A method and application for predicting compound targets based on compound-perturbed transcriptome and protein sequences

Abstract

The present invention provides a method and application for predicting compound targets based on compound-perturbed transcriptomes and protein sequences. The present invention innovatively combines compound-perturbed transcriptome features and protein target sequence features, and uses these as input features in combination with a deep learning model to achieve prediction of compound targets. Specifically, by collecting the compound's perturbed transcriptome and its active and inactive target information data, extracting transcriptome features through principal component analysis and linear discriminant analysis, and using a protein language model to represent the target protein features, the compound's transcriptome and target are converted into learnable embeddings, and a classification model is used to train and learn the potential association between the two, and finally used for compound target prediction. The present invention uses an external test set to demonstrate the high accuracy and practical applicability of the model. The present invention has the advantages of low cost, high efficiency and wide range, can significantly improve the efficiency of compound target discovery, and help accelerate the drug research and development process.

Description

Technical Field

[0001] The present invention relates to the technical field of compound target prediction in the drug discovery process, and specifically to a method for predicting compound targets based on a combination of perturbation transcriptome data, protein sequences, and deep learning. Background Art

[0002] Drug development is a crucial task in modern medicine, and it is of great significance for protecting and improving human health. However, drug development also faces many challenges and difficulties. Drug development is a long and expensive process. According to statistics, the development of a drug typically spans 10-15 years and costs billions of dollars. The drug discovery process is a complex one, involving the identification of new drug targets, the design and synthesis of compounds, and the evaluation of their efficacy and safety. Target discovery is a crucial first step in modern drug development. A drug target generally refers to the binding site of a drug in the body. However, due to the complex biological systems in the human body and the complexity of compound-protein interactions, identifying drug targets still presents many challenges.

[0003] Traditional experimental methods usually identify targets based on changes in relevant characteristics after protein binding to drugs. For example, the cellular thermal shift assay (CETSA) uses the principle that target proteins usually become more stable after binding to drug molecules; the drug affinity target stability analysis (DARTS) is based on the phenomenon that target proteins are resistant to proteases after binding to drugs, and is used to identify the binding relationship between drugs and target proteins. In addition, there are methods such as stability analysis based on protein oxidation rate. However, the above methods are time-consuming and labor-intensive. Computational target prediction methods have become an important supplementary tool for experiments due to their low cost and high throughput. Structure-based target prediction methods, such as reverse virtual screening, are based on the principle of molecular docking. For a given ligand molecule, a molecular docking program is used to dock the small molecule with the protein receptors in the database one by one, and finally identify the drug target protein. However, they are limited by protein crystal structure data and computing power requirements.

[0004] The rapid accumulation of perturbation transcriptional profile data has provided new perspectives for computational target prediction. One strategy is to link the transcriptional profile of query compounds with that of known drugs and infer targets based on similarity in gene signatures. One example is the connectivity map (cMAP) study, which compares the transcriptional profiles of known drugs and query compounds to assign new molecules to drug targets with the most similar transcriptional profiles. Another strategy is protein network-based analysis, such as ProTINA, which uses cell-type-specific dynamic models of protein-gene transcriptional regulation to infer drug targets from gene transcription profiles.

[0005] While the above strategies successfully predict compound targets using perturbed transcriptional profiles, their accuracy still needs improvement. Perturbed transcriptional profile data is subjected to multiple experiments under different perturbation conditions, which inevitably introduces data noise due to environmental factors, technical manipulation, and other factors. Furthermore, characterizing protein targets remains a challenge, making it difficult for models to learn potential associations between them. Summary of the Invention

[0006] The purpose of the present invention is to address the shortcomings of the existing technology. Principal component analysis (PCA) and linear discriminant analysis (LDA) can be used to extract features from transcriptome data with noisy background. Protein language models, which have been popular in recent years, can characterize target protein sequences. Therefore, the combined use of the two and combined with deep learning algorithms provide a method for predicting compound targets based on compound-perturbed transcriptomes and protein sequences, assisting in drug target discovery and solving the shortcomings of current prediction technologies in accuracy and efficiency.

