A high-voltage power supply DNA sequencing visual detection method and system

Abstract

This application relates to the field of image processing technology, specifically to a high-voltage power supply DNA sequencing visual inspection method and system. The method includes: acquiring a four-channel DNA electrophoresis image using Sanger sequencing technology; acquiring contours in the DNA electrophoresis grayscale image; constructing a DNA local density confidence coefficient and local DNA density parameters for each pixel based on the grayscale values ​​of each pixel within a channel and its neighboring pixels; acquiring the density run matrix for each pixel; constructing a DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters for each pixel, and based on these, constructing a sequencing segmentation probability coefficient for each pixel; acquiring a decision threshold for each pixel; using the center of each contour as a seed point, and acquiring each DNA fragment based on an adaptive threshold using a region growing method, thereby obtaining the DNA sequence. This application can improve the accuracy of detecting the base sequence of a DNA sequence.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a high-voltage power supply DNA sequencing visual inspection method and system. Background Technology

[0002] DNA sequencing is an important research method used to determine the sequence of bases in DNA molecules. Sequencing reveals the genetic information and gene function of organisms, providing data support for fields such as biology, medicine, and biotechnology. The continuous development of DNA sequencing methods has greatly promoted scientific research and laid the foundation for applications in disease diagnosis and gene therapy. By interpreting and analyzing DNA sequences, we can better understand the structure and function of organisms, bringing more opportunities for human health and the development of biotechnology. Sanger sequencing is a commonly used method for DNA sequencing. This technology is based on dideoxy chain termination, randomly terminating the DNA chain synthesis process using four different fluorescently labeled dideoxynucleotides (ddNTPs). These ddNTPs lack hydroxyl functional groups, causing the DNA chain to stop extending. After the terminated DNA fragments are separated by electrophoresis, a fluorescence detector is used to determine the length and composition of each DNA fragment.

[0003] However, in the Sanger sequencing technology, DNA fragments of different lengths obtained from ddNTPs move at different rates when subjected to an electric field during the electrophoretic separation stage. This can lead to overlap or ghosting between adjacent DNA fragments, making it impossible to accurately identify consecutive bases on the electrophoresis image. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a high-voltage power supply DNA sequencing visual inspection method and system, the specific technical solution of which is as follows:

[0005] In a first aspect, embodiments of this application provide a visual inspection method for DNA sequencing using a high-voltage power supply, the method comprising the following steps:

[0006] A high-voltage power supply was applied to both ends of the DNA sequencer, and the Sanger sequencing technology was used to obtain DNA electrophoresis images in four channels; grayscale images of DNA electrophoresis were obtained by grayscale processing.

[0007] The local DNA density parameters of each pixel are determined based on the distribution characteristics of DNA fragments in each channel in the DNA electrophoresis grayscale image; a density run-length matrix is ​​constructed based on the quantization results of the local DNA density parameters of all pixels; the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters of the pixels are obtained based on the element size in the density run-length matrix; the sequencing segmentation probability coefficient of the pixels is determined based on the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters.

[0008] The decision threshold during growth is adaptively determined based on the sequencing segmentation probability coefficient of pixels within the channel, and the base sequence of the DNA sequence is determined based on each growth region obtained by the region growth method, thus completing the visual detection of the DNA sequence.

[0009] Preferably, determining the local DNA density parameters for each pixel includes:

[0010] Treat each pixel as a pixel to be processed, and obtain a first window of a preset size for each pixel to be processed, centered on each pixel to be processed;

[0011] The DNA local density confidence coefficient of each pixel in the first window is determined based on the gray value difference between each pixel in the first window and the pixels in its eight neighboring regions.

[0012] Calculate the product of the DNA local density confidence coefficient of each pixel in the first window and the gray value of each pixel, and use the average of the accumulated results of the product on the first window of a preset size for each pixel to be processed as the local DNA density parameter of each pixel to be processed.

