Cross-hole identification method and device, storage medium and electronic device

Abstract

This application discloses a method, apparatus, storage medium, and electronic device for cross-well identification, relating to the field of bioinformatics. It involves obtaining a first matrix corresponding to each well position in a sequencing chip, translating the first matrix along at least one target direction to obtain at least one second matrix, and comparing each second matrix with the first matrix to determine the sequence similarity between each well position and its adjacent well positions along the corresponding target direction. Based on the sequence similarity, cross-well behavior is identified. Thus, by mapping the sequence data of the chip well positions into a matrix structure and utilizing matrix translation operations to achieve batch alignment of adjacent well positions, this application achieves efficient identification and quantitative evaluation of cross-well behavior, solving the problems of high computational load and low efficiency in traditional point-by-point traversal methods, and improving the accuracy of cross-well behavior identification in sequencing chips.

Description

Technical Field

[0001] This application relates to the field of bioinformatics processing technology, and in particular to a method, apparatus, storage medium and electronic device for cross-pore identification. Background Technology

[0002] In the research and application of array-based sequencing chips (such as Pattern array sequencing chips), the amplification reaction of the sequencing template is one of the core steps in achieving high-throughput sequencing. However, due to various factors such as the chip hole structure design and the characteristics of the amplification reaction system, the template sequence within a hole can easily diffuse into adjacent holes during amplification, such as... Figure 1 As shown in the figure, template sequences should typically be amplified only within their respective wells. For example, sequence clusters 1 and 2 are located in different wells. However, in practice, sequences often migrate from one well to an adjacent well and participate in amplification, resulting in highly similar sequences in adjacent wells. This phenomenon can interfere with the accuracy of sequencing data, limit the increase in sequencing throughput, and consequently affect the overall performance and application effect of the chip.

[0003] In the existing technology, there is a lack of effective evaluation methods for the above-mentioned amplification cross-pore phenomenon, making it difficult to quantify the severity of the cross-pore phenomenon. This results in a lack of reliable data support for the optimization of chip process design (such as key parameters such as hole spacing and hole size), which limits the technological upgrade and performance improvement of array-type sequencing chips. Summary of the Invention

[0004] In view of the above problems, this application provides a method, apparatus, storage medium, and electronic device for cross-pore identification, so as to achieve efficient and accurate identification of cross-pore behavior during the sequencing chip amplification process. The specific solution is as follows:

[0005] The first aspect of this application provides a method for cross-aperture identification, comprising:

[0006] A first matrix corresponding to each well position in the sequencing chip is obtained, wherein the elements of the first matrix correspond to the well positions of the sequencing chip, and the value of the element represents the sequencing sequence of the corresponding well position;

[0007] The first matrix is ​​translated along at least one target direction to obtain at least one second matrix; wherein the translation is such that each element in the second matrix corresponds to the sequencing sequence of the adjacent pore in the first matrix along the target direction at that position;

[0008] Each second matrix is ​​compared with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction;

[0009] Based on the sequence similarity, cross-pore behavior is identified.

[0010] A second aspect of this application provides a cross-aperture identification device, comprising:

[0011] The acquisition unit is used to obtain a first matrix corresponding to each well position in the sequencing chip, wherein the elements of the first matrix correspond to the well positions of the sequencing chip, and the value of the element represents the sequencing sequence of the corresponding well position.

[0012] A translation unit is used to translate the first matrix along at least one target direction to obtain at least one second matrix; wherein the translation is used to make the elements at each position in the second matrix correspond to the sequencing sequences of the adjacent pores at that position in the first matrix in the target direction;

[0013] A comparison unit is used to compare each of the second matrix with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction;

[0014] The identification unit is used to identify cross-pore behavior based on the sequence similarity.

[0015] A third aspect of this application provides a storage medium storing a computer program that, when executed by a processor, implements the cross-hole recognition method of the first aspect or any implementation thereof.

