Method and device for predicting space transcriptome miRNA activity based on batch sequencing data
By integrating Bulk RNA-seq and scRNA-seq data and establishing and optimizing machine learning models, the problem of miRNA activity detection in spatial transcriptomics technology is solved, high-precision miRNA activity prediction is achieved, and new perspectives and personalities are provided for cancer research. Chemical treatment plan.
Patent Information
- Application Number
- CN202411193364.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-26
AI Technical Summary
Existing spatial transcriptomics technologies have resolution and sensitivity limitations in detecting and analyzing microRNA (miRNA) activities, making it difficult to fully analyze miRNA activity in tissues at high resolution.
By integrating Bulk RNA-seq and scRNA-seq data, machine learning models are established and preprocessed, hyperparameter combination is optimized using grid search method, expression matrix is constructed for miRNA activity prediction, and accurate prediction of miRNA activity is achieved by combining spatial transcriptome data. .
Improve the analytical ability of miRNA activity in tissues, reveals the differences in activity of miRNAs at different spatial locations, provides a new perspective for understanding the role of miRNAs in cancer, and ensures the accuracy of predictions through multiple performance indicators and evaluation. reliability.
Smart Images

Figure CN119028451A_ABST
Abstract
Description
Technical Field
[0001] This document relates to the technical field of miRNA activity prediction, and in particular to a method and device for predicting spatial transcriptome miRNA activity based on batch sequencing data. Background Art
[0002] Spatial transcriptomics (ST) is a technology that has emerged in recent years and can analyze gene expression in the spatial context of tissue sections. The emergence of ST technology allows researchers to gain a deeper understanding of gene expression patterns in tissues from a spatial dimension. However, traditional ST technology faces limitations in resolution and sensitivity, especially in the detection and analysis of microRNA (miRNA) activity. Existing spatial transcriptomics technologies include: BulkRNA-seq technology, single-cell RNA-seq (scRNA-seq) technology, and spatial transcriptomics (ST) technology. The above technologies have the following disadvantages:
[0003] Bulk RNA-seq is a widely used technique for measuring gene expression levels in large cell populations. Although this technique can provide rich gene expression information, it is difficult to resolve the specific contributions of different cell types in complex tissues because it cannot resolve the heterogeneity of individual cells.
[0004] Single-cell RNA-seq technology can analyze gene expression in individual cells, thereby revealing heterogeneity between cells. This technology has great potential in analyzing complex biological systems and disease mechanisms. However, due to the lack of spatial information, single-cell RNA-seq is difficult to provide the spatial distribution of gene expression in tissues.
[0005] ST technology provides spatial information of gene expression by sequencing gene expression on tissue sections. However, current ST technology has challenges in detecting miRNA activity, mainly because the short sequence and low expression level of miRNA make it difficult to efficiently capture and analyze. Resolution and sensitivity limitations make it difficult to fully analyze miRNA activity in tissues at high resolution.
[0006] In summary, the existing bulk RNA-seq, scRNA-seq and ST technologies each have different limitations and deficiencies, especially in analyzing miRNA activity in spatial transcriptomes. To address these issues, a method for building a model to predict miRNA activity in spatial transcriptomes based on bulk RNA-seq and scRNA-seq data is proposed, which will help overcome the shortcomings of existing technologies and improve the ability to analyze miRNA activity in tissues. Summary of the invention
[0007] The present invention provides a method and device for predicting spatial transcriptome miRNA activity based on batch sequencing data, aiming to solve the above problems.
[0008] The embodiment of the present invention provides a method for predicting spatial transcriptome miRNA activity based on batch sequencing data, comprising:
[0009] S1, bulk sequencing data is collected and preprocessed to obtain a first data set, and spatial transcriptomics data is collected and preprocessed to obtain a second data set;
[0010] S2. Obtaining a trained machine learning model through the first data set, and obtaining an optimal hyperparameter combination of the trained machine learning model through a grid search method; and
[0011] S3. Predicting the miRNA activity in the second data set based on the trained machine learning model.
