Drug toxicity prediction method, device and equipment and readable storage medium

Abstract

The invention relates to the field and relates to a drug toxicity prediction method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining a drug data set; converting a drug molecular formula included in the drug data set into a corresponding drug molecular map to obtain first drug molecular map information; sending the first drug molecular map information to a trained neural network to obtain second drug molecular map information; calculating according to the first drug molecular map information and the second drug molecular map information to obtain first score information which is used for representing the influence degree of the functional group on the drug toxicity; according to the drug toxicity prediction method and device, the drug molecules are represented through the hypergraph technology, advanced relation features between functional groups in the drug molecules are captured more comprehensively, the structural features of the molecules are captured more accurately, and therefore the accuracy and interpretability of toxicity prediction are improved.

Description

Technical Field

[0001] The present invention relates to a field, and in particular to a drug toxicity prediction method, device, equipment and readable storage medium. Background Art

[0002] In drug development, predicting drug toxicity is a very important part, because toxicity assessment is directly related to drug safety. Existing technologies usually consider the correlation between multiple chemical toxicities and share important features between multiple toxicities during model training to discover potential correlations between different toxicity results, thereby achieving prediction of drug toxicity. However, this method not only lacks interpretability but also tends to ignore the interactions between molecular functional groups, affecting the accuracy and completeness of drug toxicity prediction. Summary of the invention

[0003] The object of the present invention is to provide a drug toxicity prediction method, device, equipment and readable storage medium to improve the above-mentioned problems.

[0004] In order to achieve the above objectives, the present application provides the following technical solutions:

[0005] In one aspect, the present invention provides a method for predicting drug toxicity, the method comprising:

[0006] Get the drug dataset;

[0007] Converting the drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information;

[0008] Sending the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information;

[0009] Calculating according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity;

[0010] The drug toxicity is predicted according to the first score information.

[0011] In a second aspect, an embodiment of the present application provides a drug toxicity prediction device, the device comprising:

[0012] An acquisition module, used to acquire drug datasets;

[0013] A first processing module, configured to convert a drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information;

[0014] A second processing module, used for sending the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information;

[0015] a calculation module, configured to calculate according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity;

[0016] A prediction module is used to predict drug toxicity based on the first score information.

[0017] In a third aspect, an embodiment of the present application provides a drug toxicity prediction device, the device comprising a memory and a processor. The memory is used to store a computer program; the processor is used to implement the steps of the above-mentioned drug toxicity prediction method when executing the computer program.

[0018] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned drug toxicity prediction method are implemented.

[0019] The beneficial effects of the present invention are:

[0020] The present invention converts the drug molecular formula included in the drug data set into the corresponding drug molecular graph to obtain the first drug molecular graph information, and then sends the first drug molecular graph information to the trained neural network to obtain the second drug molecular graph information. The influence of the functional group on the drug toxicity is calculated through the first drug molecular graph information and the second drug molecular graph information, so as to predict the drug toxicity according to the characteristics of different functional group combinations, thereby improving the accuracy and completeness of the toxicity prediction.

[0021] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 Schematic diagram of the drug toxicity prediction method described in an embodiment of the present invention.

[0024] Figure 2 Schematic diagram of the structure of the drug toxicity prediction device described in an embodiment of the present invention.

[0025] Figure 3 Schematic diagram of the structure of the drug toxicity prediction device described in an embodiment of the present invention.

[0026] Labels in the figure: 901, acquisition module; 902, first processing module; 903, second processing module; 904, calculation module; 905, prediction module; 800, drug toxicity prediction device; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0029] Embodiment 1:

[0030] This embodiment provides a method for predicting drug toxicity. It can be understood that a scenario can be set up in this embodiment, for example, a scenario in which toxicity testing of drugs is required.

[0031] See also Figure 1 , the figure shows that the method includes step S1, step S2, step S3, step S4 and step S5.

[0032] Step S1, obtaining a drug data set;

[0033] Step S2, converting the drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information;

[0034] In this step, the drug molecular formulas included in the drug dataset are converted into corresponding drug molecular graphs through the Python RDKit library.

