Cell ranger single cell rna sequencing data analysis method

Abstract

The present application relates to the technical field of single cell RNA sequencing, and discloses a CellRanger single cell RNA sequencing data analysis method, which comprises hierarchical processing of cell quality screening and a tolerant mode of gene expression screening. The method provided by the present application not only strictly guarantees the integrity of various functional indicators in the cells finally stored in the library, but also retains part of the cell information which may have biological significance due to the loss of candidate markers caused by low overall transcriptional activity through further judgment of the low expression state of the cells to be determined. Through this method, the reliability of subsequent analysis data can be ensured, and the information of low expression or rare cells missed due to strict one-size-fits-all standards can be reduced as much as possible.

Description

Technical Field

[0001] This invention relates to the field of single-cell RNA sequencing technology, specifically the CellRanger single-cell RNA sequencing data analysis method. Background Technology

[0002] The primary goal of single-cell RNA sequencing is to analyze gene expression at the individual cell level, revealing the heterogeneity and dynamic changes within a cell population. By analyzing each cell independently, this technology can not only identify different cell subtypes and rare cells but also provide a deeper understanding of the role of cells in development, differentiation, immune responses, and disease pathogenesis. Compared to population RNA sequencing, single-cell RNA sequencing provides much more refined information.

[0003] CellRanger is a complete software suite for processing and analyzing single-cell RNA sequencing data, designed to automate the conversion from raw sequencing data to single-cell expression matrices. The tool first uses mkfastq to convert BCL files to FASTQ files, then uses the count module for alignment, cell barcode recognition, and UMI error correction to accurately separate and quantify gene transcripts within each cell.

[0004] While CellRanger enables one-click, automated processing of single-cell RNA sequencing data, its stringent quality control and filtering strategies may result in the loss of information from some low-expression or rare cells, thus affecting the discovery and in-depth analysis of certain cell subpopulations. Summary of the Invention

[0005] To address the problems existing in the prior art, the purpose of this invention is to provide a CellRanger single-cell RNA sequencing data analysis method, which can improve the quality control and filtering strategies of CellRanger.

[0006] To achieve the above objectives, this invention provides the following technical solution: a CellRanger single-cell RNA sequencing data analysis method, comprising: Step 1, converting the original BCL data to FASTQ format and splitting the sequencing data; after splitting, identifying cell barcodes and unique molecular markers to filter low-quality reads; after filtering, summarizing multiple reads of the same gene to obtain the gene-cell expression matrix; after obtaining the expression matrix, using cell call identification to obtain the effective cell barcode; Step 2, obtaining the UMI number of all genes and the UMI number of mitochondrial genes in each cell; dividing the UMI number of mitochondrial genes by the UMI number of all genes to obtain the mitochondrial gene ratio; simultaneously, setting a mitochondrial gene ratio range threshold; comparing the mitochondrial gene ratio with the mitochondrial gene ratio range threshold; and based on the comparison results, identifying high-quality cells and low-quality cells; Step 3, for high-quality cells, selecting five candidate genes based on predefined annotations; detecting the presence of the five candidate genes in each high-quality cell and marking them; after marking, screening is performed; for a single high-quality cell, all five markers must be present simultaneously to be considered a high-quality cell; all other cases are considered non-high-quality cells.

[0007] In some implementations, the specific process for determining high-quality and low-quality cells based on comparison results is as follows: if the proportion of mitochondrial genes falls within the threshold range of mitochondrial gene proportions or is less than the minimum value of the threshold range, it indicates that the RNA originates from nuclear genes. In this case, the cell is normal in transcriptional activity and is marked as a high-quality cell. If the proportion of mitochondrial genes is greater than the maximum value of the threshold range, it indicates that the cell is under stress or undergoing apoptosis. In this case, it is marked as a low-quality cell.

[0008] In some implementations, for low-quality cells, a second threshold for the mitochondrial gene ratio is set, which is greater than the maximum value of the threshold range for the mitochondrial gene ratio. Under this condition, the mitochondrial gene ratio of low-quality cells is compared again with the second threshold for the mitochondrial gene ratio, and different responses are given based on the comparison results.

[0009] In some implementations, if the proportion of mitochondrial genes is less than or equal to a second threshold for the proportion of mitochondrial genes, a second assessment is performed; if the proportion of mitochondrial genes is greater than the second threshold for the proportion of mitochondrial genes, the assessment that the cell is of low quality is maintained.

