Single-cell spatial transcriptome analysis method and device based on cell segmentation
By acquiring the corrected scanned images and combining the barcode microsphere position information in the biochip, cell segmentation and statistical analysis are performed, the problem of low cell analysis efficiency and accuracy in the prior art is solved, and efficient and accurate cell position is achieved and statistical information acquisition.
Patent Information
- Application Number
- PCT/CN2024/101345
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-06-26
- Filing Date
- 2024-06-25
- Publication Date
- 2025-01-02
AI Technical Summary
The prior art fails to combine the cell segmentation results with a biochip containing barcoded microspheres after cell segmentation, resulting in low efficiency and accuracy of cell analysis.
By obtaining the corrected scanning image, cell segmentation is performed, the position information of the barcode microspheres corresponding to each cell in the cell segmentation result is extracted, the position and statistical information of the cells are determined, and the position information of the barcode microspheres in the biochip is improved. Efficiency and accuracy of cell analysis.
The efficiency and accuracy of cell analysis are improved, and the location information of each cell and the statistical information of barcoded microspheres are accurately located and identified.
Smart Images

Figure CN2024101345_02012025_PF_FP_ABST
Abstract
Description
Single-cell spatial transcriptome analysis method and device based on cell segmentation
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to Chinese patent application No. 2023107520082, filed on June 25, 2023, entitled “Information processing method, device, equipment and storage medium based on cell segmentation”, which is incorporated herein by reference in its entirety. This application claims priority to Chinese patent application No. 2023107586874, filed on June 26, 2023, entitled “Scanning image correction method, device and equipment”, which is incorporated herein by reference in its entirety. This application claims priority to Chinese patent application No. 2023107520218, filed on June 25, 2023, entitled “Scanning image processing method, device and equipment”, which is incorporated herein by reference in its entirety. This application claims priority to Chinese patent application No. 2023107520186, filed on June 25, 2023, entitled “Method for constructing a single-cell spatial transcriptome sequencing library”, which is incorporated herein by reference in its entirety. Technical Field
[0003] The present disclosure relates to the field of bioinformatics, and in particular to a single-cell spatial transcriptome analysis method and device based on cell segmentation. Background Art
[0004] With the development of computer vision and machine learning technologies, more and more automated cell segmentation methods have been proposed and applied. These methods mainly include edge detection, threshold segmentation, morphological operations, region growing, image segmentation, and deep learning.
[0005] However, related technologies do not combine the cell segmentation results with the biochip containing barcoded microspheres after cell segmentation, resulting in the inability to obtain the accurate location information of each cell and the statistical information of the corresponding barcoded microspheres, thereby resulting in low efficiency and accuracy of cell analysis.
[0006] Summary of the Invention
[0007] In response to the problem of low efficiency and accuracy of cell analysis in related technologies, the embodiments of the present disclosure provide a single-cell spatial transcriptome analysis method and device based on cell segmentation, which is used to achieve the purpose of improving the efficiency and accuracy of cell analysis.
[0008] The present disclosure provides a single-cell spatial transcriptome analysis method based on cell segmentation, comprising:
[0009] Acquiring a corrected scanned image; the scanned image is obtained based on the tissue slice attached to the biochip;
[0010] performing cell segmentation on the scanned image to obtain a cell segmentation result;
[0011] Extracting position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip;
[0012] Based on the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result, the position information of each cell and the statistical information of the corresponding barcoded microsphere are determined.
[0013] Optionally, the cell segmentation result includes a cell nucleus boundary segmentation result and a cell boundary segmentation result; performing cell segmentation on the scanned image to obtain the cell segmentation result includes:
[0014] Splitting the scanned image into multiple sub-images;
[0015] For each of the subgraphs, preprocessing the subgraph to obtain a target subgraph;
[0016] Obtaining a pre-trained model that matches the cell type of the tissue section, and using the pre-trained model to predict the cell nucleus boundary segmentation result of the target sub-image, to obtain the cell nucleus boundary segmentation result of each cell nucleus in the target sub-image;
[0017] Merging the target sub-images to obtain a target image;
[0018] Merging the cell nucleus boundary segmentation results of each cell nucleus in each target subimage to obtain the cell nucleus boundary segmentation result of each cell nucleus in the target image;
[0019] The cell nucleus boundary segmentation result of each cell nucleus in the target image is expanded to generate the cell boundary segmentation result of each cell in the target image.
[0020] Optionally, splitting the scanned image into a plurality of sub-images includes:
[0021] Splitting the scanned image into a plurality of initial sub-images using a custom sliding window;
[0022] Fine-tuning edge boundaries of the multiple initial sub-images is performed to obtain the multiple sub-images.
[0023] Optionally, preprocessing each subgraph to obtain a target subgraph includes:
[0024] For each of the sub-images, pre-processing including equalization, size adjustment, color space conversion, erosion, expansion and pixel value normalization is performed on the sub-image to obtain the target sub-image.
[0025] Optionally, performing an expansion process on the cell nucleus boundary segmentation result of each cell nucleus in the target image to generate the cell boundary segmentation result of each cell in the target image includes:
[0026] generating a binary image based on the cell nucleus boundary segmentation result of each cell nucleus in the target image;
[0027] Calculating a distance map of the binary image by distance transformation; the distance map is used to represent the distance from each pixel to the nearest cell nucleus boundary;
[0028] The distance map and the preset cell boundary threshold are processed using a watershed algorithm to obtain the cell boundary segmentation result of each cell in the target image.
[0029] Optionally, extracting the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip includes:
[0030] Calculating the central coordinate position of each barcoded microsphere based on the intrinsic position of each barcoded microsphere in the biochip;
[0031] Extracting a region corresponding to a position in the cell data of the cell segmentation result based on the center coordinate position;
[0032] Creating a circular mask, and multiplying the cell data in the area by the circular mask to obtain the cell data within the circular range;
[0033] The cell data within each circular range are counted to obtain the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result.
[0034] Optionally, it also includes:
[0035] Performing bioinformatics analysis on the position information of each cell and the statistical information of the corresponding barcoded microspheres to obtain an expression matrix of the cell corresponding to each position information;
[0036] performing clustering processing on the expression matrix corresponding to each of the cells to obtain cell clustering information;
[0037] Based on the cluster number of each cell in the cell clustering information, a cell region corresponding to each cluster number in the scanned image is drawn with a corresponding color to obtain a cluster image;
[0038] Generating contour position information of each cell in the cluster image by searching for cell contours in the cluster image;
[0039] The cluster image and the contour position information of each cell in the cluster image are visually displayed.
[0040] Optionally, obtaining the corrected scanned image includes:
[0041] With the tissue slice attached to the biochip, the sample is fixed with methanol and then bright-field stained, and then scanned and imaged under a bright-field light source to obtain multiple field-of-view scanning images of the biochip, wherein the biochip includes a substrate pattern, the substrate pattern includes multiple key points, and any adjacent field-of-view scanning images have overlapping areas;
[0042] According to the overlapping areas between any adjacent visual field scanning images, the plurality of visual field scanning images are fused to obtain a full-width scanning image of the biochip;
[0043] determining first position information of the plurality of key points in the full-scan image according to the substrate pattern;
[0044] The full-width scan image is corrected according to the first position information and the substrate pattern to obtain a target full-width scan image as the corrected scan image.
[0045] Optionally, the fusing of the plurality of visual field scanning images according to the overlapping area between any adjacent visual field scanning images to obtain the full-width scanning image of the biochip includes:
[0046] determining relative positions between any adjacent visual field scanning images according to overlapping areas between any adjacent visual field scanning images;
[0047] According to the relative positions between any adjacent visual field scanning images, the multiple visual field scanning images are fused to obtain the full-width scanning image.
[0048] Optionally, determining the relative positions between any adjacent visual field scanning images according to the overlapping areas between any adjacent visual field scanning images includes:
[0049] For any first visual field scanning image among the plurality of visual field scanning images, determining a second visual field scanning image adjacent to the first visual field scanning image;
[0050] performing a two-dimensional window sliding process on the first field of view scanning image and the second field of view scanning image according to an overlapping area between the first field of view scanning image and the second field of view scanning image to determine a matching parameter between the first field of view scanning image and the second field of view scanning image;
[0051] The relative position between the first field of view scanning image and the second field of view scanning image is determined according to the matching parameters.
[0052] Optionally, the fusing the plurality of visual field scanning images according to the relative positions between any adjacent visual field scanning images to obtain the full-width scanning image includes:
[0053] Determining a position of a reference visual field scanning image among the multiple visual field scanning images, where the reference visual field scanning image is any one of the multiple visual field scanning images;
[0054] Determining positions of other visual field scanning images according to the position of the reference visual field scanning image and the relative positions between any adjacent visual field scanning images, wherein the other visual field scanning images are visual field scanning images other than the reference visual field scanning image among the multiple visual field scanning images;
[0055] The multiple visual field scanning images are fused according to the position of the reference visual field scanning image and the positions of the other reference visual field scanning images to obtain the full-width scanning image.
[0056] Optionally, determining first position information of the plurality of key points in the full-scan image according to the substrate pattern includes:
[0057] Determining a key point recognition model corresponding to the substrate pattern;
[0058] Inputting the full-scan image into the key point recognition model to obtain the first position information;
[0059] In which, the key point recognition model is obtained by training based on multiple groups of training samples, any group of training samples includes a sample full-width scan image and label information corresponding to the sample full-width image, the sample full-width scan image includes the substrate pattern, and the label information includes the position information of multiple key points in the sample full-width scan image.
[0060] Optionally, performing correction processing on the full-width scan image according to the first position information and the substrate pattern to obtain a target full-width scan image includes:
[0061] Determining second position information of the plurality of key points in the full-scan image according to the substrate pattern and the size of the full-scan image;
[0062] Correction processing is performed on the full-width scanned image according to the first position information and the second position information to obtain the target full-width scanned image.
[0063] Optionally, performing correction processing on the full-width scanned image according to the first position information and the second position information to obtain the target full-width scanned image includes:
[0064] determining a correction transformation matrix according to the first position information and the second position information;
[0065] According to the correction conversion matrix, mapping transformation processing is performed on the full-width scanned image to obtain the target full-width scanned image.
[0066] Optionally, obtaining the corrected scanned image includes:
[0067] With the tissue slice attached to the biochip, the sample is fixed with methanol and then fluorescently stained, followed by fluorescence scanning imaging to obtain a plurality of binary channel sub-images of the biochip and fluorescence channel sub-images corresponding to each of the binary channel sub-images;
[0068] Amplifying the substrate pattern included on the biochip according to the scanning magnifications corresponding to the plurality of binary mirror channel sub-images to obtain a full-frame reference image;
[0069] Determining a correction position of each of the two-mirror channel sub-images in the full-frame reference image according to a collection order of the plurality of two-mirror channel sub-images and a plurality of key points included in the substrate pattern;
[0070] According to the corrected position of each of the two-mirror channel sub-images in the full-frame reference image, the fluorescence channel sub-images corresponding to each of the two-mirror channel sub-images are fused to obtain a full-frame fluorescence image of the biochip as the corrected scanned image.
[0071] Optionally, determining the correction position of each of the two-mirror channel sub-images in the full-frame reference image based on the acquisition order of the multiple two-mirror channel sub-images and the multiple key points included in the substrate pattern includes:
[0072] For any binary mirror channel sub-image, performing key point recognition processing on the binary mirror channel sub-image to determine first coordinate information of a first key point in the binary mirror channel sub-image, where the first key point is a key point included in the binary mirror channel sub-image among the multiple key points;
[0073] Determining the initial position of the two-mirror channel sub-image in the full-frame reference image according to the acquisition sequence and the first coordinate information;
[0074] The initial position is corrected to obtain a corrected position of the binary mirror channel sub-image in the full-frame reference image.
[0075] Optionally, performing key point recognition processing on the two-mirror channel sub-image to determine first coordinate information of a first key point in the two-mirror channel sub-image includes:
[0076] Inputting the two-mirror channel sub-image into a key point recognition model to obtain the first coordinate information;
[0077] In which, the key point recognition model is obtained by training based on multiple groups of training samples, any group of training samples includes a sample two-division mirror channel sub-image and label information corresponding to the sample two-division mirror channel sub-image, the sample two-division mirror channel sub-image includes sample key points, and the label information includes the coordinate information of the sample key points in the sample two-division mirror channel sub-image.
[0078] Optionally, determining the initial position of the binary mirror channel sub-image in the full-frame reference image according to the acquisition order and the first coordinate information includes:
[0079] determining, according to the acquisition order, second coordinate information of the first key point in the full reference image;
[0080] An initial position of the two-mirror channel sub-image in the full-frame reference image is determined according to the first coordinate information and the second coordinate information.
[0081] Optionally, the correcting the initial position to obtain the corrected position of the binary mirror channel sub-image in the full-frame reference image includes:
[0082] Determining, according to the initial position, a local sub-image corresponding to the binary mirror channel sub-image in the full-frame reference image, wherein the region corresponding to the local sub-image in the full-frame reference image includes the region corresponding to the binary mirror channel sub-image in the full-frame reference image;
[0083] Performing matching processing on the local sub-image and the dichroic channel sub-image to obtain a matching result;
[0084] According to the matching result, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
[0085] Optionally, performing matching processing on the local sub-image and the dichroic channel sub-image to obtain a matching result includes:
[0086] When the local sub-image is in a fixed state, determining a plurality of relative positions of the two-mirror channel sub-image relative to the local sub-image according to a preset step size;
[0087] performing a two-dimensional window sliding process on the two-mirror channel sub-image according to the plurality of relative positions, and determining the two-mirror channel sub-image at each of the relative positions;
[0088] The local sub-image and the two-mirror channel sub-image at each relative position are matched to obtain the matching result, which includes the similarity between the two-mirror channel sub-image and the local sub-image at each relative position.
[0089] Optionally, obtaining the corrected position of the binary mirror channel sub-image in the full-frame reference image according to the matching result includes:
[0090] Determine the relative position with the highest corresponding similarity as the target relative position of the two-mirror channel sub-image relative to the local sub-image;
[0091] According to the relative position of the target and the position of the local sub-image in the full-frame reference image, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
[0092] Optionally, a single-cell spatial transcriptome sequencing library is used for the clustering process, the tissue section is a fresh frozen tissue section, and the biochip is a spatial transcriptome chip;
[0093] The single-cell spatial transcriptome sequencing library was constructed by the following steps:
[0094] (1) attaching the fresh frozen tissue slices to the spatial transcriptome chip, fixing them with methanol, fluorescently staining the samples, performing fluorescence scanning imaging, and finally performing scanning imaging under a bright field light source to obtain an uncorrected scanning image of the single-cell spatial transcriptome;
[0095] (2) Tissue permeabilization was performed after scanning;
[0096] (3) Reverse transcription and single-strand cDNA synthesis;
[0097] (4) cDNA second-strand synthesis;
[0098] (5) cDNA amplification;
[0099] (6) cDNA purification;
[0100] (7) cDNA quality control;
[0101] (8) Sequencing on a machine.
[0102] Optionally, for animal samples, DAPI or ssDNA is used for fluorescent staining in step (1), and then fluorescent scanning imaging is performed; or, DAPI or ssDNA is used for fluorescent staining, and then fluorescent scanning imaging is performed, and then the sample is stained with HE or Masson, and finally scanned and imaged under a bright field light source;
[0103] For plant samples, in step (1), fluorescent whitening agent 28 is used for autofluorescence enhancement staining; or, toluidine blue or HE staining is performed directly without fluorescent staining.
[0104] Optionally, the thickness of the frozen tissue section in step (1) is 5-20 μm, and the chip after patching is placed in -20°C methanol for fixation for 30-60 minutes.
[0105] Optionally, after the chip after patching in step (1) is fixed with methanol, a fluorescent staining reagent is added to the sample, incubated in the dark for 1-20 minutes, and then fluorescence scanning imaging is performed under the illumination of a full-spectrum or wide-spectrum light source;
[0106] In the fluorescence channel, by removing the filter block of the scanner and retaining only the dichroic mirror, a dichroic mirror setting can be performed to simultaneously achieve cell nucleus fluorescence scanning and background microsphere matrix scanning; or by setting a filter block with a certain transmittance, cell nucleus fluorescence scanning and background microsphere matrix scanning can be achieved under full spectrum or wide spectrum conditions.
[0107] Optionally, step (2) includes: placing the chip obtained in step (1) in the tissue permeabilization solution and incubating at 37° C. for 1-30 min, removing the tissue permeabilization solution and adding 0.1% SSC solution into the wells of the chip.