[0035] Step S3, sending the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information;

[0036] In this step, the trained neural network is a generative adversarial neural network, and the second drug molecule graph is similar to the first drug molecule graph, but the second drug molecule graph includes toxic functional group features learned from the generative adversarial neural network.

[0037] The step S3 also includes step S31, step S32, step S33 and step S34, which specifically include:

[0038] Step S31, obtaining a preset first loss function;

[0039] In this step, the preset first loss function is specifically:

[0040]

[0041] In the above formula, L1 represents the preset first loss function, n represents the total number of functional group categories in all sample sets, and n d is the number of samples, K is a weight factor, x is a drug sample, G(·) is the generator, D(·) is the discriminator, E(·) is the encoder, ||·|| 2 is the squared Euclidean distance, and the purpose is to make samples from the same category closer in the latent space and samples from different categories farther away. The encoder finally obtains the unique features of drug molecules, so that it can identify the functional group information in drug data.

[0042] Step S32, training the encoder on the drug data set according to the preset first loss function to obtain a trained encoder;

[0043] In this step, the encoder is used to train on the drug dataset. The main function of the encoder is to project the real drug molecule data into the latent space.

[0044] Step S33, using the trained encoder to generate a latent vector of a real drug molecule;

[0045] In this step, the latent vector includes random noise generated from a Gaussian distribution.

[0046] Step S34: training the neural network according to the latent vector to obtain a trained neural network.

[0047] The step S34 also includes step S341, step S342, step S343 and step S344, which specifically include:

[0048] Step S341: using the potential vector, obtaining a first poison molecule graph, wherein the first poison molecule graph includes poison molecules generated by a generator;

[0049] Step S342, obtaining poison molecule information from a preset database;

[0050] In this step, a specific implementation method is to obtain targeted real poison molecule smilies formulas (such as hepatotoxicity) from the PubChem database.

[0051] Step S343, converting the poison molecule information into a second poison molecule graph;

[0052] Step S344: sending the first poison molecule graph and the second poison molecule graph to a discriminator and training them using a preset second loss function to obtain the trained neural network.

[0053] In this step, the preset second loss function is specifically:

[0054]

[0055] In the above formula, L2 represents the preset second loss function, λ is the gradient penalty coefficient, represents the gradient, ε is a random number sampled from a uniform distribution [0, 1), ||·||2 is the sum of squared gradients, and x represent the features included in the first poison molecule graph and the second poison molecule graph, respectively. During the training process, the discriminator will distinguish the difference between the poison molecule features generated by the generator and the real poison molecule features.

[0056] Step S4, performing calculation according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity;

[0057] In this step, the first score information can intuitively demonstrate the importance or influence of the functional group in drug toxicity and provide a certain degree of interpretability.

[0058] The step S4 also includes step S41 and step S42, which specifically include:

[0059] Step S41, preprocessing the first drug molecule graph information and the second drug molecule graph information respectively to obtain first functional group graph information and second functional group graph information;

[0060] In this step, the preprocessing is to segment the molecules using BRICS and Harper graph reduction methods to obtain functional groups. It should be noted that in this step, the functional groups obtained by segmenting the same molecule using multiple segmentation methods are all used as the functional group set of the molecule, providing different functional group combination perspectives to improve the model's generalization ability for different types of drug molecules.

[0061] Step S42: Calculate the number of times each functional group appears in each sample according to the first functional group map information and the second functional group map information to obtain first score information.

[0062] In this step, the specific calculation process of the first score information is:

[0063]

[0064] Among them, S f(i) represents the toxicity impact score of each functional group, i.e., the first score information, n represents the total number of functional group categories in the sample, α and β are learnable parameters, and N (Sample) represents the number of sample pairs. N represents the number of the i-th functional group in the second drug molecule graph in the j-th sample pair, j (x(i)) represents the number of the i-th functional group in the first drug molecule graph in the j-th sample pair.

[0065] Step S5: predicting drug toxicity according to the first score information.

