Neural network model training method and device, electronic equipment and storage medium

Abstract

The invention provides a network model training method and device, electronic equipment and a storage medium, and relates to the technical field of biological information. According to the specific implementation scheme, the method comprises the following steps: acquiring a data set, wherein data in the data set comprises drug target protein sequence data and non-drug target protein sequence data; according to the data set, based on an initial ensemble learning model, a target neural network model with the function of predicting the drug target protein is obtained, the initial ensemble learning model comprises a feature extraction network, at least two base learners and a total classifier, and the feature extraction network comprises an ESM2 model. According to the method and the device, the target neural network model with the function of predicting the drug target protein can be obtained through training on the basis of the integrated learning model comprising the EMS2 model, so that whether the protein corresponding to the protein sequence is the drug target protein or not can be rapidly, efficiently and accurately predicted according to the neural network model, and the prediction cost is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of bioinformatics, and in particular to a neural network model training method, device, electronic device and storage medium. Background Art

[0002] Drug targets refer to biological macromolecules in the body that have pharmacological functions and can be acted upon by drugs, such as certain proteins and nucleic acids. The design of a modern drug often requires the effective identification of drug targets from complex biological systems.

[0003] At present, the prediction of drug targets is usually obtained through biological experimental methods and computational model methods. The biological experimental method determines the drug target through a large number of biological experiments, and the computational model method predicts the drug target through a pre-designed mathematical model.

[0004] However, in the prior art, biological experimental methods have high accuracy, but are time-consuming, costly and inefficient; computational model methods have high efficiency, but their accuracy and reliability are limited by the design of the mathematical model and the quality of the data. Summary of the invention

[0005] The present disclosure provides a neural network model training method, device, electronic device and storage medium, which can obtain a target neural network model with the function of predicting drug target proteins based on an integrated learning model including an EMS2 model, so that whether the protein corresponding to a protein sequence is a drug target protein can be predicted quickly, efficiently and accurately based on the neural network model, thereby reducing the prediction cost.

[0006] According to a first aspect of the present disclosure, a neural network model training method is provided, comprising:

[0007] Acquire a data set, wherein the data in the data set include drug target protein sequence data and non-drug target protein sequence data; obtain a target neural network model with a function of predicting drug target proteins based on the data set and an initial ensemble learning model, wherein the initial ensemble learning model includes a feature extraction network, at least two base learners and a total classifier, wherein the feature extraction network includes an ESM2 model.

[0008] Furthermore, before obtaining a target neural network model having a function of predicting drug target proteins based on the initial integrated learning model according to the data set, the method further comprises:

[0009] Dividing the data set into a training set and a test set;

[0010] The method of obtaining a target neural network model having a function of predicting drug target proteins based on the data set and the initial integrated learning model comprises:

[0011] Based on the initial integrated learning model according to the training set, a target neural network model with the function of predicting drug target proteins is obtained.

[0012] Furthermore, dividing the data set into a training set and a test set includes:

[0013] According to a preset ratio, the data set is divided into the training set and the test set.

[0014] Furthermore, the base learner includes at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM.

[0015] Furthermore, the training set includes a first training set and a second training set, the total classifier is a learning-based support vector machine SVM, and the target neural network model having the function of predicting drug target proteins is obtained based on the initial ensemble learning model according to the training set, including:

[0016] According to the first training set, the at least two base learners are optimized, and according to the second training set, the overall classifier is optimized to obtain a target neural network model having the function of predicting drug target proteins.

[0017] Further, the at least two base learners are optimized according to the first training set, and the total classifier is optimized according to the second training set to obtain a target neural network model having a function of predicting drug target proteins, including:

[0018] According to the EMS2 model, feature extraction is performed on the first training set to obtain a first feature, and according to the amino acid composition ACC algorithm, feature extraction is performed on the first training set to obtain a second feature; according to the first feature and the second feature, fusion and normalization processing are performed to obtain a third feature; the at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; based on the classification results corresponding to the at least two base learners and the first training set, the loss functions corresponding to the at least two base learners are obtained; based on the loss functions corresponding to the at least two base learners, the at least two base learners are respectively optimized to obtain the optimized at least two base learners; according to the second training set and the classification results of the optimized at least two base learners for the second training set, the total classifier is optimized to obtain the optimized total trainer; based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer, the target neural network is obtained.