[0007] The present invention is achieved through the following technical solutions:

[0008] A method for predicting compound targets based on compound perturbations of transcriptome and protein sequences, comprising:

[0009] Obtain the perturbation transcriptome data of the compound to be predicted and extract the perturbation transcriptome features through principal component analysis and linear discriminant analysis;

[0010] Collect a number of protein sequence data and use protein language models to convert protein sequences into embeddings that can be used for deep learning to represent protein features;

[0011] Connect the perturbed transcriptome features one by one with the protein features to form a new feature vector;

[0012] The new feature vectors are input one by one into a trained classification model, and the classification model outputs a prediction result of whether the protein corresponding to the input new feature vector is the target of the compound.

[0013] Furthermore, the perturbed transcriptome features are extracted through principal component analysis and linear discriminant analysis as follows: First, principal component analysis is performed on the perturbed transcriptome data to calculate the principal components of the data and reduce the dimensionality, mapping the high-dimensional data to a low-dimensional space to remove noise and redundant information while retaining the main variation characteristics; then, based on the low-dimensional data after dimensionality reduction, linear discriminant analysis is used to learn a linear transformation matrix that maximizes intra-class similarity and inter-class difference; finally, the data after PCA dimensionality reduction is projected based on this linear transformation matrix and mapped into a new feature space, thereby obtaining perturbed transcriptome features that characterize the perturbation pattern. In the new feature space, perturbed transcriptome data points with the same label are clustered together, while data points with different labels are separated.

[0014] Furthermore, the perturbation transcriptome data of the compound to be predicted is derived from the CMAP LINCS2020 Level 5 dataset, and the protein sequence data is derived from UniProt.

[0015] Furthermore, the classification model adopts multi-layer perceptron, random forest, support vector machine, convolutional neural network, etc.

[0016] Furthermore, the classification model is obtained by training based on a data set with the goal of minimizing the error between the classification model output and the label.

[0017] Furthermore, each sample in the dataset includes a new feature vector and a label formed by connecting the transcriptome feature and protein feature of the corresponding compound perturbation, and the label is set in the following manner:

[0018] If the IC50, EC50, and Ki values ​​of the corresponding compound against the protein are less than or equal to 10 μM, the protein is the active target of the corresponding compound and the label is 1; if the IC50, EC50, and Ki values ​​of the corresponding compound against the protein are greater than 100 μM, the protein is the inactive target of the corresponding compound and the label is 0.

[0019] Furthermore, in the data set, among samples containing the same compound, the ratio of the number of samples with label 0 to the number of samples with label 1 is greater than 3:1.

[0020] Furthermore, the classification model also includes performing hyperparameter combination search optimization on the validation set.

[0021] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a method for predicting compound targets based on transcriptome and protein sequence perturbations by compounds is implemented.

[0022] A storage medium containing computer-executable instructions, which, when executed by a computer processor, implement a method for predicting compound targets based on compound-perturbed transcriptomes and protein sequences.

[0023] The beneficial effects of the present invention are as follows: the drug target prediction method of the present invention can theoretically be applied to any compound, with compound perturbation transcriptome data and target protein sequence as the input features of the model. The model has high accuracy and can be used to perform batch screening of a large number of compound databases. It has the advantages of low cost and high efficiency. This method has broad application prospects in the direction of auxiliary drug targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 : Flowchart for constructing the compound target prediction model in the present invention.

[0025] Figure 2 : In an optional embodiment proposed by the present invention, a graph showing the changes in F1 and AUPRC values ​​on the validation set for each different parameter combination in 20 hyperparameter searches using Optuna.

[0026] Figure 3 : In an optional embodiment proposed by the present invention, the training process of the hyperparameter training model using Trial9 is demonstrated, showing the changes in AUPRC with increasing number of training rounds on the training set (Training) and test set (Test) as well as the evaluation results of the fixed model on the test set.

[0027] Figure 4 : In an optional embodiment proposed by the present invention, a comparison result of the Top 100 accuracy of two data forms (log2FC and MODZ) in actual application of natural product target prediction.

[0028] Figure 5 : In an optional embodiment proposed by the present invention, the model predicts 4 natural products with unknown target information and displays their Top 30 predicted target results. DETAILED DESCRIPTION

[0029] In order to better illustrate the purpose, technical solutions and advantages of the present invention, the present invention is further described below in conjunction with specific embodiments.

