Plant cell type prediction method and apparatus based on interpretable machine learning

Abstract

The application provides a plant cell type prediction method and device based on an interpretable machine learning, and the method comprises the following steps: obtaining single cell data of a target organ of a target plant; preprocessing the single cell data to obtain standardized single cell data; performing feature screening on the standardized single cell data to obtain high-variable genes of the single cell data, wherein the expression level of the high-variable genes in the target cell is higher than a first expression threshold, and the expression level of the high-variable genes in other cells is lower than the first expression threshold; performing normalization on the single cell data and the high-variable genes to obtain a cell gene matrix; determining a target multi-classification model corresponding to the target organ of the target plant; inputting the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model; and the application can realize efficient and accurate prediction of cell types of various organs of various plants.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and device for predicting plant cell types based on interpretable machine learning. Background Art

[0002] Unlike higher animals, higher plant cells have high plasticity, and differentiated plants can acquire totipotency and regenerate complete fertile plants.

[0003] In single-cell data analysis, accurate cell classification is the basis of data analysis. Specific marker genes of different cell populations are identified based on statistical methods. The final experimental method requires specific fluorescent labeling of the cell type expressing the marker genes. The above method lacks accuracy and has great limitations.

[0004] It can be seen that the single-cell data analysis methods for plants in related technologies have technical problems such as low accuracy and large limitations. Summary of the Invention

[0005] The present invention provides a plant cell type prediction method and device based on interpretable machine learning, which is used to solve the technical problems of low accuracy and large limitations of single-cell data analysis methods for plants in related technologies, and to achieve efficient and accurate prediction of cell types of multiple organs of various plants.

[0006] The present invention provides a plant cell type prediction method based on interpretable machine learning, comprising the following steps.

[0007] Acquire single-cell data of a target organ of a target plant; preprocess the single-cell data to obtain standardized single-cell data; perform feature screening on the standardized single-cell data to obtain highly variable genes in the single-cell data, wherein the expression level of the highly variable genes in the target cells is higher than a first expression threshold, and the expression level of the highly variable genes in other cells is lower than the first expression threshold; perform normalization based on the single-cell data and the highly variable genes to obtain a cell gene matrix; determine a target multi-classification model corresponding to the target organ of the target plant; input the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0008] According to a plant cell type prediction method based on interpretable machine learning provided by the present invention, after inputting the cell gene matrix into the target multi-classification model to obtain the predicted cell type output by the target multi-classification model, the method further includes: performing SHAP value analysis on the target multi-classification model to obtain specific marker genes corresponding to the predicted cell type.

[0009] According to the plant cell type prediction method based on the interpretable machine learning provided by the application, the single cell data is preprocessed to obtain standardized single cell data, which comprises: determining the original single cell RNA sequencing data of the single cell data; performing quality control processing on the original single cell RNA sequencing data to obtain filtered single cell RNA sequencing data, wherein the quality control processing comprises: removing low-expression genes and removing mitochondrial genes, wherein the expression level of the low-expression genes is lower than a second expression threshold; and performing standardization processing on the sequencing depth and gene expression of the filtered single cell RNA sequencing data to obtain standardized single cell data.

[0010] According to the plant cell type prediction method based on the interpretable machine learning provided by the application, before the single cell data of the target organ of the target plant is obtained, the method further comprises: obtaining single cell data of different organs of a plurality of plant species; performing preprocessing and annotation on the single cell data to obtain a cell type label of each single cell data; performing feature screening on the single cell data based on the cell type label to obtain a target number of high-variable genes of each single cell data; performing normalization based on each single cell data and the high-variable genes of each single cell data to obtain a cell gene matrix of each single cell; classifying the cell gene matrix of each single cell according to different organs of the plurality of plant species to obtain a training data set of each organ of each plant species, wherein the training data set comprises the cell gene matrix and a cell type label corresponding to the cell gene matrix; and constructing a multi-classification model corresponding to each organ of each plant species based on the training data set of each organ of each plant species and a target machine learning algorithm.

[0011] According to the plant cell type prediction method based on the interpretable machine learning provided by the application, before the single cell data of the target organ of the target plant is obtained, the method further comprises: obtaining single cell data of different organs of a plurality of plant species; performing preprocessing and annotation on the single cell data to obtain a cell type label of each single cell data; performing feature screening on the single cell data based on the cell type label to obtain a target number of high-variable genes of each single cell data; performing normalization based on each single cell data and the high-variable genes of each single cell data to obtain a cell gene matrix of each single cell; classifying the cell gene matrix of each single cell according to different organs of the plurality of plant species to obtain a training data set of each organ of each plant species, wherein the training data set comprises the cell gene matrix and a cell type label corresponding to the cell gene matrix; and constructing a multi-classification model corresponding to each organ of each plant species based on the training data set of each organ of each plant species and a target machine learning algorithm.