[0019] According to a second aspect of the present disclosure, a neural network model training device is provided, the device comprising:

[0020] The data processing module is used to obtain a data set, wherein the data in the data set includes drug target protein sequence data and non-drug target protein sequence data.

[0021] A model training module is used to obtain a target neural network model with the function of predicting drug target proteins based on the data set and an initial integrated learning model, wherein the initial integrated learning model includes a feature extraction network, at least two base learners and a total classifier, wherein the feature extraction network includes an ESM2 model.

[0022] Furthermore, the data processing module is also used to divide the data set into a training set and a test set; the model training module is specifically used to obtain a target neural network model with the function of predicting drug target proteins based on the initial ensemble learning model according to the training set.

[0023] Furthermore, the data processing module is specifically used to: divide the data set into the training set and the test set according to a preset ratio.

[0024] Furthermore, the base learner includes at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM.

[0025] Furthermore, the training set includes a first training set and a second training set, the total classifier is a learning-based support vector machine SVM, and the model training module is specifically used to: optimize the at least two base learners according to the first training set, and optimize the total classifier according to the second training set to obtain a target neural network model with the function of predicting drug target proteins.

[0026] Further, the model training module is specifically used to: perform feature extraction on the first training set according to the EMS2 model to obtain a first feature, perform feature extraction on the first training set according to the amino acid composition ACC algorithm to obtain a second feature; perform fusion and normalization processing on the first feature and the second feature to obtain a third feature; the at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; based on the classification results corresponding to the at least two base learners and the first training set, obtain the loss function corresponding to the at least two base learners; based on the loss function corresponding to the at least two base learners, respectively optimize the at least two base learners to obtain the optimized at least two base learners; according to the second training set and the classification results of the optimized at least two base learners for the second training set, optimize the total classifier to obtain the optimized total trainer; based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer, obtain the target neural network.

[0027] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method of the first aspect.

[0028] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute the method according to the first aspect.

[0029] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to the first aspect.

[0030] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0032] Figure 1 A flowchart of a neural network model training method provided in an embodiment of the present disclosure;

[0033] Figure 2 Another flowchart of the neural network model training method provided in the embodiment of the present disclosure;

[0034] Figure 3 A schematic diagram of the composition of a neural network model training device provided in an embodiment of the present disclosure;

[0035] Figure 4 A schematic diagram of the composition of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0036] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0037] It should be understood that in the embodiments of the present disclosure, the character " / " generally indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", etc. are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0038] Drug targets refer to biological macromolecules in the body that have pharmacological functions and can be acted upon by drugs, such as certain proteins and nucleic acids. The design of a modern drug often requires the effective identification of drug targets from complex biological systems.

[0039] At present, the prediction of drug targets is usually obtained through biological experimental methods and computational model methods. The biological experimental method determines the drug target through a large number of biological experiments, and the computational model method predicts the drug target through a pre-designed mathematical model.

[0040] However, in the prior art, biological experimental methods have high accuracy, but are time-consuming, costly and inefficient; computational model methods have high efficiency, but their accuracy and reliability are limited by the design of the mathematical model and the quality of the data.

[0041] Against this background technology, the present disclosure provides a neural network model training method, which can obtain a target neural network model with the function of predicting drug target proteins based on an integrated learning model including an EMS2 model, so that the protein corresponding to the protein sequence can be quickly, efficiently and accurately predicted based on the neural network model, thereby reducing the prediction cost.

[0042] The execution subject of the neural network model training method provided by the embodiment of the present disclosure may be a computer or a server, or may also be other electronic devices with data processing capabilities; or, the execution subject of the method may also be a processor (e.g., a central processing unit (CPU)) in the above electronic device; or, the execution subject of the method may also be an application (application, APP) installed in the above electronic device that can implement the functions described in the method; or, the execution subject of the method may be a functional module or unit in the above electronic device that has the functions described in the method, etc. There is no limitation on the execution subject of the method herein.

[0043] In some embodiments, the server may be a single server, or a server cluster composed of multiple servers. In some implementations, the server cluster may also be a distributed cluster. The present disclosure does not limit the specific implementation of the server.

