A method, system, device and medium for detecting cell types
By obtaining whole-genome expression levels and marker gene information of single cells, a set of cells with significant expression was screened out. Combined with the KNN algorithm and image data, the problem of accuracy in cell type labeling in single-cell sequencing was solved, and rapid and accurate cell type determination was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- RES INST OF TSINGHUA PEARL RIVER DELTA
- Filing Date
- 2022-12-16
- Publication Date
- 2026-05-26
AI Technical Summary
In existing technologies, the accuracy of cell type labeling during single-cell sequencing is low, especially because some cells lose genes or are in the division and growth stage, resulting in low expression levels of marker genes and inaccurate detection results.
By acquiring whole-genome expression data and marker gene information, a set of cells with significant expression was screened out, and the cell type was determined by combining the K-nearest neighbor algorithm with image data. In particular, for cells with low marker gene expression, the KNN algorithm was used for cluster analysis.
It enables rapid and accurate identification of large batches of cell type information, improving the labeling accuracy and efficiency in single-cell sequencing tasks.
Smart Images
Figure CN116189771B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of bioinformatics, and in particular to a method, system, device, and medium for detecting cell types. Background Technology
[0002] Single-cell sequencing is an emerging technology that performs high-throughput sequencing analysis of the genome and transcriptome at the individual cell level. This technology is generally used to reveal heterogeneous differences between cells, thereby helping to gain a deeper understanding of cell function and crosstalk, increasing our understanding of complex ecosystems and diseases in multicellular organisms, and precisely locating pathogenic cells in complex diseases. With the generation of massive amounts of single-cell sequencing data, comprehensively integrating and analyzing this vast amount of data to mine potential information has become a consensus within the bioinformatics industry.
[0003] However, in the process of single-cell sequencing, there is currently no ideal method for efficiently and accurately labeling cell types from massive amounts of single cells. Most related technologies rely on cell marker databases like CellMarker, which utilize the high expression of marker genes to detect cell types. However, because some cells may have gene loss or be in the process of division and growth, the expression levels of their marker genes are significantly lower than normal, often resulting in low accuracy in cell type determination using this method. Summary of the Invention
[0004] The purpose of this application is to at least partially solve one of the technical problems existing in the related art.
[0005] Therefore, one object of the embodiments of this application is to provide a method, system, device and medium for detecting cell types.
[0006] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of this application include:
[0007] On one hand, embodiments of this application provide a method for detecting cell types, the method comprising:
[0008] Acquire whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested;
[0009] Based on the whole-genome expression data and the marker gene information, the first expression level of the marker gene corresponding to each of the cells to be tested is determined;
[0010] Based on the first expression level, a set of cells with significant expression is selected from the cells to be tested; the set of cells with significant expression includes a plurality of first cells;
[0011] The cell type of the first cell is determined based on the whole genome expression data and marker gene information corresponding to the first cell;
[0012] Based on the cell type corresponding to each of the first cells and the image, the cell type of the second cell is determined by the K-nearest neighbor algorithm; the second cell is another cell to be detected besides the first cell.
[0013] In addition, the cell type detection method according to the above embodiments of this application may also have the following additional technical features:
[0014] Furthermore, in one embodiment of this application, after the step of acquiring whole-genome expression data of a batch of cells to be tested, the method further includes:
[0015] The whole genome expression data is converted into matrix data; the matrix data consists of n rows and 1 column, where the numerical elements in each row correspond to the expression data of one gene, and the numerical elements in each row are arranged in a predetermined order according to the gene types, where n is a positive integer.
[0016] Furthermore, in one embodiment of this application, obtaining marker gene information includes:
[0017] Obtain the species information of the cells to be detected;
[0018] Based on the species information, obtain the marker gene information.
[0019] Furthermore, in one embodiment of this application, the step of screening a set of cells with significant expression from the cells to be detected based on the first expression level includes:
[0020] Based on the first expression level, calculate the average expression level of the marker gene corresponding to each of the cells to be tested;
[0021] Based on the first expression level and the mean expression level, a set of cells with significant expression is selected from the cells to be tested.
