Intelligent generation method and system of novel antibacterial peptide sequence and computer equipment

Abstract

The invention relates to a novel antibacterial peptide sequence intelligent generation method and system and computer equipment, and the method comprises the steps: inputting a first antibacterial peptide sequence data set containing multi-sequence alignment into a pre-trained generator, and outputting a novel first candidate antibacterial peptide sequence set according to a preset condition; inputting the first candidate antibacterial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antibacterial peptide sequence set belonging to the antibacterial peptide sequence; inputting the second candidate antibacterial peptide sequence set into a pre-trained predictor to predict the antibacterial activity of each second candidate antibacterial peptide sequence in the second candidate antibacterial peptide sequence set; and based on the antibacterial activity of each second candidate antibacterial peptide sequence, selecting at least one optimal target antibacterial peptide sequence from each second candidate antibacterial peptide sequence set. According to the application, the generator can be used for capturing sequence diversity and generating a wider-spectrum antibacterial peptide sequence, and through cascade optimization of the discriminator and the predictor, the generated antibacterial peptide sequence has higher antibacterial activity.

Description

Technical Field

[0001] The present application relates to the fields of bioinformatics and artificial intelligence, and in particular to a method, system and computer device for intelligently generating a novel antimicrobial peptide sequence. Background Art

[0002] Antimicrobial peptides (AMPs) have become the focus of antimicrobial drug research due to their broad-spectrum antimicrobial activity. However, due to the short length, high diversity, and inherent structural flexibility of AMP sequences, these characteristics pose a major challenge to traditional design methods.

[0003] Existing methods for generating antimicrobial peptide sequences often fail to effectively capture the complex relationship between sequence and antimicrobial activity and are unable to generate antimicrobial peptide sequences that meet the requirements. Summary of the invention

[0004] Based on this, it is necessary to provide a method, system, computer device and storage medium for intelligently generating a novel antimicrobial peptide sequence in response to the above technical problems.

[0005] In a first aspect, an embodiment of the present invention provides a method for intelligently generating a novel antimicrobial peptide sequence, the method comprising: Inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; Inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; Inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; Based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences, at least one optimal target antimicrobial peptide sequence is selected from each of the second candidate antimicrobial peptide sequence sets.

[0006] In some embodiments, the preset condition is used to set the length range of each first candidate antimicrobial peptide sequence in the first candidate antimicrobial peptide sequence set.

[0007] In some embodiments, the generator is based on a diffusion model of the Transformer architecture and is trained using the OpenFold dataset.

[0008] In some embodiments, the discriminator is based on an XGBoost binary classifier and is trained using a second antimicrobial peptide sequence dataset and a non-antimicrobial peptide sequence dataset.

[0009] In some embodiments, the predictor is based on an LSTM regression model and is trained using the third antimicrobial peptide sequence dataset and the non-antimicrobial peptide sequence dataset.

[0010] In some embodiments, the method further comprises: The first candidate antimicrobial peptide sequence set is subjected to physicochemical property calculation and filtering, and the first candidate antimicrobial peptide sequences with charge>0 and hydrophilic amino acid ratio between 40% and 70% are retained.

[0011] In some embodiments, the selecting at least one optimal target antimicrobial peptide sequence from each set of second candidate antimicrobial peptide sequences based on the antimicrobial activity of each second candidate antimicrobial peptide sequence comprises: In each of the second candidate antimicrobial peptide sequence sets, at least one second candidate antimicrobial peptide sequence having an antimicrobial activity greater than a set threshold is selected as the optimal target antimicrobial peptide sequence.

[0012] In a second aspect, an embodiment of the present invention provides a system for intelligently generating novel antimicrobial peptide sequences, the system comprising: A sequence generation module, used for inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; A sequence discrimination module, used for inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; A prediction module, used for inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; A sequence selection module is used to select at least one optimal target antimicrobial peptide sequence from each set of the second candidate antimicrobial peptide sequences based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences.

[0013] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps described in the first aspect.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the processor implements the steps described in the first aspect when executing the computer program.