[0044] The neural network model training method is exemplarily described below with reference to the accompanying drawings.

[0045] Figure 1 The flowchart of the neural network model training method provided by the embodiment of the present disclosure is as follows. Figure 1 As shown, the method may include:

[0046] S101. Obtain a data set.

[0047] The data in the data set include drug target protein sequence data and non-drug target protein sequence data.

[0048] For example, the sequence data of proteins that can be used as drug targets and the sequence data of proteins that cannot be used as drug targets can be collected from existing protein sequence databases to form a data set. For example, protein sequence data can be obtained from public protein sequence databases such as DRUGBANK and UniProt to obtain a data set. It is understood that the protein sequence data in the data set can come from various species (such as plants, animals, humans, and microorganisms), which will not be repeated here.

[0049] As a possible implementation, after obtaining the data set, the data in the data set may be preprocessed to ensure the quantity and high quality of the data in the data set.

[0050] For example, protein sequence data with sequence length exceeding a preset number (such as 512) can be removed to improve the training efficiency of the model; the data set can be deduplicated to remove repeated redundant data in the data set; data enhancement and processing methods can also be used, such as expanding the data set by generating pseudo samples, adding noise, etc., to increase the amount of data; oversampling, undersampling, SMOTE and other techniques can also be used to balance the ratio of drug target protein sequence data to non-drug target protein sequence data in the data set, thereby avoiding the neural network model obtained by training the data set from being biased towards the majority class and improving the accuracy of the model.

[0051] S102. Obtaining a target neural network model having a function of predicting drug target proteins according to the data set based on an initial ensemble learning model, wherein the initial ensemble learning model includes a feature extraction network, at least two base learners and an overall classifier.

[0052] Wherein, the feature extraction network includes an ESM2 model.

[0053] Exemplarily, the feature extraction network may also include other models and methods for extracting protein feature information. For example, other Transformer-based models (such as BERT, ProtBert, etc.) may also be used to extract protein sequence features.

[0054] For example, when performing feature extraction, more protein feature information can be combined, such as secondary structure, tertiary structure and functional domain information, as well as features based on the physicochemical properties of amino acids, such as hydrophobicity, charge and volume, etc., to improve the accuracy of the target neural network model obtained.

[0055] Exemplarily, the ESM2 model may be an esm2_t33_630M_UR30D model, or may be other models in the ESM2 series models, which is not limited here.

[0056] Optionally, the base learner may include at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM. It is understood that the neural network NN may be any artificial neural network, such as a neural network GNN, an ESM2 model, etc., without limitation.

[0057] Exemplarily, the base learner may also be other machine learning classifiers such as random forests and decision trees, which are not limited thereto.

[0058] Exemplarily, the overall classifier may be a support vector machine (SVM) that has pre-set corresponding weights for each base learner, or it may be a learning-based weighted voting or multi-model fusion technology, such as a hybrid expert system or a soft voting mechanism, etc., which will not be described in detail here.

[0059] Exemplarily, based on the above-mentioned data set, an integrated learning model including the ESM2 model can be used as an initial model for training (for example, training using supervised learning related techniques) to obtain a target neural network model.

[0060] Through the neural network model training method disclosed in the present invention, a target neural network model with the function of predicting drug target proteins can be obtained based on the integrated learning model training including the EMS2 model, so that whether the protein corresponding to the protein sequence is a drug target protein can be quickly, efficiently and accurately predicted based on the neural network model, thereby reducing the prediction cost.

[0061] In some possible embodiments, before obtaining a target neural network model having a function of predicting drug target proteins based on the initial integrated learning model according to the data set, the method may further include:

[0062] The data set is divided into a training set and a test set.

[0063] For example, the data set can be randomly split into two data sets, a training set and a test set. Of course, in some possible implementations, the data set can also be split in other ways to obtain a training set and a test set, which is not limited here. For example, the data set can be evenly split according to the species corresponding to the protein sequences in the data set, so that the obtained training set and test set both have protein sequences from various species.

[0064] Optionally, dividing the data set into a training set and a test set may include:

[0065] According to a preset ratio, the data set is divided into the training set and the test set.

[0066] For example, the preset ratio may be 9:1, that is, 90% of the data in the data set may be used as a training set, and the remaining 10% of the data may be used as a test set. Of course, in actual applications, other preset ratios may be set according to actual needs or requirements, and are not limited here.