[0022] Furthermore, in one embodiment of this application, the step of screening a set of cells with significant expression from the cells to be detected based on the first expression level and the mean expression level includes:
[0023] The expression reference value of the cells to be tested is calculated using the following formula:
[0024]
[0025] In the formula, Pi represents the expression reference value of the i-th cell to be tested, ki represents the first expression level of the marker gene of the i-th cell to be tested, λ represents the mean expression level, and e represents the Euler constant;
[0026] Based on the magnitude of the expression reference value, a set of cells with significant expression is selected from the cells to be tested.
[0027] Furthermore, in one embodiment of this application, determining the cell type of the second cell using the K-nearest neighbor algorithm based on the cell type corresponding to each of the first cells and the image includes:
[0028] Based on the cell type corresponding to each of the first cells and the image, a KNN model is trained, and the K value in the KNN model is determined.
[0029] From the image, identify the K first cells that are closest to the second cell;
[0030] The cell type with the largest number of cells in the K first cells is determined, and the cell type with the largest number of cells is determined as the cell type of the second cell.
[0031] Furthermore, in one embodiment of this application, determining the K first cells closest to the second cell from the image includes:
[0032] Calculate the Euclidean distance between the second cell and each of the first cells;
[0033] Sort the first cells according to their Euclidean distance, and select the K first cells with the smallest Euclidean distance.
[0034] On the other hand, embodiments of this application provide a cell type detection device, the device comprising:
[0035] The acquisition module is used to acquire whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested.
[0036] The processing module is used to determine the first expression level of the marker gene corresponding to each of the cells to be tested based on the whole genome expression data and the marker gene information;
[0037] A screening module is used to screen a set of cells with significant expression from the cells to be detected based on the first expression level; the set of cells with significant expression includes a plurality of first cells;
[0038] The first annotation module is used to determine the cell type of the first cell based on the whole genome expression data and marker gene information corresponding to the first cell;
[0039] The second annotation module is used to determine the cell type of the second cell by using the K-nearest neighbor algorithm based on the cell type corresponding to each of the first cells and the image; the second cell is another cell to be detected besides the first cell.
[0040] On the other hand, embodiments of this application provide a computer device, including:
[0041] At least one processor;
[0042] At least one memory for storing at least one program;
[0043] When the at least one program is executed by the at least one processor, the at least one processor implements the above-described method for detecting a cell type.
[0044] On the other hand, embodiments of this application also provide a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the aforementioned cell type detection method.
[0045] The advantages and beneficial effects of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application:
[0046] This application discloses a method for detecting cell types. The method includes: acquiring whole-genome expression data, marker gene information, and an image including the batch of cells to be detected; determining the first expression level of the marker gene corresponding to each cell to be detected based on the whole-genome expression data and the marker gene information; selecting a set of cells with significant expression from the cells to be detected based on the first expression level; the set of cells with significant expression includes several first cells; determining the cell type of the first cells based on the whole-genome expression data and marker gene information corresponding to the first cells; and determining the cell type of second cells using a K-nearest neighbor algorithm based on the cell type corresponding to each first cell and the image; the second cells are other cells to be detected besides the first cells. This method can quickly and accurately determine the type information of a large number of cells and can be used in single-cell type labeling, facilitating subsequent single-cell sequencing and other tasks. Attached Figure Description
[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of this application or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0048] Figure 1 This is a schematic flowchart of a cell type detection method provided in an embodiment of this application;
[0049] Figure 2 This is a schematic diagram illustrating a method for screening a set of cells that significantly express this expression, as provided in an embodiment of this application.
[0050] Figure 3 This is a schematic diagram illustrating how the cell type of a second cell is determined using the K-nearest neighbor algorithm, as provided in an embodiment of this application.
[0051] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0052] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.
[0053] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.
[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0055] Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to have perception, reasoning, and decision-making capabilities. AI technology is a comprehensive discipline involving a wide range of fields, encompassing both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, speech processing, natural language processing, and machine learning / deep learning.