[0015] Compared with the prior art, the above method, system, computer device and storage medium can capture sequence diversity and generate a broader spectrum of antimicrobial peptide sequences by using a pre-trained generator. Through the cascade optimization of the discriminator and the predictor, the generated antimicrobial peptide sequences have higher antimicrobial activity. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the process of the intelligent generation method of novel antimicrobial peptide sequences in one embodiment of the present application; Figure 2 This is a schematic diagram of model training in one embodiment of the present application; Figure 3 This is a schematic diagram of the structure of the LSTM model in one embodiment of the present application; Figure 4 This is a schematic diagram of the module connections of the intelligent generation system of the novel antimicrobial peptide sequence in one embodiment of the present application; Figure 5 It is a schematic diagram of the structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present invention. For ordinary technicians in this field, the present invention can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0018] As shown in the present invention and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0019] Although the present invention makes various references to certain modules in the system according to an embodiment of the present invention, any number of different modules can be used and run on a computing device and / or processor. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0020] It should be understood that when a unit or module is described as being "connected," "coupled" to other units, modules, or blocks, it may refer to being directly connected or coupled, or communicating with other units, modules, or blocks, or there may be intermediate units, modules, or blocks, unless the context clearly indicates otherwise. The term "and / or" as used herein may include any and all combinations of one or more of the associated listed items.

[0021] like Figure 1 As shown, the embodiment of the present invention provides a method for intelligently generating a novel antimicrobial peptide sequence, comprising the following steps: S102: inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; S104: inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; S106: inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; S108: Based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences, at least one optimal target antimicrobial peptide sequence is selected from each of the second candidate antimicrobial peptide sequence sets.

[0022] Based on the above steps S102 to S108, the pre-trained generator can capture sequence diversity and generate a broader spectrum of antimicrobial peptide sequences. Through the cascade optimization of the discriminator and the predictor, at least one optimal target antimicrobial peptide sequence is selected from each set of second candidate antimicrobial peptide sequences, and the target antimicrobial peptide sequence has higher antimicrobial activity.

[0023] The generator is based on the pre-trained model Evodiff, specifically the model Evodiff-MSA_OA_DM_MAXSUB trained on the OpenFold dataset, and uses a diffusion model with a 100M parameter MSATransformer architecture. The generator takes the first antimicrobial peptide sequence dataset (AMP-MSA dataset) as input and can output a large number of new first candidate antimicrobial peptide sequence sets that meet the preset conditions based on the input. The diffusion model is pre-trained on the OpenFold database to further enhance its generation capability.

[0024] The preset conditions are used, for example, to set the length range of each first candidate antimicrobial peptide sequence in the first candidate antimicrobial peptide sequence set. The length range of the generated sequence can be customized by the preset conditions, and precise control of sequence generation can be provided by specifying the length range of the input sequence. For example, it is defined as 15-35 amino acids to take into account the functional characteristics and synthesis cost of the antimicrobial peptide sequence.

[0025] like Figure 2As shown, a second antimicrobial peptide sequence dataset (AMP dataset) was obtained from multiple public antimicrobial peptide databases (APD, DADP, DBAASP, DRAMP, YADAMP, dbAMP, etc.), and the entries containing the minimum inhibitory concentration (MIC) values ​​in the second antimicrobial peptide sequence dataset were filtered, cleaned, and merged to obtain a third antimicrobial peptide sequence dataset containing MIC values ​​(AMP-MIC dataset). Based on the sequences in the second antimicrobial peptide sequence dataset, the search tool UniClust30 was used to retrieve the first antimicrobial peptide sequence dataset (AMP-MSA dataset) containing multiple sequence alignments (MSA).

[0026] In some embodiments, the antimicrobial peptide data is subjected to deduplication, filtering, and standardization processing in sequence, and the obtained first antimicrobial peptide sequence is a non-redundant sequence.

[0027] In some embodiments, the generated first candidate antimicrobial peptide sequence set is screened for physicochemical properties. The physicochemical properties are calculated and filtered to retain sequences with charge > 0 and a hydrophilic amino acid ratio between 40% and 70%. The threshold can be adjusted as needed.

[0028] The discriminator is based on the XGBoost binary classifier and is trained using the second antimicrobial peptide sequence dataset (AMP dataset) and the non-antimicrobial peptide sequence dataset (NonAMP dataset).

[0029] The second antimicrobial peptide sequence dataset (AMP dataset) was also obtained from the public antimicrobial peptide database as a positive sample, and the non-antimicrobial peptide sequence dataset (NonAMP dataset) was obtained from the Uniport database as a negative sample. These databases all contain public antimicrobial peptide sequences and their corresponding antimicrobial peptide activities.

[0030] The non-antimicrobial peptide sequence dataset (NonAMP dataset) retrieved antimicrobial peptide sequences of 5-65 aa from the UniProt database, and removed those containing the following 17 keywords: antimicrobial, antibiotic, antibacterial, antiviral, antifungal, antimalarial, antiparasitic, anti-protist, anticancer, defense, defensin, cathelicidin, histatin, bacteriocin, microbicidal, fungicideandtoxin.