[0030] The present invention provides a method for predicting compound targets based on transcriptome and protein sequence perturbations induced by compounds, the method comprising:

[0031] Obtain the perturbation transcriptome data of the compound to be predicted and extract the perturbation transcriptome features through principal component analysis and linear discriminant analysis;

[0032] Collect a number of protein sequence data and use protein language models to convert protein sequences into embeddings that can be used for deep learning to represent protein features;

[0033] Connect the perturbed transcriptome features one by one with the protein features to form a new feature vector;

[0034] The new feature vectors are input one by one into a trained classification model, and the classification model outputs a prediction result of whether the protein corresponding to the input new feature vector is the target of the compound.

[0035] The classification model is obtained by training based on a data set with the goal of minimizing the error between the classification model output and the label. The effect of the present invention will be further described below with reference to a specific embodiment.

[0036] First, build and train a classification model such as Figure 1 As shown, the following steps are included:

[0037] Step 1: Construct a dataset. Each sample in the dataset includes a new feature vector and label formed by connecting the transcriptome features and protein features of the corresponding compound perturbation. Specifically, the following sub-steps are included:

[0038] Step 1.1. Data collection and processing:

[0039] The CMAP LINCS 2020 Level 5 dataset was downloaded from the Clue database. This dataset contains 720,216 perturbed transcriptomes obtained from 34,419 small molecule compounds in hundreds of cell lines under various perturbation conditions. Compound and protein target affinity data were downloaded from the ChEMBL database and screened using the following criteria: only data for human targets were used, and proteins with IC50, EC50, and Ki values ​​less than or equal to 10 μM were defined as active targets of the compound, while proteins with values ​​greater than 100 μM were defined as inactive targets. This screening process ultimately yielded information on 553,347 compounds and their targets.

[0040] As a preferred embodiment, to match the real-world distribution where most proteins are not compound targets, for each compound, a 1:3 ratio of active to inactive targets was randomly selected from all inactive targets. For the few compounds with an inactive to active target ratio greater than 3:1, all target information was retained.

[0041] Furthermore, InChIKey is a standardized hash value based on the International Chemical Identifier (InChI) developed by the International Union of Pure and Applied Chemistry (IUPAC). When the InChIKeys of two compounds are exactly the same, it means that they are the same compound. Using InChIKey to match the above two database compounds, we finally obtained 3141 compounds that contain both perturbed transcriptome and target information.

[0042] Step 1.2, transcriptome feature extraction:

[0043] A previously reported paper 1 (DOI:10.1038 / s43587-021-00138-z) proposes using metric learning to extract features from transcriptome data in a noisy environment. Specifically, principal component analysis (PCA) is first used for dimensionality reduction, followed by linear discriminant analysis (LDA) to learn a transformation matrix that maximizes intra-class similarity and inter-class differences. This matrix projects the data into a new space, clustering perturbed transcriptome data points with the same label as much as possible, while data points with different labels are separated as much as possible. Perturbed transcriptome features are obtained by projecting the original perturbed transcriptome data after PCA dimensionality reduction based on the transformation matrix.

[0044] The literature also evaluated the effects of factors such as cell type, compound, time point and dose on transcriptional markers, focusing on how to minimize the interference of these factors on similarity calculations. The results show that the effect of dose on marker distribution is small, so different doses of compounds are usually regarded as the same treatment conditions. In actual experiments, when repeating experiments on the same compound under the same perturbation conditions, data noise is inevitable. Therefore, the present invention first uses PCA for dimensionality reduction (n_components = 0.98), and then learns the transformation matrix through LDA metric. Before the learning process, each transcriptome data is assigned a perturbation treatment label, including compound, cell line and treatment time information, and data with the same perturbation treatment label are considered to be the same class. PCA dimensionality reduction and LDA metric learning can extract the most significant features that distinguish different categories from the original data and project the original data into a lower-dimensional feature space. This method makes the perturbed transcriptome data points with the same label as clustered as much as possible, while the data points with different labels are as dispersed as much as possible. Finally, the present invention reduces the 12328-dimensional original features of the CMAP LINCS2020level5 dataset to a 1240-dimensional embedding vector.

[0045] Step 1.3, protein target feature extraction:

[0046] The target protein sequence was queried and downloaded through UniProt, and the protein language model pre-trained in ConPLex was used to convert the protein sequence into a 1024-dimensional embedding for subsequent training.

