AIDS molecular propagation network monitoring method and device based on deep learning classification

Abstract

The invention discloses an AIDS molecular propagation network monitoring method and device based on deep learning classification, and the method comprises the steps: collecting AIDS virus gene sequence data and social attribute data of AIDS patients, calculating a genetic distance and a genetic distance matrix, and converting the genetic distance and the genetic distance matrix into AIDS image data; then constructing an AIDS neural network model to output propagation network characteristics, obtaining virus data classification of different AIDS virus gene sequences according to the propagation network characteristics, selecting a target classification result, obtaining a propagation path and a propagation speed, and further knowing the propagation dynamic state of the AIDS virus. Predicting the transmission risk of an HIV gene sequence corresponding to the HIV image data; according to the embodiment of the invention, the transmission of the HIV not only depends on gene sequence data, but also provides more accurate and comprehensive transmission risk collection and distribution in combination with social attributes such as social relationship and gender of the propagator, and can identify higher transmission risks, so that prevention measures can be implemented in a targeted manner.

Description

Technical Field

[0001] The present invention relates to the field of medical virus transmission technology, and in particular to a method and device for monitoring AIDS molecular transmission network based on deep learning classification. Background Art

[0002] At present, in the study of AIDS virus transmission, for example, the patent with publication number CN113517030A discloses a gene sequence representation learning method based on a virus transmission network, including encoding the virus gene sequence, and inputting the sequence code into a hidden layer of a specific length to obtain a dimensionality reduction embedding; extracting a set of neighbor nodes with a high correlation with each node in the virus transmission network, and obtaining the topological information of the node from the virus transmission network; using forward LSTM and reverse LSTM to aggregate the graph context node information of the node respectively; using the attention mechanism to aggregate the sequence information of its neighbor nodes to obtain a new representation of the gene sequence of the target node; using graph context loss as the loss function to perform model training and target optimization on the gene sequence representation model. Another example is the patent with publication number CN112687344A, which discloses a human adenovirus molecular typing and tracing method and system based on metagenomics, including constructing a human adenovirus typing database including a typing hierarchical classification system, obtaining the classification unit and complete genome / marker gene sequence of human adenovirus, determining the complete genome / marker gene sequence of each classification unit of human adenovirus, obtaining the metagenomic sequencing raw data of clinical samples, pre-processing it to obtain target metagenomic data, analyzing the target metagenomic data using a preset metagenomic sequencing data double comparison annotation system, determining whether the clinical sample has a target human adenovirus classification unit, and when it is determined that the clinical sample has a target human adenovirus classification unit, determining the target association feature corresponding to the target human adenovirus classification unit, and generating a human adenovirus typing identification report. It is capable of typing and identifying the classification unit (subtype / genotype) of human adenovirus in clinical infection samples with low virus content. Summary of the invention

[0003] In response to the above-mentioned defects, an embodiment of the present invention discloses a method and device for monitoring the molecular transmission network of AIDS based on deep learning classification, which combines social attributes and viral gene sequences to provide more accurate virus transmission monitoring.

[0004] The first aspect of the embodiment of the present invention discloses a method for monitoring AIDS molecular transmission network based on deep learning classification, comprising:

[0005] Collect HIV gene sequence data and social attribute data of AIDS patients, calculate genetic distance based on HIV gene sequence data and social attribute data, construct a genetic distance matrix, and convert the genetic distance matrix into AIDS image data;

[0006] An AIDS neural network model is constructed using the AIDS image data as input data and the propagation network features as output data; the propagation network features include virus data classification;

[0007] Inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and selecting the virus data classification with the largest probability value from the at least one virus data classification as the target classification result;

[0008] The transmission risk of the HIV gene sequence corresponding to the AIDS image data is predicted according to the target classification result.

[0009] As an optional implementation, in the first aspect of the embodiment of the present invention, calculating the genetic distance based on HIV gene sequence data and social attribute data includes:

[0010] Dividing all HIV gene sequences in the HIV gene sequence data into a number of comparison groups, each comparison group including two different HIV gene sequences;

[0011] Based on the differences between each comparison group or continuous HIV gene sequence;

[0012] The genetic distance is calculated based on the difference.