[0067] In this way, it is convenient to subsequently train the neural network model according to the training set, and to test the trained neural network model according to the test set, thereby improving the training quality of the target neural network model.

[0068] Correspondingly, obtaining a target neural network model having a function of predicting drug target proteins based on the data set and the initial integrated learning model may include:

[0069] Based on the initial integrated learning model according to the training set, a target neural network model with the function of predicting drug target proteins is obtained.

[0070] In some possible implementations, the training set includes a first training set and a second training set, the total classifier is a learning-based support vector machine SVM, and obtaining a target neural network model having a function of predicting drug target proteins based on the initial ensemble learning model according to the training set may include:

[0071] According to the first training set, the at least two base learners are optimized, and according to the second training set, the overall classifier is optimized to obtain a target neural network model having the function of predicting drug target proteins.

[0072] Exemplarily, the base learner can output a classification result of whether the protein corresponding to the current feature is a drug target protein based on the protein sequence features extracted by the feature extraction network from the first training set, and obtain a loss function based on the classification result and the actual type of the protein corresponding to the current feature (i.e., drug target protein or non-drug target protein), and optimize the base learner according to the loss function.

[0073] For example, let P i =[p i1 ,p i2 ,…,p iN ], P i represents the classification probability of the ith base learner for N categories, and the overall prediction result is P = {P 1 ,P 2 ,…,P M}, M represents the number of base learners. In the traditional weighted voting mechanism, the final prediction result It can be expressed as formula (1):

[0074]

[0075] Among them, w i Represents the weight of the i-th basic model, satisfying w i It is usually manually set in advance based on preset rules or the performance of the base learner.

[0076] In this embodiment, the total classifier is a learning-based support vector machine SVM. This embodiment uses an SVM as a voter, which automatically learns weights through an optimization problem. Let x represent the combined prediction probability vector, and define x = [P 1 ,P 2 ,....,P i ], the learning-based support vector machine SVM in this embodiment determines the weight ω and the bias b by solving the following optimization problem:

[0077]

[0078] Among them, ω represents the weight learned by SVM, and b represents the bias.

[0079] Through this optimization problem, the support vector machine SVM can automatically adjust the weight ω and bias b of each base learner, avoiding the subjectivity and inaccuracy of manually setting weights in traditional voters.

[0080] Exemplarily, the second training set and the classification results of the second training set by the at least two optimized base learners can be combined with formula (2) to optimize the total classifier, so that the total classifier automatically optimizes the weights during the training process, thereby avoiding the subjectivity and inaccuracy of manually setting weights in traditional voters, thereby ensuring the optimization of the performance of the target neural network model.

[0081] In some possible implementations, optimizing the at least two base learners according to the first training set and optimizing the total classifier according to the second training set to obtain a target neural network model having a function of predicting drug target proteins may include:

[0082] S201. Perform feature extraction on the first training set according to the EMS2 model to obtain a first feature, and perform feature extraction on the first training set according to the amino acid composition ACC algorithm to obtain a second feature.

[0083] For example, taking the ESM2 model as esm2_t33_630M_UR30D model, the self-attention mechanism of the model can be expressed using formula (3):

[0084]

[0085] Where Q, K, and V represent query, key, and value matrices, respectively. K Indicates the dimension of the key vector.

[0086] According to this model, feature extraction is performed on the data in the training set to obtain the sequence representation of the protein sequence ESM2 = [e 1 ,e2 ,e 3 ,…,e i ], where e i Represents the feature of position i.

[0087] For example, AAC (amino acid composition) is a commonly used method to describe protein characteristics. It converts protein sequences into data representing the percentages of different amino acids, reflecting the amino acid composition of the protein. The principle is as follows:

[0088]

[0089] Among them, F AAC (i) represents the frequency of occurrence of amino acid i in the sequence, R i represents the total number of times amino acid i appears in the sequence, and L represents the length of the sequence.

[0090] S202: Perform fusion and normalization processing on the first feature and the second feature to obtain a third feature.

[0091] Exemplarily, the first feature and the second feature can be fused (such as splicing fusion, convolution fusion, attention fusion, etc.), and the fused features can be normalized to obtain a third feature.