[0013] Preferably, the method for determining the confidence coefficient of the local DNA density of each pixel within the first window is as follows:

[0014] Calculate the absolute value of the difference between the gray value of each pixel in the first window and the gray value of any pixel in the eight neighboring regions. Take the reciprocal of the sum of the absolute values ​​of the differences in the eight neighboring regions and the sum of the sum of the preset constant parameters as the DNA local density confidence coefficient of each pixel in the first window.

[0015] Preferably, the construction of the density run-length matrix based on the quantization results of the local DNA density parameters of all pixels includes:

[0016] The local DNA density parameters of all pixels are obtained respectively, and all the local DNA density parameters are mapped to a first preset number of density levels;

[0017] A second window of a preset size is constructed with each pixel as the center. The electrophoretic direction of the DNA fragment is used as the run direction. The density run matrix of the second window for each pixel is obtained based on the mapping result and the run direction according to the grayscale run matrix method.

[0018] Preferably, the step of obtaining the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters of pixels based on the element sizes in the density run-length matrix includes:

[0019] Construct a third window of a preset size centered on each pixel;

[0020] Obtain the maximum run length in the density run-length matrix of the second window for each pixel;

[0021] Each pixel is taken as the target pixel, and the absolute value of the difference between the maximum run length corresponding to the target pixel and the maximum run length corresponding to any non-center pixel in the third window of the target pixel is taken as the first difference.

[0022] The absolute value of the difference between the mean of all run lengths in the density run length matrix of the second window of the target pixel and the mean of all run lengths in the density run length matrix of the second window of any non-center pixel in the third window of the target pixel is taken as the second difference.

[0023] The average of the product of the first difference and the second difference accumulated over all non-center pixels in the third window of the target pixel is used as the DNA density distribution heterogeneity coefficient of the target pixel.

[0024] Calculate the mean of each column element in the density run-length matrix of the second window of the target pixel, and use the sum of the absolute values ​​of the differences between the means of two adjacent columns on the density run-length matrix as the numerator;

[0025] Calculate the sum of the elements in each row of the density run-length matrix of the second window of the target pixel, and use the sum of the absolute difference between the sums of the elements in two adjacent rows on the density run-length matrix and the preset constant parameter as the denominator.

[0026] The ratio of the numerator to the denominator is used as the evaluation parameter for the channel tailing effect of the target pixel.

[0027] Preferably, the determination of the sequencing segmentation probability coefficient of the pixel includes:

[0028] The product of the normalized result of the DNA density distribution heterogeneity coefficient of each pixel and the normalized result of the channel tailing effect evaluation parameter is used as the sequencing segmentation probability coefficient of each pixel.

[0029] Preferably, the step of adaptively determining the decision threshold during growth based on the sequencing segmentation probability coefficients of pixels within the channel includes:

[0030] The initial grayscale threshold is set to three times the standard deviation of the grayscale values ​​of all pixels in the statistical channel.

[0031] The optimal threshold is obtained by taking the sequencing segmentation probability coefficients of all pixels as input and using the Otsu thresholding method.

[0032] The decision threshold for region growing of each pixel is determined by comparing the sequencing segmentation probability coefficient of each pixel with the optimal threshold.

[0033] Preferably, determining the decision threshold for region growing for each pixel includes:

[0034] If the sequencing segmentation probability coefficient of a pixel is less than the optimal threshold, then the sum of the sequencing segmentation probability coefficient of the pixel and the sum of the sum and the initial grayscale threshold are used as the decision threshold of the pixel.

[0035] If the sequencing segmentation probability coefficient of a pixel is greater than or equal to the optimal threshold, then the product of the sequencing segmentation probability coefficient of the pixel and the initial grayscale threshold is used as the decision threshold of the pixel.

[0036] Preferably, determining the base sequence of the DNA sequence includes:

[0037] The center pixels of each contour in the DNA electrophoresis grayscale image are used as the initial seed points for the region growing method.

[0038] For any initial seed point, calculate the absolute value of the difference between the gray values ​​of each initial seed point and its neighboring pixels. If the absolute value of the difference is greater than the decision threshold of the neighboring pixels, the neighboring pixels are not added to the already grown region; if the absolute value of the difference is less than or equal to the decision threshold of the neighboring pixels, the neighboring pixels are added to the already grown region.