[0012] The embodiment of the present invention provides a device for predicting spatial transcriptome miRNA activity based on batch sequencing data, comprising:
[0013] A data collection module, used to collect bulk sequencing data and obtain a first data set after preprocessing, and to collect spatial transcriptomics data and obtain a second data set after preprocessing;
[0014] A model building module, which obtains a trained machine learning model through the first data set, and obtains an optimal hyperparameter combination of the trained machine learning model through a grid search method;
[0015] An activity prediction module is used to predict the activity of miRNA in the second data set based on the trained machine learning model.
[0016] An embodiment of the present invention further provides an electronic device, including:
[0017] processor; and,
[0018] A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps of the method for predicting spatial transcriptome miRNA activity based on bulk sequencing data as described above.
[0019] An embodiment of the present invention further provides a storage medium for storing computer executable instructions, which, when executed, implement the steps of the method for predicting spatial transcriptome miRNA activity based on batch sequencing data as described above.
[0020] By adopting the embodiment of the present invention, a machine learning model is established by integrating bulk RNA-seq data from multiple sources and applied to the prediction of miRNA activity in the spatial transcriptome, which solves the problem that spatial transcriptome sequencing technology is difficult to detect miRNA, and the effectiveness of the prediction method is verified by subsequent experiments. A set of systematic preprocessing processes is proposed, including denoising, normalization, sample alignment and batch effect elimination of bulk RNA-seq and miRNA-seq data, ensuring the high quality and consistency of the data. The evolutionary conservation principle of miRNA is used to reduce the occurrence of false positives, improve the reliability of data and the accuracy of prediction. By selecting common genes from bulk RNA-seq data and constructing an expression matrix, combining miRNA expression data, and using a machine learning model for prediction, the conversion from macro data to micro miRNA activity prediction is achieved. Five-fold cross validation and grid search optimization of multiple models improve the generalization ability and prediction accuracy of the model. Using the common gene expression matrix in the spatial transcriptome data, the pre-trained machine learning model is applied to predict miRNA activity, and accurate prediction of miRNA expression in the spatial transcriptome data is achieved. This method can reveal the differences in the activity of miRNAs at different spatial locations, providing a new perspective for understanding the role of miRNAs in cancer. The model was evaluated using multiple performance indicators such as MAE, MSE, and R2 to ensure that the selected model had the best prediction performance. The correlation of the prediction results was verified by the Spearman correlation coefficient, further ensuring the accuracy and reliability of the prediction. The proposed method has significant advantages in data preprocessing, model construction, and performance optimization, and can more accurately predict miRNA activity in spatial transcriptome data, providing a new method for studying the role of miRNAs in cancer. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0022] Figure 1 A flowchart of a method for predicting spatial transcriptome miRNA activity based on batch sequencing data according to an embodiment of the present invention;
[0023] Figure 2 A schematic diagram of the architecture and technical details of an embodiment of the present invention;
[0024] Figure 3 Integrating Bulk RNA-seq Data and Building Models for the Embodiments of the Present Invention
[0025] Figure 4 A schematic diagram of the Spearman correlation distribution of the predicted miRNA values and the true values of four cancers by the XGBoost model of an embodiment of the present invention;
[0026] Figure 5 The spatial expression distribution of the spatial transcriptome miRNA activity of breast cancer predicted in an embodiment of the present invention;
[0027] Figure 6 Schematic diagram of a device for predicting spatial transcriptome miRNA activity based on batch sequencing data according to an embodiment of the present invention. DETAILED DESCRIPTION
[0028] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.
[0029] Method Embodiment
[0030] According to an embodiment of the present invention, a method for predicting spatial transcriptome miRNA activity based on batch sequencing data is provided. Figure 1 Flow chart of the method for predicting spatial transcriptome miRNA activity based on batch sequencing data according to an embodiment of the present invention. Figure 1 As shown, the method for predicting spatial transcriptome miRNA activity based on batch sequencing data in an embodiment of the present invention specifically includes:
[0031] S1, collecting bulk sequencing data and preprocessing to obtain a first data set, collecting spatial transcriptomics data and preprocessing to obtain a second data set; S1 specifically includes:
[0032] Bulk sequencing data collection and preprocessing:
[0033] In step S1, bulk RNA-seq and miRNA-seq gene expression data from multiple sources are obtained respectively, and mRNA and miRNA data in the gene expression data are paired and aligned;
[0034] De-noising and normalizing the Bulk RNA-seq and miRNA-seq gene expression data from the multiple sources;
[0035] The rank operation is used on the normalized data to eliminate the batch effect and obtain the first data set.
