reaction technology.

[0013] The dual-labeled virus particles were imaged in real time using a total internal reflection fluorescence microscope to obtain several frames of dual-color labeled fluorescence image data of the virus particles.

[0014] An image processing program is used to adjust the coordinates and intensity of the fluorescent spots in each frame of dual-color labeled fluorescent image data, and to reconstruct the position information of the fluorescent spots in the virus particles over time to obtain dual-color labeled fluorescent trajectory data.

[0015] Optionally, according to the paired trajectory data, using a trained virus release detection model, outputting a virus release detection result specifically includes:

[0016] For the paired trajectory data, the position information of the virus envelope and the position information of the virus genome are interpolated using the cubic spline interpolation method to obtain interpolated trajectory data.

[0017] For the interpolation trajectory data, the position information of the virus envelope and the position information of the virus genome are combined by subtraction to obtain combined paired trajectory data.

[0018] The combined paired trajectory data is input into the trained virus release detection model to obtain the virus release detection result.

[0019] Optionally, the virus release detection model is a binary classification model constructed and trained using an XGBoost algorithm.

[0020] Optionally, before inputting the merged paired trajectory data into a trained virus release detection model to obtain a virus release detection result, the method further includes:

[0021] A model training data set is obtained; the model training data set includes merged paired trajectory data and virus release results.

[0022] The initial virus release detection model was constructed based on the XGBoost algorithm.

[0023] The merged paired trajectory data is used as the model input, the virus release result is used as the target output of the model, and the model training data set is used to train the initial virus release detection model to obtain a trained virus release detection model.

[0024] Optionally, the preset matching rule is that, in a number of consecutive frames, the distance between the virus envelope and the virus genome in more than half of the frames is less than a preset threshold.

[0025] In a second aspect, the present application provides a virus infection detection system based on machine learning, comprising the following modules:

[0026] The dual-color trajectory data acquisition module is used to acquire dual-color labeled fluorescence trajectory data of virus particles; the dual-color labeled fluorescence trajectory data includes the position information of the fluorescent spots of the virus envelope changing with time and the position information of the fluorescent spots of the virus genome changing with time.

[0027] The trajectory data extraction module is used to extract virus envelope trajectory data and virus genome trajectory data respectively according to the dual-color labeled fluorescence trajectory data; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames.

[0028] The trajectory data pairing module is used to match the virus envelope trajectory data and the virus genome trajectory data according to a preset matching rule to determine paired trajectory data; the paired trajectory data includes the position information of the virus envelope and the position information of the virus genome that meet the preset matching rule.

[0029] The virus release detection module is used to output the virus release detection result according to the paired trajectory data using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

[0030] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-mentioned machine learning-based virus infection detection methods.

[0031] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned machine learning-based virus infection detection methods.

[0032] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned machine learning-based virus infection detection methods.

[0033] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0034] The present application provides a method and related device for detecting virus infection based on machine learning, the method comprising: obtaining two-color labeled fluorescence trajectory data containing the position information of the fluorescent spots of the virus envelope and the virus genome changing over time, and extracting the virus envelope trajectory data and the virus genome trajectory data respectively; matching the virus envelope trajectory data and the virus genome trajectory data according to preset matching rules to determine the paired trajectory data; finally, according to the paired trajectory data, using a virus release detection model built and trained based on a machine learning algorithm, to obtain the virus release detection result. The above-mentioned scheme of the present application reduces the dependence on human eye judgment through a fully automatic trajectory extraction, pairing and detection scheme, greatly saves manpower and time consumption, and can quickly and batch detect viral nucleic acid release events based on fluorescence trajectory data, thereby quickly and large-scale calculating and evaluating the virus infection rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A flowchart of a virus infection detection method based on machine learning is provided in one embodiment of the present application.

[0037] Figure 2 This is a flowchart of step A2 in a virus infection detection method based on machine learning provided in one embodiment of the present application.

[0038] Figure 3 This is a flowchart of step A4 in a virus infection detection method based on machine learning provided in one embodiment of the present application.

[0039] Figure 4 A flowchart of model construction and training in a virus infection detection method based on machine learning provided in one embodiment of the present application.

[0040] Figure 5 A schematic diagram comparing verification results of different models in a virus infection detection method based on machine learning provided in one embodiment of the present application.

[0041] Figure 6 A schematic diagram of the interpretability of an XGBoost model used in a virus infection detection method based on machine learning provided in one embodiment of the present application.

[0042] Figure 7 A schematic diagram of functional modules of a virus infection detection system based on machine learning provided in another embodiment of the present application.