[0039] After traversing all pixels, growth is stopped, and each growth region in the DNA electrophoresis grayscale image is obtained. Each growth region is taken as a DNA fragment.

[0040] DNA fragments are decoded based on the segmented DNA electrophoresis grayscale image, and the base types of ddNTPs in the corresponding channels are read from bottom to top based on the decoding results.

[0041] Secondly, embodiments of this application also provide a high-voltage power supply DNA sequencing visual inspection system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described high-voltage power supply DNA sequencing visual inspection methods.

[0042] This application has at least the following beneficial effects:

[0043] This application calculates the local DNA density parameter of each pixel by analyzing the grayscale distribution in a DNA electrophoresis grayscale image, and analyzes the density of DNA fragments around the pixel. Based on the local DNA density parameter and the principle of grayscale run-length matrix, the density run-length matrix of the pixel is obtained. Further analysis of the texture features of the pixel distribution yields the DNA density distribution heterogeneity coefficient and the tailing effect perception coefficient of each pixel. This fully considers the similarity between the DNA density distribution of the pixel and the surrounding area, as well as the influence of the tailing effect on the pixel, improving the reliability of setting adaptive thresholds based on pixel features. Based on the DNA density distribution heterogeneity coefficient and the tailing effect perception coefficient of the pixel, the sequencing segmentation probability coefficient of the pixel is constructed. Furthermore, the decision threshold of the pixel is adaptively calculated based on the sequencing segmentation probability coefficient of the pixel, dividing different DNA fragments in the DNA electrophoresis grayscale image into different growth regions, resulting in a more accurate electrophoresis image of the DNA fragments. Based on the segmented DNA electrophoresis grayscale image, the base sequence of the DNA sequence is obtained, reducing the influence of DNA fragment overlap or trailing and improving the accuracy of detecting the base sequence of the DNA sequence. Attached Figure Description

[0044] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating the steps of a high-voltage power supply DNA sequencing visual inspection method according to one embodiment of this application. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a high-voltage power supply DNA sequencing visual inspection method and system proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0048] The following description, in conjunction with the accompanying drawings, details the specific scheme of the high-voltage power supply DNA sequencing visual inspection method and system provided in this application.

[0049] Please see Figure 1 The diagram illustrates a flowchart of a high-voltage power supply DNA sequencing visual inspection method according to an embodiment of this application. The method includes the following steps:

[0050] Step S001: Obtain DNA electrophoresis images of four channels using Sanger sequencing technology.

[0051] Sanger sequencing technology generally involves the following steps: extracting the DNA to be tested; breaking the DNA double helix into single strands through thermal denaturation or alkaline conditions; performing in vitro replication and amplification of the DNA fragments using polymerase chain reaction (PCR); obtaining DNA fragments of different lengths using fluorescently labeled dideoxynucleotides (ddNTPs); applying a high-voltage power supply to both ends of the DNA sequencer to subject the fluorescently labeled DNA fragments to a constant electric field, causing them to migrate from the negative electrode to the positive electrode in the gel. Since DNA fragments of different lengths migrate at different rates in the gel, DNA fragments will separate; imaging the fluorescently labeled DNA fragments using a CCD camera to obtain DNA electrophoresis images; and analyzing the captured fluorescence signals to determine the length of each DNA fragment and its corresponding fluorescent label, thereby inferring the base sequence of the DNA sequence.

[0052] In imaging fluorescently labeled DNA fragments, the DNA fragment is typically divided into four reaction tubes, each containing a specific ddNTP and four deoxyribonucleotides (dNTPs). ddNTPs are deoxyribonucleotides lacking 3... ’A -hydroxyl nucleotide can terminate the elongation of a DNA strand. Therefore, when a ddNTP is added to a new strand, the DNA strand elongation terminates at that position. In this way, the DNA fragments in each reaction tube are terminated with specific bases of different lengths. These terminated DNA fragments can be separated by polyacrylamide gel electrophoresis and detected by a fluorescence or radiometric detector.