[0036] Bulk sequencing data collection and preprocessing Bulk RNA-seq and miRNA-seq gene expression data of the same cancer from multiple sources are obtained respectively. Since the target gene of miRNA may be multiple mRNAs, the mRNA and miRNA data should be paired first, and the samples should be aligned. There may be some outliers in the raw data. The raw data is first denoised and normalized. Then, considering the miRNA name, its long suffix (such as -3p) is deleted, and then the duplicates of the same miRNA name are deleted and replaced with the median. When integrating data from multiple sources, a rank operation is used for the gene expression of each sample to eliminate batch effects. For example Figure 3 Shown is a schematic diagram of integrating bulk RNA-seq data and building a model;
[0037] Spatial transcriptomics sequencing data collection and preprocessing:
[0038] Gene expression data, tissue images and spatial coordinate information data of the spatial transcriptome are obtained, the gene expression data are filtered based on preset quality control indicators and then normalized, and a preset number of highly variable genes are identified in the normalized gene expression data as a second data set.
[0039] Spatial transcriptomics sequencing data collection and preprocessing By obtaining the filtered_feature_bc_matrix.h5 gene expression data, tissue images and spatial coordinate information data of the spatial transcriptome, the ST expression data was filtered based on different quality control indicators, such as total UMI counts, number of genes, etc. The data was then normalized to identify the top 2000 highly variable genes.
[0040] The data collection in S1 includes paired miRNA-mRNA RNA-seq gene expression data of breast cancer, lung cancer, ovarian cancer and prostate cancer obtained from TCGA and CCLE databases, and ST data of corresponding cancer types obtained from 10xGenomics. Data preprocessing includes denoising of raw data, ranking operations to eliminate batch effects, standardization, and screening of highly variable genes.
[0041] S2. Obtaining a trained machine learning model through the first data set, and obtaining an optimal hyperparameter combination of the trained machine learning model through a grid search method; S2 specifically includes:
[0042] A set of common genes C, X was selected from the top 2000 highly variable genes in the spatial transcriptomics data and bulk RNA-seq data. B represents the expression matrix of common genes extracted from bulk RNA-seq data, with dimensions m×|G|, where m is the number of samples and Y B Represents the corresponding miRNA expression value in bulk RNA-seq data, with a dimension of m×1. S represents the expression matrix of common genes extracted from spatial transcriptome data, with dimension n×|G|, where n is the number of spatial positions, represents the predicted miRNA expression value in the spatial transcriptome data, with a dimension of n×1, and f(·): represents a machine learning model (e.g., XGBoost).
[0043] First, extract the common gene expression matrix X from the bulk RNA-seq data B and the corresponding miRNA expression value Y B :
[0044] X B =B[G];
[0045] Among them, B[G] represents the expression value of the common gene G extracted from bulk RNA-seq data.
[0046] Second, train the machine learning model f using X B As input features, Y B As a tag:
[0047] f = TrainModel(X B ,Y B )
[0048] Here, multiple machine learning models are built. The models are optional. Five-fold cross validation is performed and the grid search method is applied to identify the optimal hyperparameter combination of the model. The performance measurement indicators include mean absolute error (MAE), mean square error MSE and determination coefficient R 2 ,After training, the Spearman correlation coefficient between the predicted miRNA ,activity and the actual value was calculated, and the model with the best ,prediction accuracy performance was selected by comparing multiple indices.
[0049] In S2, the mRNA expression data in bulk RNA-seq data was used as input features and miRNA activity was used as output variables to construct a machine learning model. The hyperparameter combination of the model was optimized by the grid search method. The model performance was evaluated on the training set and the test set, and the Spearman correlation coefficient was calculated to evaluate the prediction accuracy of the model.