[0056] Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence. Its applications span all areas of artificial intelligence. Machine learning (deep learning) typically includes techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.
[0057] Single-cell sequencing is an emerging technology that performs high-throughput sequencing analysis of the genome and transcriptome at the individual cell level. This technology is generally used to reveal heterogeneous differences between cells, thereby helping to gain a deeper understanding of cell function and crosstalk, increasing our understanding of complex ecosystems and diseases in multicellular organisms, and precisely locating pathogenic cells in complex diseases. With the generation of massive amounts of single-cell sequencing data, comprehensively integrating and analyzing this vast amount of data to mine potential information has become a consensus within the bioinformatics industry.
[0058] However, in the process of single-cell sequencing, there is currently no ideal method for efficiently and accurately labeling cell types from massive amounts of single cells. Most related technologies rely on cell marker databases like CellMarker, which utilize the high expression of marker genes to detect cell types. However, because some cells may have gene loss or be in the process of division and growth, the expression levels of their marker genes are significantly lower than normal, often resulting in low accuracy in cell type determination using this method.
[0059] In view of this, this application provides a cell type detection method. When detecting cell type, this method determines the first expression level of the marker gene in each cell by using whole-genome expression data and marker gene information. From this, it selects the first cell with significant marker gene expression and determines the cell type based on the marker gene expression. For the second cell with low marker gene expression, the K-Nearest Neighbors (KNN) algorithm from machine learning in artificial intelligence is used to cluster these cells based on their image data, thereby determining their cell type. In this way, the type information of a large number of cells can be quickly and accurately determined, which can be used in single-cell type labeling, facilitating subsequent single-cell sequencing and other tasks.
[0060] Please refer to Figure 1 , Figure 1 This is a schematic flowchart of a cell type detection method provided in an embodiment of this application, referred to... Figure 1 The detection method for this cell type includes, but is not limited to:
[0061] Step 110: Obtain whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested;
[0062] In this step, for cells whose cell types need to be detected, their whole-genome expression data, marker gene information, and images can be obtained. Commonly used gene expression data indicators include FPKM and TPM, and this application does not limit the type of indicator. The calculation and processing of gene expression data can be implemented with reference to existing technologies, and will not be elaborated here. In this embodiment, marker gene information is used to record the cell's marker genes. A marker gene, or maker gene, refers to a gene with a known function or known sequence that can play a specific marking role. For cell type detection, since the expression of marker genes in different types of cells often differs significantly, they can be distinguished based on the expression data of marker genes. It is understood that marker gene information can be collected and obtained through existing information channels, and this application does not limit this. In some embodiments, the species information of the cells to be detected can be obtained first, and then the known marker gene information of that species can be obtained based on the species information. Images of a batch of cells to be detected can be acquired using relevant microscopic observation equipment, and the images can be pre-processed such as filtering and denoising to improve the data accuracy of the images.
[0063] Step 120: Based on the whole-genome expression data and the marker gene information, determine the first expression level of the marker gene corresponding to each of the cells to be tested;
[0064] In this step, after obtaining the whole-genome expression data and marker gene information for each cell to be tested, marker genes can be selected from the whole genome, and then the corresponding expression data of the marker genes can be retrieved. Here, for each cell to be tested, there may be multiple marker genes, so the expression data corresponding to each marker gene can be accumulated to obtain the total expression level of the marker genes, which is referred to as the first expression level in this application. It is understandable that for a cell to be tested, if its overall first expression level is high, it is highly likely that the cell is in a normal gene expression state, such as not being in the division and growth stage. Conversely, if its overall first expression level is low, it indicates that the cell may have gene loss or is in the division and growth stage. In this case, if it is classified based on its marker gene expression data, there is a high probability that the classification will be inaccurate.