[0047] Step 1.4, build the dataset:

[0048] For each of the 3141 compounds collected, the corresponding transcriptome-perturbing signatures were concatenated with the corresponding protein signatures to form a new feature vector, resulting in 2264-dimensional training data. A label was assigned based on whether the protein signature was an active target of the compound, with a value of 1 indicating an active target and a value of 0 indicating an inactive target, as shown in Table 1. This ultimately resulted in the construction of the dataset.

[0049] Table 1: Example of model training input

[0050] Compound Label Compound perturbation transcriptome signature Protein targets Active Tags CHEMBL10 PBIOA016_PC3_24H:C03 CHEMBL215 1 CHEMBL10 PBIOA016_PC3_24H:C03 CHEMBL221 1 CHEMBL10 PBIOA016_PC3_24H:C03 CHEMBL335 0 CHEMBL10 LJP009_PC3_24H:G08 CHEMBL2223 0 CHEMBL423 CPD001_PC3_24H:BRD-A02759312-003-17-9:10 CHEMBL1075145 0 CHEMBL423 CPC011_PC3_24H:BRD-A02759312-003-19-5:10 CHEMBL1075145 0 CHEMBL17860 REP.A012_PC3_24H:M09 CHEMBL1075113 0 CHEMBL17860 REP.A012_PC3_24H:M09 CHEMBL3708265 0 CHEMBL17860 REP.A012_PC3_24H:M11 CHEMBL2121 0 CHEMBL17860 REP.A012_PC3_24H:M11 CHEMBL214 1

[0051] In this embodiment, the data set is divided into a training set, a validation set, and a test set in a ratio of 80%:10%:10% using the train_test_split function of the Python machine learning framework software package scikit-learn. The validation set is used for model optimization, and the test set is used to evaluate the final performance of the model.

[0052] Step 2: Train the classification model as follows:

[0053] The classification model can adopt multi-layer perceptron, random forest, support vector machine, convolutional neural network, etc. In this embodiment, a multi-layer perceptron (MLP) model is used as the classification model. Specifically, the multi-layer perceptron is an artificial neural network that learns complex patterns and functions in data through multiple neuron layers. Due to its hierarchical structure, it can learn nonlinear representations of input data. It is a powerful model commonly used for various tasks such as classification and regression. Among them, the input of the classification model is a new feature vector, and the output is the prediction result of whether the corresponding protein is the target of the corresponding compound. Each sample of the training set is used as input, and training is performed with the goal of minimizing the error between the classification model output and the label to obtain a trained classification model.

[0054] After training is complete, this example uses six metrics to evaluate the model's performance: accuracy, precision, recall, and F1 score, AUROC, and AUPRC. Their calculation formulas are as follows, where TP is true positive, TN is true negative, FP is false positive, and FN is false negative. AUROC is the area under the ROC curve, and AUPRC refers to the area under the precision-recall curve, expressed as follows:

[0055]

[0056] The MLP model is sensitive to the combination of hyperparameters. As a preferred solution, this embodiment uses optuna to perform hyperparameter combination search and optimization on the validation set. Since the compound has more inactive targets than active targets (negative samples> positive samples), the present invention selects F1 and AUPRC to evaluate the classification performance of the model. Specifically, Optuna is a hyperparameter optimization framework based on Bayesian optimization, which can find the best hyperparameter combination in the fewest number of experiments. Since the training data dimension is 2264, this embodiment comprehensively considers data characteristics, computing resources and other issues, and sets the model hidden layer search range to 1 to 4 layers, dropout to 0.1 to 0.4, the number of neurons in each layer to 100 to 200, the learning rate to 1e-4, 1e-2, the maximum number of rounds to 150, and the optimizer to be Adam or SGD, which is commonly used to solve binary classification problems. F1 is used as the optimization indicator, the number of searches is 20 times, and some results are as follows. Figure 2 and as shown in Table 2.

[0057] Table 2: Scores of six indicators for 20 parameter searches

[0058]

[0059] As shown in Table 2, in the 20 searches, the F1 and AUPRC of the model in Trial9 on the validation set were the highest. Therefore, this embodiment uses this parameter for training and fixes it as the final trained classification model. Its specific hyperparameters are hidden layer: [2264,1809], [1809,1977], [1977,990], [990,835], [835,1], Dropout: 0.26051198024596567, 0.16616269153317942, 0.1351244003622237, 0.34033060495716216, learning rate: 0.0012212088107909464, optimizer: Adam, loss function: BCELoss, and training for 149 rounds. The model is fixed and named PTPCM, and the prediction ability of the model is evaluated using the test set, such as Figure 3 As shown in the figure, the improvement of AUPRC during the training process of the model is demonstrated. The final model performs well on the test set, with F1 of 0.971, AUROC of 0.998, and AUPRC of 0.995, indicating that the model can accurately distinguish active and inactive targets and can be used for target prediction.