[0050] S3, predicting the miRNA activity in the second data set based on the trained machine learning model. S3 specifically includes:
[0051] The trained machine learning model is applied to spatial transcriptomics to predict miRNA activity. First, the expression matrix X of common genes is extracted from the spatial transcriptomics data. S :
[0052] X S = S[G];
[0053] Among them, S[G] represents the expression value of the common gene G extracted from the spatial transcriptome data;
[0054] Secondly, we use the pre-trained machine learning model f to predict miRNA expression values in spatial transcriptome data.
[0055]
[0056] In S3, overlapping genes in Bulk RNA-seq and ST data were selected, and the trained XGBoost model was applied to ST data to predict the miRNA activity in each spot.
[0057] Through the above steps, the activity of miRNA in the spatial transcriptome can be predicted, which can provide a deep understanding of the interaction between different cell types in the tumor microenvironment. MiRNA plays an important role in regulating gene expression, promoting or inhibiting tumor growth and metastasis. Knowing the spatial distribution of miRNA can help identify therapeutic targets in specific areas, develop personalized treatment plans, and improve treatment effects.
[0058] In step S1, bulk RNA-seq data from multiple sources are integrated and batch effects are eliminated to improve the accuracy and robustness of the model in predicting miRNA activity.
[0059] A specific implementation of the embodiment of the present invention is as follows:
[0060] First, paired miRNA-mRNA gene expression data of breast cancer, lung cancer, ovarian cancer and prostate cancer were collected from The Cancer Genome Atlas (TCGA) and Cancer Cell Line Encyclopedia (CCLE) databases. The original TCGA and CCLE data were denoised, and ranking operations were applied to eliminate batch effects. The names of miRNAs were processed by deleting long suffixes, deleting duplicate columns and retaining only one column, and taking the mean as a replacement. A set of comparative models, such as ridge regression, Lasso regression, random forest regression, neural network regression and XGBoost models, were constructed, using mRNA expression data in Bulk RNA-seq data as input features and miRNA activity as output variables. The five-fold cross-validation method was used to train each model, and the grid search method was used to find the optimal hyperparameters of the model. The results compared various metrics to select the model with the best predictive ability. Figure 2 The figure shows a schematic diagram of the architecture and technical details of an embodiment of the present invention.
[0061] Table 1 Performance of each model on four cancer data
[0062]
[0063] It can be seen from Table 1 that XGBoost has the best comprehensive performance in MSE, MAE, and R2, so the pre-trained XGBoost model can be selected. Figure 4 It shows that most of the miRNA values predicted by the XGBoost model are positively correlated with the true miRNA values (average correlation>0.5), and a few miRNAs are negatively correlated, which may be due to abnormal sample data. The model is not fixed and can be replaced with a model with better prediction performance.
[0064] Next, we obtained spatial transcriptome data of the same cancer from 10x Genomics, filtered it, normalized it, and screened for highly variable genes. We used the spatial transcriptome expression data of cancer as features and input them into the pre-trained XGBoost model to predict miRNA activity.
[0065] Breast cancer data was selected for case study. After predicting the miRNA activity in the spatial transcriptome of breast cancer, further downstream analysis can be performed. Figure 5The expression distribution of possible miRNA biomarkers in each cell type in breast cancer tissue is shown, which is consistent with previous research results, thereby verifying the method and results of the present invention. For example, this study determined that miR-200b is a possible marker in breast cancer malignant cells, and G.-L.Huang et al. believed that the lower expression levels of miR-200b and miR-200c in breast cancer were associated with shorter progression-free survival (PFS) and overall survival (OS), indicating that they have the potential as prognostic biomarkers. In addition, the results showed that miR-497 in plasma is a potential biomarker for breast cancer, which is consistent with the previous research of Plieskatt et al., who determined miR-497-5p in the miRNAs that were dysregulated in breast cancer, and Luo et al. emphasized the potential of miR-497 as a non-invasive diagnostic biomarker for various cancers (including breast cancer), and pointed out that its expression in tissues and circulation (serum / plasma) is positively correlated. The consistency between the prediction results of the present invention and these established studies makes it plausible that miRNAs in other cell types are promising biomarker candidates in breast cancer diagnosis.