[0065] Step 130: Based on the first expression level, select a set of cells with significant expression from the cells to be tested; the set of cells with significant expression includes a number of first cells;
[0066] In this embodiment, after determining the first expression level data corresponding to each cell to be tested, a set of cells with significant expression can be screened from the cells to be tested based on the first expression level data. Cells in this set can be denoted as the first cell, while the other cells to be tested can be denoted as the second cell. It is understood that in this embodiment, by distinguishing the cells based on the first expression level data of each cell to be tested, a set of cells in which the marker gene is significantly expressed (first cell) and a set of cells in which the marker gene is not significantly expressed (second cell) can be identified. (Refer to...) Figure 2 For example, in some embodiments, an expression level threshold can be set as a boundary, and then the first expression level of the cell to be detected can be compared with that of the expression level threshold. If the first expression level of the cell to be detected exceeds the expression level threshold, it is identified as the first cell belonging to the significant expression state; if the first expression level of the cell to be detected does not exceed the expression level threshold, it is identified as the second cell belonging to the non-significant expression state. Of course, the expression level threshold here can be set according to specific circumstances, and this application does not limit its size.
[0067] Step 140: Determine the cell type of the first cell based on the whole-genome expression data and marker gene information corresponding to the first cell;
[0068] In this step, based on the foregoing description, it is understood that for the first cell, since it can be determined that the marker genes are significantly expressed, its cell type can be determined based on known statistical information according to the specific expression of the marker genes. For example, cells of a certain cell type may have one or more marker genes with much higher expression levels than other cell types. Therefore, by comparing the expression levels of these marker genes, it can be determined whether each first cell belongs to that cell type. Alternatively, it can be a comprehensive assessment of whether the expression levels of each marker gene in the first cell fall within the range of known marker gene expression levels for different cell types to determine the most likely cell type. This process has already been applied in related technologies, and this application will not elaborate on it further.
[0069] It should be emphasized that, in this embodiment, since the first cell is already screened and has a relatively significant level of marker gene expression, the cell type determined by step 140 is relatively accurate. However, for the second cell, only the overall level of marker gene expression can be determined; it cannot be determined whether this is due to the cell type or other abnormalities. Therefore, other methods are needed to determine its cell type.
[0070] Step 150: Based on the cell type corresponding to each of the first cells and the image, determine the cell type of the second cell using the K-nearest neighbor algorithm; the second cell is any cell to be detected other than the first cell.
[0071] In this step, the positional relationships between cells in the image data can, to some extent, reflect the correlation between their categories. Generally, cells of the same type tend to be located close to each other in the image. Therefore, after determining the cell type of the first cell, the K-nearest neighbor algorithm can be used to classify and label the cell type of the second cell. For example, referring to... Figure 3 Suppose that in a region of an image, the cell types of most cells have been identified, categorized into cell type A and cell type B. Then, for a second cell of an unknown cell type, the KNN algorithm can be used to determine which group it is closer to, thus classifying it as either cell type A or cell type B. This allows for a more accurate identification of the cell type of second cells where marker genes are not significantly expressed.
[0072] In some embodiments, after the step of acquiring whole-genome expression data of a batch of cells to be tested, the method further includes:
[0073] The whole genome expression data is converted into matrix data; the matrix data consists of n rows and 1 column, where the numerical elements in each row correspond to the expression data of one gene, and the numerical elements in each row are arranged in a predetermined order according to the gene types, where n is a positive integer.