[0013] As an optional implementation manner, in the first aspect of the embodiment of the present invention, it also includes:

[0014] When the difference is greater than a set threshold, the gene sequence region with a difference greater than the set threshold is identified according to a preset algorithm, and the gene sequence region is removed.

[0015] As an optional implementation manner, in the first aspect of the embodiment of the present invention, constructing a genetic distance matrix includes:

[0016] A genetic distance matrix is ​​constructed based on the genetic distance and social attribute data.

[0017] As an optional implementation, in the first aspect of the embodiment of the present invention, inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data includes:

[0018] Inputting new AIDS image data into the AIDS neural network model to obtain the grayscale value of the AIDS image data;

[0019] Obtaining a genetic distance matrix of the AIDS image data according to a preset gray value and genetic distance correspondence relationship, and extracting image features of the AIDS image data;

[0020] At least one virus data classification is output based on the genetic distance matrix and the image features.

[0021] As an optional implementation, in the first aspect of the embodiment of the present invention, predicting the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result includes:

[0022] Calculate the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result;

[0023] The transmission risk of HIV gene sequences is predicted based on the virus transmission pathway and the virus.

[0024] As an optional implementation, in the first aspect of the embodiment of the present invention, the step of calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the HIV image data according to the target classification result includes:

[0025] Obtaining corresponding propagation rate and removal rate according to the target classification result;

[0026] Calculate the ratio between the propagation rate and the removal rate as the basic reproduction number of the virus, and calculate the propagation speed according to the basic reproduction number of the virus;

[0027] The virus propagation path corresponding to the target classification result is obtained according to the preset virus classification and propagation path relationship table.

[0028] A second aspect of an embodiment of the present invention discloses an AIDS molecular transmission network monitoring device based on deep learning classification, comprising:

[0029] Data collection module: used to collect HIV gene sequence data and social attribute data of AIDS patients, calculate genetic distance based on HIV gene sequence data and social attribute data, construct genetic distance matrix, and convert into AIDS image data according to the genetic distance matrix;

[0030] Model building module: used to build an AIDS neural network model using the AIDS image data as input data and the propagation network features as output data; the propagation network features include virus data classification;

[0031] Result output module: used for inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and selecting the virus data classification with the largest probability value from the at least one virus data classification as the target classification result;

[0032] Risk prediction module: used to predict the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result.

[0033] As an optional implementation, in the second aspect of the embodiment of the present invention, the genetic distance is calculated based on HIV gene sequence data and social attribute data, including:

[0034] Dividing all HIV gene sequences in the HIV gene sequence data into a number of comparison groups, each comparison group including two different HIV gene sequences;

[0035] Based on the differences between each comparison group or continuous HIV gene sequence;

[0036] The genetic distance is calculated based on the difference.

[0037] As an optional implementation manner, in the second aspect of the embodiment of the present invention, it also includes:

[0038] When the difference is greater than a set threshold, the gene sequence region with a difference greater than the set threshold is identified according to a preset algorithm, and the gene sequence region is removed.

[0039] As an optional implementation, in the second aspect of the embodiment of the present invention, constructing a genetic distance matrix includes:

[0040] A genetic distance matrix is ​​constructed based on the genetic distance and social attribute data.

[0041] As an optional implementation, in the second aspect of the embodiment of the present invention, inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data includes:

[0042] Inputting new AIDS image data into the AIDS neural network model to obtain the grayscale value of the AIDS image data;

[0043] Obtaining a genetic distance matrix of the AIDS image data according to a preset gray value and genetic distance correspondence relationship, and extracting image features of the AIDS image data;

[0044] At least one virus data classification is output based on the genetic distance matrix and the image features.

[0045] As an optional implementation, in the second aspect of the embodiment of the present invention, predicting the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result includes:

[0046] Calculate the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result;

[0047] The transmission risk of HIV gene sequences is predicted based on the virus transmission pathway and the virus.