[0010] In some implementations, during the reassessment, the number of gene types in low-quality cells is obtained. At the same time, a threshold for the number of gene types is set, and the number of gene types is compared with the threshold. If the number of gene types is greater than or equal to the threshold, it means that the cell has rich gene expression and is a healthy cell. In this case, the low-quality cell is remarked as a high-quality cell. If the number of gene types is less than the threshold, it means that the cell has poor gene expression and is a cell in the process of degradation. In this case, the judgment that it is a low-quality cell is maintained.

[0011] In some implementations, a threshold is set for the number of gene types that is less than the threshold for the number of gene types. When the number of gene types in a cell is less than the threshold, the number of gene types in the cell is compared with the threshold again. If the number of gene types in the cell is greater than or equal to the threshold, it means that the cell has a low degree of gene expression, and further judgment is made in this case. If the number of gene types in the cell is less than the threshold, it means that the cell has a high degree of gene expression, and the judgment that it is a low-quality cell is maintained.

[0012] In some implementations, further determination involves obtaining the number of ribosomal gene UMIs in the cell, dividing the number of ribosomal gene UMIs by the total number of UMIs of all genes to obtain the ribosomal gene ratio, and simultaneously setting a ribosomal gene ratio range threshold. The cell's ribosomal gene ratio is compared with the ribosomal gene ratio range threshold. If the cell's ribosomal gene ratio falls within the ribosomal gene ratio range threshold, it indicates that the cell is in a normal state, and in this case, low-quality cells are re-marked as high-quality cells. If the cell's ribosomal gene ratio is greater than the maximum value of the ribosomal gene ratio range threshold or less than the minimum value of the ribosomal gene ratio range threshold, it indicates that the cell has an imbalance in transcription or translation, and in this case, the determination that it is a low-quality cell is maintained.

[0013] In some implementations, among all non-compliant high-quality cells, at least one labeled high-quality cell is identified and designated as a pending high-quality cell. A high-quality cell without a label is designated as a non-pending high-quality cell. Non-pending high-quality cells are directly excluded. For pending high-quality cells, the method for determining whether they are low-expression cells is as follows: multiple cells are randomly selected from the compliant high-quality cells. The gene types in each of the selected compliant high-quality cells are obtained and averaged to obtain the average gene type of the compliant high-quality cells. Simultaneously, the gene types of each pending high-quality cell are obtained, and the gene types of each pending high-quality cell are compared with the average gene types of the compliant high-quality cells. Different responses are derived based on the comparison results.

[0014] In some implementations, if the number of gene types in a cell awaiting high quality is less than the average number of gene types in a cell meeting the criteria for high quality, it means that the number of gene types in this cell is below the average. In this case, the missing marker is a result of low expression, and this cell awaiting high quality is marked as a low-expression cell. If the number of gene types in a cell awaiting high quality is greater than or equal to the average number of gene types in a cell meeting the criteria for high quality, it means that the number of gene types in this cell is not below the average. In this case, the missing marker is not a result of low expression, and this cell awaiting high quality is directly excluded.

[0015] The present invention further provides a computer-readable storage medium storing a computer program, which is executed by a processor to implement the CellRanger single-cell RNA sequencing data analysis method described above.

[0016] The technical solution provided by this invention has the following advantages compared with the prior art:

[0017] The method provided by this invention not only rigorously ensures the integrity of various functional indicators in the final cells included in the database, but also retains some biologically significant cellular information by further judging the low expression state in the cells to be identified, even if candidate markers are missing due to low overall transcriptional activity. This method ensures the reliability of subsequent analysis data while minimizing the omission of information from low-expression or rare cells due to strict, one-size-fits-all criteria. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0021] The CellRanger single-cell RNA sequencing data analysis method provided by this invention, such as... Figure 1 As shown, it includes:

[0022] The first step involves converting raw BCL data obtained from the lab into FASTQ format using the mkfastq template, and automatically splitting the sequencing data of different samples according to a pre-defined sample table. Then, the STAR algorithm is used to align the FASTQ reads to a reference genome, identifying the cell barcode and unique molecular marker corresponding to each read. The cell barcode distinguishes each cell, and the unique molecular marker distinguishes redundant information generated by duplicate amplification, achieving molecular-level deduplication counting to filter low-quality reads. A mismatch tolerance mechanism is employed to correct the cell barcodes, reducing the impact of sequencing errors on downstream counting. After read alignment, multiple reads of the same gene are aggregated based on the unique molecular markers to avoid counting errors caused by PCR amplification bias, ensuring that the true molecules expressed by each gene in each cell are accurately counted, generating a gene × cell expression matrix. After obtaining the expression matrix, cell calling is used to identify the actual cells and background noise, obtaining the effective cell barcode.