[0031] After obtaining the second antimicrobial peptide sequence dataset (AMP dataset) and the non-antimicrobial peptide sequence dataset (NonAMP dataset), the PseKRAAC and QSOrder methods were used for feature extraction. The sequences in the second antimicrobial peptide sequence dataset (AMP dataset) were marked as 1, and the sequences in the non-antimicrobial peptide sequence dataset (NonAMP dataset) were marked as 0. Ten-fold cross validation was used as the parameter adjustment method, and the extracted features were used for training to obtain the discriminator.

[0032] The pre-trained discriminator can further determine whether the first candidate antimicrobial peptide sequence in the first candidate antimicrobial peptide sequence set is an antimicrobial peptide sequence.

[0033] The predictor is based on the LSTM regression model and is trained using the third antimicrobial peptide sequence dataset (AMP-MIC dataset) and the non-antimicrobial peptide sequence dataset (NonAMP dataset).

[0034] The third antimicrobial peptide sequence data set is the entries containing effective MIC values ​​in the second antimicrobial peptide data set.

[0035] Before training, the pre-trained protein language model ESM-2 was used to extract features from the first antimicrobial peptide sequence dataset (AMP-MSA dataset) and the non-antimicrobial peptide sequence dataset (NonAMP dataset).

[0036] The protein language model is specifically ESM2-t36-3B, which is based on the Transformer architecture and can capture the complex structural dependencies of antimicrobial peptide sequences.

[0037] The Long Short-Term Memory (LSTM) network model is a special recursive neural network that solves the gradient vanishing problem of ordinary neural networks in processing long sequence data by introducing memory units. The LSTM regression model can capture dependencies in long sequences and is a deep learning model that is very suitable for processing time series.

[0038] like Figure 3 As shown in the figure, the LSTM regression model includes two LSTM layers, a Dropout layer and an output layer constructed in sequence; the hidden state dimension of the LSTM layer is set to 256; the output layer is used to convert the hidden state dimension of the LSTM layer into a logMIC value for characterizing the activity of antimicrobial peptides. Adding a Dropout layer (Dropout rate is 0.7) after the LSTM layer can reduce the risk of model overfitting.

[0039] During the training process, the first antimicrobial peptide sequence data set and the non-antimicrobial peptide sequence data set are used as sample training sets and divided into a training set, a validation set and a test set according to a ratio; the antimicrobial peptide activity of the sample training set is used as a label, and the training set, the validation set and the test set are used to train the initial LSTM model using the first high-dimensional embedding feature to obtain a trained LSTM regression model.

[0040] During the training process, the time back propagation algorithm is used to obtain the gradient, the square error loss function is used to calculate the error, and the optimizer is used to optimize the model parameters based on the gradient and the error.

[0041] Specifically, the L2 loss function is used to calculate the error between the predicted value and the true logMIC value, the time back propagation algorithm is used to calculate the gradient, and the gradient explosion problem is avoided by gradient clipping, and the Adam optimizer is used to update the parameters based on the gradient and the error.

[0042] The predictor outputs the predicted logMIC value, which evaluates the antimicrobial peptide activity of the antimicrobial peptide sequence in the target bacterial genus (such as Escherichia coli or Staphylococcus aureus). The lower the logMIC value, the stronger the antimicrobial peptide activity.

[0043] In step S108, at least one second candidate antimicrobial peptide sequence having an antimicrobial activity greater than a set threshold is selected from each of the second candidate antimicrobial peptide sequence sets as the optimal target antimicrobial peptide sequence.

[0044] The sequence generation method of the present application is used to perform experimental verification on the generated target antimicrobial peptide sequence, and the antimicrobial peptide sequence generated for Escherichia coli is effective against E. coli The k88 effects are as follows: Table 1

[0045] According to Table 1, the target antimicrobial peptide sequence generated above has high antimicrobial activity.

[0046] It should be understood that, although the various steps in the above flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the above flowchart may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0047] In one embodiment, if Figure 4 As shown, the present invention provides a novel intelligent generation system for antimicrobial peptide sequences, the system comprising: A sequence generation module 402 is used to input a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and output a novel first candidate antimicrobial peptide sequence set according to preset conditions; A sequence discrimination module 404 is used to input the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; A prediction module 406, configured to input the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; The sequence selection module 408 is configured to select at least one optimal target antimicrobial peptide sequence from each of the second candidate antimicrobial peptide sequence sets based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences.

[0048] In some embodiments, the preset condition is used to set the length range of each first candidate antimicrobial peptide sequence in the first candidate antimicrobial peptide sequence set.

[0049] In some embodiments, the generator is based on a diffusion model of the Transformer architecture and is trained using the OpenFold dataset.