[0060] Step 3: Target prediction based on the trained classification model. This example takes the prediction of natural product targets as an example, and the details are as follows:

[0061] Natural products usually have complex chemical structures and may contain multiple active ingredients, making it difficult to determine their targets. This example collects perturbation transcriptome data and target information of 496 natural products, and uses this as an external test set to verify the actual application effect of the method of the present invention. The LINCS level 5 perturbation transcriptome uses the moderated z-score (MODZ) to represent the changes in genes. However, in real scenarios, log2FC is often used to represent the degree of change in genes, so this implementation uses log2FC for target prediction. At the same time, referring to the method reported in document 2 (DOI: 10.1007 / s13238-021-00885-0), RNAseq was processed to obtain the modulated Z value (MODZ), and target prediction was also performed, and the prediction effects of the two treatment methods were compared. After obtaining the above two types of data, they are arranged in the training data format, and the perturbation transcriptome features are obtained through the trained metric learning model and connected one by one with the protein features to form new features, and then respectively input into PTPCM, and sorted according to the output score. The higher the score, the more likely the target is to be the target of the natural product. In this embodiment, natural products with less than 50 targets are screened and collected and considered to be more credible, and finally 210 natural products with target records are obtained. The TopK accuracy is used as an evaluation indicator. Specifically, the TopK accuracy is the ratio of the number of compounds that can hit any of its recorded targets before the model predicts the TopK target to the entire predicted compound set (210 in this case). As Figure 4 As shown, both prediction data formats achieved high accuracy. The Top 100 prediction accuracy using log2FC was 0.78, while the Top 100 prediction accuracy using MODZ was 0.68. Therefore, Log2FC was used for subsequent predictions. Some prediction results are shown in Table 3. For some compounds, such as fritillin, the Top 30 targets predicted by the model hit two of its recorded targets. For strychnine, the model even predicted its only recorded target, demonstrating the high accuracy of the model.

[0062] Table 3: Prediction results of some natural products

[0063]

[0064]

[0065] For natural products without target information, this example also performed target prediction and found some reported literature support, such as Figure 5As shown, quercetin is a flavonoid found in various plants with antioxidant, anti-inflammatory, and antibacterial properties. A report in literature 3 (DOI: 10.1007 / s13238-021-00885-0) indicates that quercetin can inhibit MAPK signaling factors such as ERK1 / 2, JNK, and p38MAPK in adipocytes and macrophages. ERK2, also known as MAPK1, appears in the top 30 predicted targets of the present method.

[0066] Salvianolic acid B is a natural compound extracted from Danshen. Recent studies have shown that it has therapeutic effects on the neurodegenerative disease Alzheimer's disease, as reported in the literature4.

[0067] (DOI:10.1016 / j.tice.2023.102260) found that salvianolic acid B had a protective effect on Aβ (amyloid β protein)-induced neuroinflammation by inhibiting the activation of NLRP3 inflammasome and promoting the polarization transition of microglia from M1 (pro-inflammatory) to M2 (anti-inflammatory). Although the key factors mentioned in the article were not found in the prediction results of the present invention, CHRM2 was predicted, which is an important target in the treatment of AD. Acetylcholine can bind to it to improve cognitive function.

[0068] Osteoarthritis is a chronic degenerative joint disease affecting over 500 million people worldwide and can even lead to disability. Reference 5 (DOI: 10.16333 / j.1001-6880.2021.8.010) has demonstrated that eleutheroside E can significantly improve inflammatory cell infiltration, fibrous tissue proliferation, and cartilage surface destruction in osteoarthritis sites, while also reducing levels of inflammatory mediators (IL-1β and PGE2) as well as MMP-3 and MMP-9 in joint fluid to treat osteoarthritis. The prediction results of the present invention show the presence of multiple MMP family proteins, including MMP-9.