[0023] The second step involves, for each cell, counting the UMIs of all genes and the UMIs of mitochondrial genes. The mitochondrial gene proportion is calculated by dividing the mitochondrial gene UMI by the total UMI of all genes. Simultaneously, a threshold range for the mitochondrial gene proportion is set, and the mitochondrial gene proportion is compared to this threshold. Different responses are then determined based on the comparison results. If the mitochondrial gene proportion falls within or is less than the minimum value of the threshold range, it indicates that most RNA originates from nuclear genes. In this case, the cell has normal transcriptional activity and is marked as a high-quality cell. If the mitochondrial gene proportion is greater than the maximum value of the threshold range, it indicates that the cell is under stress or undergoing apoptosis. In this case, it is marked as a low-quality cell. For low-quality cells, a second threshold for the mitochondrial gene proportion is set, slightly larger than the maximum value of the threshold range. For example, if the threshold range is 5%-10%, the second threshold is 15%. Under this condition, the mitochondrial gene proportion of low-quality cells is again compared to the second threshold, and different responses are determined based on the comparison results. If the mitochondrial gene proportion is less than or equal to the second threshold, a further assessment is performed. If the proportion of mitochondrial genes is greater than the second threshold for mitochondrial gene proportion, the cell is still classified as a low-quality cell. During the re-evaluation process, the number of gene types in the low-quality cell is obtained. Simultaneously, a threshold for the number of gene types is set, and the number of gene types is compared to this threshold. Different actions are taken based on the comparison results. If the number of gene types is greater than or equal to the threshold, it indicates that the cell has rich gene expression and is a healthy cell; in this case, the low-quality cell is re-marked as a high-quality cell. If the number of gene types is less than the threshold, it indicates that the cell has low gene expression and is a cell undergoing degradation; in this case, the cell is still classified as a low-quality cell.

[0024] The third step, after obtaining high-quality cells, involves selecting five candidate genes based on their defined gene annotations. The presence of each of these five candidate genes is then checked in each high-quality cell. Specifically, for each high-quality cell, the expression matrix is ​​examined to determine if the candidate gene is detected. Candidate genes one, two, three, four, and five are scanned separately. High-quality cells containing candidate gene one are marked as high-quality cells with one candidate gene, those containing candidate gene two as high-quality cells with two candidate genes, those containing candidate gene three as high-quality cells with three candidate genes, those containing candidate gene four as high-quality cells with four candidate genes, and those containing candidate gene five as high-quality cells with five candidate genes. After all high-quality cells are marked, a screening process is performed. For a single high-quality cell to be considered a qualified high-quality cell, it must simultaneously contain all five markers; otherwise, it is considered an unqualified high-quality cell. Because if a cell exhibits only one, two, three, or four markers, although it may perform well in some indicators, it still does not meet the stringent standard of "full markers," as the absence of any one marker may indicate a deficiency or imbalance in the corresponding biological function. This stringent requirement is based on a comprehensive assessment of the cell's overall state, aiming to exclude samples whose abnormalities in individual indicators might affect the accuracy and reliability of subsequent analyses. Each marker represents an important detection point for a cell in a specific transcriptional, translational, metabolic, or organelle function, and therefore none can be omitted. Only by simultaneously meeting the requirements of all markers can it be ensured that the cell is in an ideal state in all key functions. It is worth mentioning that, in selecting candidate genes, we first selected key genes that are universally and stably expressed in various cell types, based on extensively experimentally validated gene sets provided by authoritative gene annotation databases. Secondly, we used high-precision alignment tools to accurately map sequencing data to a reference genome, ensuring that reads from these candidate gene regions could be reliably captured. Furthermore, we utilized UMI correction technology to eliminate redundant errors from PCR amplification, thereby improving the accuracy of gene counting. In addition, during data preprocessing, we employed rigorous cell calling and quality screening to retain only cells expressing stable, high-quality signals, ensuring that the low-noise expression of candidate genes was accurately reflected. Finally, through preliminary data exploration and multiple biological validations, we confirmed that these five candidate genes were detectably expressed in the vast majority of high-quality cells. Even with inter-individual expression differences, their presence remained highly reliable, thus serving as indicators for subsequent cell identification and functional analysis.