[0043] Figure 8 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0046] In an exemplary embodiment, Figure 1 As shown, a method for detecting virus infection based on machine learning is provided, comprising the following steps:

[0047] A1. Obtain the dual-color fluorescent trajectory data of the virus particles; the dual-color fluorescent trajectory data includes the position information of the fluorescent spots of the virus envelope and the position information of the fluorescent spots of the virus genome. Figure 2 As shown, step A1 includes the following steps:

[0048] A11. Insert biotinylated lipid molecules into the lipid layer of the viral envelope, and then use the specific binding of streptavidin-modified quantum dots to the lipid molecules to fluorescently label the viral envelope of the virus particles.

[0049] A12. Use hybridization chain reaction technology to fluorescently label the viral genome of virus particles.

[0050] A13. Use a total internal reflection fluorescence microscope to perform real-time imaging of the dual-labeled virus particles to obtain several frames of dual-color labeled fluorescence image data of the virus particles.

[0051] A14. Use an image processing program to adjust the coordinates and intensity of the fluorescent spots in each frame of dual-color labeled fluorescent image data, and reconstruct the position information of the fluorescent spots in the virus particles that change over time to obtain dual-color labeled fluorescent trajectory data.

[0052] In step A1 of the present application, a quantitative single virus tracking method with efficient dual-color labeling is used to dynamically and quantitatively study key molecular events in the process of viral membrane fusion. First, the experiment inserts biotinylated lipid molecules (DSPE-PEG2000-biotin) into the lipid layer of the viral envelope, and then uses streptavidin-modified quantum dots (SA-QDs) to specifically bind to biotin to achieve the labeling of the viral envelope. At the same time, the viral genome is labeled using hybridization chain reaction (HCR) technology, which uses HCR amplicons with fluorescent labels (such as Cy5) through a self-assembly process, which not only achieves specific labeling of the viral genome, but also significantly amplifies the fluorescence signal of the genome.

[0053] The double-labeled virus particles were then imaged in real time using a total internal reflection fluorescence (TIRF) microscope, an imaging technique that can capture the fine dynamics of virus particles on the cytoplasmic membrane. The continuously acquired imaging data was then processed, and the coordinates and intensity of the fluorescent spots representing the virus envelope and genome in each frame were adjusted using Fiji-ImageJ software. The trajectory of the virus particles in the time series was reconstructed through the coordinate and intensity information between consecutive frames, and the movement trajectories of the virus envelope and genome could eventually be collected in batches. Over time, if the viral genome enters the cell, the genome and the envelope will separate.

[0054] A2. Extract virus envelope trajectory data and virus genome trajectory data respectively according to the dual-color labeled fluorescence trajectory data; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames.

[0055] Specifically in this embodiment, step A2 reads data from an Excel file containing the trajectory of the viral envelope (green dots) and the genome (red dots) changing over time, and converts it into a Pandas DataFrame format. Then, each trajectory is stored as a NumPy array, and after processing the missing values, these trajectories are saved in two lists, corresponding to the red dot trajectory and the green dot trajectory, respectively. In this embodiment, the viral envelope is limited to a green dot and the viral genome is a red dot, which is only used as an example. In other cases, the two can be marked with other colors.

[0056] A3. Match the virus envelope trajectory data and the virus genome trajectory data according to a preset matching rule to determine paired trajectory data; the paired trajectory data includes the location information of the virus envelope and the location information of the virus genome that meet the preset matching rule. As an exemplary embodiment, the preset matching rule for trajectory pairing is that in a number of consecutive frames, the distance between the virus envelope and the virus genome is less than a preset threshold in more than half of the frames.

[0057] That is, in step A3, the trajectories of the red and green dots are matched by a pairing algorithm. The pairing criterion is: if the distance between the two points of the red and green dots is less than 5 in more than 3 frames in five consecutive frames, then the two sets of data are considered to be paired. In this pairing process, the math module is used to calculate the distance between the two points.

[0058] After the pairing is completed, the matplotlib library is used to plot the change of the pairing trajectory over time. In addition, a graphical user interface (GUI) is created using the tkinter library, which allows users to complete the pairing of trajectories with simple mouse clicks. Finally, the text file containing the pairing trajectory is extracted and an image of the change of the pairing trajectory over time is generated.

[0059] A4. According to the paired trajectory data, the virus release detection result is output using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

[0060] In this embodiment, if Figure 3 As shown, step A4 specifically includes the following steps:

[0061] A41. For the paired trajectory data, the position information of the virus envelope and the position information of the virus genome are interpolated using the cubic spline interpolation method to obtain interpolated trajectory data.

[0062] A42. For the interpolated trajectory data, the position information of the virus envelope and the position information of the virus genome are subtracted and merged to obtain the merged paired trajectory data.

[0063] A43. Input the merged paired trajectory data into the trained virus release detection model to obtain the virus release detection result.