[0108] Optionally, step (3) includes: removing the SSC solution in the well, adding the reverse transcription and cDNA single-strand synthesis reagents into the well, and incubating at 37-65° C. for 0.5-2 hours.
[0109] Optionally, step (4) includes: removing the reverse transcription and cDNA single-strand synthesis reagents in the well, adding 0.08M KOH for denaturation, washing with EB buffer, and then adding the cDNA double-strand synthesis reagent, and incubating at 45-70°C for 10-40 minutes.
[0110] Optionally, step (5) includes: removing the cDNA double-strand synthesis reagent in the well, adding EB buffer for washing, then adding 0.08M KOH and incubating at room temperature for 10 minutes, then neutralizing KOH with Tris, and adding the cDNA amplification reagent for PCR amplification.
[0111] Optionally, step (6) includes: purifying the amplified product using magnetic beads, then washing the magnetic beads with 80% ethanol, and finally eluting the amplified product from the magnetic beads with EB buffer.
[0112] Optionally, the spatial transcriptome chip used is equipped with mRNA capture probes;
[0113] Preferably, the S1000 spatial transcriptome chip is used.
[0114] The present disclosure provides a single-cell spatial transcriptome analysis device based on cell segmentation, comprising:
[0115] An acquisition module, configured to acquire a corrected scanned image; the scanned image is obtained based on a tissue slice attached to the biochip;
[0116] a segmentation module, configured to perform cell segmentation on the scanned image to obtain a cell segmentation result;
[0117] an extraction module, configured to extract position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip;
[0118] The determination module is configured to determine the position information of each cell and the statistical information of the corresponding barcoded microsphere based on the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result.
[0119] An embodiment of the present disclosure provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps of any one of the above-described methods for single-cell spatial transcriptome analysis based on cell segmentation are implemented.
[0120] An embodiment of the present disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the above-described single-cell spatial transcriptome analysis methods based on cell segmentation.
[0121] The single-cell spatial transcriptome analysis method and apparatus based on cell segmentation provided by the embodiments of the present disclosure first obtain a corrected scan image, which is obtained based on a tissue slice attached to a biochip, so that an accurate and corrected scan image can be obtained before cell segmentation; then, cell segmentation is performed on the scan image to obtain a cell segmentation result; based on the inherent position of each barcoded microsphere in the biochip, the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result is extracted; the cell segmentation result can be combined with the biochip containing the barcoded microspheres to extract the position information of the barcoded microspheres corresponding to the individual cells; finally, based on the position information of the barcoded microspheres corresponding to each cell in the cell segmentation result, the position information of each cell and the statistical information of the corresponding barcoded microspheres are determined, and the accurate position information of each cell and the statistical information of the corresponding barcoded microspheres can be statistically calculated based on the position information of the barcoded microspheres corresponding to the individual cells, thereby improving the efficiency and accuracy of cell analysis. BRIEF DESCRIPTION OF THE DRAWINGS
[0122] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0123] FIG1 is a schematic diagram of a process for a single-cell spatial transcriptome analysis method based on cell segmentation provided by an embodiment of the present disclosure;
[0124] FIG2 is a schematic diagram of the substrate pattern of the Biotron S1000 spatial transcriptome chip provided in an embodiment of the present disclosure;
[0125] FIG3 is a schematic diagram of a corrected scanned image provided by an embodiment of the present disclosure;
[0126] FIG4 is a schematic diagram of a cell nucleus boundary segmentation result provided by an embodiment of the present disclosure;
[0127] FIG5 is a schematic diagram of a cell boundary segmentation result provided by an embodiment of the present disclosure;
[0128] FIG6 is a schematic diagram of a clustered image provided by an embodiment of the present disclosure;
[0129] FIG7 is a schematic flow chart of a scanned image correction method provided by an embodiment of the present disclosure;
[0130] FIG8 is a schematic diagram of a substrate pattern provided by an embodiment of the present disclosure;
[0131] FIG9 is a schematic diagram of collecting a visual field scanning image according to an embodiment of the present disclosure;
[0132] FIG10 is a schematic diagram of a visual field scanning image provided by an embodiment of the present disclosure;
[0133] FIG11 is a schematic diagram of a fusion process provided by an embodiment of the present disclosure;
[0134] FIG12 is a schematic diagram of multiple key points provided by an embodiment of the present disclosure;
[0135] FIG13 is a schematic diagram of a flow chart of a fusion process of a visual field scanning image according to an embodiment of the present disclosure;
[0136] FIG14 is a schematic diagram of a flow chart of a correction process according to an embodiment of the present disclosure;
[0137] FIG15 is a schematic diagram of a full-width scanned image of a target provided by an embodiment of the present disclosure;
[0138] FIG16 is an enlarged schematic diagram of a full-width scanned image of a target after superimposing a base pattern according to an embodiment of the present disclosure;
[0139] FIG17 is a flow chart of a scanned image processing method according to an embodiment of the present disclosure;
[0140] FIG18 is a comparison diagram of the effects of the binary mirror channel and the fluorescent channel provided by an embodiment of the present disclosure;
[0141] FIG19 is a schematic diagram of a substrate pattern enlargement process according to an embodiment of the present disclosure;
[0142] FIG20 is a schematic diagram of a process for determining the corrected position of a two-mirror channel sub-image in a full-frame reference image according to an embodiment of the present disclosure;
[0143] FIG21 is a schematic diagram of a full-frame fluorescence image provided by an embodiment of the present disclosure;
[0144] FIG22 is a schematic diagram of fluorescent channel sub-images before and after fusion provided by an embodiment of the present disclosure;
[0145] FIG23 is an enlarged schematic diagram of a full-frame fluorescent image after superimposing a substrate pattern according to an embodiment of the present disclosure;
[0146] FIG24 is a HE-stained cell segmentation result of animal tissue provided by an embodiment of the present disclosure;
[0147] FIG25 is a seamless scanning effect diagram of toluidine blue staining of plant tissue according to an embodiment of the present disclosure;
[0148] FIG26 is a splicing diagram of the effect of toluidine blue staining of plant tissues provided by an embodiment of the present disclosure;
[0149] FIG27 is a schematic diagram of the process of constructing a single-cell spatial transcriptome sequencing library according to an embodiment of the present disclosure;
[0150] FIG28 is a schematic diagram of the structure of a single-cell spatial transcriptome analysis device based on cell segmentation provided in an embodiment of the present disclosure;
[0151] FIG29 is a schematic structural diagram of an electronic device provided by the present disclosure. DETAILED DESCRIPTION
[0152] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present disclosure without making any creative efforts shall fall within the scope of protection of the present disclosure.
[0153] In order to clearly describe the technical solutions of the embodiments of the present disclosure, in each embodiment of the present disclosure, if the words "first", "second", etc. are used to distinguish the same items or similar items with basically the same functions and effects, those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order.
[0154] The following describes the single-cell spatial transcriptome analysis method based on cell segmentation provided by the embodiments of the present disclosure in conjunction with Figures 1 to 34.
[0155] Please refer to Figure 1, which is a flow chart of a single-cell spatial transcriptome analysis method based on cell segmentation provided by an embodiment of the present disclosure. As shown in Figure 1, the method may include the following steps:
[0156] Step 101: Acquire a calibrated scan image; the scan image is obtained based on a tissue slice attached to a biochip;
[0157] Step 102: performing cell segmentation on the scanned image to obtain a cell segmentation result;
[0158] Step 103: extracting the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip;
[0159] Step 104 : Based on the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result, determine the position information of each cell and the statistical information of the corresponding barcoded microsphere.
[0160] In step 101 , the biochip may be a spatial transcriptome chip, for example, the Biotron S1000 spatial transcriptome chip shown in FIG2 , but the present embodiment is not limited thereto.
[0161] The tissue slice can be a plant tissue slice or an animal tissue slice. The tissue slice is attached to a biochip.
[0162] The scanned image may be a fluorescent scanned image or a bright field scanned image, but the embodiment is not limited thereto.
[0163] Taking a mouse testis tissue slice as an example, a scanner first acquires multiple scanned images, which are then stitched together to create an initial scanned image. Due to systematic errors in the scanner, the initial scanned image often exhibits misalignment, missing images, or overlaps. Therefore, before cell segmentation, the initial scanned image must be corrected to produce the corrected scanned image shown in Figure 3. The corrected scanned image is free of stitching errors and its position is aligned with the correct baseplate distribution, ensuring that the corrected scanned image matches the actual image.
[0164] In step 102 , cell segmentation refers to separating cells in a tissue slice from background and other cells in order to accurately measure the gene expression of each cell.
[0165] Specifically, each cell in the scanned image is separated from the background and other cells to obtain a cell segmentation result.
[0166] In step 103, the biochip contains barcoded microspheres, and each cell in the cell segmentation result is covered with multiple barcoded microspheres. Based on the inherent position of each barcoded microsphere in the biochip, the position information of the barcoded microsphere covered under each cell can be found.
[0167] In step 104, since the position information of the barcoded microspheres covering each cell is determined, the accurate position information of each cell and the statistical information of each barcoded microsphere covering each cell can be calculated.
[0168] The single-cell spatial transcriptome analysis method based on cell segmentation provided by the embodiments of the present disclosure includes the following steps: first, obtaining a corrected scan image, where the scan image is obtained based on a tissue slice attached to a biochip, and an accurate and corrected scan image can be obtained before cell segmentation; then, performing cell segmentation on the scan image to obtain a cell segmentation result; extracting the position information of the barcoded microspheres corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip; combining the cell segmentation result with the biochip containing the barcoded microspheres to extract the position information of the barcoded microspheres corresponding to the individual cells; finally, determining the position information of each cell and the statistical information of the corresponding barcoded microspheres based on the position information of the barcoded microspheres corresponding to each cell in the cell segmentation result; and calculating the accurate position information of each cell and the statistical information of the corresponding barcoded microspheres based on the position information of the barcoded microspheres corresponding to the individual cells, thereby improving the efficiency and accuracy of cell analysis.
[0169] Based on the single-cell spatial transcriptome analysis method based on cell segmentation in the embodiment corresponding to FIG1 , the cell segmentation results include the cell nucleus boundary segmentation results and the cell boundary segmentation results; step 102 may include the following sub-steps:
[0170] Step 1021: Split the scanned image into multiple sub-images;
[0171] Step 1022: For each subgraph, pre-process the subgraph to obtain a target subgraph;
[0172] Step 1023: Obtain a pre-trained model that matches the cell type of the tissue section, and use the pre-trained model to predict the cell nucleus boundary segmentation result of the target sub-image to obtain the cell nucleus boundary segmentation result of each cell nucleus in the target sub-image;
[0173] Step 1024: Merge the target sub-images to obtain the target image.
[0174] Step 1025 , merging the cell nucleus boundary segmentation results of each cell nucleus in each target sub-image to obtain the cell nucleus boundary segmentation results of each cell nucleus in the target image;
[0175] Step 1026 : Perform an expansion process on the cell nucleus boundary segmentation result of each cell nucleus in the target image to generate a cell boundary segmentation result for each cell in the target image.
[0176] In step 1021, since the size of the scanned image is too large, it is necessary to split the scanned image into multiple sub-images of appropriate sizes. The appropriate sizes are pre-configured based on the display screen parameters of the current device.
[0177] In a possible implementation, a custom sliding window is used to split the scanned image into a plurality of initial sub-images; and edge boundary lines of the plurality of initial sub-images are fine-tuned to obtain a plurality of sub-images.
[0178] Specifically, a custom sliding window is used, starting from the upper left corner (0, 0) of the image and moving from left to right and top to bottom, until the scanned image is completely segmented into several initial sub-images. These initial sub-images have no overlapping areas after segmentation and can fully restore the original image before segmentation. Because the boundary lines of the evenly segmented initial sub-images may separate single intact cells, the boundary lines of multiple initial sub-images can be fine-tuned through offset operations to obtain multiple sub-images.
[0179] The first loop of the loop nest traverses the vertical direction of the image through the y_start variable, and the step size of each iteration is self.sub_size[0]-self.overlap[0]; the second loop of the loop nest traverses the horizontal direction of the image through the x_start variable, and the step size of each iteration is self.sub_size[1]-self.overlap[1]; through the slicing operation, the vertical and horizontal slice ranges of the current sub-image are obtained according to y_start and x_start; according to the slice range, the binary mask tmp_mask of the current sub-image is extracted from cell_mask; the current sub-image img_sub is extracted from the original image img; a copy tmp_img of the current sub-image is created to prevent changes to the original image; the areas already in the mask are set to 0 in the current sub-image, so that these areas will not participate in the cell segmentation calculation; the current sub-image is segmented by calling the model.eval() function. The incoming parameters include the current sub-image tmp_img, cell diameter, number of channels and flow field threshold; obtain the mask mask_sub from the segmentation result and remove the edge part; update tmp_mask according to the value of mask mask_sub, and set the corresponding position to True; generate the cell outline according to the mask mask_sub.
[0180] In this embodiment, a customized sliding window is used for sub-image segmentation and fine-tuning, which can improve the efficiency and accuracy of sub-image segmentation.
[0181] In step 1022, in order to improve image quality, highlight or strengthen specific information in the image and weaken interference information, each sub-image is pre-processed to obtain a target sub-image, which can better adapt to subsequent image processing or analysis tasks.
[0182] In a possible implementation, for each sub-image, pre-processing including equalization, resizing, color space conversion, erosion, dilation, and pixel value normalization is performed on the sub-image to obtain a target sub-image.
[0183] Specifically, preprocessing can include the following:
[0184] 1) Equalization
[0185] By adjusting the pixel distribution of the sub-image through methods such as histogram equalization, the brightness or color in the sub-image is made more balanced, which can enhance the visual effect and feature expression of the image.
[0186] 2) Size adjustment
[0187] Adjust the size of the subgraph to meet specific requirements or adapt to specific processing algorithms or models.
[0188] 3) Color space conversion
[0189] Convert a subimage from one color space to another, for example, converting an RGB image to a grayscale image or another color space image to suit different analysis needs.
[0190] 4) Corrosion and expansion
[0191] Eroding and dilating the sub-graph can weaken the influence of the barcoded microspheres under the tissue slices in the sub-graph on the pre-trained model.
[0192] 5) Pixel value homogenization
[0193] Adjusting the pixel values of the sub-images to a uniform range can avoid misidentification caused by being too bright or too dark.
[0194] In this embodiment, by performing preprocessing on the sub-image including equalization, resizing, color space conversion, erosion, dilation and pixel value normalization, the image quality can be improved, specific information in the image can be highlighted or strengthened, and interference information can be weakened, so as to better adapt to subsequent image processing or analysis tasks.
[0195] In step 1023, before performing cell segmentation, a pre-training model corresponding to animal cells can be trained based on samples with animal cell labels, and a pre-training model corresponding to plant cells can be trained based on samples with plant cell labels.
[0196] Taking a mouse testicular tissue slice as an example, we used the open-source cell segmentation algorithm Cellpose and loaded a pre-trained model that matched the cell types in the mouse testicular tissue slice. The pre-trained model was used to predict the nucleus boundary segmentation results for the target subgraph, resulting in the nucleus boundary segmentation results for each cell nucleus in the target subgraph, as shown in Figure 4.
[0197] Specifically, the model.eval() function can be called to predict the cell nucleus boundary segmentation result of the target sub-image. The input parameters of the model.eval() function may include: the current target sub-image tmp_img, cell diameter, number of channels and flow field threshold.
[0198] To better adapt to the characteristics of various cells, this step provides a variety of segmentation parameters, such as cell diameter, segmentation color channel, inversion, and flow threshold. By providing a variety of segmentation parameter options, it can adapt to different types of cells under different staining conditions.
[0199] For example, using a cell diameter of 35 pixels, segmentation color channels of 0RED, 1GREEN, and 2BLUE as the target subimage, use the command "python SegtoBarcode.py --img_path'. / img_dist_fg7.tif' --diameter35 --inversion True --seg_channel 1" to pass in the specified image and custom segmentation parameters. Other custom parameters can be passed in via the command line or by modifying the setup.yaml configuration file.
[0200] In steps 1024 and 1025 , all target sub-images and the corresponding cell nucleus boundary segmentation results are merged to obtain the target image and the cell nucleus boundary segmentation results of each cell nucleus in the target image.