[0048] As an optional implementation, in the second aspect of the embodiment of the present invention, the step of calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the HIV image data according to the target classification result includes:

[0049] Obtaining corresponding propagation rate and removal rate according to the target classification result;

[0050] Calculate the ratio between the propagation rate and the removal rate as the basic reproduction number of the virus, and calculate the propagation speed according to the basic reproduction number of the virus;

[0051] The virus propagation path corresponding to the target classification result is obtained according to the preset virus classification and propagation path relationship table.

[0052] The third aspect of an embodiment of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the method for monitoring the molecular transmission network of AIDS based on deep learning classification disclosed in the first aspect of an embodiment of the present invention.

[0053] A fourth aspect of an embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method for monitoring the molecular transmission network of AIDS based on deep learning classification disclosed in the first aspect of an embodiment of the present invention.

[0054] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0055] In the embodiment of the present invention, HIV gene sequence data and social attribute data of AIDS patients are first collected, genetic distance and genetic distance matrix are calculated through the collected data, and the genetic distance matrix is ​​converted into AIDS image data. Then, an AIDS neural network model is constructed to output propagation network characteristics, and virus data classification of different HIV gene sequences is obtained according to the propagation network characteristics. The data classification with the highest probability is selected as the target classification result, and the propagation path and propagation speed are obtained in combination with the target classification result, so as to understand the propagation dynamics of HIV and predict the propagation risk of HIV gene sequence corresponding to the AIDS image data. The embodiment not only relies on gene sequence data for the propagation of HIV, but also combines the social attributes such as social relations and gender of the communicator to provide a more accurate and comprehensive distribution of propagation risks. Through comprehensive prediction and analysis, it is possible to identify which social groups or individuals may face higher propagation risks, so as to implement preventive measures in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 It is a flow chart of a method for monitoring AIDS molecular transmission network based on deep learning classification disclosed in an embodiment of the present invention;

[0058] Figure 2 It is a structural schematic diagram of an AIDS molecular transmission network monitoring device based on deep learning classification provided by an embodiment of the present invention;

[0059] Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] It should be noted that the terms "first", "second", "third", "fourth", etc. in the specification and claims of the present invention are used to distinguish different objects rather than to describe a specific order. The terms "including" and "having" in the embodiments of the present invention and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] The embodiment of the present invention discloses a method, device, electronic device and storage medium for monitoring AIDS molecular transmission network based on deep learning classification. In the embodiment, HIV gene sequence data and social attribute data of AIDS patients are first collected, genetic distance and genetic distance matrix are calculated through the collected data, and the genetic distance matrix is ​​converted into AIDS image data. Then, an AIDS neural network model is constructed to output transmission network characteristics, and virus data classification of different HIV gene sequences is obtained according to the transmission network characteristics, and the data classification with the highest probability is selected as the target classification result. In combination with the target classification result, the transmission path and transmission speed are obtained, so as to understand the transmission dynamics of HIV and predict the transmission risk of HIV gene sequence corresponding to the AIDS image data. The embodiment not only relies on gene sequence data for the transmission of HIV, but also combines the social attributes such as social relations and gender of the transmitter to provide a more accurate and comprehensive transmission risk distribution. Through comprehensive prediction and analysis, it can identify which social groups or individuals may face higher transmission risks, so as to implement preventive measures in a targeted manner.

[0063] Embodiment 1

[0064] See also Figure 1 , Figure 1 It is a flow chart of a method for monitoring the molecular transmission network of AIDS based on deep learning classification disclosed in an embodiment of the present invention. Among them, the execution subject of the method described in the embodiment of the present invention is an execution subject composed of software and / or hardware, and the execution subject can receive relevant information by wired or / and wireless means, and can send certain instructions. Of course, it can also have certain processing functions and storage functions. The execution subject can control multiple devices, such as a remote physical server or cloud server and related software, or it can be a local host or server and related software that performs related operations on a device placed somewhere. In some scenarios, multiple storage devices can also be controlled, and the storage devices can be placed in the same place or different places as the devices. For example Figure 1 As shown, the AIDS molecular transmission network monitoring method based on deep learning classification includes the following steps:

[0065] 101. Collect HIV gene sequence data and social attribute data of AIDS patients, calculate genetic distance based on HIV gene sequence data and social attribute data, construct a genetic distance matrix, and convert the genetic distance matrix into AIDS image data.