[0025] When the number of gene types in a cell is less than a threshold, a threshold slightly smaller than the threshold is set (e.g., if the threshold is 100 and the threshold is 90). Under this condition, the number of gene types in the cell is compared again to the threshold, and different actions are taken based on the comparison results. If the number of gene types in the cell is greater than or equal to the threshold, it indicates low gene expression, and further judgment is made. If the number of gene types in the cell is less than the threshold, it indicates high gene expression, and the cell is maintained as a low-quality cell. In the further judgment process, the UMI number of ribosomal genes in the cell is obtained. The ribosomal gene ratio is calculated by dividing the UMI number of ribosomal genes by the UMI number of all genes. Simultaneously, a threshold range for the ribosomal gene ratio is set, and the cell's ribosomal gene ratio is compared to this threshold, with different actions taken based on the comparison results. If the cell's ribosomal gene ratio falls within the threshold range, it indicates the cell is in a normal state, and the low-quality cell is re-marked as a high-quality cell. If the proportion of ribosomal genes in a cell exceeds the maximum value of the ribosomal gene proportion range threshold or falls below the minimum value, it indicates an imbalance in transcription or translation. For example, under certain stress conditions, cells tend to oversynthesize ribosomal proteins to cope with adversity, or ribosomal RNA may become excessively enriched due to biological reasons such as cell cycle arrest or abnormal proliferation. In such cases, maintaining this proportion is a sign of low-quality cells. This is because ribosomes play a crucial role in protein synthesis within the cell, and maintaining an appropriate ribosomal gene proportion helps ensure cellular homeostasis. An abnormally high ribosomal gene proportion may mean that the cell is attempting to compensate for transcriptional or translational imbalances by synthesizing large amounts of ribosomal proteins under stress or abnormal proliferation. Conversely, an abnormally low proportion may reflect suppressed overall cellular protein synthesis activity or decreased cellular metabolic activity. Both excessively high and excessively low proportions indicate abnormalities in transcriptional or translational regulation, suggesting that the cell is no longer in a normal equilibrium state, and therefore can be considered a marker of transcriptional or translational imbalance. This differs from the situation where a mitochondrial gene ratio exceeding the maximum value of the threshold range indicates that the cell is under stress or undergoing apoptosis. An excessively high mitochondrial gene ratio often indicates that the cell may be experiencing stress, apoptosis, or damage, as these conditions lead to greater loss of cytoplasmic RNA, relatively increasing the mitochondrial RNA ratio. Healthy cells, on the other hand, generally maintain a stable mitochondrial gene expression level, thus keeping the ratio within a reasonable range.On the other hand, an excessively low proportion of mitochondrial genes does not necessarily equate to abnormal cells, because differences in cell types or technology capture may lead to a lower mitochondrial RNA fraction, which in itself does not constitute a bad signal.

[0026] In the above process, a single high-quality cell is considered a high-quality cell only if all one to five markers are present simultaneously. This quality control method may lose information about low-expression or rare cells. To preserve information about low-expression or rare cells, in addition to high-quality cells that are present with all one to five markers, high-quality cells that are not high-quality cells are identified. These high-quality cells with at least one marker are marked as pending high-quality cells. Pending high-quality cells may contain one, two, three, or four markers. High-quality cells without a marker are marked as non-pending high-quality cells. In this way, high-quality cells are divided into high-quality cells that are compliant and high-quality cells that are not compliant. The non-pending high-quality cells include pending high-quality cells and non-pending high-quality cells. Non-pending high-quality cells are directly excluded. For pending high-quality cells, the method to determine whether they are low-expression cells is as follows: multiple high-quality cells are randomly selected from the compliant high-quality cells, and the gene types in each of the selected compliant high-quality cells are obtained and averaged to obtain the average gene type of the compliant high-quality cells. Simultaneously, the gene types of each potential high-quality cell were obtained, and each gene type was compared with the average gene type of qualified high-quality cells. Different responses were determined based on the comparison results. If the gene type of a potential high-quality cell was less than the average gene type of qualified high-quality cells, it meant that the gene type of this cell was below the average. In this case, marker loss indicates low expression, and this cell was marked as a low-expression cell. If the gene type of a potential high-quality cell was greater than or equal to the average gene type of qualified high-quality cells, it meant that the gene type of this cell was not below the average. In this case, marker loss was not a result of low expression, and this cell was directly excluded.