[0201] Creating an npy matrix file of the original image size of type int32 as the storage structure for the cell nucleus boundary segmentation results can ensure point-to-point correspondence at the pixel level. In addition, the int32 type can store more data information to prevent the number of cell statistics from exceeding the upper limit of the data type.
[0202] In step 1026 , the cell nucleus boundary segmentation result of each cell nucleus in the target image is expanded outwards to a specified range, thereby generating the cell boundary segmentation result of each cell in the target image as shown in FIG. 5 .
[0203] In one possible embodiment, step 1026 may include: generating a binary image based on the cell nucleus boundary segmentation results of each cell nucleus in the target image; calculating a distance map of the binary image through distance transformation; the distance map is used to represent the distance from each pixel to the nearest cell nucleus boundary; and using a watershed algorithm to process the distance map and a preset cell boundary threshold to obtain the cell boundary segmentation results of each cell in the target image.
[0204] Specifically, first, the nuclear mask file (nucleus.npy) is loaded using the Numpy.load function and scaled. The scaling factor is 1 / self.calculate_rate, which is implemented using the Ndimage.zoom function. The scaled result is stored in mask_rs. Pixel values greater than 0 in mask_rs are set to 255, and the rest are set to 0 to obtain a binary image of the cell nucleus. This can be achieved using the Numpy.where function and type conversion. The distance map of the cell nucleus boundary is calculated using the distance transform (Ndimage.distance_transform_edt). The distance map represents the distance from each pixel to the nearest cell nucleus boundary. The cell segmentation result is generated based on the set cell edge threshold (self.cell_edge) and scaling factor (self.calculate_rate). The watershed function is used to perform the watershed algorithm, with the distance map as input and the nuclear mask (mask_rs) as a marker. The cellpose.utils.masks_to_outlines function is used to convert the cell segmentation result into a cell segmentation boundary. This will produce a mask containing the cell segmentation boundary. Scale the cell segmentation results (cells) to match the original nuclear mask (masks). This uses the Ndimage.zoom function and the size ratio calculation. Save the cell segmentation results (cells) as a file called cells.npy in the directory self.save_dir. Return the original nuclear mask (masks), the cell segmentation results (cells), and the path to the cell segmentation results (cells_path). The goal of this algorithm is to generate cell segmentation results from the nuclear mask, extract cell boundaries using a distance transform and a watershed algorithm, and save the results for later use.
[0205] In this embodiment, cell segmentation can be achieved conveniently and efficiently.
[0206] Based on the single-cell spatial transcriptome analysis method based on cell segmentation corresponding to the aforementioned embodiments, step 103 may include: calculating the center coordinate position of the barcoded microsphere based on the inherent position of each barcoded microsphere in the biochip; extracting the area corresponding to the position in the cell data of the cell segmentation result based on the center coordinate position; creating a circular mask and multiplying the cell data in the area by the circular mask to obtain the cell data within the circular range; and counting the cell data within each circular range to obtain the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result.
[0207] Specifically, load the barcode template and read the intrinsic position of each barcoded microsphere in the biochip. Load the cell data and calculate the image magnification zoom_scale. Calculate the standard width std_w and standard height std_h based on the magnification. Initialize the radius radius and calculate it as half of 0.618 times the standard width. If it is less than 1, set it to 1. Traverse the intrinsic position of each barcoded microsphere in the biochip and calculate the center coordinate position of each barcoded microsphere. Based on the center coordinate position, extract the area corresponding to the position in the cell data. Create a circular mask and multiply the cell data in the area by the mask to obtain the cell data within the circular range. Count the cell data within each circular range to obtain the position information of the barcoded microsphere corresponding to each cell and add it to the all_barcode list. Traverse the all_barcode list and write the barcoded microsphere name and corresponding cell data to the all_barcode_txt list. Save the contents of the all_barcode_txt list to the file all_barcode_num.txt.
[0208] In this embodiment, the cell segmentation results can be combined with a biochip containing barcoded microspheres to extract the position information of the barcoded microspheres corresponding to individual cells, thereby accurately locating and identifying each cell.
[0209] Based on the single-cell spatial transcriptome analysis method based on cell segmentation corresponding to the above embodiments, the method further includes:
[0210] Step 105: Perform bioinformatics analysis on the position information of each cell and the statistical information of the corresponding barcoded microspheres to obtain an expression matrix of the cell corresponding to each position information;
[0211] Step 106: performing clustering processing on the expression matrix corresponding to each cell to obtain cell clustering information;
[0212] Step 107: Based on the cluster number of each cell in the cell clustering information, a corresponding color is drawn in the cell region corresponding to each cluster number in the scanned image to obtain a cluster image;
[0213] Step 108: searching for cell outlines on the cluster image to generate outline position information of each cell in the cluster image;
[0214] Step 109: Visually display the cluster image and the contour position information of each cell in the cluster image.
[0215] In step 106, a clustering dictionary is defined to map cluster numbers to color codes. Clustering is performed on the expression matrix corresponding to each cell to obtain the cluster number, and the corresponding color code is retrieved from the clustering dictionary. An array, cell_color, is created to store the cell color codes, and the corresponding color codes are assigned to the corresponding positions in the cell_color array.
[0216] In step 107, color information is loaded. If a previously saved color file exists, the saved colors are loaded. Otherwise, a random set of colors is generated and saved. The corresponding color is drawn in the cell region corresponding to each cluster number in the scanned image, resulting in the cluster image shown in Figure 6, which is then saved to a specified directory. When drawing colors, the color within the cell nucleus is different from the color of the area between the cell boundary and the cell nucleus boundary.
[0217] In step 107 , regarding cell outline search, the cell boundaries are extracted from the cell segmentation results, and the boundary coordinates are saved as a JSON file.
[0218] Specifically, first, use the Numpy.load function to load the cell segmentation results (cells_path) and store them in masks. Create an empty dictionary cell_contours to store the boundary coordinates of each cell. Use the Ndimage.find_objects function to find the position of each cell in masks. Return a list slices consisting of tissue slices. For each non-empty tissue slice si, extract the corresponding row index sr and column index sc. Create a binary mask mask, set the pixel value belonging to the current cell to 1, and the remaining pixel values to 0. This is achieved by comparing masks[sr,sc] with (i+1). Use the cv2.findContours function to find the contour of the binary mask, and the returned result is contours. Extract the point coordinates pvc and pvr of the contour and offset them according to the starting position of the tissue slice to obtain the absolute coordinates vr and vc relative to the original image. Add the cell label and the corresponding boundary coordinates to the cell_contours dictionary, with the key being the cell label and the value being a list containing the boundary coordinates. Saves the cell_contours dictionary in JSON format to the specified file, json_name.json, in the save_dir directory. Returns the cell_contours dictionary. You can extract the boundary coordinates of each cell from the cell segmentation results and save them as a JSON file for later use.
[0219] In step 109, the contour position information of each cell in the cluster image is visualized in JSON format, and the cluster image is displayed on a visualization interface, which can clearly show the distribution of different types of cells.
[0220] In this embodiment, the cluster image and the outline position information of each cell in the cluster image can be obtained through bioinformatics analysis, clustering processing and cell outline search, which is convenient for visualization.
[0221] Optionally, for correction of a bright field scan image, as shown in FIG7 , step 101 may include:
[0222] S11, with the tissue slice attached to the biochip, fix the sample with methanol and perform bright field staining, and then scan and image under a bright field light source to obtain multiple field of view scanning images of the biochip, the biochip including a substrate pattern, the substrate pattern including multiple key points, and there is an overlapping area between any adjacent field of view scanning images.
[0223] Tissue slices are slices made from biological tissue, such as pathological tissue slices. After obtaining the tissue slices, they are switched and attached to the biochip according to a professional slicing process. Then, they are processed with appropriate reagents and fixed for scanning.
[0224] A biochip is a device used to capture genetic expression information in cells. It has a specific substrate pattern on it. The substrate pattern can be the same or different on different biochips. The substrate pattern includes, but is not limited to, circular arrays, square arrays, and more.
[0225] FIG8 is a schematic diagram of a base pattern provided in an embodiment of the present disclosure. As shown in FIG8 , a biochip includes a base pattern. In FIG8 , a region 821 of the biochip is magnified as region 822, and a subregion 823 within region 822 is magnified as subregion 824. It should be noted that the aforementioned magnification process refers to observation through microscopy and does not necessarily imply substantial magnification.
[0226] As can be seen in FIG8 , the biochip includes alternating horizontal and vertical lines, which together constitute the substrate pattern on the biochip. It should be noted that the substrate pattern on the biochip 1820 in FIG18 is merely an example. The substrate pattern on the biochip in the disclosed embodiment can be set as needed and is not limited in this embodiment.
[0227] The substrate pattern includes multiple key points. The number of these key points can be set as needed, and the positions of the key points in the substrate pattern can also be set as needed. For example, in Figure 8, the intersections of alternating horizontal and vertical lines are used as key points in the substrate pattern. In Figure 8, points A, B, C, and D are all key points in the substrate pattern.
[0228] The biochip includes multiple biological barcodes, which may also be called barcode dots, DNA barcodes, etc. In FIG8 , the small dots illustrated in the sub-area 824 are the biological barcodes in the biochip.
[0229] The function of a biological barcode is to capture the gene expression information of a cell. In the disclosed embodiments, the biological barcode on the biochip is used to capture the gene expression information of each cell in a tissue slice. Before the tissue slice is attached to the biochip, the biological barcode on the biochip has already captured the gene expression information of each cell in the tissue slice.
[0230] After the tissue slice is attached to the biochip and fixed, multiple field of view scanning images of the biochip can be collected, where the fields of view of different field of view scanning images are different. The field of view of any field of view scanning image is a local area of the biochip. The fields of view of multiple field of view scanning images together constitute the global area of the entire biochip. There is an overlapping area between any adjacent field of view scanning images.
[0231] Specifically, after the tissue slice is attached to the biochip and fixed, a specialized scanner is controlled to scan the tissue slice to obtain multiple field of view scan images. During the scanning process, the position of the scanner changes accordingly, and as the position of the scanner changes, the field of view of the scanned field of view scan images also changes accordingly. Optionally, the field of view scan images are brightfield digital microscopy scan images.
[0232] The following is an exemplary introduction to the implementation process of a scanner scanning a tissue slice with reference to FIG9 .
[0233] FIG9 is a schematic diagram of collecting a visual field scanning image provided by an embodiment of the present disclosure. As shown in FIG9 , it includes a biochip 931 and a scanner (not shown in the figure).
[0234] Initially, the scanner scans local area 9311 on biochip 931, obtaining a field of view scan image corresponding to local area 9311. The scanner then moves to scan local area 9312 on biochip 931, obtaining a field of view scan image corresponding to local area 9312, and so on. Local areas 9311 and 9312 are adjacent, and there is an overlapping region 9300 between the field of view scan image corresponding to local area 9311 and the field of view scan image corresponding to local area 9312.
[0235] The scanning order of the scanner can be preset. The scanner scans different local areas in the biochip 931 according to the preset scanning order to obtain corresponding field of view scanning images.
[0236] S12, performing fusion processing on the multiple visual field scanning images according to the overlapping areas between any adjacent visual field scanning images to obtain a full-width scanning image of the biochip.
[0237] Figure 10 is a schematic diagram of the field of view scanning image provided by an embodiment of the present disclosure. As shown in Figure 10, multiple field of view scanning images can be obtained after scanning by a scanner. Then, these multiple field of view scanning images can be rearranged according to the order of scanning by the scanner and numbered according to the position so that the position of the field of view scanning image corresponds to its actual spatial position on the biochip.
[0238] For any two adjacent visual field scanning images among the multiple visual field scanning images, since there is an overlapping area between the two visual field scanning images, the two visual field scanning images can be fused according to the overlapping area to obtain a fused image.
[0239] The fusion process can be seen in Figure 11.
[0240] Figure 11 is a schematic diagram of the fusion processing provided by an embodiment of the present disclosure. As shown in Figure 11, an example of the fusion process between four field of view scanning images is illustrated. The four field of view scanning images are image 1151, image 1152, image 1153 and image 1154.
[0241] Among them, images 1151 and 1152 are adjacent, and the right side of image 1151 overlaps with the left side of image 1152; images 1151 and 1153 are adjacent, and the bottom side of image 1151 overlaps with the top side of image 1153; images 1153 and 1154 are adjacent, and the right side of image 1153 overlaps with the left side of image 1154; and images 1152 and 1154 are adjacent, and the bottom side of image 1152 overlaps with the top side of image 1154. Each overlapping area is illustrated in FIG. 11 .
[0242] The fusion process combines the individual field of view scan images into a single image based on the overlapping areas. After fusion, the overlapping areas are combined into a single region. For example, in Figure 11, images 1151, 1152, 1153, and 1154 are fused to produce image 1155.
[0243] The aforementioned implementation method can be used to fuse adjacent visual field scan images in the plurality of visual field scan images. After all visual field scan images are fused, a corresponding full-frame scan image can be obtained. The visual field of the full-frame scan image is the entire area of the biochip, including the complete tissue slice.
[0244] S13: Determine first position information of a plurality of key points in the full-width scanned image according to the substrate pattern.
[0245] In one possible implementation, for the obtained full-width scan image, the position information of the substrate pattern in the full-width scan image can be identified, and then the first position information of the multiple key points in the full-width scan image can be determined by combining the positions of the multiple key points relative to the substrate pattern.
[0246] In a possible implementation, the first position information of multiple key points in the full-width scanned image may also be determined by using a key point recognition model.
[0247] Specifically, a key point recognition model corresponding to the substrate pattern is first determined, wherein the key point recognition model is pre-trained using multiple sets of training samples, where any set of training samples includes a sample full-width scan image and label information corresponding to the sample full-width scan image.
[0248] The sample full-width scan image is an image fused from a sample tissue slice scanned by a scanner. The sample tissue slice is attached to a sample biochip, which includes the substrate pattern. Therefore, the sample full-width scan image also includes the substrate pattern. The substrate pattern includes multiple key points, and the label information includes the position information of multiple key points in the sample full-width scan image. The label information can be in the form of an image, data, or both. The image form refers to an image obtained by annotating the key points in the sample full-width scan image, while the data form refers to the coordinate data used to annotate the positions of multiple key points in the sample full-width scan image.
[0249] Figure 12 is a schematic diagram of multiple key points provided by an embodiment of the present disclosure. As shown in Figure 12, multiple key points on a substrate pattern are illustrated. In Figure 12, the key points are the intersections of the lines, i.e., the black points in Figure 12.
[0250] In Figure 12, an example is shown of an image obtained by annotating a full-width scanned image of a sample (a part of an image), which can be directly used as label information of the full-width scanned image of the sample, or the coordinate information of multiple key points can be determined based on the annotated image, and the coordinate information of the multiple key points can be used as label information of the full-width scanned image of the sample. The embodiments of the present disclosure are not limited to this.
[0251] S14 , performing correction processing on the full-width scanned image according to the first position information and the substrate pattern to obtain a target full-width scanned image as a corrected scanned image.
[0252] The first position information is the position information of multiple key points actually determined on the full-width scan image. The substrate pattern includes multiple key points. Based on the design of the substrate pattern, various information such as the locations where the key points exist and the number of key points can be determined. Based on the information of multiple key points fed back from the design of the substrate pattern and the first position information of the multiple key points in the full-width scan image, the full-width scan image is corrected to obtain a target full-width scan image. Since the target full-width scan image is obtained by correcting the full-width scan image based on the first position information and the substrate pattern, the target full-width scan image and the biochip are aligned, thereby achieving accurate alignment of the gene expression information captured by the biochip and the area represented by the full-width scan image.
[0253] Based on any of the above embodiments, the solutions of the embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.
[0254] FIG13 is a schematic diagram of a flow chart of a fusion process of a visual field scanning image provided by an embodiment of the present disclosure, as shown in FIG13 , including:
[0255] S71 : Determine relative positions between any adjacent visual field scanning images according to overlapping areas between any adjacent visual field scanning images.
[0256] For any first field of view scan image among the multiple field of view scan images, a second field of view scan image adjacent to the first field of view scan image is first determined, wherein the number of second field of view scan images adjacent to the first field of view scan image may be one or more. In one possible implementation, the second field of view scan image may be all field of view scan images adjacent to the first field of view scan image, or may be a portion of field of view scan images adjacent to the first field of view scan image, for example, including field of view scan images located to the left and / or above the first field of view scan image, although this is not limited in the present embodiment.
