Single-cell transcriptome data preprocessing method, electronic device, and storage medium

Abstract

The present application provides a single-cell transcriptome data preprocessing method, an electronic device, and a storage medium. The method comprises: acquiring a first data set uploaded by a user and generated based on single-cell omics; filtering the first data set on the basis of metric data used for filtering cells, so as to obtain a second data set, wherein the metric data comprises threshold ranges respectively corresponding to UMI count, feature gene count, mitochondrial gene percentage, and β-actin expression; calling a batch effect removal tool to perform batch effect removal on the second data set to obtain a third data set; and performing doublet removal on the third data set by means of a DoubletFinder tool so as to obtain a result data set obtained by preprocessing. The present solution supports uploading and preprocessing of data in various formats such as tsv, txt, csv, RDS, and HDF5, and can improve the efficiency of preprocessing and the validity of processed data sets.

Description

Single-cell transcriptome data preprocessing methods, electronic devices and storage media Technical Field

[0001] This invention relates to the field of cell data processing technology, and more specifically, to a method for preprocessing single-cell transcriptome data, an electronic device, and a storage medium. Background Technology

[0002] In multicellular organisms, differences typically exist between cells, and these differences vary between different cell populations. These differences are not only reflected in morphology but also in genetic information, such as genomic information and gene expression levels. With the deepening and refinement of single-cell RNA sequencing (scRNA-seq) applications, it is often necessary to perform single-cell sequencing on complex organs; simply sequencing a few cells no longer meets research needs. In other words, large-scale single-cell RNA sequencing has become a powerful way to break down the heterogeneity of individual cells. Currently, although large-scale single-cell RNA sequencing analysis platforms exist (such as GranatumX and Cellxgene), their functions are relatively limited, lacking preprocessing of single-cell transcriptome data, thus affecting the flexibility and accuracy of single-cell RNA sequencing. For example, existing analysis platforms lack batch effect elimination and double-cell knockout functions, and cell filtering is inflexible.

[0003] Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method, electronic device and storage medium for preprocessing single-cell transcriptome data, which can improve the existing analysis platform's lack of batch effect elimination, double cell removal and other functions, as well as the problem of inflexible cell filtering in the preprocessing of single-cell transcriptome data.

[0005] To achieve the above technical objectives, the technical solution adopted in this application is as follows:

[0006] In a first aspect, embodiments of this application provide a method for preprocessing single-cell transcriptome data, the method comprising:

[0007] Obtain the first dataset generated based on single-cell omics uploaded by the user. The data formats in the first dataset include tsv, txt, csv, RDS and HDF5 formats.

[0008] Upon receiving a preprocessing instruction, the first dataset is filtered based on the user-defined indicator data for filtering cells to obtain a second dataset. The indicator data includes a first threshold range for UMI counts, a second threshold range for feature gene counts, a third threshold range for mitochondrial gene percentages, and a fourth threshold range corresponding to β-action expression.

[0009] The batch effect elimination tool was used to eliminate batch effects on the second dataset to obtain the third dataset.

[0010] Using the DoubletFinder tool, the third dataset is subjected to double-cell removal to obtain the fourth dataset, which is then used as the preprocessed result dataset.

[0011] In conjunction with the first aspect, in some optional implementations, the first dataset is filtered based on user-defined indicator data for filtering cells to obtain a second dataset, including:

[0012] Based on the first threshold range of UMI counts set by the user via the first bar slide button, filter out data in the first dataset whose UMI counts are not within the first threshold range;

[0013] Based on the second threshold range of the feature gene count set by the user via the second bar slider button, data in the first dataset whose feature gene count is not within the second threshold range are filtered out;

[0014] Based on the third threshold range of the percentage of mitochondrial genes set by the user via the third bar slider button, data in the first dataset whose percentage of mitochondrial genes is not within the third threshold range are filtered out;

[0015] Based on the fourth threshold range corresponding to the β-action expression set by the user through the fourth bar slide button, data in the first dataset whose β-action expression is not within the fourth threshold range are filtered out;

[0016] The first dataset, filtered by UMI count, feature gene count, mitochondrial gene percentage, and β-action expression, is used as the second dataset.

[0017] In conjunction with the first aspect, in some optional implementations, the third dataset is subjected to double-cell removal using the DoubletFinder tool to obtain a fourth dataset, including:

[0018] The DoubletFinder tool is used to randomly fuse artificial twin cells from single-cell data uploaded by the user in advance.