[0066] In this embodiment, since the prior art usually considers the correlation between multiple chemical toxicities, important features between multiple toxicities are shared during the model training process to discover the potential correlation between different toxicity results, thereby achieving drug toxicity prediction. However, this method cannot clearly understand how the model uses input features for prediction, and which features play an important role in the prediction, and lacks interpretability. Therefore, the present invention provides interpretability for the model by calculating the first score information to characterize the importance or influence of the functional group in drug toxicity.

[0067] The step S5 also includes step S51, step S52, step S53, step S54 and step S55, which specifically include:

[0068] Step S51, determining atoms constituting functional groups in the drug molecule graph according to the first drug molecule graph information, and obtaining atomic information;

[0069] Step S52, determining nodes in the hypergraph according to the atomic information, and obtaining node position information;

[0070] Step S53, determining the hyperedges between nodes according to the first score information to obtain hyperedge information;

[0071] Step S54, constructing a hypergraph according to the node position information and the hyperedge information to obtain at least one sub-hypergraph information, wherein one sub-hypergraph information corresponds to one functional group;

[0072] Step S55: predicting drug toxicity based on the sub-hypergraph information.

[0073] In this embodiment, each atom constituting a functional group in the first drug molecule graph information is connected by a hyperedge to construct a sub-hypergraph information, and each sub-hypergraph information corresponds to a functional group. The weight of the hyperedge is the first score information, and these scores reflect the importance or influence of the functional group in the toxicity of the drug molecule. The first drug molecule graph information is processed by the hypergraph technology, and the functional groups in the drug molecule are displayed in the form of a network, and the toxic effects of these functional groups are expressed by weights, so that the characteristics of different functional group combinations can be better obtained, and the high-level connection characteristics between the functional groups in the drug molecule can be more comprehensively captured. The structural characteristics of the molecule are more accurately captured, thereby improving the accuracy and completeness of toxicity prediction.

[0074] The step S55 also includes step S551, step S552, step S553, step S554, step S555 and step S556, which specifically include:

[0075] Step S551: Send each sub-hypergraph information to the graph attention network to obtain at least two feature vector information;

[0076] In this step, the graph neural network is used to obtain the feature vector information corresponding to each sub-hypergraph information.

[0077] Step S552: Calculate the attention score between the two feature vector information to obtain second score information;

[0078] In this step, the calculation process of the second score information is specifically as follows:

[0079] e ij =LeakyRuLU(a T [W*h i ,W*h j ])

[0080] Among them, e ij Indicates the second score information, h i and h jare the feature vector information corresponding to the i-th and j-th sub-hypergraph information, respectively, T Represents the weight parameters of the neural network, W is the weight matrix, and LeakyReLU is an activation function.

[0081] Step S553: ​​Calculate the attention coefficient according to the second score information to obtain attention coefficient information;

[0082] In this step, calculating the attention coefficient is a technical solution well known to those skilled in the art, so it will not be described here in detail.

[0083] Step S554, updating the feature vector information according to the attention coefficient information to obtain an updated feature vector;

[0084] In this step, the attention coefficients of the two sub-hypergraph information can reflect the toxic impact score of the two functional group combinations. Then, the attention coefficient information is used to further update the feature vector of the sub-hypergraph information to obtain an updated feature vector.

[0085] Step S555, pooling the updated feature vector to obtain a pooled feature vector;

[0086] Step S556: Send the pooled feature vector to the prediction layer to obtain a prediction result.

[0087] Embodiment 2:

[0088] like Figure 2 As shown, this embodiment provides a drug toxicity prediction device, which includes an acquisition module 901, a first processing module 902, a second processing module 903, a calculation module 904 and a prediction module 905, which specifically include:

[0089] An acquisition module 901 is used to acquire a drug data set;

[0090] A first processing module 902 is used to convert the drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information;

[0091] The second processing module 903 is used to send the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information;

[0092] A calculation module 904, configured to perform calculation according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity;

[0093] The prediction module 905 is used to predict the drug toxicity according to the first score information.