[0257] After determining the second field of view scan image, a two-dimensional window sliding process is performed on the first field of view scan image and the second field of view scan image based on the overlapping area between the first field of view scan image and the second field of view scan image to determine matching parameters between the first field of view scan image and the second field of view scan image. The matching parameters are used to indicate the optimal matching position between the first field of view scan image and the second field of view scan image.
[0258] The process of two-dimensional window sliding processing, for example, can be to overlap the first field of view scanning image and the second field of view scanning image first, and then slide the second field of view scanning image according to a preset step size, so that the relative position of the first field of view scanning image and the second field of view scanning image changes. Then, the first field of view scanning image and the second field of view scanning image are matched with the part of the first field of view scanning image and the second field of view scanning image in the same area (that is, the overlapping area of the first field of view scanning image and the second field of view scanning image), and the similarity between the part of the first field of view scanning image in the area and the part of the second field of view scanning image in the area is determined. When the similarity exceeds the preset threshold, the matching position at this time is determined as the best matching position between the first field of view scanning image and the second field of view scanning image. According to the position of the first field of view scanning image and the position of the second field of view scanning image at this time, the relative position between the first field of view scanning image and the second field of view scanning image can be determined.
[0259] The relative position can be indicated in the form of relative coordinates. For example, the upper left corner of the first field of view scanning image can be used as a reference, and the relative coordinate value of the upper left corner of the second field of view scanning image relative to the upper left corner of the first field of view scanning image can be determined based on the relative position between the first field of view scanning image and the second field of view scanning image. After the position of the first field of view scanning image is determined, the position of the second field of view scanning image can be determined based on the relative coordinate value. For example, the upper right corner of the first field of view scanning image can be used as a reference, and the relative coordinate value of the upper left corner of the second field of view scanning image relative to the upper right corner of the first field of view scanning image can be determined based on the relative position between the first field of view scanning image and the second field of view scanning image; the lower left corner of the first field of view scanning image can be used as a reference, and the relative coordinate value of the lower left corner of the second field of view scanning image relative to the lower left corner of the first field of view scanning image can be determined based on the relative position between the first field of view scanning image and the second field of view scanning image, and so on. The presently disclosed embodiments do not limit this.
[0260] S72 , performing fusion processing on the multiple visual field scanning images according to the relative positions between any adjacent visual field scanning images to obtain a full-width scanning image.
[0261] Among the multiple visual field scanning images, the position of a reference visual field scanning image is first determined, wherein the reference visual field scanning image is any one of the multiple visual field scanning images, for example, the first, second, third, etc. of the multiple visual field scanning images. The position of the reference visual field scanning image may also be represented by the coordinates of the reference visual field scanning image, and the coordinates of the reference visual field scanning image may be represented by the coordinates of a certain point on the reference visual field scanning image, for example, the coordinates of the upper left corner, the upper right corner, the lower left corner, the lower right corner, the coordinates of the middle point, etc. of the reference visual field scanning image.
[0262] After determining the position of the reference field of view scan image, since the relative positions between any adjacent field of view scan images have been determined before, the positions of other field of view scan images can be determined based on the position of the reference field of view scan image and the relative positions between any adjacent field of view scan images. The positions of other field of view scan images can also be represented by the coordinates of other field of view scan images.
[0263] Taking the reference visual field scanning image as the first visual field scanning image as an example, the visual field scanning images adjacent to the first visual field scanning image are the second visual field scanning image and the fifth visual field scanning image, and the second visual field scanning image is adjacent to the third visual field scanning image, and the fifth visual field scanning image is adjacent to the fourth visual field scanning image and the sixth visual field scanning image. After determining the absolute coordinates of the first visual field scanning image, the absolute coordinates of the second visual field scanning image can be determined according to the relative positional relationship between the first visual field scanning image and the second visual field scanning image, and the absolute coordinates of the fifth visual field scanning image can be determined according to the relative positional relationship between the first visual field scanning image and the fifth visual field scanning image.
[0264] Furthermore, after determining the absolute coordinates of the second visual field scanning image, the absolute coordinates of the third visual field scanning image can be determined based on the relative positional relationship between the second and third visual field scanning images. After determining the absolute coordinates of the fifth visual field scanning image, the absolute coordinates of the fourth visual field scanning image can be determined based on the relative positional relationship between the fifth and fourth visual field scanning images. The absolute coordinates of the sixth visual field scanning image can be determined based on the relative positional relationship between the fifth and sixth visual field scanning images. Furthermore, the absolute coordinates of all visual field scanning images can be determined.
[0265] In summary, after determining the position of the reference field of view scan image, the positions of the other field of view scan images can be determined by combining the relative positions of any adjacent field of view scan images. Then, based on the position of the reference field of view scan image and the positions of the other reference field of view scan images, multiple field of view scan images are fused to obtain the full scan image.
[0266] After obtaining the full-width scan image, first position information of multiple key points in the full-width scan image is determined according to the substrate pattern. The specific implementation process can be found in the relevant introduction in S13 and will not be repeated here.
[0267] After obtaining the first position information of multiple key points in the full-scan image, the full-scan image is corrected according to the first position information and the substrate pattern on the biochip to obtain a target full-scan image, which is described below in conjunction with Figure 14.
[0268] FIG14 is a flow chart of a correction process according to an embodiment of the present disclosure, as shown in FIG14 , including:
[0269] S81 , determining second position information of a plurality of key points in the full-scan image according to the substrate pattern and the size of the full-scan image.
[0270] The first position information is the position information of multiple key points obtained by identifying and processing the actual full-width scanned image. It should be noted that due to various reasons such as a certain degree of distortion in the scanning process of the scanner and errors in the fusion process, there are certain errors between the obtained full-width scanned image and the theoretical full-width scanned image of the biochip. Therefore, the first position information and the second position information are not exactly the same, so correction processing is required.
[0271] Based on the substrate pattern on the biochip and the size of the full-scan image, second position information of multiple key points in the full-scan image can be determined. The second position information represents the theoretical positions of the multiple key points in the full-scan image. Specifically, the full-scan image is scaled and processed based on the size of the full-scan image and the size of the biochip, so that the size of the full-scan image matches the size of the biochip. Then, based on the design of the substrate pattern and the degree of scaling and processing of the full-scan image (which can be represented by a matrix), the second position information of the multiple key points in the full-scan image is determined. For example, based on the size of the full-scan image, it is determined that scaling and processing of the full-scan image are required to obtain an image that matches the biochip. The processed full-scan image is denoted as image A. Assuming that point B on the biochip has a key point based on the design of the substrate pattern, the corresponding point C on image A can be determined based on the position of point B on the biochip. The position of point C on image A is the same as the position of point B on the biochip. Since image A is obtained by scaling the full-frame scan image, image A is reversely processed to obtain the original full-frame scan image, and the position of point C on the full-frame scan image can be determined, that is, the second position information of the key point.
[0272] S82: Correct the full-width scanned image according to the first position information and the second position information to obtain a target full-width scanned image.
[0273] After obtaining the first position information and the second position information of the plurality of key points, a correction conversion matrix may be determined based on the first position information and the second position information. The correction conversion matrix is used to achieve conversion between the first position information and the second position information.
[0274] Then, based on the correction transformation matrix, the full-scan image is mapped and transformed to obtain a final, correct full-scan image. The correction transformation matrix indicates the transformation from first position information to second position information. The first position information is the position information of multiple key points determined based on the fused full-scan image, and the second position information is the theoretical position information of multiple key points in the full-scan image. Correction of multiple key points is achieved from the first position information to the second position information. The correction transformation matrix, obtained based on the first position information and the second position information, represents the transformation parameters from the first position information to the second position information. Therefore, mapping the full-scan image based on the correction transformation matrix can correct the full-scan image and obtain a corrected target full-scan image. The mapping transformation performed on the full-scan image based on the correction transformation matrix essentially multiplies the pixel matrix in the full-scan image by the correction transformation matrix. This may include scaling or translating the full-scan image, etc., which is not limited in the present embodiment.
[0275] Figure 15 is a schematic diagram of a full-width scanned image of a target provided by an embodiment of the present disclosure, and Figure 16 is an enlarged schematic diagram of a full-width scanned image of a target after superimposing a base pattern provided by an embodiment of the present disclosure. As shown in Figure 15, after the correction process, a complete full-width scanned image of the target can be obtained. As shown in Figure 16, in order to evaluate the accuracy of the correction, a layer of base pattern of a biochip is superimposed on the obtained full-width scanned image of the target. The left example in Figure 16 shows an image obtained without adopting the scanning image correction method of the embodiment of the present disclosure. It can be seen that without correction, there will be ghosting or errors in the image, and the accuracy is low. The right example in Figure 16 shows an image obtained by adopting the scanning image correction method of the embodiment of the present disclosure. It can be seen that when correction is performed using this solution, the images will completely overlap, there will be no ghosting, and the accuracy is high.
[0276] The scanning image correction method provided by the embodiment of the present disclosure is to collect multiple field of view scanning images of the biochip while the tissue slice is attached to the biochip. The biochip includes a substrate pattern, and the substrate pattern includes multiple key points. There is an overlapping area between any adjacent field of view scanning images. Then, based on the overlapping area between any adjacent field of view scanning images, the multiple field of view scanning images are fused to obtain a full-width scanning image of the biochip, thereby achieving the stitching of the field of view scanning images. Furthermore, based on the substrate pattern, the first position information of the multiple key points in the full-width scanning image is determined, and the full-width scanning image is corrected based on the first position information and the substrate pattern to obtain a target full-width scanning image. The first position information and second position information of the multiple key points in the full-width scanning image can be used to correct the generated image errors, thereby achieving accurate alignment of the gene expression information captured by the biochip and the area represented by the full-width scanning image to meet subsequent high-precision alignment requirements.
[0277] Optionally, for correction of the fluorescence scanning image, as shown in FIG17 , step 101 may include:
[0278] S110, with the tissue slice attached to the biochip, fix the sample with methanol and then perform fluorescence staining, and then perform fluorescence scanning imaging to obtain multiple binary channel sub-images of the biochip and fluorescence channel sub-images corresponding to each binary channel sub-image.
[0279] Tissue slices are slices made from biological tissue, such as pathological tissue slices. After obtaining the tissue slices, they are switched and attached to the biochip according to a professional slicing process. Then, they are processed with appropriate reagents and fixed for scanning.
[0280] A biochip is a device used to capture genetic expression information in cells. It has a specific substrate pattern on it. The substrate pattern can be the same or different on different biochips. The substrate pattern includes, but is not limited to, circular arrays, square arrays, and more.
[0281] For example, the substrate pattern on the biochip can be understood in conjunction with FIG8 , which will not be described in detail here.
[0282] In the disclosed embodiment, the biobarcode on the biochip is used to capture the gene expression information of each cell in the tissue slice. Before the tissue slice is attached to the biochip, the biobarcode on the biochip has already captured the gene expression information of each cell in the tissue slice.
[0283] After the tissue slice is attached to the biochip and fixed, multiple binary channel sub-images of the biochip and fluorescent channel sub-images corresponding to each binary channel sub-image can be collected.
[0284] Optionally, use a professional scanner that can scan multiple channels. The multi-channel professional scanner includes multiple spectrometers. Replace one of the spectrometers with a binary mirror. Scan the biochip based on the binary mirror to obtain the corresponding binary mirror channel sub-image. Scan the biochip based on the spectrometer to obtain the fluorescence channel sub-image.
[0285] The biochip is scanned by controlling a professional scanner. For each different field of view of the biochip, it is scanned by the binary mirror in the professional scanner to obtain the corresponding binary mirror channel sub-image. By scanning it by the spectrometer in the professional scanner, the corresponding fluorescence channel sub-image can be obtained. The field of view of this set of binary mirror channel sub-images and fluorescence channel sub-images is the same.
[0286] During the scanning process, the position of the professional scanner will change accordingly, and with this change in position, the field of view of the professional scanner will also change. However, during a single scan, the field of view of the resulting set of binary channel sub-images and fluorescence channel sub-images is identical. This means that the binary channel sub-images and fluorescence channel sub-images correspond to the same region within the biochip. It's just that because the binary channel and the fluorescence channel are two different channels, the imaging effects differ. Fluorescence imaging is a passive imaging method. For tissue sections, only the luminescent portion has a corresponding brightness. Since the substrate pattern on the biochip does not actively emit light, it cannot be displayed in the fluorescence channel sub-image. However, the binary channel is not a passive imaging method. The substrate pattern on the biochip can be imaged in the binary channel, so it can be displayed in the binary channel sub-image.
[0287] The imaging effects of the bisection mirror channel and the fluorescence channel can be understood in conjunction with Figure 18. Figure 18 is a comparison diagram of the imaging effects of the bisection mirror channel and the fluorescence channel provided by an embodiment of the present disclosure. As shown in Figure 18, the image 1831 on the left is a sub-image of the bisection mirror channel obtained by scanning a local area on the biochip, and the image 1832 on the right is a sub-image of the fluorescence channel obtained by scanning the local area.
[0288] Images 1831 and 1832 scan the same area, but the imaging results are quite different. Image 1831 is a sub-image obtained using the binocular channel. The biochip's substrate pattern is imaged in image 1831, along with some visible cellular tissue. Image 1832 is a sub-image obtained using the fluorescence channel. Because the substrate pattern does not actively emit light, the substrate pattern is not visible in image 1832.
[0289] It can also be seen from Figure 18 that since the fluorescence channel is based on the principle of passive imaging, most of the imaged areas are dark or black, which also leads to insufficient feature points in the fluorescence channel sub-image, making it difficult to accurately align it with the biochip.
[0290] It should be noted that the fields of view of different binary channel sub-images are different. The field of view of any binary channel sub-image is a local area of the biochip. The fields of view of multiple binary channel sub-images together constitute the global area of the entire biochip. Adjacent binary channel sub-images may or may not overlap. Correspondingly, the fields of view of different fluorescent channel sub-images are different. The field of view of any fluorescent channel sub-image is a local area of the biochip. The fields of view of multiple fluorescent channel sub-images together constitute the global area of the entire biochip. Adjacent fluorescent channel sub-images may or may not overlap.
[0291] S120 , amplifying the substrate pattern included on the biochip according to the scanning magnifications corresponding to the plurality of binary mirror channel sub-images to obtain a full-width reference image.
[0292] For any set of bisection channel sub-images and fluorescence channel sub-images, their sizes are identical, even for different bisection channel sub-images. Therefore, the scanning magnification corresponding to the bisection channel sub-image is also the same as the scanning magnification of the fluorescence channel sub-image. The scanning magnification corresponding to the bisection channel sub-image is a pre-determined value that represents the relationship between the size of a local region in the biochip and the size of the corresponding bisection channel sub-image. Specifically, the size obtained by magnifying the size of the local region by a certain factor is the size corresponding to the corresponding bisection channel sub-image, and the magnification factor is the scanning magnification.
[0293] The process of the amplification process can be understood, for example, in conjunction with Figure 19. Figure 19 is a schematic diagram of the substrate pattern amplification process provided by an embodiment of the present disclosure. As shown in Figure 19, for a biochip 1940, the biochip 1940 includes a substrate pattern, and the length of the biochip 1940 is a and the width is b.
[0294] Taking a scanning magnification of 5 times as an example, after the substrate pattern on the biochip 1940 is magnified 5 times, the size of the obtained full-width reference image is: length 5a, width 5b.
[0295] Since the scanning magnification is also the magnification used by the professional scanner to scan the different fields of view of biochip 1940, for a local region corresponding to any field of view in biochip 1940, the size of the bisection channel sub-image and the fluorescence channel sub-image obtained after the professional scanner scans that local region is also 5 times the size of the local region (taking a scanning magnification of 5 as an example). Therefore, the size of the full-frame bisection image obtained by fusing all scanned bisection channel sub-images should be the same as the size of the full-frame reference image, and the size of the full-frame fluorescence image obtained by fusing all scanned fluorescence channel sub-images should also be the same as the size of the full-frame reference image.
[0296] S130 , determining a correction position of each of the two-mirror channel sub-images in the full-frame reference image according to the acquisition order of the plurality of two-mirror channel sub-images and the plurality of key points included in the substrate pattern.