[0074] In this embodiment, after obtaining the whole-genome expression data of the cell to be tested, the whole-genome expression data can be converted into matrix data for easier data processing. Specifically, the matrix data is in the form of n rows and 1 column, where n is a positive integer, the same as the number of data points in the whole-genome expression data, i.e., the same as the number of genes in the whole genome of the cell to be tested. In the matrix data, each row contains a numerical element, which corresponds to the expression data of a gene; and the gene types corresponding to the numerical elements in each row can be pre-assigned in a predetermined order according to species information. For example, suppose that a species has 4 genes in its whole genome, and the expression data of these genes differs in various cells of that species. In this embodiment, for that species, a gene type arrangement order can be pre-set, such as sorting by the first letter of the gene name or by function; this application does not limit the specific arrangement order. After the sorting is completed, for a cell to be tested in that species, after obtaining its whole-genome expression data, the data can be processed into matrix data according to the sorted order. For example, suppose the whole genome contains four genes: A, B, C, and D. The resulting whole genome expression levels are: gene A expression level is 0.94, gene B expression level is 0, gene C expression level is 2.13, and gene D expression level is 1.36. The predetermined gene order in the whole genome is CBAD. When converting the whole genome expression data into matrix data, the matrix data consists of 4 rows and 1 column. The first row contains 2.13, the second row contains 0, the third row contains 0.94, and the fourth row contains 1.36. It is understood that in this embodiment, converting the whole genome expression data into matrix data according to the pre-defined gene order facilitates a unified data structure. This ensures that data at the same position in the whole genome expression data converted to matrix data within the same species has the same meaning, which is beneficial for subsequent processing.
[0075] In some embodiments, the step of screening a set of cells with significant expression from the cells to be detected based on the first expression level includes:
[0076] Based on the first expression level, calculate the average expression level of the marker gene corresponding to each of the cells to be tested;
[0077] Based on the first expression level and the mean expression level, a set of cells with significant expression is selected from the cells to be tested.
[0078] In this embodiment of the application, when screening a set of cells with significant expression from the cells to be tested, the average expression level of the marker gene corresponding to each cell to be tested can be calculated based on the first expression level data, and then the first cell can be determined based on the average expression level. For example, the average expression level can be used as the aforementioned expression level threshold to distinguish between the first cell and the second cell.
[0079] In other embodiments, the step of screening a set of cells with significant expression from the cells to be tested based on the first expression level and the mean expression level includes:
[0080] The expression reference value of the cells to be tested is calculated using the following formula:
[0081]
[0082] In the formula, Pi represents the expression reference value of the i-th cell to be tested, ki represents the first expression level of the marker gene of the i-th cell to be tested, λ represents the mean expression level, and e represents the Euler constant;
[0083] Based on the magnitude of the expression reference value, a set of cells with significant expression is selected from the cells to be tested.
[0084] In this embodiment, the expression level reference value can also be used to distinguish between first cells and second cells. Specifically, for the expression reference value of a cell to be tested, if it is less than a certain value (a pre-defined comparison threshold that can be flexibly set), it indicates that the marker gene of the cell to be tested is significantly expressed, meaning that the cell to be tested belongs to the first cell. In other words, in this embodiment, the smaller the expression reference value, the higher the probability that the cell to be tested belongs to the first cell. Therefore, the probability of it belonging to the first cell can be determined based on a negative correlation, and then the cells with higher probabilities are classified into the set of significantly expressed cells. This application does not limit the specific functional relationship between the two.
[0085] In some embodiments, determining the cell type of the second cell using the K-nearest neighbor algorithm based on the cell type corresponding to each of the first cells and the image includes:
[0086] Based on the cell type corresponding to each of the first cells and the image, a KNN model is trained, and the K value in the KNN model is determined.
[0087] From the image, identify the K first cells that are closest to the second cell;
[0088] The cell type with the largest number of cells in the K first cells is determined, and the cell type with the largest number of cells is determined as the cell type of the second cell.
[0089] In this embodiment, when using the K-nearest neighbor algorithm, a KNN model can be trained based on the positional relationship of the first cell of a known cell type in the image. The range of values for K is predefined, and the optimal neighbor value, i.e., the K value, is found through training. The specific formula is K∈{n..Mn}, where n represents the number of possible cell types and M represents the number of cells.
[0090] Next, when performing type detection on the second cell, the K closest first cells to the second cell in the image can be identified. For example, the Euclidean distance between the second cell and each of the first cells can be calculated, and then the first cells can be sorted according to their Euclidean distances, selecting the K cells with the smallest Euclidean distances. Then, the cell type of the second cell can be determined based on the cell type with the largest proportion among these K first cells. For example, if K is 10, for a certain second cell, 7 of its 10 closest first cells belong to cell type A, 2 to cell type B, and 1 to cell type C. Since cell type A has the largest proportion, it means that the second cell's location closely matches the region where cell type A is located, and therefore, the cell type of the second cell can be determined as cell type A.