[0069] Berberine, a plant alkaloid extracted from Coptis chinensis, possesses a variety of pharmacological activities, including hypoglycemic, anti-obesity, and anti-inflammatory properties. In recent years, numerous studies have confirmed its anticancer and antidepressant effects. Reference 6 (DOI:10.1186 / s11658-024-00607-4) has demonstrated that berberine can enhance EGFR gene transcription, thereby upregulating the RAS–RAF1–MEK1 / 2–ERK1 / 2 signaling pathway, inducing autophagy in tumor cells and inhibiting tumor cell proliferation and metastasis. Notably, EGFR appears among the top 30 targets predicted by this method. This evidence demonstrates the high accuracy and reliability of the target prediction method presented herein, demonstrating its practical applicability.

[0070] Corresponding to the aforementioned embodiment of a method for predicting compound targets based on compound perturbation of transcriptome and protein sequence, the present invention also provides an electronic device comprising one or more processors for implementing the aforementioned embodiment of a method for predicting compound targets based on compound perturbation of transcriptome and protein sequence.

[0071] The electronic device of the present invention may be any device with data processing capability, and the device with data processing capability may be a device or apparatus such as a computer.

[0072] The device embodiments may be implemented through software, hardware, or a combination of software and hardware. Taking software implementation as an example, as a logically defined device, a processor of any device with data processing capabilities reads the corresponding computer program instructions from a non-volatile memory into the memory and executes them. From a hardware perspective, this includes a processor, memory, a network interface, and a non-volatile memory. In addition, any device with data processing capabilities in which the device in the embodiments is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0073] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0074] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present invention. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0075] An embodiment of the present invention also provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, a method for predicting compound targets based on compound-perturbed transcriptomes and protein sequences in the above embodiment is implemented.

[0076] The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of any device with data processing capabilities and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0077] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications derived therefrom remain within the scope of protection of the present invention.

Claims

1. A method for predicting compound targets based on compound perturbation transcriptome and protein sequence, characterized in that: include: Obtain the perturbation transcriptome data of the compound to be predicted and extract the perturbation transcriptome features through principal component analysis and linear discriminant analysis; Collect a number of protein sequence data and use protein language models to convert protein sequences into embeddings that can be used for deep learning to represent protein features; Connect the perturbed transcriptome features one by one with the protein features to form a new feature vector; The new feature vectors are input one by one into a trained classification model, and the classification model outputs a prediction result of whether the protein corresponding to the input new feature vector is the target of the compound; The perturbation transcriptome features are extracted by principal component analysis and linear discriminant analysis, as follows: Principal component analysis is used to reduce the dimensionality of the perturbed transcriptome data and extract the main feature components. Then, linear discriminant analysis is used to learn a linear transformation matrix that can maximize the intra-class similarity and inter-class difference. Finally, the linear transformation matrix is ​​used to project the reduced dimensionality data and map it to a new feature space to obtain the perturbation transcriptome features that characterize the perturbation pattern.

2. The method according to claim 1, characterized in that The perturbation transcriptome data of the compound to be predicted is derived from the CMAP LINCS 2020 Level 5 dataset, and the protein sequence data is derived from UniProt.

3. The method according to claim 1, characterized in that The classification model adopts a multi-layer perceptron, a random forest, a support vector machine or a convolutional neural network.

4. The method according to claim 1, wherein The classification model is trained based on a data set with the goal of minimizing the error between the classification model output and the label.

5. The method according to claim 4, characterized in that Each sample in the dataset includes a new feature vector and a label formed by connecting the transcriptome feature and protein feature of the corresponding compound perturbation. The label is set as follows: If the IC50, EC50, and Ki values ​​of the corresponding compound against the protein are less than or equal to 10µM, the protein is the active target of the corresponding compound and the label is 1; if the IC50, EC50, and Ki values ​​of the corresponding compound against the protein are greater than 100µM, the protein is the inactive target of the corresponding compound and the label is 0.

6. The method according to claim 5, characterized in that In the dataset, among samples containing the same compound, the ratio of the number of samples with label 0 to the number of samples with label 1 is greater than 3:

1.

7. The method according to claim 4, characterized in that The classification model also includes performing hyperparameter combination search optimization on a validation set.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for predicting compound targets based on compound-perturbed transcriptomes and protein sequences as described in any one of claims 1 to 7 is implemented.

9. A storage medium comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a computer processor, implement the method of predicting compound targets based on compound perturbation of transcriptome and protein sequence according to any one of claims 1 to 7.

Images