[0297] The acquisition sequence should be a pre-set sequence, and the corresponding acquisition sequence should be determined for different fields of view on the biochip. Therefore, based on the acquisition sequence of multiple binary channel sub-images, the approximate positions of the multiple binary channel sub-images in the full-frame reference image can be roughly estimated. The approximate relative positional relationship between the multiple binary channel sub-images can also be determined, and which other binary channel sub-images any one binary channel sub-image is adjacent to, etc., and the same applies to the fluorescence channel sub-images. For example, the pre-set acquisition sequence is to divide the field of view of the biochip into 5 rows and 5 columns, and for each row, sequentially acquire from left to right, and after acquiring one row, then acquire the second row. Based on the acquisition sequence, it can be determined that the 1st to 5th binary channel sub-images are sequentially adjacent, the 6th to 10th binary channel sub-images are sequentially adjacent, the 1st and 6th binary channel sub-images are adjacent, and so on. The same applies to the fluorescence channel sub-images.
[0298] The substrate pattern includes multiple key points, so the full-frame reference image obtained by magnifying the substrate pattern also includes these key points. Any bisection channel sub-image should also include at least one key point. In one possible implementation, the corrected position of the bisection channel sub-image in the full-frame reference image can be determined based on the matching of key points in the bisection channel sub-image with key points in the full-frame reference image.
[0299] In a possible implementation, the corrected position can be obtained by first determining the initial position of the binary mirror channel sub-image in the full-frame reference image and correcting the initial position. This implementation is described below in conjunction with FIG20 .
[0300] FIG20 is a schematic diagram of a process for determining the corrected position of a two-mirror channel sub-image in a full-frame reference image according to an embodiment of the present disclosure, as shown in FIG20 , including:
[0301] S510, for any binary mirror channel sub-image, perform key point recognition processing on the binary mirror channel sub-image to determine first coordinate information of a first key point in the binary mirror channel sub-image, where the first key point is a key point included in the binary mirror channel sub-image among multiple key points.
[0302] In one possible implementation, the key point recognition model can be used to identify the first coordinate information of the first key point in the two-mirror channel sub-image. It should be noted that in the disclosed embodiment, any one of the multiple two-mirror channel sub-images is used as an example to describe how to determine the corrected position of the two-mirror channel sub-image in the full-frame reference image. The methods for determining the corrected position of different two-mirror channel sub-images in the full-frame reference image are similar, and the disclosed embodiment only uses any one of them as an example for description.
[0303] Specifically, a corresponding key point recognition model is first determined. The key point recognition model is pre-trained based on multiple sets of training samples, where any set of training samples includes a sample bisection channel sub-image and label information corresponding to the sample bisection channel sub-image.
[0304] The sample bisection channel sub-image is an image obtained by scanning a sample biochip using a specialized scanner. The sample bisection channel sub-image includes sample keypoints, and the label information includes the coordinate information of the sample keypoints within the sample bisection channel sub-image. The label information can be in the form of an image, data, or both. The image form refers to an image obtained by labeling the sample keypoints within the sample bisection channel sub-image, while the data form refers to the coordinate data used to label the positions of the sample keypoints within the sample bisection channel sub-image.
[0305] After the key point recognition model is trained, it is able to recognize key points. By inputting the binary channel sub-image into the key point recognition model, the first coordinate information of the first key point in the binary channel sub-image can be obtained.
[0306] S520: Determine the initial position of the binary mirror channel sub-image in the full-frame reference image according to the acquisition sequence and the first coordinate information.
[0307] Specifically, first, according to the acquisition order, the second coordinate information of the first key point in the full-frame reference image is determined.
[0308] As described in the above embodiments, the approximate position of the bisection channel sub-image within the full-frame reference image can be roughly determined based on the acquisition sequence, and thus the first key point included in the bisection channel can also be determined. Since the substrate pattern includes all key points, once the first key point included in the bisection channel sub-image is determined, the second coordinate information of the first key point within the full-frame reference image can be determined.
[0309] Then, the initial position of the two-mirror channel sub-image in the full-frame reference image is determined based on the first coordinate information and the second coordinate information. The first coordinate information is the coordinate information of the first key point in the two-mirror channel sub-image, and the second coordinate information is the coordinate information of the first key point in the full-frame reference image. The first coordinate information and the second coordinate information both indicate the position of the first key point, and the position of the first key point is constant. Therefore, the initial position of the two-mirror channel sub-image in the full-frame reference image can be determined based on the first coordinate information and the second coordinate information. This initial position is a position where the coordinates indicated by the first coordinate information and the second coordinate information are the same.
[0310] S530: Correct the initial position to obtain the corrected position of the binary mirror channel sub-image in the full-frame reference image.
[0311] It should be noted that, due to limitations on the mechanical accuracy of professional scanners and other reasons, the initial position determined in S52 may have a certain deviation, and therefore the initial position needs to be corrected.
[0312] Specifically, after determining the initial position of the bisection mirror channel sub-image in the full-frame reference image, it is necessary to determine the local sub-image corresponding to the bisection mirror channel sub-image in the full-frame reference image based on the initial position. The local sub-image is part of the full-frame reference image and must meet the condition that the area corresponding to the local sub-image in the full-frame reference image includes the area corresponding to the bisection mirror channel sub-image in the full-frame reference image. For example, if the area corresponding to the bisection mirror channel sub-image in the full-frame reference image is region A, and the area corresponding to the local sub-image corresponding to the bisection mirror channel sub-image in the full-frame reference image is region B, then the condition that region B must meet is that region A is a sub-region of region B.
[0313] After the local sub-image corresponding to the bisection mirror channel sub-image is determined, a matching process is performed on the local sub-image and the bisection mirror channel sub-image to obtain a corresponding matching result.
[0314] In one possible implementation, matching can be performed as follows:
[0315] While the local sub-image is fixed, multiple relative positions of the bisection mirror channel sub-image relative to the local sub-image are determined based on a preset step size. A two-dimensional window sliding process is then performed on the bisection mirror channel sub-image based on these multiple relative positions to determine the bisection mirror channel sub-image at each relative position. The local sub-image is then matched with the bisection mirror channel sub-image at each relative position to obtain a corresponding matching result, which includes the similarity between the bisection mirror channel sub-image and the local sub-image at each relative position.
[0316] For any relative position, when the bisection channel sub-image and the local sub-image are at that relative position, the bisection channel sub-image and the local sub-image are matched to determine their similarity. By using the above method, the similarity between the bisection channel sub-image and the local sub-image at different relative positions can be obtained.
[0317] After obtaining the similarities between the bisection mirror channel sub-image and the local sub-image at different relative positions, the relative position with the highest similarity is determined as the target relative position of the bisection mirror channel sub-image relative to the local sub-image.
[0318] After obtaining the target relative position, the corrected position of the bisection mirror channel sub-image in the full-frame reference image is obtained based on the target relative position and the position of the local sub-image in the full-frame reference image. Since the position of the local sub-image in the full-frame reference image is known, and the target relative position indicates the relative position of the local sub-image and the bisection mirror channel sub-image, the corrected position of the bisection mirror channel sub-image in the full-frame reference image can be determined based on the target relative position and the position of the local sub-image in the full-frame reference image.
[0319] The above embodiment introduces how to determine the correction position of any binary mirror channel sub-image in the full-frame reference image. For each binary mirror channel sub-image, the above method can be used to determine its correction position in the full-frame reference image, thereby obtaining the correction position of the binary mirror channel sub-image corresponding to different fields of view of the biochip in the full-frame reference image.
[0320] S140, performing fusion processing on the fluorescence channel sub-images corresponding to the respective binocular channel sub-images according to the corrected positions of the respective binocular channel sub-images in the full-frame reference image, to obtain the full-frame fluorescence image of the biochip as the corrected scanned image.
[0321] The corrected position of the binocular channel sub-image in the full-frame reference image is the precise position of the binocular channel sub-image in the full-frame reference image. For any set of binocular channel sub-images and corresponding fluorescence channel sub-images, the corrected position of the binocular channel sub-image in the full-frame reference image is identical to the corrected position of the corresponding fluorescence channel sub-image in the full-frame reference image. Therefore, based on the corrected position of each binocular channel sub-image in the full-frame reference image, the corrected position of each fluorescence channel sub-image in the full-frame reference image can be determined. Then, based on the corrected position of each fluorescence channel sub-image in the full-frame reference image, the corresponding fluorescence channel sub-images of each binocular channel sub-image are fused to obtain the full-frame fluorescence image of the biochip. The fusion process involves arranging and splicing the fluorescence channel sub-images according to their corrected positions in the full-frame reference image.
[0322] Figure 21 is a schematic diagram of a full-width fluorescence image provided by an embodiment of the present disclosure, and Figure 22 is a schematic diagram of a fluorescence channel sub-image before and after fusion provided by an embodiment of the present disclosure. As shown in Figure 21, after fusion, a complete full-width fluorescence image can be obtained. As shown in Figure 22, the example on the left side of Figure 22 is a fluorescence channel sub-image that has not been corrected, and the example on the right side of Figure 22 is a fluorescence channel sub-image that has been corrected. According to the comparison between the two, it can be seen that after correction, the previous problem of being unable to align is corrected, and a full-width fluorescence image with better accuracy is obtained.
[0323] After determining the corrected positions of each bisection channel sub-image within the full-frame reference image, the bisection channel sub-images can be fused to obtain the full-frame bisection image shown in Figure 3. The fusion process is similar to that for the fluorescence channel sub-images. As can be seen, the fused full-frame bisection image and the full-frame fluorescence image in Figure 21 have the same content, but exhibit different imaging effects.
[0324] Figure 23 is an enlarged schematic diagram of a full-frame fluorescence image after superimposing a base pattern, as provided by an embodiment of the present disclosure. As shown in Figure 23, to evaluate the accuracy of the correction, a layer of the biochip base pattern was superimposed on the obtained full-frame fluorescence image. Full-frame fluorescence images obtained without the scanning image processing method of the present disclosure would have ghosting or errors, resulting in lower accuracy. Full-frame fluorescence images obtained using the scanning image processing method of the present disclosure would completely overlap, without ghosting, and thus have higher accuracy.
[0325] The scanning image processing method provided by the embodiment of the present disclosure collects multiple binary mirror channel sub-images of the biochip and fluorescence channel sub-images corresponding to each binary mirror channel sub-image when the tissue slice is attached to the biochip, and amplifies the substrate pattern included in the biochip according to the scanning magnifications corresponding to the multiple binary mirror channel sub-images to obtain a full-width reference image; then, based on the collection order of the multiple binary mirror channel sub-images and multiple key points included in the substrate pattern, the correction position of each binary mirror channel sub-image in the full-width reference image is determined; finally, based on the correction position of each binary mirror channel sub-image in the full-width reference image, the fluorescence channel sub-image corresponding to each binary mirror channel sub-image is fused to obtain a full-width fluorescence image of the biochip. When there is a large positional deviation between the full-frame fluorescence image and the biochip due to insufficient feature points in the fluorescence channel sub-image, sufficient feature points, that is, multiple key points in the substrate pattern, are captured by synchronously acquiring the binary channel sub-image corresponding to the fluorescence channel sub-image. The resulting positional error is then corrected by combining multiple key points with the acquisition sequence of multiple binary channel sub-images, achieving accurate alignment between the gene expression information captured by the biochip and the area represented by the full-frame fluorescence image to meet subsequent high-precision alignment requirements.
[0326] Optionally, a single-cell spatial transcriptome sequencing library is used for clustering processing, the tissue section is a fresh frozen tissue section, and the biochip is a spatial transcriptome chip.
[0327] In the disclosed embodiment, fresh frozen tissue sections are fixed on a gene expression S1000 chip. For animal samples, fluorescent staining of the cell nucleus is performed, such as DAPI or ssDNA staining, followed by bright field staining, such as HE staining, Masson staining, etc.; for plant samples, FB28 tissue autofluorescence enhancement staining can be performed first, or toluidine blue or HE staining can be performed directly; after the scan is completed, the chip is placed in a card holder, and the tissue is permeabilized using the configured permeabilization enzyme. The tissue permeabilization solution is then removed, and a single-strand cDNA synthesis mixture is added for reverse transcription. After completion, the RNA is denatured and then the second-strand cDNA is synthesized. Finally, the cDNA sample is obtained by denaturing and melting the cDNA, and amplified to obtain a sample that can be used to construct a second-generation library.
[0328] The embodiments of the present disclosure adopt the following technical solutions:
[0329] (1) Tissue fixation and staining: Place the frozen sections in -20℃ methanol and fix them for 30-60 minutes. Then, perform DAPI staining on animal samples and perform fluorescence scanning imaging and cell nucleus fluorescence scanning under full-spectrum light source. Then perform bright-field HE staining. For plant samples, perform FB28 tissue autofluorescence enhancement staining, or directly perform toluidine blue staining and scan under bright-field light source to obtain uncorrected scanning images of single-cell spatial transcriptome.
[0330] (2) After scanning, tissue permeabilization is performed: the chip with the fixed and stained tissue is placed in the matching card holder, and the tissue permeabilization reagent is prepared and incubated according to the permeabilization time determined by tissue optimization. After permeabilization, the tissue permeabilization reagent is removed and 0.1% SSC solution is added to the well.
[0331] (3) cDNA single-strand synthesis: Remove the 0.1% SSC solution in the wells, add the prepared single-strand synthesis reagent into the wells, and incubate at 42°C for 1.5 hours.
[0332] (4) cDNA second-strand synthesis: remove the first-strand synthesis reagent, add 0.08 M KOH for denaturation, wash with EB buffer, add the second-strand synthesis reagent, and incubate at 65°C for 30 minutes.
[0333] (5) cDNA melting and amplification: Remove the second-strand synthesis reagent, add EB buffer for washing, then add 0.08 M KOH and incubate at room temperature for 10 min. After neutralizing the KOH with cDNA with a certain proportion of Tris, take 1 μl of sample for qPCR detection to determine the cycle number, and then use the amplification enzyme to amplify the cDNA of the remaining sample.
[0334] (6) cDNA purification: The amplified product was purified using SPRIselect magnetic beads, washed with 80% ethanol, and eluted with EB buffer.
[0335] (7) cDNA quantification and quality control: The concentration of the purified cDNA was detected using Qubit. 1 μl was diluted to 2 ng / μl and the length distribution of the final cDNA was detected using the highly sensitive Agilent 2100 detection chip.
[0336] The following examples are used to illustrate the embodiments of the present disclosure, but are not intended to limit the scope of the embodiments of the present disclosure. Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art, and the raw materials used are all commercially available products.
[0337] The scanner used in the following examples was a Pannoramic Scan II, purchased from 3D HISTECH. In the fluorescence channel, by removing the scanner's filter cube and retaining only the dichroic mirror, a dichroic mirror setup was performed, enabling simultaneous scanning of cell nuclei and background microsphere arrays.
[0338] The chip used was the S1000 spatial transcriptome chip, purchased from Beijing Biomarker Biotechnology Co., Ltd.
[0339] Example 1: Construction of a Single-Cell Spatial Transcriptome Sequencing Library (Animal Tissue Samples)
[0340] 1. Staining methods for animal tissues
[0341] 1.1 Tissue Fixation
[0342] 1.1.1 Place 40 ml of methanol in a 50 ml centrifuge tube and pre-cool to -20°C;
[0343] 1.1.2 Set the PCR instrument to 37°C and the heated lid to 37°C or 50°C, and place the adapter to equilibrate the temperature.
[0344] 1.1.3 Remove the chip with the mouse testis tissue (10 μm slice thickness) from -80°C (shipped on dry ice) and incubate it tissue-side up on a 37°C adapter for 1 minute (do not cover with the hot cover). This step should also be repeated if the tissue has just been mounted in a cryostat.
[0345] 1.1.4 Place the chip in pre-cooled methanol and incubate at -20°C for 30 minutes;
[0346] During this time, reagents required for tissue staining can be prepared.
[0347] 1.2 Fluorescent staining
[0348] 1.2.1 Wipe off excess methanol from the fixed chip and place it on a clean surface of the platform;
[0349] 1.2.2 Prepare the fluorescent staining mix according to Table 1:
[0350] Table 1
[0351] 1.2.3 Add the fluorescent staining mix to the tissue surface and evenly cover it. Incubate in the dark for 5 minutes.
[0352] 1.2.4 Tilt the chip to absorb the fluorescent staining mix;
[0353] 1.2.5 Add 100 μl of prepared 0.1× SSC buffer to the wells and evenly cover the matrix;
[0354] 1.2.6 Tilt the chip to draw up 0.1× SSC buffer, open the card holder, and wipe off excess liquid. Gently shake the chip to evaporate excess liquid on the tissue.