[0012] According to the plant cell type prediction method based on an interpretable machine learning algorithm provided by the present application, the target machine learning algorithm is a LightGBM algorithm.

[0013] The present application also provides a plant cell type prediction device based on an interpretable machine learning algorithm, comprising the following modules: an acquisition module for acquiring single cell data of a target organ of a target plant; a preprocessing module for preprocessing the single cell data to obtain standardized single cell data; a screening module for performing feature screening on the standardized single cell data to obtain high variable genes of the single cell data, wherein the expression level of the high variable genes in a target cell is higher than a first expression threshold, and the expression level of the high variable genes in other cells is lower than the first expression threshold; a normalization module for normalizing the single cell data and the high variable genes to obtain a cell gene matrix; a determination module for determining a target multi-classification model corresponding to the target organ of the target plant; and a prediction module for inputting the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0014] The present application also provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the plant cell type prediction method based on an interpretable machine learning algorithm as described above.

[0015] The present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable by a processor to implement the plant cell type prediction method based on an interpretable machine learning algorithm as described above.

[0016] The present application also provides a computer program product comprising a computer program, wherein the computer program is executable by a processor to implement the plant cell type prediction method based on an interpretable machine learning algorithm as described above.

[0017] The plant cell type prediction method and device based on explainable machine learning provided by the application can eliminate noise and bias in data and improve the accuracy and reliability of data by preprocessing single-cell data, can identify high-variable genes with significant differences in expression level in single-cell data, which is of great significance for distinguishing different cell types, and can further filter out genes that are highly expressed only in target cells by setting a first expression threshold, thereby improving the accuracy and pertinence of feature selection, can eliminate expression differences between different samples based on normalization processing of single-cell data and high-variable genes, so that the data is more comparable and easy to analyze, can input cell gene matrix into the model in a targeted manner by determining a target multi-classification model corresponding to the target organ of the target plant, so as to more efficiently and accurately classify cell types and identify predicted cell types in the target organ of the target plant, and further solve the technical problems of low accuracy and great limitations in the related art of plant single-cell data analysis methods. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description one by one. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 is a flowchart of the plant cell type prediction method based on explainable machine learning provided by the application.

[0020] Figure 2 is a whole flowchart of the plant cell type prediction method based on explainable machine learning provided by the application.

[0021] Figure 3 is a comparison diagram of the prediction effect of the model based on LightGBM algorithm provided by the application.

[0022] Figure 4 is a structure diagram of the plant cell type prediction device based on explainable machine learning provided by the application.

[0023] Figure 5 is a physical structure diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0025] Single-cell sequencing technology is developing rapidly and has become a powerful method for studying gene expression in complex multicellular organisms. Compared with bulk sequencing, single-cell sequencing technology can identify rare cell populations and reveal cell state transitions at different developmental stages, which is difficult to capture with traditional methods. Therefore, solving scientific problems at the single-cell level is currently the focus of research.

[0026] Unlike higher animals, higher plant cells have high plasticity in cell fate, and differentiated plant bodies can regain totipotency and regenerate complete fertile plants. Now there are many single-cell sequencing technologies applied to model organisms such as Arabidopsis, rice, and corn. These studies are particularly important for plant research. In single-cell data analysis, accurate classification of cells is the basis for data analysis. Unsupervised clustering algorithms are used to classify cells, and specific marker genes for different cell populations are identified based on statistical methods. Finally, the experimental method needs to express cell type-specific fluorescent markers for marker genes. These methods lack accuracy and have great limitations.

[0027] Therefore, starting from various single-cell omics and machine learning methods, standard and unified processing of data, modeling analysis of multi-omics data of different single-cell populations, and identification of new cell subpopulations and marker genes are of great significance for analyzing cell heterogeneity.

[0028] The main technical problem to be solved by the present application is to intelligently predict plant single-cell data and mine new plant cell type-specific marker genes.

[0029] In order to solve the above technical problems, the present application integrates existing plant single-cell data, and after re-integrating and analyzing the data, a machine learning model is constructed for each tissue of each species to accurately predict cell types. According to the feature importance ranking in the model, cell type-specific marker genes are screened out for subsequent experimental verification.