[0019] The simulated twin cells and the cells in the third dataset are mixed to obtain mixed cell data;

[0020] The proportion of artificial k nearest neighbor pANNs for each unit is found using PCA dimensionality reduction or PCA distance matrix in the DoubletFinder tool, where each unit is a bin divided from the mixed cell data, and each bin includes multiple feature genes.

[0021] The mixed cell data is sorted based on a preset number of doublets, and a threshold for the pANN value is determined.

[0022] Based on the threshold of the pANN value, twin-cell data are determined from the mixed cell data, and the twin-cell data are filtered out.

[0023] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0024] The resulting dataset is classified and statistically analyzed according to preset categories, and the classification results are displayed on a page. The preset categories include the number of cells and the number of feature genes.

[0025] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0026] Based on a pre-created temporary project repository, the first dataset, the second dataset, the third dataset, the result dataset, and the classification results are stored.

[0027] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0028] When a query instruction for retrieving the second dataset is received, the second dataset is retrieved from the temporary project repository and displayed in a violin diagram format.

[0029] In conjunction with the first aspect, in some alternative implementations, the method further includes:

[0030] Upload your personalized genome to a temporary project repository using a pre-created upload interface;

[0031] Alternatively, a specified genome can be deleted from the temporary project repository based on a pre-created deletion interface.

[0032] In conjunction with the first aspect, in some alternative implementations, the batch effect elimination tool includes any one of RPCA, FastMNN, Harmony, scVI, and svANVI in Seurat v4.

[0033] Secondly, embodiments of this application also provide an electronic device, which includes a processor and a memory coupled to each other. The memory stores a computer program, and when the computer program is executed by the processor, the electronic device performs the above-described method.

[0034] Thirdly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the above-described method.

[0035] The invention employing the above technical solution has the following advantages:

[0036] The technical solution provided in this application supports the uploading and processing of data in multiple formats, including TSV, TXT, CSV, RDS, and HDF5, for datasets generated from single-cell omics. This solution integrates multiple preprocessing methods, enabling functions such as cell filtering, batch effect elimination, and double-cell removal, thus improving preprocessing efficiency. Users can flexibly set the indicator data used for filtering cells to achieve flexible cell filtering. Furthermore, the batch effect elimination tool allows for batch effect elimination of multiple datasets, improving the dataset's effectiveness. The DoubletFinder tool enables double-cell removal, further enhancing the dataset's effectiveness and reliability. Attached Figure Description

[0037] This application can be further illustrated by the non-limiting embodiments given in the accompanying drawings. It should be understood that the following drawings only illustrate some embodiments of this application and should not be considered as limiting the scope. For those skilled in the art, other related drawings can be obtained from these drawings without any inventive effort.

[0038] Figure 1 is a flowchart illustrating the single-cell transcriptome data preprocessing method provided in the embodiments of this application.

[0039] Figure 2 is a schematic diagram of the bar-shaped sliding button provided in an embodiment of this application.

[0040] Figure 3 is a schematic diagram of the classification results obtained after the result dataset provided in the embodiment of this application has been classified and statistically analyzed. Detailed Implementation

[0041] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that similar or identical parts are referred to by the same reference numerals in the drawings or description. Implementations not shown or described in the drawings are forms known to those skilled in the art. In the description of this application, terms such as "first" and "second" are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0042] This application provides an electronic device. The electronic device may include a processing module and a storage module. The storage module stores a computer program, which, when executed by the processing module, enables the electronic device to perform the corresponding steps in the following single-cell transcriptome data preprocessing method.

[0043] The electronic device can be, but is not limited to, a personal computer or a network server. The electronic device may be equipped with an analysis platform for preprocessing single-cell transcriptome data, or the analysis platform may be deployed on a cloud server, and the electronic device can access the analysis platform via the web to perform preprocessing of single-cell transcriptome data.

[0044] Understandably, this analysis platform can be used to preprocess datasets from single-cell RNA sequencing (scRNA-seq) to filter out interfering data and improve the validity of the datasets. Single-cell transcriptome data, which is a dataset generated based on single-cell omics, can be called scRNA-seq data. The preprocessed dataset is typically used for subsequent cell analysis (such as cell type annotation, isotype intercellular communication, etc.), thus improving the accuracy and reliability of subsequent cell analyses.

[0045] Referring to Figure 1, this application also provides a method for preprocessing single-cell transcriptome data, which can be applied to the aforementioned electronic device, and the electronic device executes or implements the steps of the method. The single-cell transcriptome data preprocessing method may include the following steps:

[0046] Step 110: Obtain the first dataset generated based on single-cell omics uploaded by the user. The data formats in the first dataset include tsv, txt, csv, RDS and HDF5 formats.