[0355] 1.3 Fluorescence scanning imaging
[0356] 1.3.1 Place the chip on the carrier and load it into the instrument, setting up the dichroic channel and the FAM channel;
[0357] 1.3.2 After the system pre-acquires the image, place the cursor on the tissue image for autofocus;
[0358] 1.3.3 Select the folder where you want to save the image and name it;
[0359] 1.3.4 Frame the tissue location and click Scan.
[0360] 1.4 Tissue staining
[0361] 1.4.1 Prepare three 1L beakers in advance and add at least 800ml of ultrapure water to each beaker;
[0362] 1.4.2 Prepare eosin mix according to Table 2:
[0363] Table 2
[0364] Hematoxylin staining solution must be filtered through a 0.2 μm filter membrane before use;
[0365] 1.4.4 Add isopropanol to the tissue surface and evenly cover it. Incubate at room temperature for 1 minute.
[0366] 1.4.5 Tilt the chip to remove the isopropyl alcohol and wipe off any excess liquid. Gently shake the chip to evaporate any excess isopropyl alcohol on the tissue. This process should not exceed 5 minutes to avoid over-drying.
[0367] 1.4.6 Add filtered hematoxylin stain to the tissue surface and cover evenly. Incubate at room temperature for 1-7 minutes (depending on the tissue).
[0368] 1.4.7 Tilt the chip to remove the hematoxylin stain. Place the chip in the first cup of ultrapure water and gently rinse it about 15 times. Then place it in the second cup of ultrapure water and gently rinse it about 15 times. Wipe off any excess liquid.
[0369] 1.4.8 Add indigo dye to the tissue surface and evenly cover it. Incubate at room temperature for 30 seconds to 2 minutes (depending on the tissue).
[0370] 1.4.9 Tilt the chip to remove the indigo stain, then gently rinse it in a second cup of ultrapure water about five times, wiping off any excess liquid.
[0371] 1.4.10 Add the prepared eosin mix to the tissue surface and evenly cover it. Incubate at room temperature for 30 seconds to 1 minute (depending on the tissue).
[0372] 1.4.11 Tilt the chip to remove the eosin mix, then gently rinse it about 20 times in a third cup of ultrapure water, wiping off any excess liquid.
[0373] 1.4.12 Place the chip in a clean 50ml centrifuge tube and centrifuge at 250g for 30 seconds.
[0374] 1.4.13 Remove the chip and incubate it on a 37°C adapter for 1 minute.
[0375] 1.5 Brightfield tissue imaging
[0376] 1.5.1 Place the chip on the carrier and load it into the instrument;
[0377] 1.5.2 After the system pre-acquires the image, place the cursor on the tissue image for autofocus;
[0378] 1.5.3 Adjust the focus value to ±50, select a folder to save the image and name it;
[0379] 1.5.4 Frame the tissue location and click Scan.
[0380] 2. Tissue Permeabilization
[0381] 2.1 Preheat the permeabilization enzyme (0.1% pepsin) in a 37°C metal bath for 10 min.
[0382] 2.2 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0383] 2.3 Install the chip into the chip holder and add 70 μl of preheated permeabilization enzyme along the side of the well to ensure that the enzyme evenly covers the entire matrix (do not introduce bubbles);
[0384] 2.4 Seal with sealing film, place on the adapter in the PCR instrument, cover with the hot cover, and incubate at 37°C for the determined time, using a timer;
[0385] 2.5 Dilute 20×SSC to 0.1×SSC, vortex to mix, and centrifuge briefly;
[0386] 2.6 Permeabilization time varies depending on the tissue and should be determined through tissue permeabilization experiments. When finished, remove the cartridge, tear open the seal, aspirate the permeabilization enzyme along the edge of the matrix, add 100 μL of the prepared 0.1× SSC buffer to the well, evenly covering the matrix (avoiding the introduction of air bubbles), and reseal.
[0387] 3. cDNA single-strand synthesis
[0388] 3.1 Prepare the following cDNA single-strand synthesis mix on ice:
[0389] Vortex mix and centrifuge briefly;
[0390] 3.2 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0391] 3.3 Tear open the seal, aspirate the buffer along the side of the well, add the first-strand synthesis mix (cover evenly and avoid bubbles), seal again, place the card holder on the adapter, cover with the hot cover, skip the preheating step, and start the first-strand synthesis.
[0392] 4. cDNA Second-Strand Synthesis and Denaturation
[0393] 4.1 Remove the card holder from the adapter, tear open the seal, and aspirate a strand of synthetic mix along the side of the hole;
[0394] 4.2 Add 100 μl of 0.08 M KOH to the wells and incubate at room temperature for 5 minutes.
[0395] 4.3 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the 0.08M KOH;
[0396] 4.4 Add 100 μl of EB buffer to the wells and incubate at room temperature for 2 minutes;
[0397] 4.5 Prepare the following cDNA second-strand synthesis mix on ice:
[0398] Second Strand Primer sequence: 5′-TTTCTGTTGGTGCTGATATTGC-3′.
[0399] Vortex mix and centrifuge briefly;
[0400] 4.6 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0401] 4.7 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the EB buffer;
[0402] 4.8 Add cDNA second-strand synthesis mix to the well (avoid bubbles), place it on the adapter, skip preheating, and start cDNA second-strand synthesis.
[0403] 4.9 After the cDNA second-strand synthesis is completed, remove the second-strand synthesis mix from the well;
[0404] 4. Add 100 μl of EB buffer to each well and incubate at room temperature for 2 minutes.
[0405] 4.11 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the EB buffer;
[0406] 4. Add 45 μl of 0.08 M KOH to each well, ensuring that it covers the entire matrix. Incubate at room temperature for 10 minutes, pipetting twice every two minutes.
[0407] 4.13 Pipette out 42 μl of 0.08 M KOH and 6 μl of Tris (1 M, pH 7.0) into a 0.2 ml low-adsorption tube, mix well, centrifuge briefly, and place on ice.
[0408] 5. cDNA Amplification
[0409] 5.1 qPCR cycle number determination
[0410] 5.1.1 Prepare the following qPCR mix on ice:
[0411] Vortex mix and centrifuge briefly;
[0412] 5.1.2 Add 9 μl of qPCR mix to each well of the qPCR plate (including two replicate samples, a negative control, and a positive control. The qPCR mix system can be increased proportionally).
[0413] 5.1.3 Pipette the sample from step 4.13 and add 1 μl of the sample to each replicate well. Add 1 μl of water to the negative control well. If a positive control is added, add 1 μl of the quality-controlled cDNA sample to the well.
[0414] 5.1.4 Perform qPCR reaction according to the following procedure:
[0415] 5.1.5 Record the obtained Cq value.
[0416] 5.2 cDNA amplification
[0417] 5.2.1 Prepare the following cDNA amplification mix on ice:
[0418] The amp mix was KAPA HiFi HotStar DNA polymerase.
[0419] The cDNA Primer sequences are as follows:
[0420] Primer F: 5′-CTACACGACGCTCTTCCGATCT-3′
[0421] Primer R: 5′-TTTCTGTTGGTGCTGATATTGC-3′
[0422] Vortex mix and centrifuge briefly;
[0423] 5.2.2 Add 51 μl of cDNA amplification mix to the sample prepared in 4.13, mix thoroughly by pipetting, and centrifuge briefly to avoid air bubbles.
[0424] 5.2.3 Set the following program on the PCR instrument for cDNA amplification:
[0425] 5.2.4 Quality Control Node: Qubit can be used to measure the concentration of post-PCR products.
[0426] 6. cDNA Purification
[0427] 6.1 Vortex the SPRIselect magnetic beads to mix thoroughly. Add 60 μl of SPRIselect magnetic beads (0.6X) to the sample (100 μl) and pipette to mix 15 times.
[0428] 6.2 Incubate at room temperature for 5 minutes;
[0429] 65.3 Place on a magnetic stand until the solution becomes clear;
[0430] 6.4 Remove the supernatant;
[0431] 6.5 Add 200 μl of 80% ethanol to the tube and wait for 30 seconds;
[0432] 6.6 Remove ethanol;
[0433] 6.7 Repeat steps 6.5 and 6.6 for a total of 2 washes;
[0434] 6.8 Centrifuge briefly and transfer to a magnetic rack;
[0435] 6.9 Remove residual alcohol and dry at room temperature for 2 minutes (do not exceed 2 minutes, otherwise the elution efficiency will be reduced);
[0436] 6.10 Remove the tube from the magnetic rack, add 40.5 μl of EB buffer, and mix by pipetting 15 times;
[0437] 6.11 Incubate at room temperature for 2 minutes;
[0438] 6.12 Place the test tube on a magnetic rack until the solution becomes clear;
[0439] 6.13 Pipette 40 μl of liquid into a new test tube;
[0440] 6.14 Store at 4°C for 72 hours, or at -20°C for 4 weeks, or proceed directly to the next step.
[0441] 7. cDNA quantification and quality control
[0442] 7.1 Use Qubit to determine the concentration of the purified cDNA product and dilute it to 2 ng / ul;
[0443] 7.2 Use the High Sensitivity Agilent Technologies 2100 Bioanalyzerd to determine the peak shape of the cDNA product.
[0444] The experimental results showed that through the construction of a single-cell spatial transcriptome library, spatial transcriptome analysis with single-cell resolution was achieved, and mouse testicular tissue can truly reflect the distribution of various cell types in the tissue.
[0445] The original image of the animal tissue binaural scan setup is shown in Figure 3.
[0446] The original scanned image of fluorescently stained cell nuclei in animal tissue is shown in Figure 21.
[0447] The original image of the HE staining effect of animal tissue is shown in Figure 15.
[0448] The results of cell segmentation in the fluorescent channel of animal tissue are shown in Figure 4.
[0449] The results of HE-stained cell segmentation of animal tissue are shown in Figure 24.
[0450] The clustering diagram after animal tissue cell segmentation is shown in Figure 6.
[0451] Example 2: Construction of a Single-Cell Spatial Transcriptome Sequencing Library (Plant Tissue Samples)
[0452] 1. Staining method of plant tissue (tomato callus)
[0453] 1.1 Tissue Fixation
[0454] 1.1.1 Place 40 ml of methanol in a 50 ml centrifuge tube and pre-cool to -20°C;
[0455] 1.1.2 Set the PCR instrument to 37°C and the heated lid to 37°C or 50°C, and place the adapter to equilibrate the temperature.
[0456] 1.1.3 Remove the chip with the tissue attached (10 μm slice thickness) from -80°C (shipped on dry ice) and incubate it on a 37°C adapter with the tissue side facing up for 1 minute (do not cover with the hot cover). This step should also be performed if the tissue has just been attached in the cryostat.
[0457] 1.1.4 Place the chip in pre-cooled methanol and incubate at -20°C for 60 minutes;
[0458] During this time, reagents required for tissue staining can be prepared.
[0459] 1.2 Tissue staining
[0460] 1.2.1 Wipe off excess methanol from the fixed chip and place it on a clean surface of the platform;
[0461] 1.2.2 Stain with 0.05% toluidine blue dye for 5-20 seconds;
[0462] 1.2.3 Gently rinse in ultrapure water for about 5 times and wipe off excess liquid;
[0463] 1.2.4 Place the chip in a clean 50ml centrifuge tube and centrifuge at 250g for 30 seconds.
[0464] 1.2.5 Remove the chip and incubate it on a 37°C adapter for 1 minute.
[0465] 1.3 Brightfield tissue imaging
[0466] 1.3.1 Place the chip on the carrier and load it into the instrument;
[0467] 1.3.2 After the system pre-acquires the image, place the cursor on the tissue image for autofocus;
[0468] 1.3.3 Adjust the focus value to ±50, select a folder to save the image and name it;
[0469] 1.3.4 Frame the tissue location and click Scan.
[0470] 2. Tissue Permeabilization
[0471] 2.1 Preheat the permeabilization enzyme (0.1% pepsin) in a 37°C metal bath for 10 min.
[0472] 2.2 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0473] 2.3 Install the chip into the chip holder and add 70 μl of preheated permeabilization enzyme along the side of the well to ensure that the enzyme evenly covers the entire matrix (do not introduce bubbles);
[0474] 2.4 Seal with sealing film, place on the adapter in the PCR instrument, cover with the hot cover, and incubate at 37°C for the determined time, using a timer;
[0475] 2.5 Dilute 20×SSC to 0.1×SSC, vortex to mix, and centrifuge briefly;
[0476] 2.6 Permeabilization time varies depending on the tissue and should be determined through tissue permeabilization experiments. When finished, remove the cartridge, tear open the seal, aspirate the permeabilization enzyme along the edge of the matrix, add 100 μL of the prepared 0.1× SSC buffer to the well, evenly covering the matrix (avoiding the introduction of air bubbles), and reseal.
[0477] 3. cDNA single-strand synthesis
[0478] 3.1 Prepare the following cDNA single-strand synthesis mix on ice:
[0479] Vortex mix and centrifuge briefly;
[0480] 3.2 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0481] 3.3 Tear open the seal, aspirate the buffer along the side of the well, add the first-strand synthesis mix (cover evenly and avoid bubbles), seal again, place the card holder on the adapter, cover with the hot cover, skip the preheating step, and start the first-strand synthesis.
[0482] 4. cDNA Second-Strand Synthesis and Denaturation
[0483] 4.1 Remove the card holder from the adapter, tear open the seal, and aspirate a strand of synthetic mix along the side of the hole;
[0484] 4.2 Add 100 μl of 0.08 M KOH to the wells and incubate at room temperature for 5 minutes.
[0485] 4.3 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the 0.08M KOH;
[0486] 4.4 Add 100 μl of EB buffer to the wells and incubate at room temperature for 2 minutes;
[0487] 4.5 Prepare the following cDNA second-strand synthesis mix on ice:
[0488] Second Strand Primer sequence: 5′-TTTCTGTTGGTGCTGATATTGC-3′.
[0489] Vortex mix and centrifuge briefly;
[0490] 4.6 Place the adapter in the PCR instrument and set the PCR instrument to the following program:
[0491] 4.7 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the EB buffer;
[0492] 4.8 Add cDNA second-strand synthesis mix to the well (avoid bubbles), place it on the adapter, skip preheating, and start cDNA second-strand synthesis.
[0493] 4.9 After the cDNA second-strand synthesis is completed, remove the second-strand synthesis mix from the well;
[0494] 4. Add 100 μl of EB buffer to each well and incubate at room temperature for 2 minutes.
[0495] 4.11 Pipette slowly and evenly along the side of the well 5 times (do not pipette violently) and discard the EB buffer;
[0496] 4. Add 45 μl of 0.08 M KOH to each well, ensuring that it covers the entire matrix. Incubate at room temperature for 10 minutes, pipetting twice every two minutes.
[0497] 4.13 Pipette out 42 μl of 0.08 M KOH and 6 μl of Tris (1 M, pH 7.0) into a 0.2 ml low-adsorption tube, mix well, centrifuge briefly, and place on ice.
[0498] 5. cDNA Amplification
[0499] 5.1 qPCR cycle number determination
[0500] 5.1.1 Prepare the following qPCR mix on ice:
[0501] Vortex mix and centrifuge briefly;
[0502] 5.1.2 Add 9 μl of qPCR mix to each well of the qPCR plate (including two replicate samples, a negative control, and a positive control. The qPCR mix system can be increased proportionally).
[0503] 5.1.3 Pipette the sample from step 4.13 and add 1 μl of the sample to each replicate well. Add 1 μl of water to the negative control well. If a positive control is added, add 1 μl of the quality-controlled cDNA sample to the well.
[0504] 5.1.4 Perform qPCR reaction according to the following procedure:
[0505] 5.1.5 Record the obtained Cq value.
[0506] 5.2 cDNA amplification
[0507] 5.2.1 Prepare the following cDNA amplification mix on ice:
[0508] The amp mix was KAPA HiFi HotStar DNA polymerase.
[0509] The cDNA Primer sequences are as follows:
[0510] Primer F: 5′-CTACACGACGCTCTTCCGATCT-3′
[0511] Primer R: 5′-TTTCTGTTGGTGCTGATATTGC-3′
[0512] Vortex mix and centrifuge briefly;
[0513] 5.2.2 Add 51 μl of cDNA amplification mix to the sample prepared in 4.13, mix thoroughly by pipetting, and centrifuge briefly to avoid air bubbles.
[0514] 5.2.3 Set the following program on the PCR instrument for cDNA amplification:
[0515] 5.2.4 Quality Control Node: Qubit can be used to measure the concentration of post-PCR products.