[0092] For example, minimum-maximum normalization may be applied to the feature data to ensure consistency and comparability of the feature data, and the principle is as follows:

[0093]

[0094] Among them, X new represents the features obtained by normalization, X max and X min They represent the largest eigenvalue and the smallest eigenvalue respectively, and X represents the feature data that needs to be normalized.

[0095] S203. The at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature.

[0096] S204: Based on the classification results corresponding to the at least two base learners and the first training set, obtain loss functions corresponding to the at least two base learners.

[0097] Exemplarily, the loss functions of the at least two base learners can be obtained respectively according to the classification results of the at least two base learners and the actual types of proteins corresponding to the at least two classification results in the first training set.

[0098] S205. Based on the loss functions corresponding to the at least two base learners, respectively optimize the at least two base learners to obtain at least two optimized base learners.

[0099] For example, the corresponding base learners may be optimized based on the aforementioned loss function, thereby obtaining at least two optimized base learners.

[0100] S206. Optimize the total classifier according to the second training set and the classification results of the at least two optimized base learners for the second training set to obtain an optimized total classifier.

[0101] For example, this embodiment can refer to the specific implementation methods in the aforementioned embodiments, which will not be described in detail here.

[0102] S207. Obtain the target neural network based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer.

[0103] Exemplarily, a neural network model can be constructed based on a feature extraction network, at least two optimized base learners, and an optimized overall trainer.

[0104] In this embodiment, the first feature is obtained by extracting features from the first training set according to the EMS2 model, and the second feature is obtained by extracting features from the first training set according to the amino acid composition ACC algorithm; the third feature is obtained by fusing and normalizing the first feature and the second feature; the at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; the loss functions corresponding to the at least two base learners are obtained based on the classification results corresponding to the at least two base learners and the first training set; the at least two base learners are respectively optimized based on the loss functions corresponding to the at least two base learners to obtain the optimized at least two base learners; the total classifier is optimized according to the second training set and the classification results of the optimized at least two base learners for the second training set to obtain the optimized total trainer; the target neural network is obtained based on the feature extraction network, the optimized at least two base learners and the optimized total trainer, so that the target neural network model with the function of predicting drug target proteins can be obtained quickly and accurately.

[0105] It should be noted that many current models are weak in the interpretability of prediction results, making it difficult to provide researchers with in-depth model analysis. The present disclosure can use t-SNE (t-distributed Stochastic Neighbor Embedding) and SHAP (Shapley Additive Explanations) technology to enhance the interpretability of the model, allowing researchers to understand the model more clearly, improving the transparency of the model and the basis for scientific decision-making. The following specifically introduces t-SNE and SHAP technology.

[0106] t-SNE first calculates any two points x in the high-dimensional space i and x j The conditional probability P between i|j , which reflects the probability of point x i For point x j Importance:

[0107]

[0108] Among them, ||x i -x j || 2 It is point x i and x j The square of the Euclidean distance between i is around the point x i The standard deviation of the Gaussian distribution of , which is dynamically determined by the perplexity of the data.

[0109] To simplify the model, t-SNE defines a symmetric joint probability P by averaging the conditional probabilities:

[0110]

[0111] Where N is the total number of points in the dataset.

[0112] In low-dimensional space, point y i and point y j The similarity between is represented by a t-distribution, which has heavier tails at smaller distances:

[0113]

[0114] Among them, ||y i -y j || 2 It is point y i and point y j Squared Euclidean distance in low-dimensional space.

[0115] The goal of the algorithm is to minimize the Kullback-Leibler divergence between the high-dimensional and low-dimensional probability distributions:

[0116]

[0117] Among them, C is the loss function, P and Q represent the probability distribution in high-dimensional and low-dimensional space respectively.

[0118] The SHAP value measures the contribution of each feature to a single prediction by treating the model prediction task as a cooperative game in which each feature is a player. This approach reveals the absolute contribution of each feature to the prediction and maintains consistency and local accuracy, meaning that for any given prediction, the sum of the SHAP values ​​is equal to the difference between the actual model prediction and the baseline prediction. When the SHAP value φ is calculated for each feature i i , all possible subsets S except i in the feature set N are also considered. The specific calculation formula is as follows:

[0119]

[0120] Where |S| is the size of subset S, |N| is the total number of features, v(S) is the prediction without feature i, and v(S∪{i}) is the prediction with feature i. This formula quantifies the marginal contribution of feature i by considering the effect of adding different subsets S to the predicted output.