[0066] First, it is necessary to collect HIV gene sequence data and social attribute data of AIDS patients. Gene sequence data provides genetic information of the virus itself, while social attribute data (such as age, gender, sexual behavior, etc.) provides the social background of virus transmission. Based on these data, the genetic distance between different virus samples is calculated, which helps to understand the degree of variation and relationship between viruses. The calculated genetic distance is organized into a matrix form to facilitate subsequent data processing and conversion. The genetic distance matrix is ​​converted into image data form, which is a data type that deep learning models can handle. This conversion may involve mapping matrix data to image pixel values, or encoding genetic information into image features by other means.

[0067] First, the HIV gene sequence data collected can be collected through multiple research institutions, laboratories and databases around the world. These data come from various sources, including but not limited to public bioinformatics databases, clinical case reports and data shared by scientific research projects. At the same time, the embodiment collects social attribute information of relevant infected individuals as much as possible, including gender, sexual behavior pattern, age, education level, place of residence, occupation, etc.

[0068] After the data is collected, the obtained HIV gene sequences are compared in detail, and the genetic distances between different strains are calculated to form a genetic distance matrix. This includes dividing all HIV gene sequences in the HIV gene sequence data into several comparison groups, each of which includes two different HIV gene sequences; based on the difference between each comparison group or between HIV gene sequences; and calculating the genetic distance according to the difference.

[0069] The purpose of dividing the alignment groups is to compare and analyze the variation and similarity between HIV gene sequences. All HIV gene sequences need to be divided into several alignment groups. Each alignment group contains two different HIV gene sequences. These alignment groups can be randomly selected or selected based on a certain strategy (such as geographical location, transmission chain, time, etc.) to more comprehensively reflect the natural variation and social transmission characteristics of the virus. Use sequence alignment algorithms (such as ClustalW, MAFFT, etc.) to align the two gene sequences, find out the similarities and differences between them, record the mismatched sites (i.e., different nucleotide positions in the sequence) and insertion / deletion (indel) events in the alignment results, and calculate the total number or ratio of these differences as the difference degree. Common genetic distance metrics include nucleotide difference rate (p-distance), Kimura two-parameter distance, Tamura-Nei distance, etc. Substitute the difference degree into the selected genetic distance metric formula to calculate the genetic distance of each alignment group. It should be noted that, before calculating the genetic distance, the embodiment needs to pre-process the gene sequence data, including removing low-quality sequences, filling missing data, etc., to ensure that the two gene sequences in the comparison group have similar lengths, or at least to compare their common parts to avoid the influence of length differences on the genetic distance calculation.

[0070] The method further includes: when the difference is greater than a set threshold, identifying a gene sequence region whose difference is greater than the set threshold according to a preset algorithm, and eliminating the gene sequence region.

[0071] Embodiment sets reasonable difference degree threshold value to distinguish between normal compilation and highly compiled gene sequence regions, which can be fixed values ​​or calculated based on statistical methods (such as median, mean plus a certain multiple of standard deviation). The comparison result generated using the sequence alignment algorithm is marked, all difference sites are marked, the total number or ratio of difference sites in each comparison group is calculated, and compared with the set threshold value. If the difference degree of a certain region (such as a continuous nucleotide sequence) in a certain comparison group exceeds the threshold value, the region is marked as a highly variable region. For each comparison group, if highly variable regions are identified, these regions can be selected to be excluded, and only relatively conservative parts are retained for the calculation of genetic distance. It can be a simple deletion of the nucleotide sequence of the highly variable region, or it can be a smoothing process using a certain algorithm (such as a sliding window method) to reduce its impact on the overall genetic distance.