[0047] Step 120: Upon receiving the preprocessing instruction, the first dataset is filtered based on the indicator data set by the user for filtering cells to obtain the second dataset. The indicator data includes a first threshold range for UMI counts, a second threshold range for feature gene counts, a third threshold range for mitochondrial gene percentages, and a fourth threshold range corresponding to β-action expression.

[0048] Step 130: Use the batch effect elimination tool to perform batch effect elimination on the second dataset to obtain the third dataset;

[0049] Step 140: Using the DoubletFinder tool, perform doublet removal on the third dataset to obtain the fourth dataset, and use the fourth dataset as the preprocessed result dataset.

[0050] The steps of the single-cell transcriptome data preprocessing method will be described in detail below:

[0051] In step 110, the user can upload the first dataset containing data in multiple formats to the analysis platform via an electronic device. Understandably, regarding data upload, the user can customize the creation of project folders and the relevant information for entering data, loading various data and formats generated by single-cell transcriptomics. For example, data formats may include tab-delimited values / text (tsv / txt) format and comma-delimited values ​​(csv) format downloaded directly from the GEO database, the normal RDS format from the Seurat package as input, and the Hd5 format generated from the 10x Cellranger package. That is, the electronic device can be compatible with the upload and preprocessing of single-cell transcriptomics data in multiple formats.

[0052] Referring to Figure 2, in step 120, the first dataset is filtered based on the user-defined indicator data for filtering cells to obtain the second dataset, which may include:

[0053] Based on the first threshold range of UMI counts set by the user via the first bar slide button, filter out data in the first dataset whose UMI counts are not within the first threshold range;

[0054] Based on the second threshold range of the feature gene count set by the user via the second bar slider button, data in the first dataset whose feature gene count is not within the second threshold range are filtered out;

[0055] Based on the third threshold range of the percentage of mitochondrial genes set by the user via the third bar slider button, data in the first dataset whose percentage of mitochondrial genes is not within the third threshold range are filtered out;

[0056] Based on the fourth threshold range corresponding to the β-action expression set by the user through the fourth bar slide button, data in the first dataset whose β-action expression is not within the fourth threshold range are filtered out;

[0057] The first dataset, filtered by UMI count, feature gene count, mitochondrial gene percentage, and β-action expression, is used as the second dataset.

[0058] Referring to Figure 2, the first slider button (UMI Count), the second slider button (Gene Count), the third slider button [MT-Genes(%)], and the fourth slider button (ACTB Activity) can be as shown in Figure 2. Different lengths of a single slider button correspond to different values. Users can drag one or both ends of the slider button to flexibly select the corresponding threshold range. In step 120, all indicator data (such as the first threshold range, second threshold range, third threshold range, and fourth threshold range) can be flexibly set by the user by dragging the corresponding slider button. That is, users can independently set the corresponding threshold range according to their individual needs, thus improving the flexibility of threshold setting and the flexibility of cell data filtering.

[0059] In this embodiment, the first threshold range can be set based on the total UMI count in each cell, and the second threshold range can be set based on the number of characteristic gene counts. Using a slider button, the user can select an appropriate range for the maximum values ​​of "UMI Count" and "Gene Count" for each cell, serving as the corresponding threshold range. By utilizing the first and second threshold ranges, low-quality cells can be filtered out.

[0060] In this embodiment, by using the corresponding bar slider button, the user can select an appropriate percentage range of mitochondrial genes among the total detected genes, and the relative expression level of β-actin, which is typically highly expressed in living cells, as the third and fourth threshold ranges, respectively. Using the third and fourth threshold ranges, potentially dead cells in the first dataset can be filtered out.

[0061] Once the user sets the appropriate threshold range, the electronic device can automatically filter out data in the first dataset that is not within the threshold range, thus obtaining the second dataset.

[0062] Understandably, raw single-cell transcriptome data acquired at different times, by different operators, using different reagents, and with different instruments (i.e., the first dataset may contain data from multiple batches) may lead to experimental errors, which are reflected in gene expression levels as batch effects. Single-cell transcriptome data from different batches typically contain some error; directly using this data for downstream analysis without any processing may affect the accuracy of the results. In step 130, by using a batch effect elimination tool to eliminate batch effects on the first dataset, the validity of the data can be improved, while a valid second dataset is beneficial for improving the accuracy and reliability of subsequent cell analyses.