[0066] The embodiments of the present invention have the following beneficial effects:
[0067] 1. Technical effect: By integrating bulk RNA-seq data from multiple sources to establish a machine learning model, and applying it to the prediction of miRNA activity in the spatial transcriptome, the problem that spatial transcriptome sequencing technology is difficult to detect miRNA is solved, and the effectiveness of the prediction method is verified by subsequent experiments. A set of systematic preprocessing processes is proposed, including denoising, normalization, sample alignment and batch effect elimination of bulk RNA-seq and miRNA-seq data, ensuring the high quality and consistency of the data. The evolutionary conservation principle of miRNA is used to reduce the occurrence of false positives, improve the reliability of the data and the accuracy of the prediction. By selecting common genes from bulk RNA-seq data and constructing an expression matrix, combined with miRNA expression data, and using machine learning models for prediction, the conversion from macro data to micro miRNA activity prediction is achieved. Five-fold cross-validation and grid search optimization of multiple models improve the generalization ability and prediction accuracy of the model. Using the common gene expression matrix in the spatial transcriptome data, the pre-trained machine learning model is applied to predict miRNA activity, achieving accurate prediction of miRNA expression in the spatial transcriptome data. This method can reveal the differences in the activity of miRNAs at different spatial locations, providing a new perspective for understanding the role of miRNAs in cancer. 2The model was evaluated by multiple performance indicators, ensuring that the selected model had the best prediction performance. The correlation of the prediction results was verified by the Spearman correlation coefficient, further ensuring the accuracy and reliability of the prediction. The proposed method has significant advantages in data preprocessing, model building, and performance optimization, and can more accurately predict miRNA activity in spatial transcriptome data, providing a new method for studying the role of miRNA in cancer.
[0068] 2. Economic Effect
[0069] Cost saving: By integrating the existing Bulk RNA-seq and ST data resources, the existing data can be fully utilized to reduce the cost of repeated experiments. Machine learning technology can be used to improve the efficiency of data processing and analysis, reducing labor and time costs.
[0070] Wide application: This method is applicable to many types of cancer, including breast cancer, lung cancer, ovarian cancer and prostate cancer, and has broad application prospects, which can promote research and development in related fields;
[0071] 3. Social Effect
[0072] Promote cancer research: It provides a high-precision miRNA activity prediction method to help researchers deeply understand the role and regulatory network of miRNA in cancer and promote the study of cancer molecular mechanisms. It can provide a deep understanding of the interaction between different cell types in the tumor microenvironment. miRNA plays an important role in regulating gene expression and promoting or inhibiting tumor growth and metastasis.
[0073] Assisting precision medicine: This technology can provide cancer patients with personalized miRNA activity maps and cell type-specific information, which helps to formulate precise treatment plans and improve treatment efficacy and patient survival rates. Understanding the spatial distribution of miRNA under different pathological conditions can help reveal the molecular mechanism of the disease and promote the development of new treatment strategies;
[0074] Environmentally friendly: Through efficient data integration and analysis methods, the use and waste of experimental materials are reduced, which helps to reduce environmental pollution in the scientific research process and promote green scientific research.
[0075] Device Example 1
[0076] According to an embodiment of the present invention, a device for predicting spatial transcriptome miRNA activity based on batch sequencing data is provided. Figure 6 Schematic diagram of a device for predicting spatial transcriptome miRNA activity based on batch sequencing data according to an embodiment of the present invention. Figure 6 As shown, the device for predicting spatial transcriptome miRNA activity based on batch sequencing data in an embodiment of the present invention specifically includes:
[0077] A data collection module 60 is used to collect bulk sequencing data and perform preprocessing to obtain a first data set, and to collect spatial transcriptomics data and perform preprocessing to obtain a second data set;
[0078] A model building module 62, which obtains a trained machine learning model through the first data set, and obtains an optimal hyperparameter combination of the trained machine learning model through a grid search method;
[0079] The activity prediction module 64 is used to predict the activity of miRNA in the second data set based on the trained machine learning model.