[0016] A fourth aspect of this application provides an electronic device, comprising:

[0017] Memory, used to store computer programs;

[0018] A processor is configured to execute a computer program stored in the memory, wherein the computer program, when executed by the processor, implements the cross-hole recognition method of the first aspect or any implementation thereof.

[0019] By employing the above technical solution, this application provides a method, apparatus, storage medium, and electronic device for cross-well identification. This involves obtaining a first matrix corresponding to each well position in a sequencing chip, translating the first matrix along at least one target direction to obtain at least one second matrix, comparing each second matrix with the first matrix to determine the sequence similarity between each well position and its adjacent well positions in the corresponding target direction, and identifying cross-well behavior based on the sequence similarity. Thus, this application maps the sequence data of the chip well positions into a matrix structure and utilizes matrix translation operations to achieve batch comparison of adjacent well positions, achieving efficient identification and quantitative evaluation of cross-well behavior. This solves the problems of high computational load and low efficiency in traditional point-by-point traversal methods, improving the accuracy of cross-well behavior identification in sequencing chips. Attached Figure Description

[0020] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0021] Figure 1 A schematic diagram illustrating a transpore amplification phenomenon provided in an embodiment of this application;

[0022] Figure 2 A schematic flowchart illustrating a cross-hole identification method provided in an embodiment of this application;

[0023] Figure 3 This is a schematic diagram illustrating the filling of a cellular array mapped as a matrix, provided in an embodiment of this application.

[0024] Figure 4 This is a schematic diagram showing the proximity of aperture positions in a honeycomb array chip, provided in an embodiment of this application.

[0025] Figure 5 This is a schematic diagram of a matrix translation operation provided in an embodiment of this application;

[0026] Figure 6 An example diagram of sequence similarity calculation provided in this application embodiment;

[0027] Figure 7 This is a schematic diagram of a score matrix corresponding to a target direction provided in an embodiment of this application;

[0028] Figure 8 This application provides a schematic diagram illustrating the number of cross-hole points in a target direction, as part of an embodiment of the present application.

[0029] Figure 9 A schematic diagram illustrating the cumulative number of cross-hole points provided in this application embodiment;

[0030] Figure 10 This application provides a probability distribution diagram of the same number of bases under random conditions in an embodiment of the present application;

[0031] Figure 11 This is a schematic diagram of a rectangular array of aperture positions provided in an embodiment of this application;

[0032] Figure 12 This is a schematic diagram of a cross-hole recognition device provided in an embodiment of this application. Detailed Implementation

[0033] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0034] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0035] The terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0036] To facilitate understanding and description of the embodiments of this application, the relevant technical terms are explained as follows:

[0037] Sequencing chips: Sequencing chips include random surface sequencing chips and array-type sequencing chips (including patterned arrays and microwell arrays). This application primarily addresses the issue of cross-well amplification in microwell array chips. For example, the sequencing chip in this application embodiment is a nanowell pattern chip. A nanowell pattern chip includes an array of nanowells for fixing nucleic acid templates. A nanowell, also known as a nanopore, micropore, or pore, refers to a nanoscale porous structure formed on the surface of a solid support. The shape of a nanowell or nanopore is not strictly limited; it can be circular, elliptical, triangular, quadrilateral, or other polygonal, or even have no specific shape. In some embodiments, the nanowell or nanopore is configured with a symmetrical cross-section. For example, the cross-section of a nanowell or nanopore is circular or a regular polygon.

[0038] Well sites: In this embodiment, well sites refer to all microwells on the surface of the sequencing chip. Ideally, each well site is fixed with one type of nucleic acid template for independent amplification and sequencing reactions, producing the corresponding nucleic acid template. However, in practical applications, due to factors such as template loading and the amplification process, there may be a very small number of well sites that are not fixed with nucleic acid templates or where one well site is fixed with two different nucleic acid templates. When the same well site contains two different nucleic acid templates, that well site may generate two detectable signals in one round of sequencing. Such situations may interfere with the identification of sequencing signals and affect sequencing accuracy. The cross-well phenomenon involved in this embodiment is one of the situations that leads to the appearance of abnormal nucleic acid templates within a well site. That is, when the template sequence diffuses from the original well site to adjacent well sites, it forms a state similar to "one template occupying two adjacent well sites," thereby affecting the accurate identification of sequencing signals.