[0094] In a specific implementation of the present disclosure, the second processing module further includes a first acquisition unit, a first training unit, a first processing unit and a second processing unit, which specifically include:

[0095] A first acquisition unit, used to acquire a preset first loss function;

[0096] A first training unit, used for training the encoder on the drug data set according to the preset first loss function to obtain a trained encoder;

[0097] A first processing unit, configured to generate a latent vector of a real drug molecule using the trained encoder;

[0098] The second processing unit is used to train the neural network according to the potential vector to obtain a trained neural network.

[0099] In a specific implementation of the present disclosure, the second processing unit further includes a third processing unit, a second acquisition unit, a fourth processing unit and a second training unit, which specifically include:

[0100] A third processing unit is used to obtain a first poison molecule graph using the potential vector, wherein the first poison molecule graph includes poison molecules generated by the generator;

[0101] A second acquisition unit is used to acquire poison molecule information from a preset database;

[0102] A fourth processing unit, used for converting the poison molecule information into a second poison molecule map;

[0103] The second training unit is used to send the first poison molecule graph and the second poison molecule graph to the discriminator and train them using a preset second loss function to obtain the trained neural network.

[0104] In a specific implementation of the present disclosure, the calculation module further includes a fifth processing unit and a first calculation unit, which specifically include:

[0105] a fifth processing unit, configured to preprocess the first drug molecule graph information and the second drug molecule graph information respectively to obtain first functional group graph information and second functional group graph information;

[0106] The first calculation unit is used to calculate the number of times each functional group appears in each sample according to the first functional group map information and the second functional group map information to obtain first score information.

[0107] In a specific implementation of the present disclosure, the prediction module further includes a sixth processing unit, a seventh processing unit, an eighth processing unit, a ninth processing unit and a tenth processing unit, which specifically include:

[0108] a sixth processing unit, configured to determine atoms constituting functional groups in the drug molecule graph according to the first drug molecule graph information, and obtain atomic information;

[0109] A seventh processing unit, configured to determine nodes in the hypergraph according to the atomic information, and obtain node position information;

[0110] An eighth processing unit, configured to determine a hyperedge between nodes according to the first score information to obtain hyperedge information;

[0111] A ninth processing unit, configured to construct a hypergraph according to the node position information and the hyperedge information, and obtain at least one sub-hypergraph information, wherein one sub-hypergraph information corresponds to one functional group;

[0112] The tenth processing unit is used to predict drug toxicity according to the sub-hypergraph information.

[0113] In a specific implementation of the present disclosure, the tenth processing unit also includes an eleventh processing unit, a second computing unit, a third computing unit, a twelfth processing unit, a pooling unit, and a prediction unit, which specifically include:

[0114] an eleventh processing unit, configured to send each sub-hypergraph information to a graph attention network to obtain at least two feature vector information;

[0115] A second calculation unit is used to calculate the attention score between the two feature vector information to obtain second score information;

[0116] A third calculation unit, configured to calculate an attention coefficient according to the second score information to obtain attention coefficient information;

[0117] A twelfth processing unit, used to update the feature vector information according to the attention coefficient information to obtain an updated feature vector;

[0118] A pooling unit, used for pooling the updated feature vector to obtain a pooled feature vector;

[0119] The prediction unit is used to send the pooled feature vector to the prediction layer to obtain a prediction result.

[0120] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0121] Embodiment 3:

[0122] Corresponding to the above method embodiment, this embodiment further provides a drug toxicity prediction device. The drug toxicity prediction device described below and the drug toxicity prediction method described above can refer to each other.

[0123] Figure 3 FIG. 8 is a block diagram of a drug toxicity prediction device 800 according to an exemplary embodiment. Figure 3 As shown, the drug toxicity prediction device 800 may include: a processor 801 and a memory 802. The drug toxicity prediction device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0124] The processor 801 is used to control the overall operation of the drug toxicity prediction device 800 to complete all or part of the steps in the above-mentioned drug toxicity prediction method. The memory 802 is used to store various types of data to support the operation of the drug toxicity prediction device 800, and these data may include, for example, instructions for any application or method operating on the drug toxicity prediction device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component can include a microphone, and the microphone is used to receive external audio signals. The received audio signal can be further stored in the memory 802 or sent through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the drug toxicity prediction device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0125] In an exemplary embodiment, the drug toxicity prediction device 800 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned drug toxicity prediction method.