[0030] Alternatively, the plant cell type prediction method based on interpretable machine learning of the embodiments of the present application can be executed by a server, or by a terminal device, or by both a server and a terminal device. Taking the plant cell type prediction method based on interpretable machine learning in the embodiments of the present application executed by a server as an example.

[0031] Figure 1 is a flowchart of a plant cell type prediction method based on an interpretable machine learning provided by the present application, as shown in Figure 1 The method comprises the following steps:

[0032] Step 101, obtaining single cell data of a target organ of a target plant.

[0033] It should be noted that the single cell data is extracted from the target organ of the target plant, and it can be understood that the target plant and the target organ are predetermined.

[0034] In the embodiments of the present application, single cell RNA sequencing (scRNA-seq) data from a specific organ of a target plant is collected; the cells of the target organ are sequenced using single cell sequencing technology such as 10x Genomics, SMART-seq2, etc., to obtain single cell data containing a large amount of cell gene expression information.

[0035] Step 102, preprocessing the single cell data to obtain standardized single cell data.

[0036] The main purpose of standardizing the single cell data is to eliminate technical bias, batch effect, and improve the consistency and comparability of the data.

[0037] In the embodiments of the present application, based on indicators such as sequencing depth and number of gene detections, low-quality cells are identified and removed; genes with extremely low expression in all cells are removed, which may be generated due to sequencing noise.

[0038] Since the sequencing depth of different cells may be different, the gene expression of each cell needs to be divided by the sequencing depth (i.e. total reads or UMI number) of the cell to eliminate the influence of sequencing depth on expression; the normalized expression is logarithmically transformed (such as log2 or log10) to reduce the influence of extreme values and make the data more close to normal distribution.

[0039] If the data comes from different experimental batches, a specific algorithm is used to correct the batch effect.

[0040] In some embodiments, the expression of each gene in the single cell data is converted into a Z-score, i.e. subtracting the average value of the gene in all cells and then dividing by the standard deviation, to eliminate the expression difference between genes.

[0041] Step 103, performing feature screening on the standardized single cell data to obtain high-variable genes of the single cell data, wherein the expression level of the high-variable gene in the target cell is higher than a first expression threshold, and the expression level of the high-variable gene in other cells is lower than the first expression threshold.

[0042] Feature screening is performed on the standardized single-cell data to identify high-variable genes (i.e., genes that are expressed significantly differently in a particular cell type than in other cell types).

[0043] A first expression threshold is set based on the overall distribution of the data or the expression characteristics of a particular cell type. The first expression threshold is used to distinguish between highly expressed and lowly expressed genes.

[0044] For each gene in the standardized single-cell data, the expression level of the gene in the target cell and in other cells is compared, and a gene whose expression level in the target cell is higher than the first expression threshold and whose expression level in other cells is lower than the first expression threshold is selected as a high-variable gene.

[0045] Step 104: Based on the single-cell data, normalization is performed on the high-variable genes to obtain a cell gene matrix.

[0046] In an embodiment of the present application, expression data of the high-variable genes in all cells is extracted to obtain high-variable gene expression data.

[0047] The high-variable gene expression data is normalized and organized into a cell gene matrix; wherein the rows of the cell gene matrix represent genes, the columns represent cells, and the values in the matrix represent the expression amount of the genes in the cells.

[0048] Here, the purpose of normalization is to eliminate the differences in expression amounts among different genes, so that the expression data of different genes can be compared on the same scale.

[0049] For example, commonly used normalization methods include Z-score standardization, min-max normalization, or quantile normalization, etc.

[0050] Step 105: Determine a target multi-classification model corresponding to the target organ of the target plant.

[0051] In order to more efficiently and accurately classify cell types, from a plurality of pre-set multi-classification models for different plant organs of different plant species, a target multi-classification model corresponding to the target organ of the target plant is determined.

[0052] Step 106: Input the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0053] In the embodiment of the present application, the cell gene matrix is transmitted as input data to the target multi-classification model, for example, a prediction function of the multi-classification model is called, and the cell gene matrix is taken as a parameter of the function; the target multi-classification model takes the received cell gene matrix as input, performs internal calculation including feature extraction, classification decision and the like, and outputs a prediction result, which includes a vector or a list of predicted cell types corresponding to the single cell data.

[0054] Through the embodiment of the present application, single cell data of a target organ of a target plant is acquired; the single cell data is preprocessed to obtain standardized single cell data; the standardized single cell data is subjected to feature screening to obtain high-variable genes of the single cell data, wherein the expression level of the high-variable genes in the target cells is higher than a first expression threshold, and the expression level of the high-variable genes in other cells is lower than the first expression threshold; the single cell data and the high-variable genes are normalized based on the single cell data to obtain a cell gene matrix; a target multi-classification model corresponding to the target organ of the target plant is determined; and the cell gene matrix is input into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model; the cell gene matrix can be input into the model for cell type classification more efficiently and more accurately, and the predicted cell type in the target organ of the target plant is identified; thereby solving the technical problem of the single cell data analysis method for plants in the related art, which has low accuracy and large limitations.