[0063] In step 130, the batch effect elimination tool can include any one of RPCA, FastMNN, Harmony, scVI, and svANVI in Seurat v4. When batch effect elimination is required, users can flexibly select the appropriate batch effect elimination tool according to the actual situation; of course, electronic devices can also automatically select the appropriate batch effect elimination tool. The batch effect elimination tool can be encapsulated in the Nextflow analysis workflow to achieve simplified and efficient data processing.

[0064] Understandably, single-cell RNA sequencing expects only one real cell per barcode tag. However, in actual data, there may be two or more cells sharing a single barcode. A twin cell refers to a droplet or microwell containing two or more cells. Cells within the same droplet carry the same cell barcode in subsequent analysis, thus being mistaken for a single cell (pseudo-cell). Therefore, a twin cell is not only a mixture of two cells but may also be multiple cells confused as one. The main characteristic of these pseudo-cells is that the number of UMIs and genes detected is often twice or more than that of normal cells. The presence of twin cells can significantly impact cell analysis results, such as affecting cell type identification.

[0065] In this embodiment, the DoubletFinder tool in step 140 can be used to remove twins from the third dataset to reduce the impact of twins on subsequent cell analysis results.

[0066] Step 140, which uses the DoubletFinder tool to remove twins from the third dataset to obtain the fourth dataset, may include:

[0067] The DoubletFinder tool is used to randomly fuse artificial twin cells from single-cell data uploaded by the user in advance.

[0068] The simulated twin cells and the cells in the third dataset are mixed to obtain mixed cell data;

[0069] The proportion of artificial k nearest neighbor pANNs for each unit is found using PCA dimensionality reduction or PCA distance matrix in the DoubletFinder tool, where each unit is a bin divided from the mixed cell data, and each bin includes multiple feature genes.

[0070] The mixed cell data is sorted based on a preset number of doublets, and a threshold for the pANN value is determined.

[0071] Based on the threshold of the pANN value, twin-cell data are determined from the mixed cell data, and the twin-cell data are filtered out.

[0072] Understandably, DoubletFinder can generate artificially simulated doublets and incorporate them into the original single-cell expression data. In principle, these simulated doublets will be closer to the real doublets in the third dataset. Based on this, by calculating the proportion of artificially simulated doublets in each cell's K nearest neighbor cells (pANN), the probability of doublets in each sample of the third dataset can be ranked according to the pANN value. Furthermore, based on the statistical principle of the Poisson distribution, the number of doublets in each sample can be calculated. Combining this with the previous cell pANN value ranking, doublet filtering can be achieved.

[0073] The preset number of doublets and the threshold for pANN values ​​can be flexibly set according to the actual situation. The feature genes included in each bin can be divided from the mixed cell data according to the actual situation.

[0074] Referring to Figure 3, as an optional implementation, the method may further include:

[0075] The resulting dataset is classified and statistically analyzed according to preset categories, and the classification results are displayed on a page. The preset categories include the number of cells and the number of feature genes.

[0076] Understandably, in Figure 3, "Used.Cells" refers to the number of cells after preprocessing, and "Used.Features" refers to the number of feature genes. After filtering and other preprocessing, the electronic device can interactively display the statistical results after preprocessing, allowing users to view the results data more intuitively.

[0077] As an optional implementation, the method may further include:

[0078] Based on a pre-created temporary project repository, the first dataset, the second dataset, the third dataset, the result dataset, and the classification results are stored.

[0079] Understandably, by providing users with a temporary project repository, datasets generated at each stage of the preprocessing process can be stored, allowing users to easily view or retrieve the datasets from the temporary project repository as needed.

[0080] As an optional implementation, the method may further include:

[0081] When a query instruction for retrieving the second dataset is received, the second dataset is retrieved from the temporary project repository and displayed in a violin diagram format.

[0082] Understandably, when a user needs to view data in a temporary project repository, they can enter a query command through the web interface of their electronic device. The device can then retrieve the desired dataset from the temporary project repository via an analytics platform and display it as a violin plot. For example, a user can view violin plots of UMI counts and feature counts, as well as violin plots of mitochondrial gene percentages and β-action expression, in the second dataset through the plotting and visualization area on the web page.

[0083] As an optional implementation, the method may further include:

[0084] Upload your personalized genome to a temporary project repository using a pre-created upload interface;

[0085] Alternatively, a specified genome can be deleted from the temporary project repository based on a pre-created deletion interface.