[0072] In the embodiment, a genetic distance matrix is ​​constructed, including: constructing a genetic distance matrix according to the genetic distance and social attribute data. The genetic distance between each pair of gene sequences is obtained by comparing the gene sequences and calculating the difference, and social attribute data related to AIDS patients, such as age, gender, geographic location, sexual behavior, etc., are collected. These data may exist in the form of a table or a database, and it is determined how many different HIV gene sequences need to be included in the matrix. According to the number of gene sequences, an N×N matrix is ​​constructed, where N is the number of gene sequences. Each row and column of the matrix represents a gene sequence, and each element in the matrix represents the genetic distance between the two gene sequences of the corresponding row and column. According to the genetic distance data calculated before, the corresponding position of the matrix is ​​filled with the corresponding genetic distance value. For a symmetric matrix (i.e., the genetic distance is bidirectional), only the upper or lower triangle part needs to be filled, and then copied to the symmetric position. Social attribute data can be used as additional information to enhance the practicality of the matrix. A table can be provided next to or below the matrix to list the social attribute data corresponding to each gene sequence. If the social attribute data is quantifiable (such as age can be converted into a numerical value), it can be added to the matrix as an additional dimension to form a higher dimensional matrix.

[0073] 102. Construct an AIDS neural network model using the AIDS image data as input data and the propagation network features as output data; the propagation network features include virus data classification.

[0074] This step builds a neural network model suitable for processing image data, such as a convolutional neural network (CNN), which can automatically extract features from images and use them for classification tasks.

[0075] 103. Input new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and select the virus data classification with the largest probability value from the at least one virus data classification as the target classification result.

[0076] Input the new AIDS image data into the trained neural network model, and the model will output at least one virus data classification, which represents the model's prediction results for the new data. The classification with the largest probability value is selected from multiple classification results as the target classification result, which is the most likely virus type considered by the model.

[0077] Specifically, input new AIDS image data to the AIDS neural network model to obtain the grayscale value of the AIDS image data; obtain the genetic distance matrix of the AIDS image data according to the preset grayscale value and genetic distance correspondence, and extract the image features of the AIDS image data; output at least one virus data classification based on the genetic distance matrix and image features. Perform image data preprocessing on the input new AIDS image data, including denoising, size adjustment, etc., to ensure that it meets the input requirements of the neural network model. Preset a grayscale value and genetic distance correspondence table or function, calculate the genetic distance corresponding to each pixel according to the extracted grayscale value using the preset correspondence, and organize the calculated genetic distance into a matrix form, in which each element represents the genetic distance between a certain position in the image and other positions. Extract useful image features from the grayscale image and the genetic distance matrix, which may include texture features, shape features, spatial relationship features, etc. Use image processing technology (such as edge detection, filtering, segmentation, etc.) and machine learning algorithms (such as PCA, LDA, etc.) to extract features. The extracted image features are used as input data and input into the trained AIDS neural network model. The neural network model performs calculations based on the input feature data and outputs at least one virus data classification result. If the model outputs multiple classification results, the most likely classification can be selected as the target classification result based on the classification probability value or other indicators.

[0078] 104. Predict the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result.

[0079] According to the target classification results, the transmission risk of HIV gene sequences is predicted. Different virus types may have different transmission capabilities and mutation rates, so the classification results are crucial for predicting transmission risks. Combined with social attribute data and virus type information, the transmission risk of the virus in specific social groups can be further evaluated to provide a basis for formulating prevention and control strategies.

[0080] The embodiment combines multiple technical means such as deep learning, image processing and genetic distance calculation to provide a new and efficient method for predicting the transmission risk of HIV.

[0081] Specifically, this step includes calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result; and predicting the transmission risk of the HIV gene sequence according to the virus transmission path and virus. The step of calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result includes: obtaining the corresponding transmission rate and removal rate according to the target classification result; calculating the ratio between the transmission rate and the removal rate as the basic reproduction number of the virus, and calculating the transmission speed according to the basic reproduction number of the virus; and obtaining the virus transmission path corresponding to the target classification result according to a preset relationship table between virus classification and transmission path.

[0082] The basic reproduction number (R0) of the virus refers to how many people an infected person can infect on average during his or her infectious period without any intervention measures. It is an important indicator for evaluating the virus's ability to spread. The basic reproduction number (R0) of the virus can be obtained by calculating the ratio between the transmission rate and the removal rate. The speed of virus transmission is usually related to the R0 value. In a simple model, the transmission speed can be regarded as a function of the R0 value, which may also be affected by other factors, such as population mobility, contact frequency, etc.