[0027] The entire process of this application first uses mkfastq to convert the raw BCL data into a FASTQ file, then uses the STAR alignment algorithm to map reads onto a reference genome. Cells are identified using cell barcodes and UMIs, and molecular-level deduplication counting is performed to construct a gene × cell expression matrix. Cell call analysis is then used to remove background noise and obtain effective cell barcodes. Next, the number of UMIs for all genes and mitochondrial genes in each cell is calculated to obtain the mitochondrial gene ratio. An appropriate threshold range is set to distinguish high-quality cells (where RNA mainly originates from nuclear genes) from low-quality cells (those under stress or apoptosis). For cells initially marked as low-quality, a second threshold for the mitochondrial ratio is set, and the number of gene types in each cell is compared, combined with indicators such as the ribosomal gene ratio, to further assess their status. Cells with abundant overall expression are re-marked as high-quality cells. After obtaining high-quality cells, five candidate marker genes are selected based on predefined gene annotations, and each... The presence of these five markers in each cell is scanned. Only cells that simultaneously detect all five candidate genes are considered high-quality cells that meet the criteria, as each marker represents a key detection point in transcription, translation, metabolism, and specific organelle functions. However, this strict full-marker standard may miss some cells that only show partial markers due to low expression or rarity. Therefore, high-quality cells that do not meet the full-marker requirement are further divided into two categories: undetermined high-quality cells and non-undetermined high-quality cells. Cells that detect at least one candidate gene are classified as undetermined, while those completely lacking candidate genes are directly excluded. Subsequently, samples are randomly drawn from high-quality cells that meet the criteria, and the number of gene types is counted and averaged. The number of gene types in each undetermined cell is then compared with this average: if the number of gene types in an undetermined cell is lower than the average, it indicates that the absence of candidate markers may be due to low expression, and the cell is marked as a low-expression cell; if the number of gene types is not lower than the average, it is considered that the absence is not caused by low expression and the cell is directly excluded. This process ensures that high-quality cells are in ideal condition in all key functions, while also preserving information from some low-expression or rare cells while maintaining data rigor.

[0028] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. Embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination thereof.

[0029] 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 the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0030] Those skilled in the art should understand that the above description is only a specific embodiment of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application.

Claims

1. A CellRanger single-cell RNA sequencing data analysis method characterized in that, The application relates to a method for screening high-quality cells from a single-cell RNA sequencing data set. The method comprises the following steps: Step 1: converting original BCL data into FASTQ format, splitting the sequencing data, identifying cell barcodes and unique molecular labels, filtering low-quality reads, and collecting multiple reads of the same gene to obtain a gene-cell expression matrix; and obtaining effective cell barcodes by using cell calling; Step 2: obtaining the UMI number of all genes and the UMI number of mitochondrial genes of each cell, dividing the UMI number of mitochondrial genes by the UMI number of all genes to obtain a mitochondrial gene proportion, comparing the mitochondrial gene proportion with a mitochondrial gene proportion range threshold, and obtaining high-quality cells and low-quality cells based on the comparison result; 2. The CellRanger single-cell RNA sequencing data analysis method of claim 1, wherein, Step 3: for high-quality cells, five candidate genes are selected based on defined annotations, and it is detected whether the five candidate genes exist in each high-quality cell and are marked; after marking, screening is performed; for a single high-quality cell, all the labels from one to five must exist simultaneously, and the high-quality cell is determined to be a qualified high-quality cell; otherwise, the high-quality cell is determined to be an unqualified high-quality cell; high-quality cells in which at least one label exists are obtained from all the unqualified high-quality cells and are recorded as pending high-quality cells; high-quality cells in which no label exists are recorded as non-pending high-quality cells; the non-pending high-quality cells are directly excluded; and whether the pending high-quality cells are low-expression cells is determined by the following method: a plurality of qualified high-quality cells are randomly selected from the qualified high-quality cells, the gene types in the selected qualified high-quality cells are obtained, and the gene types are averaged to obtain a gene type average value of the qualified high-quality cells; the gene types of each pending high-quality cell are obtained, and the gene types of each pending high-quality cell are compared with the gene type average value of the qualified high-quality cells; and different responses are obtained based on the comparison result.