[0053] Step S002: Determine the local DNA density parameter of each pixel based on the distribution characteristics of DNA fragments in each channel in the DNA electrophoresis grayscale image; construct a density run-length matrix based on the quantization results of the local DNA density parameters of all pixels; obtain the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameter of the pixel based on the element size in the density run-length matrix; determine the sequencing segmentation probability coefficient of the pixel based on the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameter.

[0054] In DNA sequencing, the reaction is typically performed in four reaction tubes. When using PCR to amplify single-stranded DNA, different amounts of ddNPT are added to each of the four reaction tubes. Since DNA molecules cannot extend further if they are bound to ddNPT during amplification, each reaction tube will yield DNA fragments of different lengths ending with specific bases.

[0055] Finally, an electric field is generated by connecting a power source to both ends of the reaction tube after amplification. Under the influence of this electric field, the linear double-stranded DNA molecules migrate in a polyacrylamide gel of a certain concentration. The larger the DNA molecule, i.e., the longer the DNA fragment, the greater the resistance during migration and the more difficult it is to move within the gel pores, thus migrating more slowly. Based on this characteristic, the sequence of DNA molecules can be effectively identified.

[0056] However, during DNA fragment migration, fragments of similar length migrate only a small distance, resulting in blurred or overlapping bands on the electrophoresis image, making it difficult to distinguish between DNA fragments. Therefore, image processing of the sampled DNA electrophoresis images is necessary to obtain clearer images and achieve more accurate sequencing.

[0057] For DNA electrophoresis images, DNA fragments of different lengths migrate different distances from the negative to the positive electrode of the power source. To more accurately distinguish DNA fragments of different lengths, the DNA electrophoresis image is first converted into a grayscale image. Then, the Canny edge detection operator is used to obtain the edge information of the grayscale image, resulting in a binary image where the edge information is highlighted and the background is muted. Next, the findContours function is used to extract a list of contours from the binary image, where each contour is a set of pixels. The use of the Canny edge detection operator and the findContours function is a well-known technique and will not be elaborated upon in this embodiment.

[0058] In DNA sequencing, the reaction is usually carried out in four reaction tubes, including adenine (A), guanine (G), thymine (T), and cytosine (C). Each reaction tube corresponds to the electrophoretic status of one base. In this example, the adenine (A) channel is analyzed.

[0059] During electrophoresis, due to the different migration rates of DNA fragments of different lengths, two DNA fragments with trailing or overlapping appear blurry at the boundary and are difficult to distinguish. Therefore, this embodiment uses the region growing method to determine the boundary between different DNA fragments. The region growing method is a well-known technique and will not be described in detail in this embodiment.

[0060] Based on the contour information returned by the `findContours` function, the center pixel of each contour is selected as the seed pixel for the region growing method. When formulating the growth criteria, the DNA fragment growth coefficient for each pixel is calculated based on the texture information in the image, as follows:

[0061] For adenine (A) channels, different DNA fragments exhibit different pixel distribution textures. This is because fluorescently labeled DNA fragments aggregate at different locations during electrophoretic migration under an electric field. The more DNA fragments aggregated in a region, the more labeled the region, and the higher its brightness; that is, the denser the DNA fragment distribution, the higher its grayscale value. Specifically, each pixel is used as the center to construct... The window that appears first is designated as the first window. The implementer can choose an odd number from the range [5, 13] in this embodiment. The value is 5, and this application does not impose any special restrictions on this. First, taking pixel i as an example, the local density confidence coefficient of DNA is calculated based on the gray values ​​of all pixels within the first window centered on pixel i. The expression is:

[0062]

[0063] In the formula, This represents the confidence coefficient of the local DNA density of pixel j within the first window of pixel i in channel A; This represents the sum of the absolute values ​​of the differences in grayscale values ​​between the j-th pixel and its eight neighboring pixels. This represents a preset constant parameter used to prevent the denominator from being 0. The range of values ​​for is (0, 1), here The value is taken as an empirical value of 0.01, in order to minimize the impact on the calculation results while ensuring that the denominator is not zero.