[0121] In addition to SHAP and t-SNE, other interpretability and visualization tools such as LIME (Local Interpretable Model-Effect Explanations), PCA (Principal Component Analysis), or UMAP (Uniform Manifold Approximation and Projection) can be used.

[0122] The above mainly introduces the solution provided by the embodiment of the present disclosure from the perspective of method. In order to achieve the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. It should be easy to realize that the technical goals in this field, combined with the units and algorithm steps of each example described in the embodiment disclosed in this article, can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional technical goals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0123] In an exemplary embodiment, the present disclosure also provides a neural network model training device, which can be used to implement the neural network model training method as described in the above embodiments.

[0124] Figure 3A schematic diagram of the composition of a neural network model training device provided in an embodiment of the present disclosure. Figure 3 As shown, the device may include: a data processing module 301 and a model training module 302.

[0125] A data processing module 301 is used to obtain a data set, wherein the data in the data set includes drug target protein sequence data and non-drug target protein sequence data;

[0126] The model training module 302 is used to obtain a target neural network model having the function of predicting drug target proteins based on the data set and an initial integrated learning model, wherein the initial integrated learning model includes a feature extraction network, at least two base learners and a total classifier, wherein the feature extraction network includes an ESM2 model.

[0127] In some possible implementations, the data processing module 301 is also used to divide the data set into a training set and a test set; the model training module 302 is specifically used to obtain a target neural network model having the function of predicting drug target proteins based on the initial ensemble learning model according to the training set.

[0128] In some possible implementations, the data processing module 301 is specifically used to: divide the data set into the training set and the test set according to a preset ratio.

[0129] In some possible implementations, the base learner includes at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM.

[0130] In some possible embodiments, the training set includes a first training set and a second training set, the overall classifier is a learning-based support vector machine SVM, and the model training module 302 is specifically used to: optimize the at least two base learners according to the first training set, and optimize the overall classifier according to the second training set to obtain a target neural network model with the function of predicting drug target proteins.

[0131] In some possible implementations, the model training module 302 is specifically used to: perform feature extraction on the first training set according to the EMS2 model to obtain a first feature, and perform feature extraction on the first training set according to the amino acid composition ACC algorithm to obtain a second feature; perform fusion and normalization processing on the first feature and the second feature to obtain a third feature; the at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; based on the classification results corresponding to the at least two base learners and the first training set, obtain the loss function corresponding to the at least two base learners; based on the loss function corresponding to the at least two base learners, respectively optimize the at least two base learners to obtain the optimized at least two base learners; according to the second training set and the classification results of the optimized at least two base learners for the second training set, optimize the total classifier to obtain the optimized total trainer; based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer, obtain the target neural network.

[0132] It should be noted that Figure 3 The division of modules in the above is schematic and is only a logical function division. There may be other division methods in actual implementation. For example, two or more functions may be integrated into one processing module. The embodiments of the present disclosure are not limited to this. The above integrated modules may be implemented in the form of hardware or in the form of software function modules.

[0133] In the technical solution disclosed in the present invention, the acquisition, storage and application of protein sequence data involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0134] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0135] In an exemplary embodiment, an electronic device includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in the above embodiment. The electronic device may be the above-mentioned computer or server.

[0136] In an exemplary embodiment, the readable storage medium may be a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to the above embodiments.

[0137] In an exemplary embodiment, a computer program product includes a computer program, and when the computer program is executed by a processor, the method according to the above embodiments is implemented.

[0138] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0139] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0140] Multiple components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0141] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the neural network model training method. For example, in some embodiments, the neural network model training method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the neural network model training method described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the neural network model training method in any other appropriate manner (e.g., by means of firmware).

[0142] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0143] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0144] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0145] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0146] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0147] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0148] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0149] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A neural network model training method, characterized in that: The method comprises: Acquire a data set, wherein the data in the data set includes drug target protein sequence data and non-drug target protein sequence data; According to the data set and based on the initial integrated learning model, a target neural network model with the function of predicting drug target proteins is obtained, wherein the initial integrated learning model includes a feature extraction network, at least two base learners and a total classifier, wherein the feature extraction network includes an ESM2 model.