3. The CellRanger single-cell RNA sequencing data analysis method of claim 2, wherein, The specific process of obtaining high-quality cells and low-quality cells based on the comparison result is as follows: if the mitochondrial gene proportion falls within the mitochondrial gene proportion range threshold or is less than the minimum value of the mitochondrial gene proportion range threshold, it is determined that the RNA comes from a nuclear gene, the cell is normal in transcription activity, and the cell is marked as a high-quality cell; if the mitochondrial gene proportion is greater than the maximum value of the mitochondrial gene proportion range threshold, it is determined that the cell is in a stress state or is undergoing apoptosis, and the cell is marked as a low-quality cell.

4. The CellRanger single-cell RNA sequencing data analysis method of claim 3, wherein, For the low-quality cells, a second threshold of the mitochondrial gene proportion greater than the maximum value of the mitochondrial gene proportion range threshold is set; the mitochondrial gene proportion of the low-quality cells is compared with the second threshold of the mitochondrial gene proportion again, and different responses are obtained based on the comparison result. If the mitochondrial gene proportion is less than or equal to the second threshold of the mitochondrial gene proportion, the judgment is re-performed; and if the mitochondrial gene proportion is greater than the second threshold of the mitochondrial gene proportion, the judgment that the cell is a low-quality cell is maintained.

5. The CellRanger single-cell RNA sequencing data analysis method of claim 4, wherein, In the re-determination, the number of gene types in the low-quality cell is obtained, and a gene type number threshold is set. The number of gene types is compared with the gene type number threshold. If the number of gene types is greater than or equal to the gene type number threshold, it means that the gene expression of the cell is rich, and the cell is a healthy cell. In this case, the low-quality cell is re-labeled as a high-quality cell. If the number of gene types is less than the gene type number threshold, it means that the gene expression of the cell is not rich, and the cell is in degradation. In this case, the determination that the cell is a low-quality cell is maintained.

6. The CellRanger single-cell RNA sequencing data analysis method of claim 5, wherein, A gene type number proximity threshold less than the gene type number threshold is set. When the number of gene types of the cell is less than the gene type number threshold, the number of gene types of the cell is compared with the gene type number proximity threshold again. If the number of gene types of the cell is greater than or equal to the gene type number proximity threshold, it means that the degree of non-rich gene expression of the cell is low. In this case, further determination is performed. If the number of gene types of the cell is less than the gene type number proximity threshold, it means that the degree of non-rich gene expression of the cell is high. In this case, the determination that the cell is a low-quality cell is maintained.

7. The CellRanger single-cell RNA sequencing data analysis method of claim 6, wherein, In the further determination, the number of ribosome gene UMIs in the cell is obtained. The number of ribosome gene UMIs is divided by the number of UMIs of all genes to obtain a ribosome gene proportion. Meanwhile, a ribosome gene proportion range threshold is set. The ribosome gene proportion of the cell is compared with the ribosome gene proportion range threshold. If the ribosome gene proportion of the cell falls within the ribosome gene proportion range threshold, it means that the cell is in a normal state. In this case, the low-quality cell is re-labeled as a high-quality cell. If the ribosome gene proportion of the cell is greater than the maximum value of the ribosome gene proportion range threshold or less than the minimum value of the ribosome gene proportion range threshold, it means that the cell has imbalance in transcription or translation. In this case, the determination that the cell is a low-quality cell is maintained.

8. The CellRanger single-cell RNA sequencing data analysis method of claim 7, wherein, If the number of gene types of the to-be-determined high-quality cell is less than the average number of gene types of the consistent high-quality cells, it means that the number of gene types of the to-be-determined high-quality cell is lower than the average value. In this case, the missing marker is the result of low expression, and the to-be-determined high-quality cell is labeled as a low-expression cell. If the number of gene types of the to-be-determined high-quality cell is greater than or equal to the average number of gene types of the consistent high-quality cells, it means that the number of gene types of the to-be-determined high-quality cell is not lower than the average value. In this case, the missing marker is not the result of low expression, and the to-be-determined high-quality cell is directly excluded.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the CellRanger single-cell RNA sequencing data analysis method of any one of claims 1-8.

Images