[0064] The greater the difference in grayscale value between pixel j within the first window of pixel i and its eight neighboring pixels, the greater the difference in local DNA density around pixel j, indicating the possible presence of noise. Therefore, the confidence coefficient of local DNA density of pixel j should be smaller. Conversely, the greater the difference in grayscale value between pixel j and its eight neighboring pixels, the more stable the distribution of local DNA around pixel j, and the less likely there is to be noise. Therefore, the confidence coefficient of local DNA density of pixel j should be larger.

[0065] Furthermore, based on the confidence coefficients of the local DNA density of the pixels within the first window and their grayscale values, the local DNA density parameters of each pixel are constructed, expressed as follows:

[0066]

[0067] In the formula, This represents the local DNA density parameter of pixel i within channel A. This represents the number of pixels in the first window centered at pixel i within channel A. This represents the grayscale value of pixel j within the first window centered on pixel i in channel A, and characterizes the local density of pixel j.

[0068] Furthermore, based on the local DNA density parameters of each pixel within the current channel, the DNA density texture distribution characteristics of each pixel are analyzed to obtain the DNA density distribution heterogeneity coefficient. The specific implementation process is as follows:

[0069] First, the density run-length matrix is ​​obtained using the principle of grayscale run-length matrix. Specifically, for pixel i, a density run-length matrix is ​​constructed centered on pixel i. The window that appears is called the second window. The implementer can choose an odd number from the range [5, 13] in this embodiment. The value is 7, and this application does not impose any special restrictions on this. The density run-length matrix of pixel i is calculated based on the second window. Since the electrophoretic direction of DNA fragments is from top to bottom, the density change in the vertical direction is relatively large. Therefore, the run direction of the density run matrix in this embodiment is selected... In the vertical direction, the density level can be set to a positive integer between [5, 10]. The more density levels there are, the higher the accuracy of the density run-length matrix, but the computational cost is high. On the other hand, the fewer density levels there are, the easier it is for different local DNA density parameters to be mapped to the same level, resulting in a poor run-length matrix effect. In this embodiment, the density level is set to 8, that is, the local DNA density parameters are mapped to 8 levels from 1 to 8.

[0070] The specific calculation process for the density run-length matrix of pixels is the same as that for the grayscale run-length matrix. This is a well-known technique in the field of image processing, and will not be described again in this embodiment.

[0071] Then, the heterogeneity coefficient of DNA density distribution is calculated based on the density run-length matrix of each pixel combined with the principle of nonlocal mean (NLM). Specifically, a matrix is ​​constructed centered on each pixel. This window is referred to as the third window. The implementer can choose an odd number within the range of [7, 15]. In order to ensure that a certain amount of data is used to construct the density run-length matrix, in this embodiment... The value is set to 11; this application does not impose any special restrictions on this. The DNA density distribution heterogeneity coefficient of pixel i within channel A is calculated using the following expression:

[0072]

[0073] In the formula, This represents the heterogeneity coefficient of DNA density distribution at pixel i within channel A. This represents the number of pixels other than pixel i within the third window centered at pixel i. This represents the maximum run length of the density run matrix for pixel i within channel A. This represents the maximum run length of the density run length matrix of pixel j within the third window centered on pixel i in channel A. This represents the mean run length of the density run matrix for pixel i within channel A; This represents the mean run length of the density run matrix of pixel j within the third window centered at pixel i in channel A.

[0074] The greater the difference between the maximum run length of the density run length matrix of pixel i in channel A and the other pixels in the third window centered on pixel i, the greater the first difference. The larger the value, the greater the difference in the mean of the run length, i.e., the second difference. A larger value indicates a greater inconsistency between the DNA density distribution at pixel i and the surrounding area; hence, the DNA density distribution heterogeneity coefficient... The larger the coefficient, the more consistent the DNA density distribution is between the location of pixel i and the surrounding area; conversely, the smaller the coefficient, the more consistent the DNA density distribution heterogeneity coefficient is with the surrounding area. The smaller the value, the greater the heterogeneity coefficient of the DNA density distribution of the pixel, indicating a higher probability that the pixel is at the boundary of a DNA fragment.