[0080] The device for predicting spatial transcriptome miRNA activity based on batch sequencing data in the embodiment of the present invention is the device claim corresponding to the above method embodiment. For its specific implementation, please refer to the method embodiment and will not be repeated here.
[0081] Device Example 2
[0082] An embodiment of the present invention provides an electronic device, including:
[0083] processor; and,
[0084] A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps as described in the above method embodiment.
[0085] Device Example 3
[0086] An embodiment of the present invention provides a storage medium for storing computer-executable instructions, which, when executed, implement the steps described in the above method embodiment.
[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for predicting spatial transcriptome miRNA activity based on bulk sequencing data, characterized in that include: S1, bulk sequencing data is collected and preprocessed to obtain a first data set, and spatial transcriptomics data is collected and preprocessed to obtain a second data set; S2. Obtaining a trained machine learning model through the first data set, and obtaining an optimal hyperparameter combination of the trained machine learning model through a grid search method; and S3. Predicting the miRNA activity in the second data set based on the trained machine learning model.
2. The method according to claim 1, characterized in that: The step of collecting bulk sequencing data and performing preprocessing to obtain a first data set specifically includes: Obtain bulk RNA-seq and miRNA-seq gene expression data from multiple sources, pair and align the mRNA and miRNA data in the gene expression data; De-noising and normalizing the Bulk RNA-seq and miRNA-seq gene expression data from the multiple sources; The rank operation is used on the normalized data to eliminate the batch effect and obtain the first data set.
3. The method according to claim 1, characterized in that The collecting of spatial transcriptomics data and preprocessing to obtain a second data set specifically includes: Gene expression data, tissue images and spatial coordinate information data of the spatial transcriptome are obtained, the gene expression data are filtered based on preset quality control indicators and then normalized, and a preset number of highly variable genes are identified in the normalized gene expression data as a second data set.
4. The method according to claim 1, characterized in that The S2 specifically includes: The first data set is used as input features and miRNA activity is used as output variables to construct a machine learning model, specifically including: Extract the expression matrix X of common genes in the first data set B and the corresponding miRNA expression value Y B : X B =B[G] Formula 1; Wherein, B[G] represents the expression value of the common gene G extracted from the first data set; Train a machine learning model f using X B As input features, Y B As a tag: f = TrainModel(X B ,Y B ) Formula 2; Optimize the hyperparameter combination of machine learning models through grid search method.
5. The method according to claim 1, characterized in that The method for predicting spatial transcriptome miRNA activity based on batch sequencing data further includes calculating the Spearman correlation coefficient to evaluate the prediction accuracy of the machine learning model.
6. The method according to claim 1, characterized in that The method for predicting spatial transcriptome miRNA activity based on batch sequencing data further includes: using mean absolute error, mean square error and determination coefficient to evaluate the machine learning model.
7. The method according to claim 1, characterized in that: The S3 specifically includes: Extract the common gene expression matrix X from the second dataset S : X S =S[G] Formula 3; Where S[G] represents the expression value of the common gene G extracted from the second data set; Use the trained machine learning model f to predict the miRNA expression values in the second dataset 8. A device for predicting spatial transcriptome miRNA activity based on batch sequencing data, characterized in that: include: A data collection module, used to collect bulk sequencing data and obtain a first data set after preprocessing, and to collect spatial transcriptomics data and obtain a second data set after preprocessing; A model building module, which obtains a trained machine learning model through the first data set, and obtains an optimal hyperparameter combination of the trained machine learning model through a grid search method; An activity prediction module is used to predict the activity of miRNA in the second data set based on the trained machine learning model.
9. An electronic device, comprising: processor; as well as, A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps of the method for predicting spatial transcriptome miRNA activity based on bulk sequencing data as described in any one of claims 1 to 7.
10. A storage medium for storing computer executable instructions, which, when executed, implement the steps of the method for predicting spatial transcriptome miRNA activity based on bulk sequencing data as described in any one of claims 1 to 7.
Citation Information
Cited By
Transcription sequencing data processing method, device, equipment, medium and product
CN120727093A
Methods, devices, equipment, media and products for processing transcription sequencing data
CN120727093B