[0055] According to the plant cell type prediction method based on interpretable machine learning provided in the present application, after the cell gene matrix is input into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model, the method further includes:

[0056] The target multi-classification model is subjected to SHAP value analysis to obtain specific marker genes corresponding to the predicted cell type.

[0057] In the embodiment of the present application, based on the analysis of the feature importance of the target multi-classification model and the SHAP importance, marker genes specific to each cell type are screened.

[0058] In some embodiments, the SHAP value calculation analysis is performed on each constructed model to identify genes with high importance for distinguishing cell types, and the contribution of each gene in the prediction of a specific cell type is quantified according to the SHAP value corresponding to each cell type. The final result is output in the form of a table.

[0059] SHAP (SHapley Additive exPlanations) value analysis is a method for explaining the prediction results of machine learning models, suitable for both classification and regression tasks. Through SHAP value analysis, the contribution of each feature (here, gene) to the model's prediction results can be understood, and specific marker genes corresponding to the predicted cell type can be identified.

[0060] For example, according to the type of the target multi-classification model (such as tree-based models, linear models, etc.), a suitable SHAP explainer is selected for initialization; using the initialized SHAP explainer, SHAP value calculation is performed on the cell gene matrix. This will generate a SHAP value for each gene of each cell sample, indicating the contribution of the gene to the model's prediction results.

[0061] By analyzing the size and positive / negative of the SHAP value, specific marker genes corresponding to the predicted cell type can be identified. Generally, genes with larger and more explicit positive / negative SHAP values are more likely to be specific marker genes.

[0062] Through the embodiments of the present application, SHAP value analysis is used to explain the target multi-classification model and identify specific marker genes corresponding to the predicted cell type, which are used to understand the biological characteristics of the cell type.

[0063] According to the plant cell type prediction method based on interpretable machine learning provided by the present application, the single cell data is preprocessed to obtain standardized single cell data, including:

[0064] Determine the original single cell RNA sequencing data of the single cell data;

[0065] Perform quality control processing on the original single cell RNA sequencing data to obtain filtered single cell RNA sequencing data, wherein the quality control processing includes: removing low expression genes and removing mitochondrial genes, wherein the expression level of the low expression genes is lower than a second expression threshold;

[0066] Standardize the sequencing depth and gene expression of the filtered single cell RNA sequencing data to obtain standardized single cell data.

[0067] The second expression threshold is pre-set to distinguish between low expression genes and high expression genes, for example, by comparing the gene expression with the second expression threshold, retaining genes with expression higher than the threshold and removing low expression genes with expression lower than the threshold.

[0068] In the embodiments of the present application, data collection and preprocessing includes collecting single cell data and marker gene data of multiple plant species.

[0069] Quality control was performed on the collected raw single-cell RNA sequencing data to remove low-quality cells, lowly expressed genes, and mitochondrial genes, ensuring the reliability of the subsequent analysis data. For the filtered data, standardization was performed to exclude technical errors and make the gene expression of cells with different sequencing depths and library sizes comparable, thus better presenting the differences at the gene level.

[0070] Highly variable genes that are highly expressed in certain cells and lowly expressed in others were screened. These genes are usually most important for distinguishing cell types, so they are used for subsequent analysis. Further normalization analysis of the data was performed using highly variable genes, and the resulting matrix was used for subsequent modeling analysis.

[0071] Meanwhile, for single-cell data of different species, we integrated them at the organ level. Different sources of single-cell data of the same organ from different species were combined and batch corrected using Harmony with data source as a variable.

[0072] Finally, the integrated data was re-annotated for cell types. Based on known marker genes and cell characteristics in the literature, each cell was assigned an appropriate cell type label. This process may include manual review and manual correction to ensure the accuracy of the annotation.