[0075] Secondly, during electrophoresis, in addition to the overlap between two adjacent DNA fragments, particle diffusion can also cause trailing. Therefore, this embodiment constructs channel trailing effect evaluation parameters for each pixel using a density run-length matrix, and determines the sequencing segmentation probability coefficient for each pixel within the current channel by combining the DNA density distribution heterogeneity coefficient of the pixel. The expression is as follows:

[0076]

[0077]

[0078] In the formula, Let m and n represent the parameters for evaluating the channel trailing effect of pixel i in channel A, where m and n represent the number of rows and columns in the density run-length matrix of pixel i, respectively. , Let represent the mean values ​​of all elements in the p-th and q-th columns of the density run-length matrix for pixel i, respectively. , Let represent the sums of all elements in the k-th and (k+1)-th rows of the grayscale run-length matrix for pixel i, respectively. Indicates a preset constant parameter;

[0079] This represents the sequencing segmentation probability coefficient of pixel i within channel A. This represents the normalization function; here, the maximum value normalization function is used. This represents the heterogeneity coefficient of DNA density distribution at pixel i within channel A.

[0080] Among these, the greater the probability that the second window centered on pixel i is located in the tailing effect region within channel A, the more consistent the DNA density variation characteristics of each pixel in the same row within the second window centered on pixel i, and the closer the total run lengths of each DNA density level in the density run matrix of pixel i are. , The closer their sizes are, the better. The smaller the value, the more pronounced the tailing effect within the second window centered on pixel i. The more uniform the distribution of DNA density levels within this second window, the more they decrease from top to bottom according to the tailing direction. The run lengths of each DNA density level in the grayscale run-length matrix of pixel i will not be uniformly distributed; that is, the non-zero run lengths of each DNA density level are almost concentrated in one column, and the sum of the elements in that column is much greater than the sum of the elements in the other columns. The larger the value, the better. The larger the value, the better; that is, The larger the value, the greater the influence of the tailing effect on pixel i in the current channel in the electrophoresis image, and the more uneven the DNA density distribution in the local area where pixel i is located. The larger the channel tailing effect evaluation parameter of pixel i, the greater the probability that pixel i is located at the boundary of the DNA fragment.

[0081] Step S003: Adaptively determine the decision threshold during growth based on the sequencing segmentation probability coefficient of the pixels in the channel, and determine the base sequence of the DNA sequence based on each growth region obtained by the region growth method, thus completing the visual detection of the DNA sequence.

[0082] Specifically, the decision threshold for subsequent region growth is adjusted based on the sequencing segmentation probability coefficient. The adjustment principle is as follows: when the sequencing segmentation probability coefficient of pixel i in the channel is larger, it indicates that the current pixel is more likely to be at the boundary of a DNA fragment. The region growth method should be more sensitive to boundary information at this point, and a smaller decision threshold should be set to distinguish subtle differences. Conversely, the smaller the sequencing segmentation probability coefficient, the less likely the current pixel is to be at the boundary of a DNA fragment. The region growth method should be less sensitive to boundary information at this point, and a larger decision threshold should be set.

[0083] The specific process is as follows: The region growing method sets a decision threshold, and then determines whether the grayscale difference between the current seed pixel and the pixels in its 8-neighborhood exceeds the threshold. Neighborhood pixels that do not exceed the threshold are added to the already grown region; otherwise, they are not added.

[0084] First, three times the standard deviation of the gray values ​​of all pixels within the statistical channel is used as the initial gray threshold y. Then, following the above procedure, the sequencing segmentation probability coefficients for each pixel are obtained, and the optimal threshold for all sequencing segmentation probability coefficients is obtained using the Otsu thresholding method. The Otsu threshold method is a well-known technique, and its specific process will not be described in detail here.