[0126] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned drug toxicity prediction method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the drug toxicity prediction device 800 to complete the above-mentioned drug toxicity prediction method.

[0127] Embodiment 4:

[0128] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the drug toxicity prediction method described above can refer to each other.

[0129] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the drug toxicity prediction method of the above method embodiment.

[0130] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0131] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0132] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for predicting drug toxicity, characterized in that: include: Get the drug dataset; Converting the drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information; Sending the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information; Calculating according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity; The drug toxicity is predicted according to the first score information.

2. The drug toxicity prediction method according to claim 1, characterized in that: Sending the first drug molecule graph information to the trained neural network includes: Obtain a preset first loss function; Training the encoder on the drug dataset according to the preset first loss function to obtain a trained encoder; generating a latent vector of a real drug molecule using the trained encoder; The neural network is trained according to the potential vector to obtain a trained neural network.

3. The drug toxicity prediction method according to claim 2, characterized in that: Training a neural network according to the latent vector includes: Using the latent vector, a first poison molecule graph is obtained, wherein the first poison molecule graph includes poison molecules generated by a generator; Obtain poison molecule information from a preset database; Converting the poison molecule information into a second poison molecule graph; The first poison molecule graph and the second poison molecule graph are sent to a discriminator and trained using a preset second loss function to obtain the trained neural network.

4. The drug toxicity prediction method according to claim 1, characterized in that: Calculating according to the first drug molecule graph information and the second drug molecule graph information includes: Preprocessing the first drug molecule graph information and the second drug molecule graph information respectively to obtain first functional group graph information and second functional group graph information; The number of times each functional group appears in each sample is calculated according to the first functional group map information and the second functional group map information to obtain first score information.

5. A drug toxicity prediction device, characterized in that: include: An acquisition module, used to acquire drug datasets; A first processing module, configured to convert a drug molecular formula included in the drug data set into a corresponding drug molecular graph to obtain first drug molecular graph information; A second processing module, used for sending the first drug molecule graph information to the trained neural network to obtain second drug molecule graph information; a calculation module, configured to calculate according to the first drug molecule graph information and the second drug molecule graph information to obtain first score information, wherein the first score information is used to indicate the degree of influence of the functional group on the drug toxicity; A prediction module is used to predict drug toxicity based on the first score information.

6. The drug toxicity prediction device according to claim 5, characterized in that: The second processing module comprises: A first acquisition unit, used to acquire a preset first loss function; A first training unit, used for training the encoder on the drug data set according to the preset first loss function to obtain a trained encoder; A first processing unit, configured to generate a latent vector of a real drug molecule using the trained encoder; The second processing unit is used to train the neural network according to the potential vector to obtain a trained neural network.

7. The drug toxicity prediction device according to claim 6, characterized in that: The second processing unit comprises: A third processing unit is used to obtain a first poison molecule graph using the potential vector, wherein the first poison molecule graph includes poison molecules generated by the generator; A second acquisition unit is used to acquire poison molecule information from a preset database; A fourth processing unit, used for converting the poison molecule information into a second poison molecule map; The second training unit is used to send the first poison molecule graph and the second poison molecule graph to the discriminator and train them using a preset second loss function to obtain the trained neural network.

8. The drug toxicity prediction device according to claim 5, characterized in that: The computing module comprises: a fifth processing unit, configured to preprocess the first drug molecule graph information and the second drug molecule graph information respectively to obtain first functional group graph information and second functional group graph information; The first calculation unit is used to calculate the number of times each functional group appears in each sample according to the first functional group map information and the second functional group map information to obtain first score information.

9. A drug toxicity prediction device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the drug toxicity prediction method according to any one of claims 1 to 4 when executing the computer program.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the drug toxicity prediction method according to any one of claims 1 to 4 are implemented.

Images