[0091] It is understood that the cell type detection method provided in this application embodiment determines the first expression level of the marker gene in each cell by using the whole-genome expression data and marker gene information of the cells to be detected, and then screens out the first cells with significant marker gene expression, determining the cell type of the first cells based on the expression of the marker gene. For the second cells with low marker gene expression, the K-Nearest Neighbors (KNN) algorithm from machine learning in artificial intelligence is used to cluster them based on the image data of these cells, thereby determining their cell type. This method can improve the accuracy of cell type detection and labeling, and is beneficial to improving the efficiency and accuracy of subsequent cell data analysis and processing.
[0092] This application embodiment also provides a cell type detection device, the device comprising:
[0093] The acquisition module is used to acquire whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested.
[0094] The processing module is used to determine the first expression level of the marker gene corresponding to each of the cells to be tested based on the whole genome expression data and the marker gene information;
[0095] A screening module is used to screen a set of cells with significant expression from the cells to be detected based on the first expression level; the set of cells with significant expression includes a plurality of first cells;
[0096] The first annotation module is used to determine the cell type of the first cell based on the whole genome expression data and marker gene information corresponding to the first cell;
[0097] The second annotation module is used to determine the cell type of the second cell by using the K-nearest neighbor algorithm based on the cell type corresponding to each of the first cells and the image; the second cell is another cell to be detected besides the first cell.
[0098] Understandable Figure 1 The content of the cell type detection method embodiment shown is applicable to the cell type detection device embodiment. The specific functions implemented by the cell type detection device embodiment are the same as those shown in the embodiment. Figure 1 The method for detecting one cell type shown is the same as the embodiment described, and the beneficial effects achieved are the same. Figure 1 The beneficial effects achieved by the cell type detection method embodiment shown are also the same.
[0099] Reference Figure 4 This application also discloses a computer device, including:
[0100] At least one processor 201;
[0101] At least one memory 202 is used to store at least one program;
[0102] When at least one program is executed by at least one processor 201, such that at least one processor 201 performs as follows: Figure 1 An example of a cell type detection method is shown.
[0103] It is understandable that, such as Figure 1 The content of the cell type detection method embodiment shown is applicable to the computer device embodiment, and the specific functions implemented by the computer device embodiment are the same as those shown. Figure 1 The method for detecting a cell type shown is the same as in the embodiment described above, and the beneficial effects achieved are the same as those described above. Figure 1 The beneficial effects achieved by the cell type detection method embodiment shown are also the same.
[0104] This application also discloses a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to implement, for example... Figure 1 An example of a cell type detection method is shown.
[0105] It is understandable that, such as Figure 1 The content of the cell type detection method embodiment shown is applicable to the embodiments of this computer-readable storage medium. The specific functions implemented by the embodiments of this computer-readable storage medium are the same as those shown below. Figure 1 The method for detecting a cell type shown is the same as in the embodiment described above, and the beneficial effects achieved are the same as those described above. Figure 1 The beneficial effects achieved by the cell type detection method embodiment shown are also the same.
[0106] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.
[0107] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated into a single physical system and / or software module, or one or more functions and / or features may be implemented in a separate physical system or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the system disclosed herein, the actual implementation of the module will be understood within the scope of ordinary skill of an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.
[0108] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0109] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, system, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, system, or device). For the purposes of this specification, "computer-readable medium" can mean any system that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, system, or device.
[0110] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections with one or more wires (electronic systems), portable computer disk drives (magnetic systems), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic systems, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0111] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0112] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0113] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.
[0114] The foregoing has provided a detailed description of the preferred embodiments of this application. However, this application is not limited to these embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.
[0115] In the description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0116] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.