2. The method according to claim 1, characterized in that: Before obtaining a target neural network model having a function of predicting drug target proteins based on the initial integrated learning model according to the data set, the method further comprises: Dividing the data set into a training set and a test set; The method of obtaining a target neural network model having a function of predicting drug target proteins based on the data set and the initial integrated learning model comprises: Based on the initial integrated learning model according to the training set, a target neural network model with the function of predicting drug target proteins is obtained.

3. The method according to claim 2, characterized in that The step of dividing the data set into a training set and a test set comprises: According to a preset ratio, the data set is divided into the training set and the test set.

4. The method according to claim 3, characterized in that: The base learner includes at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM.

5. The method according to claim 4, characterized in that The training set includes a first training set and a second training set, the total classifier is a learning-based support vector machine SVM, and the target neural network model having the function of predicting drug target proteins is obtained based on the initial integrated learning model according to the training set, including: According to the first training set, the at least two base learners are optimized, and according to the second training set, the overall classifier is optimized to obtain a target neural network model having the function of predicting drug target proteins.

6. The method according to claim 5, characterized in that The at least two base learners are optimized according to the first training set, and the total classifier is optimized according to the second training set to obtain a target neural network model having a function of predicting drug target proteins, including: Extracting features from the first training set according to the EMS2 model to obtain a first feature, and extracting features from the first training set according to the amino acid composition ACC algorithm to obtain a second feature; According to the first feature and the second feature, a fusion and normalization process is performed to obtain a third feature; The at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; Based on the classification results corresponding to the at least two base learners and the first training set, obtaining loss functions corresponding to the at least two base learners; Based on the loss functions corresponding to the at least two base learners, respectively optimizing the at least two base learners to obtain at least two optimized base learners; According to the second training set and the classification results of the at least two optimized base learners for the second training set, optimizing the total classifier to obtain an optimized total trainer; The target neural network is obtained based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer.

7. A neural network model training device, characterized in that: The device comprises: A data processing module, used for acquiring a data set, wherein the data in the data set includes drug target protein sequence data and non-drug target protein sequence data; A model training module is used to obtain a target neural network model with the function of predicting drug target proteins based on the data set and an initial integrated learning model, wherein the initial integrated learning model includes a feature extraction network, at least two base learners and a total classifier, wherein the feature extraction network includes an ESM2 model.

8. The device according to claim 7, characterized in that The data processing module is further used to divide the data set into a training set and a test set; The model training module is specifically used to obtain a target neural network model with the function of predicting drug target proteins based on the initial integrated learning model according to the training set.

9. The device according to claim 8, characterized in that The data processing module is specifically used for: According to a preset ratio, the data set is divided into the training set and the test set.

10. The device according to claim 9, characterized in that The base learner includes at least two of a neural network NN, a CatBoost model, an XGBoost model, and a support vector machine SVM.

11. The device according to claim 10, characterized in that The training set includes a first training set and a second training set, the overall classifier is a learning-based support vector machine SVM, and the model training module is specifically used for: According to the first training set, the at least two base learners are optimized, and according to the second training set, the overall classifier is optimized to obtain a target neural network model having the function of predicting drug target proteins.

12. The device according to claim 11, characterized in that The model training module is specifically used for: Extracting features from the first training set according to the EMS2 model to obtain a first feature, extracting features from the first training set according to the amino acid composition ACC algorithm to obtain a second feature; fusing and normalizing the first feature and the second feature to obtain a third feature; The at least two base learners respectively output classification results of whether the protein corresponding to the current feature is a drug target protein based on the third feature; Based on the classification results corresponding to the at least two base learners and the first training set, a loss function corresponding to the at least two base learners is obtained; based on the loss functions corresponding to the at least two base learners, the at least two base learners are optimized respectively to obtain the optimized at least two base learners; according to the second training set and the classification results of the at least two optimized base learners for the second training set, the total classifier is optimized to obtain an optimized total trainer; The target neural network is obtained based on the feature extraction network, the optimized at least two base learners, and the optimized total trainer.

13. An electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to any one of claims 1 to 6.

Images