[0073] According to the plant cell type prediction method based on interpretable machine learning provided by the application, before obtaining the single-cell data of the target organ of the target plant, the method further comprises:

[0074] Obtaining single-cell data of different organs of multiple plant species;

[0075] Pretreating and annotating the single-cell data to obtain a cell type label of each single-cell data;

[0076] Based on the cell type label, performing feature screening on the single-cell data to obtain a target number of highly variable genes of each single-cell data;

[0077] Based on each single-cell data and the highly variable genes of each single-cell data, performing normalization to obtain a cell gene matrix of each single cell;

[0078] Classifying the cell gene matrix of each single cell according to different organs of multiple plant species to obtain a training data set of each organ of each plant species, wherein the training data set comprises a cell gene matrix and a cell type label corresponding to the cell gene matrix;

[0079] Based on the training data set of each organ of each plant species and a target machine learning algorithm, constructing a multi-classification model corresponding to each organ of each plant species, respectively.

[0080] In embodiments of the present application, single-cell RNA sequencing data is collected from different organs of multiple plant species. These data are usually obtained by high-throughput sequencing technology and contain gene expression information of each cell.

[0081] The raw single-cell data is pre-processed, including removing noise, correcting batch effects, removing low-quality cells, etc., and each single-cell data is annotated with cell type using known gene markers or databases to obtain a cell type label for each cell. For example, by comparing gene expression profiles with characteristic gene expression profiles of known cell types.

[0082] Based on the cell type label, the single-cell data of each organ of each plant species is screened for features to identify highly variable genes (i.e., with significant expression differences between different cell types) in each organ. For example, this can be achieved by differential expression analysis, analysis of variance, etc.

[0083] According to the analysis requirements, the number of highly variable genes that need to be retained for each single-cell data is determined. The target number can be determined according to the characteristics of the data, the purpose of the analysis, and the limitations of the computing resources.

[0084] Based on each single-cell data and its highly variable genes, normalization processing is performed to eliminate differences between different gene expression amounts, making the data more suitable for subsequent machine learning algorithms. Normalization methods include but are not limited to z-score normalization, min-max normalization, etc.

[0085] The normalized data is arranged in the form of a cell gene matrix, where each row represents a gene and each column represents a cell, and the value in the matrix represents the expression amount of the gene in the cell.

[0086] According to the different organs of multiple plant species, the cell gene matrix is classified to obtain a training data set for each plant species and each organ. The data set includes a cell gene matrix and a corresponding cell type label.

[0087] Select a suitable machine learning algorithm, such as support vector machine (SVM), random forest, gradient boosting tree (GBDT), or deep learning algorithm, etc.

[0088] Using the training data set of each plant species and each organ, a multi-classification model corresponding to each organ is trained to enable the multi-classification model to predict the type of cells based on the cell gene matrix.

[0089] The performance of the model is evaluated by cross-validation, accuracy, recall rate, F1 score, etc. and the model is optimized according to the evaluation results to improve the prediction accuracy of the model.

[0090] Through the embodiments of the present application, a multi-classification model can be constructed for each organ of each plant species, thereby realizing accurate classification of single-cell data and accurate prediction of cell types.

[0091] According to the plant cell type prediction method based on an interpretable machine learning provided by the present application, before constructing a multi-classification model corresponding to each organ of each plant species respectively based on a training data set of each organ of each plant species and a target machine learning algorithm, the method further comprises:

[0092] Obtaining a test data set for plant cell type prediction, wherein the test data set comprises a test cell gene matrix and a cell type label corresponding to the test cell gene matrix;

[0093] Inputting the test cell gene matrix into a plurality of preset machine learning algorithms respectively to obtain a test predicted cell type output by each machine learning algorithm in the plurality of machine learning algorithms and a test time;

[0094] Comparing the test predicted cell type output by each machine learning algorithm with the cell type label respectively to obtain a test accuracy of each machine learning algorithm;

[0095] Determining the target machine learning algorithm based on the test time of each machine learning algorithm, the test accuracy of each machine learning algorithm, and a model robustness of each machine learning algorithm.

[0096] In the embodiments of the present application, a test data set for plant cell type prediction is collected. The test data set comprises a test cell gene matrix, i.e., gene expression data of each test cell, and a cell type label corresponding to the gene matrix, i.e., the true cell type of each test cell.

[0097] A series of preset machine learning algorithms are selected, such as a support vector machine (SVM), a random forest, a gradient boosting tree (such as XGBoost, LightGBM), a neural network (such as a deep learning model), etc. The above algorithms have been preliminarily trained or optimized.

[0098] The test cell gene matrix is input into these preset machine learning algorithms respectively. Each algorithm will make a prediction according to the input data and output a test predicted cell type.

[0099] At the same time of inputting the test data, the test time required for each algorithm to complete the prediction is recorded. The test time is used to evaluate the calculation efficiency of the algorithm and the feasibility in actual application.