[0083] Embodiment 2

[0084] See also Figure 2 , Figure 2 Schematic diagram of the structure of a device for monitoring the molecular transmission network of AIDS based on deep learning classification disclosed in an embodiment of the present invention. Figure 2 As shown, the AIDS molecular transmission network monitoring device based on deep learning classification may include: a data collection module 201, a model construction module 202, a result output module 203 and a risk prediction module 204, wherein the data collection module 201 is used to collect HIV gene sequence data and social attribute data of AIDS patients, and calculate the genetic distance based on the HIV gene sequence data and the social attribute data, construct a genetic distance matrix, and convert it into AIDS image data according to the genetic distance matrix; the model construction module 202 is used to construct an AIDS neural network model with the AIDS image data as input data and the transmission network characteristics as output data; the transmission network characteristics include virus data classification; the result output module 203 is used to input new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and select the virus data classification with the largest probability value from the at least one virus data classification as the target classification result; the risk prediction module 204 is used to predict the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result.

[0085] In the data collection module 201, the genetic distance is calculated based on the HIV gene sequence data and the social attribute data, including: dividing all HIV gene sequences in the HIV gene sequence data into a plurality of comparison groups, each comparison group including two different HIV gene sequences; based on the difference between each comparison group or between the HIV gene sequences; and calculating the genetic distance according to the difference. When the difference is greater than a set threshold, the gene sequence region with a difference greater than the set threshold is identified according to a preset algorithm, and the gene sequence region is eliminated.

[0086] The construction of the genetic distance matrix includes: constructing a genetic distance matrix according to the genetic distance and social attribute data.

[0087] In the result output module 203, new AIDS image data is input into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, including: inputting new AIDS image data into the AIDS neural network model to obtain the gray value of the AIDS image data; obtaining the genetic distance matrix of the AIDS image data according to a preset gray value and genetic distance correspondence relationship, and extracting the image features of the AIDS image data; and outputting at least one virus data classification based on the genetic distance matrix and the image features.

[0088] 8. In the risk prediction module 204, the transmission risk of the HIV gene sequence corresponding to the HIV image data is predicted based on the target classification result, including: calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the HIV image data based on the target classification result; predicting the transmission risk of the HIV gene sequence based on the virus transmission path and virus. Further, the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the HIV image data based on the target classification result include: obtaining the corresponding transmission rate and removal rate based on the target classification result; calculating the ratio between the transmission rate and the removal rate as the basic reproduction number of the virus, and calculating the transmission speed based on the basic reproduction number of the virus; obtaining the virus transmission path corresponding to the target classification result according to a preset virus classification and transmission path relationship table.

[0089] Embodiment 3

[0090] See also Figure 3 , Figure 3 Schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device may be a computer, a server, etc. Of course, in certain circumstances, it may also be a smart device such as a mobile phone, a tablet computer, a monitoring terminal, and an image acquisition device with processing functions. Figure 3 As shown, the electronic device may include:

[0091] A memory 301 storing executable program codes;

[0092] a processor 302 coupled to the memory 301;

[0093] The processor 302 calls the executable program code stored in the memory 501 to execute part or all of the steps in the method for monitoring the molecular transmission network of AIDS based on deep learning classification in the first embodiment.

[0094] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute part or all of the steps in the method for monitoring AIDS molecular transmission network based on deep learning classification in embodiment one.

[0095] The embodiment of the present invention also discloses a computer program product, wherein when the computer program product runs on a computer, the computer executes part or all of the steps in the method for monitoring AIDS molecular transmission network based on deep learning classification in embodiment one.

[0096] An embodiment of the present invention also discloses an application publishing platform, wherein the application publishing platform is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer executes part or all of the steps in the method for monitoring the molecular transmission network of AIDS based on deep learning classification in Example 1.