[0039] Adjacent apertures: In a chip aperture array structure, these are apertures that are geometrically adjacent to the target aperture. Depending on the aperture array structure, each aperture may have a different number of adjacent apertures; for example, in a honeycomb array, each aperture typically has six adjacent apertures.

[0040] Sequencing sequence: refers to the nucleic acid sequence information read from each well of the sequencing chip. It is usually represented by a string of four bases: A, T, C, and G. Each sequence corresponds to one well on the chip.

[0041] Sequence similarity: A quantitative indicator used to measure the degree of similarity between two sequencing sequences. In this application, it specifically refers to the number or percentage of bases of the same type at the same position in the two sequences.

[0042] Matrix translation: refers to the operation of moving the constructed matrix as a whole along a specified direction. Through translation, the correspondence between each element in the matrix and its adjacent elements can be established, thereby realizing batch adjacent comparison.

[0043] The pore-crossing identification method provided in the embodiments of this application can be applied to the research, development, production, and application processes of sequencing chips. For example, during the chip design stage, this pore-crossing identification method can be used to evaluate the impact of design parameters such as different pore array structures, pore spacing, and sizes on amplified pores, guiding the optimization direction of chip processes. During the manufacturing stage, it can be used as a quality inspection indicator to evaluate the pore-crossing degree of each batch of chips, enabling online monitoring of production quality and batch-to-batch consistency verification. During the sequencing application stage, sequencing data can be labeled and filtered based on the pore-crossing identification results, reducing the impact of pore-crossing on sequencing throughput and accuracy. Furthermore, long-term monitoring of pore-crossing indicators can promptly detect abnormal fluctuations in the production process, providing a quantitative basis for optimizing biochemical reaction conditions.

[0044] See Figure 2The diagram illustrates a flowchart of a cross-aperture identification method provided in an embodiment of this application. The method may include the following steps:

[0045] S101. Obtain the first matrix corresponding to each well position in the sequencing chip.

[0046] The sequencing chip has multiple wells, each yielding a sequencing sequence after sequencing. To facilitate batch processing of large amounts of well data, in this embodiment, step S101 maps these wells and their corresponding sequencing sequences into a matrix, resulting in a first matrix. The elements of this first matrix correspond to the wells on the sequencing chip, and the value of each element represents the sequencing sequence of that well. For example, the first matrix can be a three-dimensional data structure, where the first and second dimensions correspond to the rows and columns of the wells on the sequencing chip, respectively, used to locate the wells within the chip; its element type is integer index. The third dimension stores the sequencing sequence of the corresponding well, and its element type is string, composed of the four base characters A, T, C, and G, representing the base sequence information obtained after sequencing at that well. Each element in the matrix has a one-to-one correspondence with a well on the chip, and the value of the element is the sequencing sequence obtained at that well. For locations on the chip where there are no actual holes (e.g., blank areas filled to maintain the integrity of the matrix structure when mapping a non-rectangular hole array structure to a rectangular matrix), the corresponding matrix elements can be set to null values ​​or special markers for easier subsequent processing. Through this matrix representation, the originally scattered hole data on the chip is organized into a regular data structure, laying the foundation for subsequent batch processing.

[0047] S102. Translate the first matrix along at least one target direction to obtain at least one second matrix.

[0048] After obtaining the first matrix, this step establishes the correspondence between the well position and its adjacent well positions through matrix translation operation, and obtains at least one second matrix with a target translation direction corresponding to the first matrix. The translation makes the elements at each position in the second matrix correspond to the sequencing sequences of the adjacent well positions in the first matrix in the target direction.