[0064] After successfully pairing the trajectories, use the Pythonos module to read all the paired trajectories and convert them into a three-dimensional NumPy array, where the three dimensions are different paired trajectories that meet the matching rules, different frame numbers, and the corresponding red dot position information and green dot position information of each pair of trajectories in each frame; for the data of known virus release, read its label and convert it into a NumPy array. Then use the interpolate module in the SciPy library to interpolate. The interpolation method is cubic spline interpolation. The coordinates of the trajectory are extended to the interval (0,100), and the starting point of the trajectory is set to t=1, and the end point is set to t=100. Then the interpolated red dot trajectory and green dot trajectory are merged by subtraction and saved in the NumPy array in the form of [X1-X2, Y1-Y2]. Finally, the paired trajectory data after subtraction is input into the model to output the corresponding virus release detection results.

[0065] As an exemplary embodiment, the virus release detection model used is a binary classification model constructed and trained using the XGBoost algorithm.

[0066] Before the model is used, that is, before step A43, the method also includes a process of constructing and training the model, such as Figure 4 As shown, the following steps are included:

[0067] B1. Obtain a model training data set, which includes merged paired trajectory data and virus release results.

[0068] B2. Build an initial virus release detection model based on the XGBoost algorithm.

[0069] B3. Using the model training data set, the initial virus release detection model is trained. Specifically, the initial virus release detection model is trained using the merged paired trajectory data as the model input and the virus release result as the target output of the model to obtain a trained virus release detection model.

[0070] In order to train the classification model, a large number of samples with known virus release conditions were used as model training data sets, in which the virus release conditions of each sample were manually annotated and used as labels. On this basis, the model used the XGBClassifier in the XGBoost algorithm for model training. Specifically, in XGBClassifier, several hyperparameters were set: first, 20 trees were specified to train the model (n_estimators = 20), secondly, the maximum depth of each tree was specified to be 2 (max_depth = 2), and in addition, in order to ensure the repeatability of the experimental results, the random seed was set to 42 (random_state = 42). After these settings, the model was successfully trained and had classification capabilities. For the convenience of subsequent use, the dump function in the pickle module was used to save the trained model in a pkl file, so that the model can be directly loaded and applied in the future without retraining. In practical applications, a large amount of red and green point trajectory data can be directly input into the system. After the two-step processing of pairing and classification prediction in the above steps, the classification results of whether the viral nucleic acid is released can be obtained, thereby achieving the goal of fast and large-scale calculation of the virus infection rate.

[0071] In order to better illustrate the effect of the technology in this application, the experimentally measured data set is divided into a training set and a test set using the train_test_split function in the scikit-learn library. The training set data is used to train the XGBoost model and various other classification models, and the verification and comparison are performed on the test set. The verification and comparison results are shown in Figure 2. Figure 5 As shown, both the XGBoost model and the LightGBM model have good results. In this example, the XGBoost model is selected. Figure 6 For the interpretability of the XGBoost model, we can see that the information that contributes most to trajectory classification comes from information close to the beginning and end of the trajectory (i.e., normalized time close to 0 or 100). This shows that the important criteria for XGBoost trajectory classification come from two points: 1: the distance between the green trajectory and the red trajectory when they appear; 2: the distance between the green trajectory and the red trajectory before they disappear. The XGBoost algorithm has high interpretability, and what it learns is well consistent with human knowledge.

[0072] The above embodiment of the present application provides a method for detecting virus infection based on machine learning. First, the two-color labeled fluorescence trajectory data containing the position information of the fluorescent spots of the virus envelope and the virus genome changing over time is obtained, and the virus envelope trajectory data and the virus genome trajectory data are extracted respectively; the virus envelope trajectory data and the virus genome trajectory data are matched according to the preset matching rules to determine the paired trajectory data; finally, according to the paired trajectory data, a virus release detection model constructed and trained based on the machine learning algorithm is used to obtain the virus release detection result. The scheme provided by the above embodiment of the present application reduces the dependence on human eye judgment through a fully automatic trajectory extraction, pairing and detection scheme, greatly saves manpower and time consumption, and can quickly and batch detect viral nucleic acid release events based on fluorescent trajectory data, thereby quickly and large-scale calculating and evaluating the virus infection rate.

[0073] Based on the same inventive concept, the embodiment of the present application also provides a system for implementing the above-mentioned method for detecting virus infection based on machine learning. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more system embodiments provided below can refer to the limitations of the method for detecting virus infection based on machine learning above, and will not be repeated here.

[0074] As an exemplary embodiment, Figure 7 As shown, a virus infection detection system based on machine learning is provided, including the following modules:

[0075] The dual-color trajectory data acquisition module is used to acquire dual-color labeled fluorescence trajectory data of virus particles; the dual-color labeled fluorescence trajectory data includes the position information of the fluorescent spots of the virus envelope changing with time and the position information of the fluorescent spots of the virus genome changing with time.

[0076] The trajectory data extraction module is used to extract virus envelope trajectory data and virus genome trajectory data respectively according to the dual-color labeled fluorescence trajectory data; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames.