[0097] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the processes does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0099] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0100] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several requests for a computer device (which can be a personal computer, a server or a network device, etc., specifically a processor in a computer device) to perform some or all of the steps of the method described in each embodiment of the present invention.

[0101] In the embodiments provided by the present invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0102] A person of ordinary skill in the art can understand that some or all of the steps in the various methods of the embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0103] The above is a detailed introduction to the AIDS molecular transmission network monitoring method, device, electronic device and storage medium based on deep learning classification disclosed in the embodiments of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for monitoring AIDS molecular transmission network based on deep learning classification, characterized in that: include: Collect HIV gene sequence data and social attribute data of AIDS patients, calculate genetic distance based on HIV gene sequence data and social attribute data, construct a genetic distance matrix, and convert the genetic distance matrix into AIDS image data; An AIDS neural network model is constructed using the AIDS image data as input data and the propagation network features as output data; the propagation network features include virus data classification; Inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and selecting the virus data classification with the largest probability value from the at least one virus data classification as the target classification result; The transmission risk of the HIV gene sequence corresponding to the AIDS image data is predicted according to the target classification result.

2. The method for monitoring AIDS molecular transmission network according to claim 1, characterized in that: Genetic distance is calculated based on HIV gene sequence data and social attribute data, including: Dividing all HIV gene sequences in the HIV gene sequence data into a number of comparison groups, each comparison group including two different HIV gene sequences; Based on the differences between each comparison group or continuous HIV gene sequence; The genetic distance is calculated based on the difference.

3. The method for monitoring AIDS molecular transmission network according to claim 2, characterized in that: Also includes: When the difference is greater than a set threshold, the gene sequence region with a difference greater than the set threshold is identified according to a preset algorithm, and the gene sequence region is removed.

4. The method for monitoring AIDS molecular transmission network according to claim 2, characterized in that: The construction of the genetic distance matrix comprises: A genetic distance matrix is ​​constructed based on the genetic distance and social attribute data.

5. The method for monitoring AIDS molecular transmission network according to claim 1, characterized in that: Inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data includes: Inputting new AIDS image data into the AIDS neural network model to obtain the grayscale value of the AIDS image data; Obtaining a genetic distance matrix of the AIDS image data according to a preset gray value and genetic distance correspondence relationship, and extracting image features of the AIDS image data; At least one virus data classification is output based on the genetic distance matrix and the image features.

6. The method for monitoring AIDS molecular transmission network according to claim 1, characterized in that: Predicting the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result includes: Calculate the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result; The transmission risk of HIV gene sequences is predicted based on the virus transmission pathway and the virus.

7. The method for monitoring AIDS molecular transmission network according to claim 6, characterized in that: The step of calculating the virus transmission path and virus transmission speed of the HIV gene sequence corresponding to the AIDS image data according to the target classification result includes: Obtaining corresponding propagation rate and removal rate according to the target classification result; Calculate the ratio between the propagation rate and the removal rate as the basic reproduction number of the virus, and calculate the propagation speed according to the basic reproduction number of the virus; The virus propagation path corresponding to the target classification result is obtained according to the preset virus classification and propagation path relationship table.

8. A device for monitoring AIDS molecular transmission network based on deep learning classification, characterized in that: include: Data collection module: used to collect HIV gene sequence data and social attribute data of AIDS patients, calculate genetic distance based on HIV gene sequence data and social attribute data, construct genetic distance matrix, and convert into AIDS image data according to the genetic distance matrix; Model building module: used to build an AIDS neural network model using the AIDS image data as input data and the propagation network features as output data; the propagation network features include virus data classification; Result output module: used for inputting new AIDS image data into the AIDS neural network model to obtain at least one virus data classification corresponding to the AIDS image data, and selecting the virus data classification with the largest probability value from the at least one virus data classification as the target classification result; Risk prediction module: used to predict the transmission risk of the HIV gene sequence corresponding to the AIDS image data according to the target classification result.

9. An electronic device, characterized in that: include: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the AIDS molecular transmission network monitoring method based on deep learning classification as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the method for monitoring the molecular transmission network of AIDS based on deep learning classification as described in any one of claims 1 to 7.

Images