[0100] The test prediction cell type output by each machine learning algorithm is compared with the cell type label in the test data set, for example, by calculating evaluation indexes such as accuracy, recall rate, F1 score, etc.; based on the comparison result, the test accuracy of each machine learning algorithm is calculated.

[0101] It should be noted that in addition to the test accuracy, the test time (i.e. the calculation efficiency) and the model robustness (i.e. the stability of the model under different data sets and conditions) of each machine learning algorithm also need to be considered. According to the comprehensive evaluation results of the test accuracy, the test time and the model robustness, the optimal target machine learning algorithm is determined.

[0102] Through the embodiments of the present application, the performance of multiple machine learning algorithms on the plant cell type prediction task can be systematically evaluated, and the target machine learning algorithm is determined by combining the comprehensive evaluation of accuracy, test time and model robustness.

[0103] According to the plant cell type prediction method based on interpretable machine learning provided by the present application, the target machine learning algorithm is LightGBM algorithm.

[0104] The following describes an example of the plant cell type prediction method based on interpretable machine learning provided by the present application in an actual application scenario, which specifically includes the following steps.

[0105] Reference Figure 2 , Figure 2 is the overall flowchart of the plant cell type prediction method based on interpretable machine learning provided by the present application, which includes data processing (Data Processing), model training (Train Model) and downstream analysis (Downstream Analysis).

[0106] As shown in Figure 2 , the data processing includes inputting single cell RNA sequencing, i.e. scRNA (Genes), then performing feature selection (Feature selection) on the input data to obtain top genes with highly variable expression, and constructing a cell gene matrix based on the top genes and cells, wherein the rows of the cell gene matrix represent genes (Genes), the columns represent cells (Cells), and the values in the matrix represent the expression amount of the genes in the cells.

[0107] The model training specifically includes the following modules: Logistic Regression, Support Vector Machine, CatBoost, Neural Networks, and LightGBM.

[0108] The downstream analysis specifically includes predicting cell types and determining marker genes.

[0109] Step 1: Data collection and preprocessing: Collect single-cell data and marker gene data for multiple plant species.

[0110] Quality control is performed on the collected raw single-cell RNA sequencing data to remove low-quality cells, lowly expressed genes, and mitochondrial genes, ensuring the reliability of the data for subsequent analysis.

[0111] For the filtered data, standardization is performed to exclude technical errors, making the gene expression of cells with different sequencing depths and library sizes comparable, and thus better presenting the differences at the gene level.

[0112] Highly variable genes that are highly expressed in some cells and lowly expressed in others are screened. These genes are usually most important for distinguishing cell types, and are therefore used for subsequent analysis.

[0113] Further normalization analysis of the data is performed using highly variable genes, and the resulting matrix is used for subsequent modeling analysis.

[0114] At the same time, single-cell data for different species is integrated at the organ level. Different sources of single-cell data for the same organ in different species are combined and batch corrected using Harmony with data source as a variable.

[0115] Finally, we re-annotate the cell types of the integrated data. According to the known marker genes and cell characteristics in the literature, we assign appropriate cell type labels to each cell. This process may include manual review and manual correction to ensure the accuracy of the annotation.

[0116] Step 2: Calculate 5000 highly variable genes for each data based on the upstream standard process as the input features of the model.

[0117] The upstream standard process here is the data collection and preprocessing of Step 1 described above.

[0118] Step 3, multi-classification learning model construction: by comparing the classification effects of multiple machine learning models, the optimal model is selected based on calculation accuracy, calculation time cost and model robustness, and finally the LightGBM algorithm is adopted, which can effectively reduce overfitting and improve the accuracy and generalization ability of the model. Model construction for cell type prediction for each organ of each species.

[0119] Reference Figure 3 , Figure 3 is a comparison diagram of the prediction effect of the model based on the LightGBM algorithm provided by the application, which includes comparing the area under the curve (AUC) of the model between different cell types (Model AUC Comparison in each Cell Type). The abscissa includes: Seurat prediction (prediction_Seurat), singleR prediction (prediction_singleR), scPred prediction (prediction_scPred) and scMarker prediction (prediction_scMarker).

[0120] As Figure 3 shown, the cell types include: root endodermis, sclerenchyma, pericycle, root cortex, meristematic cell, exodermis, root hair, root epidermis, root cap, root stele, protoxylem, phloem and metaxylem.

[0121] meristematic cell, exodermis, root hair, epidermal cell, root cap, main axis and secondary xylem and phloem

[0122] Modeling the single-cell data (matrix of cells x genes) of each organ of each species, classifying to obtain the cell type of each cell (for example: epidermis, stele, outer epidermis and other cell types), and the marker genes of each cell type in different organs are obtained according to the importance of multi-classification model feature sorting, and the obtained genes can be used for downstream experimental verification and other work.