[0086] Understandably, the upload and delete interfaces can provide users with the ability to upload personalized genomes for genome deletion, thus facilitating downstream cell analysis.

[0087] It should be noted that in electronic devices, the processing module can be an integrated circuit chip with signal processing capabilities. The aforementioned processing module can be a general-purpose processor. For example, the processor can be a Central Processing Unit (CPU), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0088] The storage module can be, but is not limited to, random access memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, etc. In this embodiment, the storage module can be used to store a first dataset, a second dataset, a third dataset, a result dataset, etc. Of course, the storage module can also be used to store a program, which the processing module executes after receiving an execution instruction.

[0089] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the electronic device described above can be referred to the corresponding steps in the aforementioned method, and will not be elaborated further here.

[0090] This application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the single-cell transcriptome data preprocessing method as described in the above embodiments.

[0091] Based on the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product. This software product can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, electronic device, or network device, etc.) to execute the methods described in the various implementation scenarios of this application.

[0092] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system and method embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or part of code, which includes one or more executable instructions for implementing a specified logical function. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0093] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for preprocessing single-cell transcriptome data, characterized in that, The method includes: Obtain the first dataset generated based on single-cell omics uploaded by the user. The data formats in the first dataset include tsv, txt, csv, RDS and HDF5 formats. Upon receiving a preprocessing instruction, the first dataset is filtered based on the user-defined indicator data for filtering cells to obtain a second dataset. The indicator data includes a first threshold range for UMI counts, a second threshold range for feature gene counts, a third threshold range for mitochondrial gene percentages, and a fourth threshold range corresponding to β-action expression. The batch effect elimination tool was used to eliminate batch effects on the second dataset to obtain the third dataset. Using the DoubletFinder tool, the third dataset is subjected to double-cell removal to obtain the fourth dataset, which is then used as the preprocessed result dataset.

2. The method according to claim 1, characterized in that, Based on the user-defined cellular indicator data, the first dataset is filtered to obtain the second dataset, which includes: Based on the first threshold range of UMI counts set by the user via the first bar slide button, filter out data in the first dataset whose UMI counts are not within the first threshold range; Based on the second threshold range of the feature gene count set by the user via the second bar slider button, data in the first dataset whose feature gene count is not within the second threshold range are filtered out; Based on the third threshold range of the percentage of mitochondrial genes set by the user via the third bar slider button, data in the first dataset whose percentage of mitochondrial genes is not within the third threshold range are filtered out; Based on the fourth threshold range corresponding to the β-action expression set by the user through the fourth bar slide button, data in the first dataset whose β-action expression is not within the fourth threshold range are filtered out; The first dataset, filtered by UMI count, feature gene count, mitochondrial gene percentage, and β-action expression, is used as the second dataset.

3. The method according to claim 1, characterized in that, Using the DoubletFinder tool, the third dataset was subjected to double-cell removal to obtain the fourth dataset, which includes: The DoubletFinder tool is used to randomly fuse artificial twin cells from single-cell data uploaded by the user in advance. The simulated twin cells and the cells in the third dataset are mixed to obtain mixed cell data; The proportion of artificial k nearest neighbor pANNs for each unit is found using PCA dimensionality reduction or PCA distance matrix in the DoubletFinder tool, where each unit is a bin divided from the mixed cell data, and each bin includes multiple feature genes. The mixed cell data is sorted based on a preset number of doublets, and a threshold for the pANN value is determined. Based on the threshold of the pANN value, twin-cell data are determined from the mixed cell data, and the twin-cell data are filtered out.

4. The method according to claim 1, characterized in that, The method further includes: The resulting dataset is classified and statistically analyzed according to preset categories, and the classification results are displayed on a page. The preset categories include the number of cells and the number of feature genes.

5. The method according to claim 4, characterized in that, The method further includes: Based on a pre-created temporary project repository, the first dataset, the second dataset, the third dataset, the result dataset, and the classification results are stored.

6. The method according to claim 5, characterized in that, The method further includes: When a query instruction for retrieving the second dataset is received, the second dataset is retrieved from the temporary project repository and displayed in a violin diagram format.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Upload your personalized genome to a temporary project repository using a pre-created upload interface; Alternatively, a specified genome can be deleted from the temporary project repository based on a pre-created deletion interface.

8. The method according to any one of claims 1-6, characterized in that, The batch effect elimination tools include any one of RPCA, FastMNN, Harmony, scVI, and svANVI in Seurat v4.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory coupled together, the memory storing a computer program that, when executed by the processor, causes the electronic device to perform the method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1-8.

Images