[0077] The trajectory data pairing module is used to match the virus envelope trajectory data and the virus genome trajectory data according to a preset matching rule to determine paired trajectory data; the paired trajectory data includes the position information of the virus envelope and the position information of the virus genome that meet the preset matching rule.

[0078] The virus release detection module is used to output the virus release detection result according to the paired trajectory data using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

[0079] certainly, Figure 7 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different functions. Figure 7 One or at least two components of the system shown.

[0080] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a virus infection detection method based on machine learning is implemented.

[0081] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0082] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0083] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0084] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0085] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0086] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0087] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0088] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0089] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for detecting virus infection based on machine learning, characterized in that: include: Obtaining dual-color labeled fluorescence trajectory data of virus particles; the dual-color labeled fluorescence trajectory data includes position information of the fluorescent spots of the virus envelope changing over time and position information of the fluorescent spots of the virus genome changing over time; According to the two-color labeled fluorescence trajectory data, virus envelope trajectory data and virus genome trajectory data are extracted respectively; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames; Matching the virus envelope trajectory data and the virus genome trajectory data according to a preset matching rule to determine paired trajectory data; The pairing trajectory data includes the position information of the virus envelope and the position information of the virus genome that meet the preset matching rules; According to the paired trajectory data, the virus release detection result is output using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

2. The method for detecting virus infection based on machine learning according to claim 1, characterized in that: Obtain dual-color fluorescent trajectory data of virus particles, including: The biotinylated lipid molecules are inserted into the lipid layer of the viral envelope, and then the viral envelope of the virus particles is fluorescently labeled by utilizing the specific binding of streptavidin-modified quantum dots to the lipid molecules; Using hybridization chain reaction technology, the viral genome of the virus particles is fluorescently labeled; Using a total internal reflection fluorescence microscope to perform real-time imaging of the dual-labeled virus particles, several frames of dual-color labeled fluorescence image data of the virus particles are obtained; An image processing program is used to adjust the coordinates and intensity of the fluorescent spots in each frame of dual-color labeled fluorescent image data, and to reconstruct the position information of the fluorescent spots in the virus particles over time to obtain dual-color labeled fluorescent trajectory data.

3. The method for detecting virus infection based on machine learning according to claim 1, characterized in that: According to the paired trajectory data, the virus release detection results are output using the trained virus release detection model, specifically including: For the paired trajectory data, the position information of the virus envelope and the position information of the virus genome are interpolated using a cubic spline interpolation method to obtain interpolated trajectory data; For the interpolation trajectory data, the position information of the viral envelope and the position information of the viral genome are combined by subtraction to obtain combined paired trajectory data; The combined paired trajectory data is input into the trained virus release detection model to obtain the virus release detection result.

4. The method for detecting virus infection based on machine learning according to claim 3, characterized in that: The virus release detection model is a binary classification model constructed and trained using the XGBoost algorithm.

5. The method for detecting virus infection based on machine learning according to claim 4, characterized in that: Before inputting the combined paired trajectory data into the trained virus release detection model to obtain the virus release detection result, the method further includes: Obtaining a model training data set; the model training data set includes merged paired trajectory data and virus release results; The initial virus release detection model was constructed based on the XGBoost algorithm; The merged paired trajectory data is used as the model input, the virus release result is used as the target output of the model, and the model training data set is used to train the initial virus release detection model to obtain a trained virus release detection model.

6. The method for detecting virus infection based on machine learning according to claim 1, characterized in that: The preset matching rule is that, in a number of consecutive frames, the distance between the virus envelope and the virus genome in more than half of the frames is less than a preset threshold.

7. A virus infection detection system based on machine learning, characterized in that: include: A two-color trajectory data acquisition module is used to acquire two-color labeled fluorescence trajectory data of virus particles; the two-color labeled fluorescence trajectory data includes the position information of the fluorescent spots of the virus envelope changing with time and the position information of the fluorescent spots of the virus genome changing with time; A trajectory data extraction module, used to extract virus envelope trajectory data and virus genome trajectory data respectively according to the dual-color labeled fluorescence trajectory data; the virus envelope trajectory data includes the position information of each virus envelope in a number of consecutive frames; the virus genome trajectory data includes the position information of each virus genome in a number of consecutive frames; A trajectory data pairing module, used to match the virus envelope trajectory data and the virus genome trajectory data according to a preset matching rule to determine paired trajectory data; The pairing trajectory data includes the position information of the virus envelope and the position information of the virus genome that meet the preset matching rules; The virus release detection module is used to output the virus release detection result according to the paired trajectory data using the trained virus release detection model; the virus release prediction model is a binary classification model established based on a machine learning method.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the machine learning-based virus infection detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting virus infection based on machine learning described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for detecting virus infection based on machine learning described in any one of claims 1 to 6 is implemented.

Images