[0123] It should be noted that the single cell technology is applied to a small number of species, about 30 species, and only a small number of species (rice, corn, Arabidopsis) have been studied in multiple organs, and most of the other species have only one organ. In the embodiments of the present application, 25 species and about 2.5 million cells are involved. "Each species" referred to here refers to the collected species data (i.e. a total of 25).

[0124] Step 4, marker gene mining: based on the analysis of model-based feature importance and SHAP importance, the marker genes specific to each cell type are screened out.

[0125] SHAP value calculation analysis is performed on each constructed model to identify genes with high importance for cell type differentiation, and the contribution of each gene in the prediction of a specific cell type is quantified according to the SHAP value corresponding to each cell type. The final result is output in the form of a table to the local environment.

[0126] Through the above method, the present application can realize the prediction of cell types of multiple organs of multiple plants, and has the mining of cell type-specific marker genes and good cross-species applicability, providing a new technical means for plant single cell research.

[0127] The plant cell type prediction device based on interpretable machine learning provided by the present application is described below, and the plant cell type prediction device based on interpretable machine learning described below can be mutually corresponding to the plant cell type prediction method based on interpretable machine learning described above.

[0128] Reference Figure 4 , Figure 4 is a structural schematic diagram of the plant cell type prediction device based on interpretable machine learning provided by the present application.

[0129] The acquisition module 401 is configured to acquire single cell data of a target organ of a target plant;

[0130] The preprocessing module 402 is configured to preprocess the single cell data to obtain standardized single cell data;

[0131] The screening module 403 is configured to perform feature screening on the standardized single cell data to obtain high variable genes of the single cell data, wherein the expression level of the high variable genes in the target cell is higher than a first expression threshold, and the expression level of the high variable genes in other cells is lower than the first expression threshold;

[0132] The normalization module 404 is configured to normalize the single cell data and the high variable genes based on the single cell data and the high variable genes to obtain a cell gene matrix;

[0133] The determination module 405 is configured to determine a target multi-classification model corresponding to the target organ of the target plant;

[0134] The prediction module 406 is configured to input the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0135] Specifically, the plant cell type prediction device based on the interpretable machine learning provided by the present application can realize all the method steps of the plant cell type prediction method based on the interpretable machine learning, and achieve the same technical effects. The same parts and beneficial effects of the method embodiment in this embodiment will not be described in detail here.

[0136] Figure 5 is the entity structure schematic diagram of the electronic device provided by the present application, as shown in Figure 5 The electronic device can include a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other through the communications bus 540. The processor 510 can call the logical instructions in the memory 530 to execute the plant cell type prediction method based on the interpretable machine learning, which includes: obtaining single cell data of a target organ of a target plant; preprocessing the single cell data to obtain standardized single cell data; performing feature screening on the standardized single cell data to obtain high variable genes of the single cell data, wherein the expression level of the high variable genes in the target cell is higher than a first expression threshold, and the expression level of the high variable genes in other cells is lower than the first expression threshold; performing normalization on the single cell data and the high variable genes to obtain a cell gene matrix; determining a target multi-classification model corresponding to the target organ of the target plant; inputting the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0137] In addition, the logic instructions in the memory 530 described above can be implemented in the form of a software functional unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0138] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the plant cell type prediction method based on the interpretable machine learning provided by the above-mentioned methods, the method comprising: obtaining single cell data of a target organ of a target plant; preprocessing the single cell data to obtain standardized single cell data; performing feature screening on the standardized single cell data to obtain high variable genes of the single cell data, wherein the expression level of the high variable genes in the target cells is higher than a first expression threshold, and the expression level of the high variable genes in other cells is lower than the first expression threshold; performing normalization on the single cell data and the high variable genes to obtain a cell gene matrix; determining a target multi-classification model corresponding to the target organ of the target plant; inputting the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0139] In another aspect, the present application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the plant cell type prediction method based on the interpretable machine learning provided by the above-mentioned methods, the method comprising: obtaining single cell data of a target organ of a target plant; preprocessing the single cell data to obtain standardized single cell data; performing feature screening on the standardized single cell data to obtain high variable genes of the single cell data, wherein the expression level of the high variable genes in the target cells is higher than a first expression threshold, and the expression level of the high variable genes in other cells is lower than the first expression threshold; performing normalization on the single cell data and the high variable genes to obtain a cell gene matrix; determining a target multi-classification model corresponding to the target organ of the target plant; inputting the cell gene matrix into the target multi-classification model to obtain a predicted cell type output by the target multi-classification model.