[0085] The decision threshold for region growing of each pixel is adjusted based on the comparison between the sequencing segmentation probability coefficient of each pixel and the optimal threshold. The decision threshold of pixel i is represented as y(i):

[0086]

[0087] When the sequencing segmentation probability coefficient of pixel i is less than the optimal threshold, it indicates that the pixel is less likely to be located at the boundary of a DNA fragment. In this case, it can be utilized... Increase the initial grayscale threshold as the decision threshold for pixel i; however, the larger the sequencing segmentation probability coefficient of pixel i, the greater the possibility that the pixel is located at the boundary of DNA fragments, so the initial grayscale threshold should be reduced to obtain the decision threshold for pixel i.

[0088] During region growing, the center pixels of each contour in the DNA electrophoresis grayscale image are used as various sub-pixels in the region growing method. The absolute value of the difference between the grayscale value of each sub-pixel and the grayscale value of the adjacent pixels is calculated. When the absolute value of the difference is greater than the adaptive threshold of the adjacent pixels, the adjacent pixels are not added to the already grown region. After traversing all pixels, the growing method stops. Each grown region in the DNA electrophoresis grayscale image can be obtained according to the region growing method, and each grown region is used as a DNA fragment.

[0089] Finally, DNA fragment decoding is performed based on the grayscale images of the segmented DNA obtained through electrophoresis. Since the last nucleotide of each of the four channels (adenine (A), guanine (G), thymine (T), and cytosine (C)) corresponds to a different ddNTP, and each ddNTP pairs according to the complementary base pairing principle, and each channel contains DNA fragments of different lengths—longer fragments move closer together during electrophoresis, while shorter fragments move farther—the base types of the ddNTPs in the corresponding channels can be read from bottom to top. This allows for the visual detection of DNA sequences.

[0090] Based on the same inventive concept as the above method, this application also provides a high-voltage power supply DNA sequencing visual inspection system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described high-voltage power supply DNA sequencing visual inspection methods.

[0091] It should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

[0092] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A high-voltage power supply DNA sequencing visual inspection method, characterized in that, The method includes the following steps: A high-voltage power supply was applied to both ends of the DNA sequencer, and the Sanger sequencing technology was used to obtain DNA electrophoresis images in four channels; grayscale images of DNA electrophoresis were obtained by grayscale processing. The local DNA density parameters of each pixel are determined based on the distribution characteristics of DNA fragments in each channel in the DNA electrophoresis grayscale image; a density run-length matrix is ​​constructed based on the quantization results of the local DNA density parameters of all pixels; the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters of the pixels are obtained based on the element size in the density run-length matrix; the sequencing segmentation probability coefficient of the pixels is determined based on the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters. The decision threshold during growth is adaptively determined based on the sequencing segmentation probability coefficient of pixels within the channel, and the base sequence of the DNA sequence is determined based on each growth region obtained by the region growth method, thus completing the visual detection of the DNA sequence. The determination of the local DNA density parameters for each pixel includes: Treat each pixel as a pixel to be processed, and obtain a first window of a preset size for each pixel to be processed, centered on each pixel to be processed; The DNA local density confidence coefficient of each pixel in the first window is determined based on the gray value difference between each pixel in the first window and the pixels in its eight neighboring regions. Calculate the product of the DNA local density confidence coefficient of each pixel in the first window and the gray value of each pixel, and use the average of the accumulated results of the product on the first window of a preset size for each pixel to be processed as the local DNA density parameter of each pixel to be processed. The method of obtaining the DNA density distribution heterogeneity coefficient and channel tailing effect evaluation parameters of pixels based on the element sizes in the density run-length matrix includes: Construct a third window of a preset size centered on each pixel; Obtain the maximum run length in the density run-length matrix of the second window for each pixel; Each pixel is taken as the target pixel, and the absolute value of the difference between the maximum run length corresponding to the target pixel and the maximum run length corresponding to any non-center pixel in the third window of the target pixel is taken as the first difference. The absolute value of the difference between the mean of all run lengths in the density run length matrix of the second window of the target pixel and the mean of all run lengths in the density run length matrix of the second window of any non-center pixel in the third window of the target pixel is taken as the second difference. The average of the product of the first difference and the second difference accumulated over all non-center pixels in the third window of the target pixel is used as the DNA density distribution heterogeneity coefficient of the target pixel. Calculate the mean of each column element in the density run-length matrix of the second window of the target pixel, and use the sum of the absolute values ​​of the differences between the means of two adjacent columns on the density run-length matrix as the numerator; Calculate the sum of the elements in each row of the density run-length matrix of the second window of the target pixel, and use the sum of the absolute difference between the sums of the elements in two adjacent rows on the density run-length matrix and the preset constant parameter as the denominator. The ratio of the numerator to the denominator is used as an evaluation parameter for the channel tailing effect of the target pixel. The sequencing segmentation probability coefficient for determining the pixel includes: The product of the normalized result of the DNA density distribution heterogeneity coefficient of each pixel and the normalized result of the channel tailing effect evaluation parameter is used as the sequencing segmentation probability coefficient of each pixel. Determining the base sequence of a DNA sequence includes: The center pixels of each contour in the DNA electrophoresis grayscale image are used as the initial seed points for the region growing method. For any initial seed point, calculate the absolute value of the difference between the gray values ​​of each initial seed point and its neighboring pixels. If the absolute value of the difference is greater than the decision threshold of the neighboring pixels, the neighboring pixels are not added to the already grown region; if the absolute value of the difference is less than or equal to the decision threshold of the neighboring pixels, the neighboring pixels are added to the already grown region. After traversing all pixels, growth is stopped, and each growth region in the DNA electrophoresis grayscale image is obtained. Each growth region is taken as a DNA fragment. DNA fragments are decoded based on the segmented DNA electrophoresis grayscale image, and the base types of ddNTPs in the corresponding channels are read from bottom to top based on the decoding results.