[0049] In determining the target orientation, the outgoing directions of all well positions are first determined based on the needle structure of the sequencing chip. The outgoing direction refers to the direction from each well position to its adjacent well positions, reflecting the adjacency relationship between well positions. Taking a honeycomb array as an example... Figure 4 The red pore represents the pore currently being analyzed, surrounded by adjacent pores (such as...). Figure 4 The green holes surrounding the red holes correspond to six outgoing directions: left, right, upper left, lower left, upper right, and lower right.

[0050] Since aperture array structures typically exhibit central symmetry, the comparison results obtained by performing translation operations in opposite directions are numerically equivalent. Taking the left-right direction as an example, regardless of whether the matrix is ​​translated to the left or right, the matrix region ultimately involved in the comparison is the part that overlaps with the original matrix after translation, and the calculation results are completely consistent. Therefore, by selecting only one direction from each pair of opposite directions as the target direction for translation, the cross-aperture information of all adjacent directions can be obtained, avoiding redundant calculations. Based on this, a subset satisfying the condition of non-oppositeness is selected from all outgoing directions as the target direction set. That is, for any direction in the target direction set, its opposite direction is not in this set. For example, for a honeycomb array, the right, upper right, and lower right directions can be selected from the six outgoing directions as target directions, while their opposite directions, left, lower left, and upper left, are not calculated repeatedly. A global translation operation is performed on the first matrix along each selected target direction to obtain the second matrix corresponding to the target direction. Through this direction selection strategy, the number of translation operations is reduced by half while ensuring a comprehensive evaluation of the cross-aperture situation in each direction, significantly reducing the computational load.

[0051] For any given well location, there may be multiple target directions. For example, for a matrix array, the target direction could be right, down, etc.; for a cellular array, the target direction could be upper right, right, lower right, etc. For instance, in the case of a sequencing chip that is a cellular array chip, for any well location S and its adjacent well locations S1, S2, S3, S4, S5, S6, the outgoing directions include: S→S1, S→S2, S→S3, S→S4, S→S5, and S→S6. Figure 9 As shown, the target direction may include one of S→S1 and S→SX (the outgoing direction that is completely opposite to the S→S1 direction), one of S→S2 and S→SX (the outgoing direction that is completely opposite to the S→S2 direction), and one of S→S3 and S→SX (the outgoing direction that is completely opposite to the S→S3 direction).

[0052] Then, the first matrix is ​​translated as a whole along the selected target direction. Taking a rightward translation as an example, as follows... Figure 9 Assume that well S1 in the first matrix is ​​located at position (i,j), and its sequencing sequence is stored at that position. After shifting the entire first matrix one unit to the right, a second matrix is ​​obtained. In the second matrix, the element with the same position coordinates (i,j) as the original well S1 actually corresponds to the sequencing sequence of well S at position (i,j-1) in the first matrix, i.e., the sequence of the well adjacent to S1 to the left. Through this shift operation, the elements at each position in the second matrix establish a correspondence with the adjacent wells in the first matrix in the opposite direction of the shift, thereby enabling batch alignment of sequences between wells and their adjacent wells.

[0053] Translation operations can be performed along multiple target directions, each yielding a corresponding second matrix. For example, if three target directions are selected, the translation will result in three second matrices, which are used to evaluate the similarity between a hole location and its three adjacent holes in different directions.

[0054] S103. Compare each second matrix with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction.

[0055] A position-by-position alignment is performed on the first matrix and the second matrix. That is, for each identical matrix position, the sequencing sequence stored in the first matrix (i.e., the first sequencing sequence) and the sequencing sequence stored in the second matrix (i.e., the second sequencing sequence) at that position are extracted. The two sequencing sequences are treated as strings and compared position by position. The number of bases with the same type at the same position is counted, thereby calculating the sequence similarity between each well position and its adjacent well positions in the target direction.