[0140] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0141] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0142] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A plant cell type prediction method based on interpretable machine learning, characterized in that: include: Acquire single-cell data of target organs of target plants; Preprocessing the single-cell data to obtain standardized single-cell data; Performing feature screening on the standardized single-cell data to obtain highly variable genes in the single-cell data, wherein the expression level of the highly variable genes in the target cells is higher than a first expression threshold, and the expression level of the highly variable genes in other cells is lower than the first expression threshold; Normalizing the single cell data and the highly variable genes to obtain a cell gene matrix; determining a target multi-classification model corresponding to a target organ of the target plant; The cell gene matrix is ​​input into the target multi-classification model to obtain the predicted cell type output by the target multi-classification model.

2. The plant cell type prediction method based on interpretable machine learning according to claim 1, characterized in that After inputting the cell gene matrix into the target multi-classification model to obtain the predicted cell type output by the target multi-classification model, the method further includes: SHAP value analysis is performed on the target multi-classification model to obtain specific marker genes corresponding to the predicted cell types.

3. The plant cell type prediction method based on interpretable machine learning according to claim 1, characterized in that The preprocessing of the single cell data to obtain standardized single cell data includes: Determine original single-cell RNA sequencing data of the single-cell data; Performing quality control processing on the raw single-cell RNA sequencing data to obtain filtered single-cell RNA sequencing data, wherein the quality control processing includes: removing low-expression genes and removing mitochondrial genes, wherein the expression level of the low-expression genes is lower than a second expression threshold; The sequencing depth and gene expression level of the filtered single-cell RNA sequencing data are normalized to obtain standardized single-cell data.

4. The plant cell type prediction method based on interpretable machine learning according to claim 1, characterized in that Before obtaining the single cell data of the target organ of the target plant, the method further includes: Obtain single-cell data from different organs of multiple plant species; Preprocessing and annotating the single-cell data to obtain a cell type label for each single-cell data; Performing feature screening on the single-cell data based on the cell type label to obtain a target number of highly variable genes in each single-cell data; Normalizing each single cell data and the highly variable genes of each single cell data to obtain a cell gene matrix for each single cell; Classifying the cell gene matrix of each single cell according to different organs of the multiple plant species to obtain a training data set for each organ of each plant species, wherein the training data set includes: the cell gene matrix and a cell type label corresponding to the cell gene matrix; Based on the training data set of each organ of each plant species and the target machine learning algorithm, a multi-classification model corresponding to each organ of each plant species is constructed.

5. The plant cell type prediction method based on interpretable machine learning according to claim 4, characterized in that: Before constructing a multi-classification model corresponding to each organ of each plant species based on the training data set of each organ of each plant species and the target machine learning algorithm, the method further includes: Acquire a test data set for plant cell type prediction, wherein the test data set includes a test cell gene matrix and a cell type label corresponding to the test cell gene matrix; Inputting the test cell gene matrix into a plurality of preset machine learning algorithms respectively to obtain the test predicted cell type and test time output by each of the plurality of machine learning algorithms; Comparing the test predicted cell types output by each machine learning algorithm with the cell type labels to obtain the test accuracy of each machine learning algorithm; A target machine learning algorithm is determined based on the test time of each machine learning algorithm, the test accuracy of each machine learning algorithm, and the model robustness of each machine learning algorithm.

6. The plant cell type prediction method based on interpretable machine learning according to claim 4, characterized in that: The target machine learning algorithm is the LightGBM algorithm.

7. A plant cell type prediction device based on interpretable machine learning, characterized in that: include: an acquisition module, used for acquiring single-cell data of a target organ of a target plant; A preprocessing module, configured to preprocess the single-cell data to obtain standardized single-cell data; a screening module, configured to perform feature screening on the standardized single-cell data to obtain highly variable genes in the single-cell data, wherein the expression level of the highly variable genes in the target cells is higher than a first expression threshold, and the expression level of the highly variable genes in other cells is lower than the first expression threshold; a normalization module, configured to perform normalization based on the single cell data and the highly variable genes to obtain a cell gene matrix; a determination module, configured to determine a target multi-classification model corresponding to a target organ of the target plant; A prediction module is used to input the cell gene matrix into the target multi-classification model to obtain the predicted cell type output by the target multi-classification model.

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 plant cell type prediction method based on interpretable machine learning as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the plant cell type prediction method based on interpretable machine learning as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the plant cell type prediction method based on interpretable machine learning as claimed in any one of claims 1 to 6 is implemented.

Images