2. The high-voltage power supply DNA sequencing visual inspection method as described in claim 1, characterized in that, The method for determining the confidence coefficient of the local DNA density of each pixel within the first window is as follows: Calculate the absolute value of the difference between the gray value of each pixel in the first window and the gray value of any pixel in the eight neighboring regions. Take the reciprocal of the sum of the absolute values ​​of the differences in the eight neighboring regions and the sum of the sum of the preset constant parameters as the DNA local density confidence coefficient of each pixel in the first window.

3. The high-voltage power supply DNA sequencing visual inspection method as described in claim 1, characterized in that, The density run-length matrix is ​​constructed based on the quantization results of the local DNA density parameters of all pixels, including: The local DNA density parameters of all pixels are obtained respectively, and all the local DNA density parameters are mapped to a first preset number of density levels; A second window of a preset size is constructed with each pixel as the center. The electrophoretic direction of the DNA fragment is used as the run direction. The density run matrix of the second window for each pixel is obtained based on the mapping result and the run direction according to the grayscale run matrix method.

4. The high-voltage power supply DNA sequencing visual inspection method as described in claim 1, characterized in that, The adaptive determination of the decision threshold during growth based on the sequencing segmentation probability coefficient of pixels within the channel includes: The initial grayscale threshold is set to three times the standard deviation of the grayscale values ​​of all pixels in the statistical channel. The optimal threshold is obtained by taking the sequencing segmentation probability coefficients of all pixels as input and using the Otsu thresholding method. The decision threshold for region growing of each pixel is determined by comparing the sequencing segmentation probability coefficient of each pixel with the optimal threshold.

5. The high-voltage power supply DNA sequencing visual inspection method as described in claim 4, characterized in that, The determination of the decision threshold for region growing for each pixel includes: If the sequencing segmentation probability coefficient of a pixel is less than the optimal threshold, then the sum of the sequencing segmentation probability coefficient of the pixel and the sum of the sum and the initial grayscale threshold are used as the decision threshold of the pixel. If the sequencing segmentation probability coefficient of a pixel is greater than or equal to the optimal threshold, then the product of the sequencing segmentation probability coefficient of the pixel and the initial grayscale threshold is used as the decision threshold of the pixel.

6. A high-voltage power supply DNA sequencing visual inspection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the high-voltage power supply DNA sequencing visual inspection method as described in any one of claims 1-5.

Images