[0050] In some embodiments, the discriminator is based on an XGBoost binary classifier and is trained using a second antimicrobial peptide sequence dataset and a non-antimicrobial peptide sequence dataset.

[0051] In some embodiments, the predictor is based on an LSTM regression model and is trained using the third antimicrobial peptide sequence dataset and the non-antimicrobial peptide sequence dataset.

[0052] In some embodiments, the system further comprises: calculating and filtering the first candidate antimicrobial peptide sequence set through physicochemical property calculation, and retaining first candidate antimicrobial peptide sequences with charge>0 and hydrophilic amino acid ratio between 40% and 70%.

[0053] In some embodiments, the sequence selection module selects at least one second candidate antimicrobial peptide sequence having an antimicrobial activity greater than a set threshold from each of the second candidate antimicrobial peptide sequence sets as the optimal target antimicrobial peptide sequence.

[0054] The specific definition of the intelligent generation system of novel antimicrobial peptide sequences can be found in the definition of the automatic tool calling method in the above text, which will not be repeated here. Each module in the above-mentioned intelligent generation system of novel antimicrobial peptide sequences can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0055] In one embodiment, the present invention provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. 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 database of the computer device is used to store action detection data. The network 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, the steps in any of the above-mentioned embodiments of the intelligent generation method of the novel antimicrobial peptide sequence are implemented.

[0056] The method comprises: Inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; Inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; Inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; Based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences, at least one optimal target antimicrobial peptide sequence is selected from each of the second candidate antimicrobial peptide sequence sets.

[0057] Those skilled in the art will understand that Figure 5 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.

[0058] In one embodiment, the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in any of the above-mentioned intelligent generation method embodiments of novel antimicrobial peptide sequences are implemented.

[0059] The method comprises: Inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; Inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; Inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; Based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences, at least one optimal target antimicrobial peptide sequence is selected from each of the second candidate antimicrobial peptide sequence sets.

[0060] 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 memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0061] The technical features of the above embodiments may be combined arbitrarily. 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.

[0062] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for intelligently generating novel antimicrobial peptide sequences, characterized in that: The method comprises: Inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; Inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; Inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; Based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences, at least one optimal target antimicrobial peptide sequence is selected from each of the second candidate antimicrobial peptide sequence sets.

2. The method according to claim 1, characterized in that The preset condition is used to set the length range of each first candidate antimicrobial peptide sequence in the first candidate antimicrobial peptide sequence set.

3. The method according to claim 2, characterized in that The generator is based on a diffusion model of the Transformer architecture and is trained using the OpenFold dataset.

4. The method according to claim 1, characterized in that The discriminator is based on an XGBoost binary classifier and is trained using a second antimicrobial peptide sequence data set and a non-antimicrobial peptide sequence data set.

5. The method according to claim 1, characterized in that: The predictor is based on an LSTM regression model and is trained using a third antimicrobial peptide sequence data set and a non-antimicrobial peptide sequence data set.

6. The method according to claim 1, characterized in that The method further comprises: The first candidate antimicrobial peptide sequence set is subjected to physicochemical property calculation and filtering, and the first candidate antimicrobial peptide sequences with charge>0 and hydrophilic amino acid ratio between 40% and 70% are retained.

7. The method according to claim 1, characterized in that The selecting at least one optimal target antimicrobial peptide sequence from each set of the second candidate antimicrobial peptide sequences based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences comprises: In each of the second candidate antimicrobial peptide sequence sets, at least one second candidate antimicrobial peptide sequence having an antimicrobial activity greater than a set threshold is selected as the optimal target antimicrobial peptide sequence.

8. An intelligent generation system for novel antimicrobial peptide sequences, characterized in that: The system comprises: A sequence generation module, used for inputting a first antimicrobial peptide sequence data set including a multiple sequence alignment into a pre-trained generator, and outputting a novel first candidate antimicrobial peptide sequence set according to preset conditions; A sequence discrimination module, used for inputting the first candidate antimicrobial peptide sequence set into a pre-trained discriminator to discriminate a second candidate antimicrobial peptide sequence set belonging to the antimicrobial peptide sequence; A prediction module, used for inputting the second candidate antimicrobial peptide sequence set into a pre-trained predictor to predict the antimicrobial activity of each second candidate antimicrobial peptide sequence in the second candidate antimicrobial peptide sequence set; A sequence selection module is used to select at least one optimal target antimicrobial peptide sequence from each set of the second candidate antimicrobial peptide sequences based on the antimicrobial activity of each of the second candidate antimicrobial peptide sequences.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images