[0516] 6. cDNA Purification
[0517] 6.1 Vortex the SPRIselect magnetic beads to mix thoroughly. Add 60 μl of SPRIselect magnetic beads (0.6X) to the sample (100 μl) and mix by pipetting 15 times.
[0518] 6.2 Incubate at room temperature for 5 minutes;
[0519] 65.3 Place on a magnetic stand until the solution becomes clear;
[0520] 6.4 Remove the supernatant;
[0521] 6.5 Add 200 μl of 80% ethanol to the tube and wait for 30 seconds;
[0522] 6.6 Remove ethanol;
[0523] 6.7 Repeat steps 6.5 and 6.6 for a total of 2 washes;
[0524] 6.8 Centrifuge briefly and transfer to a magnetic rack;
[0525] 6.9 Remove residual alcohol and dry at room temperature for 2 minutes (do not exceed 2 minutes, otherwise the elution efficiency will be reduced);
[0526] 6.10 Remove the tube from the magnetic rack, add 40.5 μl of EB buffer, and mix by pipetting 15 times;
[0527] 6.11 Incubate at room temperature for 2 minutes;
[0528] 6.12 Place the test tube on a magnetic rack until the solution becomes clear;
[0529] 6.13 Pipette 40 μl of liquid into a new test tube;
[0530] 6.14 Store at 4°C for 72 hours, or at -20°C for 4 weeks, or proceed directly to the next step.
[0531] 7. cDNA quantification and quality control
[0532] 7.1 Use Qubit to determine the concentration of the purified cDNA product and dilute it to 2 ng / ul;
[0533] 7.2 Use the High Sensitivity Agilent Technologies 2100 Bioanalyzerd to determine the peak shape of the cDNA product.
[0534] The experimental results show that through the construction of a single-cell spatial transcriptome library, spatial transcriptome analysis with single-cell resolution was achieved, and plant tissues can truly reflect the distribution of various cell types in the tissue.
[0535] The seamless scanning effect of toluidine blue staining of plant tissue is shown in Figure 25.
[0536] The splicing effect of toluidine blue staining of plant tissue is shown in Figure 26.
[0537] The method for constructing a single-cell spatial transcriptome sequencing library provided by the present disclosure (see Figure 27 for the process) utilizes a spatial transcriptome biochip with a specific matrix and combines it with a scanning scheme to realize transcript analysis at the single-cell level. Using this method, the tissue sample is attached to the S1000 spatial transcriptome chip, and then background matrix scanning, fluorescence scanning, and bright field scanning are performed. The spatial transcriptome information is captured through a series of biochemical reactions, and finally the spatial transcriptome analysis of the tissue sample at a single-cell resolution is realized by combining the sequencing results and the bioinformatics algorithm. Compared with the traditional method, the present disclosure solves the common image splicing problems of conventional scanners on the market, and can provide more accurate single-cell level analysis in terms of spatial transcriptome analysis. This method is mainly used for single-cell resolution research of spatial transcriptomes.
[0538] The single-cell spatial transcriptome analysis device based on cell segmentation provided by the present disclosure is described below. The single-cell spatial transcriptome analysis device based on cell segmentation described below and the single-cell spatial transcriptome analysis method based on cell segmentation described above can be referenced to each other.
[0539] Please refer to Figure 28, which is a schematic diagram of the structure of a single-cell spatial transcriptome analysis device based on cell segmentation provided by an embodiment of the present disclosure. As shown in Figure 28, the device may include:
[0540] An acquisition module 10 is used to acquire a corrected scan image; the scan image is obtained based on the tissue slice attached to the biochip;
[0541] The segmentation module 20 is used to perform cell segmentation on the scanned image and obtain a cell segmentation result;
[0542] An extraction module 30 is used to extract the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip;
[0543] The determination module 40 is configured to determine the position information of each cell and the statistical information of the corresponding barcoded microspheres based on the position information of the barcoded microspheres corresponding to each cell in the cell segmentation result.
[0544] Optionally, the segmentation module 20 includes:
[0545] A splitting unit, used for splitting the scanned image into multiple sub-images;
[0546] A preprocessing unit, configured to preprocess each subgraph to obtain a target subgraph;
[0547] A prediction unit is used to obtain a pre-trained model that matches the cell type of the tissue section, and use the pre-trained model to predict the cell nucleus boundary segmentation result of the target sub-image to obtain the cell nucleus boundary segmentation result of each cell nucleus in the target sub-image;
[0548] A first merging unit is used to merge the target sub-images to obtain a target image;
[0549] The second merging unit is used to merge the cell nucleus boundary segmentation results of each cell nucleus in each target sub-image to obtain the cell nucleus boundary segmentation results of each cell nucleus in the target image;
[0550] The expansion unit is used to perform expansion processing on the cell nucleus boundary segmentation result of each cell nucleus in the target image to generate a cell boundary segmentation result of each cell in the target image.
[0551] Optionally, the splitting unit is specifically used to:
[0552] A custom sliding window is used to split the scanned image into multiple initial sub-images;
[0553] The edge boundaries of multiple initial sub-graphs are fine-tuned to obtain multiple sub-graphs.
[0554] Optionally, the preprocessing unit is specifically configured to:
[0555] For each sub-image, the sub-image is pre-processed including equalization, size adjustment, color space conversion, corrosion, expansion and pixel value normalization to obtain the target sub-image.
[0556] Optionally, the external expansion unit is specifically used to:
[0557] Generate a binary image based on the nucleus boundary segmentation result of each cell nucleus in the target image;
[0558] The distance map of the binary image is calculated by distance transformation; the distance map is used to represent the distance from each pixel to the nearest cell nucleus boundary;
[0559] The watershed algorithm is used to process the distance map and the preset cell boundary threshold to obtain the cell boundary segmentation results of each cell in the target image.
[0560] Optionally, the extraction module 30 is specifically configured to:
[0561] Based on the intrinsic position of each barcoded microsphere in the biochip, the center coordinate position of the barcoded microsphere is calculated;
[0562] Extract the area corresponding to the position in the cell data of the cell segmentation result based on the center coordinate position;
[0563] Create a circular mask and multiply the cell data in the area with the circular mask to obtain the cell data within the circular range;
[0564] The cell data within each circular range are counted to obtain the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result.
[0565] Optionally, the device further includes a visualization module, which is specifically configured to:
[0566] Perform bioinformatics analysis on the location information of each cell and the statistical information of the corresponding barcoded microspheres to obtain the expression matrix of the cells corresponding to each location information;
[0567] Perform clustering processing on the expression matrix corresponding to each cell to obtain cell clustering information;
[0568] Based on the cluster number of each cell in the cell clustering information, the cell area corresponding to each cluster number in the scanned image is painted with a corresponding color to obtain a cluster image;
[0569] Generating contour position information of each cell in the cluster image by searching for cell contours in the cluster image;
[0570] The cluster image and the contour position information of each cell in the cluster image are visually displayed.
[0571] Optionally, the acquisition module 10 includes:
[0572] an acquisition unit, configured to fix the tissue slice with methanol, perform bright field staining on the sample, and then perform scanning imaging under a bright field light source to obtain multiple field of view scan images of the biochip, wherein the biochip includes a substrate pattern, the substrate pattern includes multiple key points, and any adjacent field of view scan images have overlapping areas;
[0573] a first processing unit, configured to fuse the plurality of visual field scanning images according to an overlapping area between any adjacent visual field scanning images to obtain a full-width scanning image of the biochip;
[0574] a determining unit, configured to determine first position information of the plurality of key points in the full-scan image according to the substrate pattern;
[0575] A correction unit is used to perform correction processing on the full-width scan image according to the first position information and the substrate pattern to obtain a target full-width scan image.
[0576] Optionally, the first processing unit is specifically configured to:
[0577] determining relative positions between any adjacent visual field scanning images according to overlapping areas between any adjacent visual field scanning images;
[0578] According to the relative positions between any adjacent visual field scanning images, the multiple visual field scanning images are fused to obtain the full-width scanning image.
[0579] Optionally, the first processing unit is specifically configured to:
[0580] For any first visual field scanning image among the plurality of visual field scanning images, determining a second visual field scanning image adjacent to the first visual field scanning image;
[0581] performing a two-dimensional window sliding process on the first field of view scanning image and the second field of view scanning image according to an overlapping area between the first field of view scanning image and the second field of view scanning image to determine a matching parameter between the first field of view scanning image and the second field of view scanning image;
[0582] The relative position between the first field of view scanning image and the second field of view scanning image is determined according to the matching parameters.
[0583] Optionally, the first processing unit is specifically configured to:
[0584] Determining a position of a reference visual field scanning image among the multiple visual field scanning images, where the reference visual field scanning image is any one of the multiple visual field scanning images;
[0585] Determining positions of other visual field scanning images according to the position of the reference visual field scanning image and the relative positions between any adjacent visual field scanning images, wherein the other visual field scanning images are visual field scanning images other than the reference visual field scanning image among the multiple visual field scanning images;
[0586] The multiple visual field scanning images are fused according to the position of the reference visual field scanning image and the positions of the other reference visual field scanning images to obtain the full-width scanning image.
[0587] Optionally, the determining unit is specifically configured to:
[0588] Determining a key point recognition model corresponding to the substrate pattern;
[0589] Inputting the full-scan image into the key point recognition model to obtain the first position information;
[0590] In which, the key point recognition model is obtained by training based on multiple groups of training samples, any group of training samples includes a sample full-width scan image and label information corresponding to the sample full-width image, the sample full-width scan image includes the substrate pattern, and the label information includes the position information of multiple key points in the sample full-width scan image.
[0591] Optionally, the correction unit is specifically configured to:
[0592] Determining second position information of the plurality of key points in the full-scan image according to the substrate pattern and the size of the full-scan image;
[0593] Correction processing is performed on the full-width scanned image according to the first position information and the second position information to obtain the target full-width scanned image.
[0594] Optionally, the correction unit is specifically configured to:
[0595] determining a correction transformation matrix according to the first position information and the second position information;
[0596] According to the correction conversion matrix, mapping transformation processing is performed on the full-width scanned image to obtain the target full-width scanned image.
[0597] Optionally, the acquisition module 10 includes:
[0598] An acquisition unit is used to fix the tissue slice with methanol, perform fluorescence staining on the sample, and then perform fluorescence scanning imaging to obtain multiple binary channel sub-images of the biochip and fluorescence channel sub-images corresponding to each of the binary channel sub-images;
[0599] an amplification unit, configured to amplify the substrate pattern included on the biochip according to the scanning magnifications corresponding to the plurality of binary mirror channel sub-images to obtain a full-frame reference image;
[0600] a second processing unit, configured to determine a correction position of each of the two-mirror channel sub-images in the full-frame reference image according to an acquisition order of the plurality of two-mirror channel sub-images and a plurality of key points included in the substrate pattern;
[0601] The fusion unit is used to fuse the fluorescence channel sub-images corresponding to each of the binary mirror channel sub-images according to the corrected position of each of the binary mirror channel sub-images in the full-frame reference image to obtain the full-frame fluorescence image of the biochip.
[0602] Optionally, the second processing unit is specifically configured to:
[0603] For any binary mirror channel sub-image, performing key point recognition processing on the binary mirror channel sub-image to determine first coordinate information of a first key point in the binary mirror channel sub-image, where the first key point is a key point included in the binary mirror channel sub-image among the multiple key points;
[0604] Determining the initial position of the two-mirror channel sub-image in the full-frame reference image according to the acquisition sequence and the first coordinate information;
[0605] The initial position is corrected to obtain a corrected position of the binary mirror channel sub-image in the full-frame reference image.
[0606] Optionally, the second processing unit is specifically configured to:
[0607] Inputting the two-mirror channel sub-image into a key point recognition model to obtain the first coordinate information;
[0608] In which, the key point recognition model is obtained by training based on multiple groups of training samples, any group of training samples includes a sample two-division mirror channel sub-image and label information corresponding to the sample two-division mirror channel sub-image, the sample two-division mirror channel sub-image includes sample key points, and the label information includes the coordinate information of the sample key points in the sample two-division mirror channel sub-image.
[0609] Optionally, the second processing unit:
[0610] determining, according to the acquisition order, second coordinate information of the first key point in the full reference image;
[0611] An initial position of the two-mirror channel sub-image in the full-frame reference image is determined according to the first coordinate information and the second coordinate information.
[0612] Optionally, the second processing unit:
[0613] Determining, according to the initial position, a local sub-image corresponding to the binary mirror channel sub-image in the full-frame reference image, wherein the region corresponding to the local sub-image in the full-frame reference image includes the region corresponding to the binary mirror channel sub-image in the full-frame reference image;
[0614] Performing matching processing on the local sub-image and the dichroic channel sub-image to obtain a matching result;
[0615] According to the matching result, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
[0616] Optionally, the second processing unit:
[0617] When the local sub-image is in a fixed state, determining a plurality of relative positions of the two-mirror channel sub-image relative to the local sub-image according to a preset step size;
[0618] performing a two-dimensional window sliding process on the two-mirror channel sub-image according to the plurality of relative positions, and determining the two-mirror channel sub-image at each of the relative positions;
[0619] The local sub-image and the two-mirror channel sub-image at each relative position are matched to obtain the matching result, which includes the similarity between the two-mirror channel sub-image and the local sub-image at each relative position.
[0620] Optionally, the second processing unit:
[0621] Determine the relative position with the highest corresponding similarity as the target relative position of the two-mirror channel sub-image relative to the local sub-image;
[0622] According to the relative position of the target and the position of the local sub-image in the full-frame reference image, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
[0623] It should be noted here that the device provided in this embodiment can implement all the method steps that can be implemented in the above method embodiment, and can achieve the same beneficial effects. The same contents and beneficial effects in the embodiment of this device and the above method embodiment will not be repeated here.
[0624] FIG29 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG29 , the electronic device may include: a processor 1110, a communication interface 1120, a memory 1130, and a communication bus 1140. The processor 1110, the communication interface 1120, and the memory 1130 communicate with each other via the communication bus 1140. The processor 1110 may invoke logic instructions in the memory 1130 to execute a single-cell spatial transcriptome analysis method based on cell segmentation. The method includes: acquiring a calibrated scan image; the scan image is obtained based on a tissue section attached to a biochip; performing cell segmentation on the scan image to obtain a cell segmentation result; extracting the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip; and determining the position information of each cell and statistical information of the corresponding barcoded microsphere based on the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result.
[0625] In addition, the logic instructions in the above-mentioned memory 1130 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0626] On the other hand, the present disclosure also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the single-cell spatial transcriptome analysis method based on cell segmentation provided in the above-mentioned embodiments, the method including: obtaining a corrected scan image; the scan image is obtained based on a tissue section attached to a biochip; performing cell segmentation on the scan image to obtain a cell segmentation result; based on the inherent position of each barcode microsphere in the biochip, extracting the position information of the barcode microsphere corresponding to each cell in the cell segmentation result; based on the position information of the barcode microsphere corresponding to each cell in the cell segmentation result, determining the position information of each cell and the statistical information of the corresponding barcode microsphere.
[0627] On the other hand, the present disclosure also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the single-cell spatial transcriptome analysis method based on cell segmentation provided in the above-mentioned embodiments, the method comprising: acquiring a corrected scanned image; the scanned image is obtained based on a tissue section attached to a biochip; performing cell segmentation on the scanned image to obtain a cell segmentation result; based on the inherent position of each barcode microsphere in the biochip, extracting the position information of the barcode microsphere corresponding to each cell in the cell segmentation result; based on the position information of the barcode microsphere corresponding to each cell in the cell segmentation result, determining the position information of each cell and the statistical information of the corresponding barcode microsphere.
[0628] In the several embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit.
[0629] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.
[0630] In addition, the functional units in the various embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of software functional units.
[0631] If the integrated unit is implemented in the form of 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, all or part of the technical solution can be embodied in the form of a computer software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present disclosure. The computer storage medium is a non-transitory medium, including: flash memory, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk, etc., which can store program code.
[0632] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present disclosure. Industrial Applicability
[0633] The present invention provides a single-cell spatial transcriptome analysis method and apparatus based on cell segmentation, relating to the field of bioinformatics. The method comprises: obtaining a calibrated scanned image; the scanned image is obtained based on a tissue slice attached to a biochip; performing cell segmentation on the scanned image to obtain a cell segmentation result; extracting the positional information of the barcoded microspheres corresponding to each cell in the cell segmentation result based on the inherent position of each barcoded microsphere in the biochip; and determining the positional information of each cell and statistical information of the corresponding barcoded microspheres based on the positional information of the barcoded microspheres corresponding to each cell in the cell segmentation result. The present invention can improve the efficiency and accuracy of cell analysis and has good economic value and application prospects.