[0056] Specifically, for each well position in the first matrix, the sequencing sequence at that position in the first matrix (referred to as the first sequencing sequence) is obtained, and simultaneously, the sequencing sequence at the same position in the second matrix (referred to as the second sequencing sequence) is obtained. Due to the correspondence of the translation operation, the second sequencing sequence is essentially the sequencing sequence of the adjacent well positions of the target well position in the target direction. Then, the first sequencing sequence and the second sequencing sequence are compared, and the number of identical base types at the same positions is counted. For example, if two sequences are both 100 bp in length, and the bases at positions 1, 3, 5, etc., are completely identical, the total number of identical sites is counted, and this total number is the sequence similarity of that well position in the target direction. The higher the similarity, the closer the two sequences are, and the greater the possibility of cross-well occurrence between that well position and its adjacent well positions.

[0057] S104. Identify cross-pore behavior based on sequence similarity.

[0058] After obtaining the sequence similarity of each aperture location in each target direction, this step identifies cross-aperture behavior based on this similarity information.

[0059] For each pore location, the sequence similarity calculated in each target direction is compared with a preset condition to determine whether cross-pore behavior exists in that direction. The preset condition can be a similarity threshold set based on sequence length and statistical regularities. When the similarity between two sequences reaches or exceeds this threshold, it indicates that the similarity between the two sequences exceeds expectations under random conditions, which is highly likely due to template molecules migrating from one pore location to an adjacent pore location, thus determining the existence of cross-pore behavior. By statistically analyzing the determination results for all pore locations in all target directions, the overall cross-pore situation of the chip can be obtained, such as the total number of pore locations with cross-pores and the directional distribution of cross-pore occurrences. These statistical results can be output as cross-pore identification results for subsequent chip quality assessment or process optimization.

[0060] As an optional implementation, the sequence similarity of each pore site in each target direction can be compared with a preset threshold. If the similarity exceeds the threshold, it is determined that the pore site exhibits cross-pore behavior in that direction. For example, for a 100bp sequence, a similarity threshold of 80 can be set; when two sequences have more than 80 identical bases, cross-pore behavior is considered to have occurred.

[0061] By statistically analyzing the determination results of all vias in all target directions, the overall via situation of the chip can be obtained, such as the total number of vias with vias and the directional distribution of via occurrences. These statistical results can be output as via identification results for subsequent chip quality assessment or process optimization.

[0062] This application provides a method for cross-well identification. It obtains a first matrix corresponding to each well in a sequencing chip, translates the first matrix along at least one target direction to obtain a second matrix, and uses position-by-position comparison between the second matrix and the first matrix to determine the sequence similarity between each well and its adjacent wells. Based on this sequence similarity, cross-well behavior is identified. By employing matrix representation and overall translation operations, it enables batch parallel comparison of the sequence similarity between all wells on the chip and their adjacent wells. This solves the tedious process of point-by-point traversal and repeated calculation of adjacent relationships required in traditional methods, reducing computational complexity and improving the processing efficiency of cross-well identification. Simultaneously, through matrix-based data organization and the geometric correspondence of translation operations, it accurately establishes the association between each well and its adjacent wells in a specific direction, ensuring the accuracy and repeatability of cross-well identification. This provides an efficient, reliable, and quantifiable technical means for process optimization and quality control of sequencing chips.

[0063] The technical features of the cross-via identification method according to the embodiments of this application will be described in detail below, taking into account specific application scenarios. The specific implementation methods of each step in the method flow will be explained below. These implementation methods can be flexibly applied to the relevant steps of the cross-via identification method in the aforementioned embodiments to jointly achieve the purpose of improving the accuracy of cross-via identification and optimizing chip process parameters.

[0064] In some embodiments of this application, obtaining the first matrix corresponding to each well position in the sequencing chip includes: step S201 constructing an initial matrix framework based on the actual well position arrangement characteristics of the sequencing chip, wherein the initial matrix framework has positions that correspond one-to-one with the actual well positions; step S202 filling the corresponding positions of each well position of the sequencing chip with the sequencing sequences of the sequencing chip to obtain the first matrix.