Claims
1. A method for detecting a cell type, characterized by, The method includes: Acquire whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested; Based on the whole-genome expression data and the marker gene information, the first expression level of the marker gene corresponding to each of the cells to be tested is determined; Based on the first expression level, a set of cells with significant expression is selected from the cells to be tested; the set of cells with significant expression includes a plurality of first cells; The cell type of the first cell is determined based on the whole genome expression data and marker gene information corresponding to the first cell; Based on the cell type corresponding to each of the first cells and the image, the cell type of the second cell is determined by the K-nearest neighbor algorithm; the second cell is any cell to be detected other than the first cell. The step of determining the cell type of the second cell based on the cell type corresponding to each of the first cells and the image using the K-nearest neighbor algorithm includes: Based on the cell type corresponding to each of the first cells and the image, a KNN model is trained, and the K value in the KNN model is determined. From the image, identify the K first cells that are closest to the second cell; The cell type with the largest number of cells in the K first cells is determined, and the cell type with the largest number of cells is determined as the cell type of the second cell.
2. The method for detecting cell types according to claim 1, characterized in that, After the step of acquiring whole-genome expression data of a batch of cells to be tested, the method further includes: The whole genome expression data is converted into matrix data; the matrix data consists of n rows and 1 column, where the numerical elements in each row correspond to the expression data of one gene, and the numerical elements in each row are arranged in a predetermined order according to the gene types, where n is a positive integer.
3. The method for detecting cell types according to claim 1, characterized in that, Obtain marker gene information, including: Obtain the species information of the cells to be detected; Based on the species information, obtain the marker gene information.
4. The method for detecting cell types according to claim 1, characterized in that, The step of screening a set of cells with significant expression from the cells to be tested based on the first expression level includes: Based on the first expression level, calculate the average expression level of the marker gene corresponding to each of the cells to be tested; Based on the first expression level and the mean expression level, a set of cells with significant expression is selected from the cells to be tested.
5. The method for detecting cell types according to claim 4, characterized in that, The step of screening a set of cells with significant expression from the cells to be tested based on the first expression level and the mean expression level includes: The expression reference value of the cells to be tested is calculated using the following formula: ； In the formula, This represents the expression reference value for the i-th cell to be tested. This represents the first expression level of the marker gene in the i-th cell to be tested. This represents the mean expression level, where e represents the Euler constant. Based on the magnitude of the expression reference value, a set of cells with significant expression is selected from the cells to be tested.
6. The method for detecting cell types according to claim 1, characterized in that, The step of determining the K first cells that are closest to the second cell from the image includes: Calculate the Euclidean distance between the second cell and each of the first cells; Sort the first cells according to their Euclidean distance, and select the K first cells with the smallest Euclidean distance.
7. A device for detecting cell types, characterized in that, The device includes: The acquisition module is used to acquire whole-genome expression data, marker gene information, and images including the batch of cells to be tested from a batch of cells to be tested. The processing module is used to determine the first expression level of the marker gene corresponding to each of the cells to be tested based on the whole genome expression data and the marker gene information; A screening module is used to screen a set of cells with significant expression from the cells to be detected based on the first expression level; the set of cells with significant expression includes a plurality of first cells; The first annotation module is used to determine the cell type of the first cell based on the whole genome expression data and marker gene information corresponding to the first cell; The second annotation module is used to determine the cell type of the second cell based on the cell type corresponding to each of the first cells and the image, using the K-nearest neighbor algorithm; the second cell is another cell to be detected besides the first cell. The step of determining the cell type of the second cell based on the cell type corresponding to each of the first cells and the image using the K-nearest neighbor algorithm includes: Based on the cell type corresponding to each of the first cells and the image, a KNN model is trained, and the K value in the KNN model is determined. From the image, identify the K first cells that are closest to the second cell; The cell type with the largest number of cells in the K first cells is determined, and the cell type with the largest number of cells is determined as the cell type of the second cell.
8. A computer device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a cell type detection method as described in any one of claims 1-6.
9. A computer-readable storage medium storing a processor-executable program, characterized in that: The processor-executable program, when executed by the processor, is used to implement a cell type detection method as described in any one of claims 1-6.