Claims
1. A single-cell spatial transcriptome analysis method based on cell segmentation, characterized in that: include: Acquire a corrected scanned image; the scanned image is obtained based on a tissue slice attached to a biochip; Performing cell segmentation on the scanned image to obtain a cell segmentation result; Based on the inherent position of each barcode microsphere in the biochip, extracting the position information of the barcode microsphere corresponding to each cell in the cell segmentation result; Based on the position information of the barcoded microsphere corresponding to each cell in the cell segmentation result, the position information of each cell and the statistical information of the corresponding barcoded microsphere are determined.
2. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 1, characterized in that: The cell segmentation result includes a cell nucleus boundary segmentation result and a cell boundary segmentation result; the cell segmentation is performed on the scanned image to obtain the cell segmentation result, including: Splitting the scanned image into a plurality of sub-images; For each of the subgraphs, preprocess the subgraph to obtain a target subgraph; Acquire a pre-trained model that matches the cell type of the tissue section, and use the pre-trained model to predict the cell nucleus boundary segmentation result of the target sub-image to obtain the cell nucleus boundary segmentation result of each cell nucleus in the target sub-image; Merging the target sub-images to obtain a target image; Merging the cell nucleus boundary segmentation results of each cell nucleus in each of the target sub-images to obtain the cell nucleus boundary segmentation results of each cell nucleus in the target image; The cell nucleus boundary segmentation result of each cell nucleus in the target image is expanded to generate the cell boundary segmentation result of each cell in the target image.
3. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 2, characterized in that: The step of splitting the scanned image into a plurality of sub-images comprises: Using a custom sliding window to split the scanned image into a plurality of initial sub-images; The edge boundary lines of the multiple initial sub-images are fine-tuned to obtain the multiple sub-images.
4. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 2, characterized in that: The step of preprocessing each subgraph to obtain a target subgraph includes: For each of the sub-images, pre-processing including equalization, size adjustment, color space conversion, corrosion, expansion and pixel value normalization is performed on the sub-image to obtain the target sub-image.
5. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 2, characterized in that: The step of performing an expansion process on the cell nucleus boundary segmentation result of each cell nucleus in the target image to generate the cell boundary segmentation result of each cell in the target image includes: Generate a binary image based on the cell nucleus boundary segmentation result of each cell nucleus in the target image; Calculating a distance map of the binary image by distance transformation; the distance map is used to characterize the distance from each pixel to the nearest cell nucleus boundary; The distance map and the preset cell boundary threshold are processed by using a watershed algorithm to obtain the cell boundary segmentation result of each cell in the target image.
6. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 1 to 5, characterized in that: The step of extracting the position information of the barcode microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcode microsphere in the biochip includes: Calculating the central coordinate position of each barcode microsphere in the biochip based on the intrinsic position of each barcode microsphere; Extracting a region corresponding to a position in the cell data of the cell segmentation result based on the center coordinate position; Create a circular mask, and multiply the cell data in the area by the circular mask to obtain the cell data within the circular range; The cell data within each circular range are counted to obtain the position information of the barcode microsphere corresponding to each cell in the cell segmentation result.
7. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 1 to 5, characterized in that: Also includes: Performing bioinformatics analysis on the position information of each of the cells and the statistical information of the corresponding barcoded microspheres to obtain an expression matrix of the cells corresponding to each of the position information; Performing clustering processing on the expression matrix corresponding to each of the cells to obtain cell clustering information; Based on the cluster number of each cell in the cell clustering information, a cell region corresponding to each cluster number in the scanned image is drawn with a corresponding color to obtain a cluster image; Generating contour position information of each cell in the cluster image by searching for cell contours in the cluster image; The cluster image and the contour position information of each cell in the cluster image are visualized and displayed.
8. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 1, characterized in that: The step of obtaining the corrected scanned image comprises: In a state where the tissue slice is attached to the biochip, the sample is fixed with methanol and then bright-field stained, and then scanned and imaged under a bright-field light source to obtain a plurality of field-of-view scanning images of the biochip, wherein the biochip includes a substrate pattern, the substrate pattern includes a plurality of key points, and there is an overlapping area between any adjacent field-of-view scanning images; According to the overlapping area between any adjacent visual field scanning images, the plurality of visual field scanning images are fused to obtain a full-frame scanning image of the biochip; Determining first position information of the plurality of key points in the full-scan image according to the substrate pattern; According to the first position information and the substrate pattern, the full-width scanned image is corrected. A target full-width scanned image is obtained as the corrected scanned image.
9. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 8, characterized in that: The step of fusing the plurality of visual field scanning images according to the overlapping area between any adjacent visual field scanning images to obtain the full-frame scanning image of the biochip comprises: Determining the relative positions between any adjacent visual field scanning images according to the overlapping areas between any adjacent visual field scanning images; According to the relative positions between any adjacent visual field scanning images, the multiple visual field scanning images are fused to obtain the full-frame scanning image.
10. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 9, characterized in that: Determining the relative positions between any adjacent visual field scanning images according to the overlapping areas between any adjacent visual field scanning images comprises: For any first field of view scanning image among the plurality of field of view scanning images, determining a second field of view scanning image adjacent to the first field of view scanning image; According to the overlapping area between the first field of view scanning image and the second field of view scanning image, performing two-dimensional window sliding processing on the first field of view scanning image and the second field of view scanning image to determine the matching parameters between the first field of view scanning image and the second field of view scanning image; The relative position between the first field of view scanning image and the second field of view scanning image is determined according to the matching parameters.
11. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 9, characterized in that: The fusing process of the plurality of visual field scanning images according to the relative positions between any adjacent visual field scanning images to obtain the full-frame scanning image comprises: Determine a position of a reference visual field scanning image among the multiple visual field scanning images, where the reference visual field scanning image is any one of the multiple visual field scanning images; Determine the positions of other visual field scanning images according to the position of the reference visual field scanning image and the relative positions between any adjacent visual field scanning images, wherein the other visual field scanning images are visual field scanning images other than the reference visual field scanning image among the multiple visual field scanning images; According to the position of the reference field of view scanning image and the positions of the other reference field of view scanning images, the multiple field of view scanning images are fused to obtain the full-frame scanning image.
12. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 8 to 11, characterized in that: The step of determining first position information of the plurality of key points in the full-frame scan image according to the substrate pattern includes: Determine a key point recognition model corresponding to the substrate pattern; Inputting the full-scan image into the key point recognition model to obtain the first position information; The key point recognition model is obtained by training based on multiple sets of training samples. The sample includes a sample full-width scan image and label information corresponding to the sample full-width scan image, the sample full-width scan image includes the substrate pattern, and the label information includes position information of multiple key points in the sample full-width scan image.
13. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 8 to 11, characterized in that: The step of correcting the full-width scanned image according to the first position information and the substrate pattern to obtain a target full-width scanned image includes: Determining second position information of the plurality of key points in the full-frame scan image according to the substrate pattern and the size of the full-frame scan image; The full-width scanned image is corrected according to the first position information and the second position information to obtain the target full-width scanned image.
14. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 13, characterized in that: The step of performing correction processing on the full-frame scanned image according to the first position information and the second position information to obtain the target full-frame scanned image includes: Determine a correction transformation matrix according to the first position information and the second position information; According to the correction conversion matrix, the full-frame scanned image is subjected to mapping transformation processing to obtain the target full-frame scanned image.
15. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 1, characterized in that: The step of obtaining the corrected scanned image comprises: In a state where the tissue slice is attached to the biochip, the sample is fixed with methanol and then fluorescently stained, and then fluorescent scanning imaging is performed to obtain a plurality of binary channel sub-images of the biochip and fluorescent channel sub-images corresponding to each of the binary channel sub-images; According to the scanning magnifications corresponding to the plurality of binary mirror channel sub-images, amplifying the substrate pattern included on the biochip to obtain a full-frame reference image; Determining a correction position of each of the two-mirror channel sub-images in the full-frame reference image according to the acquisition order of the plurality of two-mirror channel sub-images and a plurality of key points included in the substrate pattern; According to the corrected position of each of the binary mirror channel sub-images in the full-frame reference image, the fluorescence channel sub-images corresponding to each of the binary mirror channel sub-images are fused to obtain a full-frame fluorescence image of the biochip as the corrected scanned image.
16. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 15, characterized in that: Determining the correction position of each of the two-mirror channel sub-images in the full-frame reference image according to the acquisition order of the multiple two-mirror channel sub-images and the multiple key points included in the substrate pattern includes: For any binary mirror channel sub-image, perform key point recognition processing on the binary mirror channel sub-image to determine first coordinate information of a first key point in the binary mirror channel sub-image, where the first key point is a key point included in the binary mirror channel sub-image among the multiple key points; Determine the initial position of the binary mirror channel sub-image in the full-frame reference image according to the acquisition sequence and the first coordinate information; The initial position is corrected to obtain the corrected position of the binary mirror channel sub-image in the full-frame reference image.
17. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 16, characterized in that: The step of performing key point recognition processing on the binary mirror channel sub-image to determine first coordinate information of a first key point in the binary mirror channel sub-image includes: Inputting the binary mirror channel sub-image into a key point recognition model to obtain the first coordinate information; Among them, the key point recognition model is obtained by training based on multiple groups of training samples, any group of training samples includes a sample binary mirror channel sub-image and label information corresponding to the sample binary mirror channel sub-image, the sample binary mirror channel sub-image includes sample key points, and the label information includes coordinate information of the sample key points in the sample binary mirror channel sub-image.
18. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 16, characterized in that: Determining the initial position of the binary mirror channel sub-image in the full-frame reference image according to the acquisition sequence and the first coordinate information includes: Determining second coordinate information of the first key point in the full reference image according to the acquisition order; An initial position of the binary mirror channel sub-image in the full-frame reference image is determined according to the first coordinate information and the second coordinate information.
19. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 16 to 18, characterized in that: The correcting process is performed on the initial position to obtain the corrected position of the binary mirror channel sub-image in the full-frame reference image, including: According to the initial position, determining a local sub-image corresponding to the binary mirror channel sub-image in the full-frame reference image, wherein the area corresponding to the local sub-image in the full-frame reference image includes the area corresponding to the binary mirror channel sub-image in the full-frame reference image; Performing matching processing on the local sub-image and the binary mirror channel sub-image to obtain a matching result; According to the matching result, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
20. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 19, characterized in that: The matching process of the local sub-image and the two-mirror channel sub-image to obtain a matching result includes: When the local sub-image is in a fixed state, determining a plurality of relative positions of the binary mirror channel sub-image relative to the local sub-image according to a preset step length; According to the plurality of relative positions, performing a two-dimensional window sliding process on the two-mirror channel sub-image, and determining the two-mirror channel sub-image at each of the relative positions; The local sub-image and the binary mirror channel sub-image at each relative position are matched to obtain the matching result, which includes the similarity between the binary mirror channel sub-image and the local sub-image at each relative position.
21. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 20, characterized in that: The step of obtaining the corrected position of the binary mirror channel sub-image in the full-frame reference image according to the matching result includes: Determine the corresponding relative position with the highest similarity as the target relative position of the binary mirror channel sub-image relative to the local sub-image; According to the relative position of the target and the position of the local sub-image in the full-frame reference image, the corrected position of the binary mirror channel sub-image in the full-frame reference image is obtained.
22. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 7, characterized in that: The single-cell spatial transcriptome sequencing library is used for the clustering process, the tissue section is a fresh frozen tissue section, and the biochip is a spatial transcriptome chip; The single-cell spatial transcriptome sequencing library was constructed by the following steps: (1) attaching the fresh frozen tissue slices to the spatial transcriptome chip, fixing the samples with methanol, fluorescently staining the samples, and then performing fluorescent scanning imaging, and finally performing scanning imaging under a bright field light source to obtain an uncorrected scanning image of the single-cell spatial transcriptome; (2) Tissue permeabilization after scanning; (3) Reverse transcription and single-strand cDNA synthesis; (4) cDNA second-strand synthesis; (5) cDNA amplification; (6) cDNA purification; (7) cDNA quality control; (8) Sequencing on a machine.
23. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 22, characterized in that: For animal samples, in step (1), fluorescent staining is performed using DAPI or ssDNA, followed by fluorescent scanning imaging; or, fluorescent staining is performed using DAPI or ssDNA, followed by fluorescent scanning imaging, and then the sample is stained with HE or Masson's stain, and finally scanned and imaged under a bright field light source; For plant samples, in step (1), fluorescent brightener 28 is used for autofluorescence enhancement staining; or, toluidine blue or HE staining is directly performed without fluorescent staining.
24. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 22, characterized in that: The thickness of the frozen tissue slices in step (1) is 5-20 μm, and the chip after bonding is placed in -20°C methanol for fixation for 30-60 minutes.
25. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 22, characterized in that: After the chip after pasting in step (1) is fixed with methanol, a fluorescent staining reagent is added to the sample, incubated in the dark for 1-20 minutes, and then fluorescence scanning imaging is performed under the irradiation of a full-spectrum or wide-spectrum light source; In the fluorescence channel, by removing the filter block of the scanner and retaining only the dichroic mirror, the dichroic mirror setting can be performed to simultaneously achieve cell nucleus fluorescence scanning and background microsphere matrix scanning; or by setting a filter block with a certain transmittance, cell nucleus fluorescence scanning and background microsphere matrix scanning can be achieved under full spectrum or wide spectrum conditions.
26. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 22, characterized in that: Step (2) comprises: placing the chip obtained in step (1) in the tissue permeabilization solution and incubating at 37° C. for 1-30 minutes, removing the tissue permeabilization solution and adding 0.1% SSC solution into the wells of the chip.
27. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 26, characterized in that: Step (3) comprises: removing the SSC solution in the well, adding the reverse transcription and cDNA single-strand synthesis reagents into the well, and incubating at 37-65° C. for 0.5-2 hours.
28. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 27, characterized in that: Step (4) comprises: removing the reverse transcription and cDNA single-strand synthesis reagents in the well, adding 0.08M KOH for denaturation, washing with EB buffer, and then adding the cDNA double-strand synthesis reagent, and incubating at 45-70°C for 10-40 minutes.
29. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 28, characterized in that: Step (5) includes: removing the cDNA double-strand synthesis reagent in the well, adding EB buffer for washing, then adding 0.08M KOH and incubating at room temperature for 10 minutes, then neutralizing KOH with Tris, and adding the cDNA amplification reagent for PCR amplification.
30. The single-cell spatial transcriptome analysis method based on cell segmentation according to claim 29, characterized in that: Step (6) comprises: purifying the amplified product with magnetic beads, then washing the magnetic beads with 80% ethanol, and finally eluting the amplified product from the magnetic beads with EB buffer.
31. The single-cell spatial transcriptome analysis method based on cell segmentation according to any one of claims 22 to 30, characterized in that: The spatial transcriptome chip used was equipped with mRNA capture probes; Preferably, the S1000 spatial transcriptome chip is used.
32. A single-cell spatial transcriptome analysis device based on cell segmentation, characterized in that: include: An acquisition module, used to acquire a corrected scanned image; the scanned image is obtained based on a tissue slice attached to the biochip; A segmentation module, used to perform cell segmentation on the scanned image to obtain a cell segmentation result; An extraction module, used for extracting the position information of the barcode microsphere corresponding to each cell in the cell segmentation result based on the inherent position of each barcode microsphere in the biochip; The determination module is used to determine the first position information of each cell and the statistical information of the corresponding barcode microsphere based on the position information of the barcode microsphere corresponding to each cell in the cell segmentation result.
33. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the single-cell spatial transcriptome analysis method based on cell segmentation are implemented as described in any one of claims 1 to 21.
34. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the single-cell spatial transcriptome analysis method based on cell segmentation are implemented.
Citation Information
Patent Citations
Space transcriptome detection chip and method
CN109762728A
Method for screening cell strain capable of expressing expected product
CN112037853A
Cell segmentation method and device based on microscopic image, equipment and storage medium
CN112330690A
Construction method of single cell space transcriptome sequencing library
CN116814742A
Scanning image processing method, device and equipment
CN116823607A
Cited By
Multi-immunofluorescence image automatic analysis method, medium and program product
CN119941715A