[0065] After obtaining the sequencing data from the sequencing chip in step S201, the dispersed well data on the chip needs to be organized into a matrix structure that facilitates batch processing. To this end, this step constructs an initial matrix framework corresponding to the well distribution based on the actual arrangement characteristics of the wells on the chip. The well arrangement of a sequencing chip can take many forms, such as common rectangular arrays, square arrays, and honeycomb arrays used to increase the well density per unit area. These arrangement methods determine the geometric positional relationships between the wells, thus affecting how they are mapped into a matrix structure.

[0066] When constructing the initial matrix framework, it is necessary to ensure that each position in the framework corresponds one-to-one with an actual aperture on the chip. The construction method differs for chips with different layout characteristics, which will be explained in detail in subsequent embodiments. Figure 3 As shown, taking a honeycomb array as an example, the holes are arranged in a hexagonal, tightly packed pattern, with a half-hole offset between adjacent rows. To map this arrangement into a matrix form, a matrix framework that can accommodate all the holes needs to be constructed, where the rows and columns of the matrix correspond to the row and column positions on the chip according to certain rules.

[0067] After constructing the initial matrix framework in step S202, the sequence data of each well obtained from sequencing is filled into the corresponding positions in the framework. Specifically, for each well on the chip, its row and column indices in the initial matrix framework are determined based on its position coordinates in the actual arrangement, and then the sequencing sequence of that well is filled into the corresponding position in the matrix as an element value.

[0068] For locations on the chip with actual wells, the values ​​of their matrix elements represent the sequencing sequence for that well, typically expressed as strings, such as "ATCGCTGA...". For locations on the chip without actual wells (e.g., blank areas filled to maintain matrix structure integrity), their corresponding matrix elements can be set to null or special markers (e.g., -1, None). In the Python implementation of this application, the null marker uses the nan (Not a Number) type from the NumPy library. This is ignored in subsequent similarity calculations. The matrix obtained after filling is the first matrix, its dimensions determined by the initial matrix framework, and its element values ​​contain the sequencing information for all wells on the chip, along with necessary null markers. Figure 3 As shown, after the honeycomb array is filled, the originally misaligned holes are organized into a regular 5x3 matrix, with blank positions filled with null values.

[0069] Through the above steps, this embodiment transforms the originally scattered hole position data on the chip into a regular matrix form, laying the data structure foundation for subsequent translation operations and batch comparisons. This matrix representation method enables the processing of massive hole position data to fully utilize the parallelism and efficiency of matrix operations, avoiding the frequent coordinate calculations and conditional judgments in the traditional point-by-point traversal method.

[0070] In some embodiments of this application, the well arrangement of the sequencing chip has diverse geometric structures, and different arrangement characteristics determine the specific way to construct the initial matrix framework. This embodiment divides the well arrangement into two cases based on whether it can be directly mapped to a matrix form. That is, based on the actual well arrangement characteristics of the sequencing chip, constructing the initial matrix framework includes at least two cases. Case 1: In the case where the actual well arrangement characteristics are the first type of arrangement where the well arrangement can be directly mapped to a matrix form, a matrix framework with the same row and column dimensions is determined as the initial matrix framework. Case 2: In the case where the actual well arrangement characteristics are the second type of arrangement where the well arrangement cannot be directly mapped to a matrix form but has a regular geometric structure, an initial matrix architecture capable of accommodating all wells is constructed. In this case, the initial matrix architecture sets null value markers at positions where no actual wells exist.

[0071] In Case 1, the first type of arrangement refers to a situation where the geometric arrangement of the holes already forms a regular row and column structure, which can be directly represented as a matrix. For example... Figure 11As shown, the core characteristic of this type of arrangement is that the holes are strictly aligned in both the row and column directions, with no offset between adjacent rows, forming a standard rectangular grid. In this structure, the row and column indices of the holes have a direct one-to-one correspondence with their actual physical locations on the chip, without the need for coordinate offsetting or correction. For the first type of arrangement, the specific method for constructing the initial matrix framework is as follows: based on the actual number of rows and columns of the chip holes, create an empty matrix framework with the same row and column dimensions. The row index of the matrix corresponds to the row number on the chip, and the column index corresponds to the column number on the chip. For example, as shown... Figure 11 The 5x10 sequencing chip shown can be directly constructed into a 5x10 matrix as the initial matrix framework. Position (i,j) in the matrix corresponds to the actual well position in the i-th row and j-th column of the chip. Because the well positions are regularly arranged, each position in the matrix framework corresponds to an actual well position. There is no need to fill in null values ​​at non-well positions; the sequencing sequences for each well position can be directly filled into the corresponding matrix position to complete the construction of the first matrix. The second type of arrangement in scenario two refers to well positions that, although possessing a regular geometric structure, cannot be directly mapped to a regular row-column matrix. Null values ​​need to be filled during matrix construction to maintain structural integrity. This type of arrangement is characterized by well positions that may be aligned in the row direction, but there may be misalignment between adjacent rows, or the well positions may exhibit a special distribution such as a ring or arc shape.

[0072] For the second type of arrangement, when constructing the initial matrix frame, it is necessary to first determine the minimum rectangular range that can accommodate all holes, and then create the matrix frame within this range, setting null values ​​for positions where no actual holes exist. For example... Figure 3 and Figure 4 Taking the honeycomb array shown as an example, the holes are arranged in a tight hexagonal pattern, with adjacent rows staggered by half a hole spacing. To map this arrangement into a matrix form, a matrix framework needs to be constructed, where the number of rows corresponds to the number of rows in the chip, and the number of columns is determined by the number of holes in the widest row. For example... Figure 3 As shown, the honeycomb array is filled into a 5x3 matrix. Some matrix positions (such as row 1, column 1, row 5, column 3, etc.) do not have corresponding holes on the actual chip. These positions are marked with null values ​​in the matrix to ensure the matrix's regularity. For example, for a chip arranged in a ring, the holes are distributed along concentric circles. This can be mapped into a matrix according to angles and radii. The rows of the matrix correspond to the radius layers, and the columns correspond to the angle positions. Null values ​​are set for the angle positions where no holes exist.

[0073] By differentiating and processing the two scenarios described above, this embodiment can adapt to sequencing chips with different geometric structures, transforming various regularly arranged well positions into a regular matrix form, laying the foundation for subsequent translation operations and batch alignment.

[0074] Further, in this embodiment, the first type of arrangement refers to a layout where the aperture positions can be directly mapped to a matrix, mainly including the following types: rectangular array or square array. A rectangular array refers to aperture positions arranged neatly in rows and columns, with rows and columns aligned to form a rectangular grid structure. The number of rows and columns in a rectangular array can be equal or unequal, and the aperture spacing can be the same or different in the row and column directions. This arrangement is common in traditional microplates, such as 96-well plates (8 rows × 12 columns) and 384-well plates (16 rows × 24 columns). A square array is a special case of a rectangular array where the number of rows and columns are equal, and the row spacing is the same as the column spacing, forming a square grid structure. This arrangement can be applied in some high-density chip designs. In this embodiment, for the first type of arrangement, since the aperture positions themselves already constitute a regular row and column structure, no null value filling is needed when constructing the matrix framework; the row and column dimensions of the chip can be directly used as the dimensions of the matrix.

[0075] The second type of arrangement refers to arrangements where the holes, although possessing a regular geometric structure, cannot be directly mapped into a matrix form and require filling with empty values ​​when constructing the matrix. This mainly includes the following types: honeycomb array, triangular array, staggered array, arc or ring array, and sparse or irregular distribution array. Among these, a honeycomb array refers to holes arranged in a tightly packed hexagonal pattern, with each hole surrounded by six adjacent holes, and adjacent rows staggered by half a hole spacing, forming a honeycomb-like structure. For example... Figure 4As shown, in a honeycomb array, non-edge holes (red) have six adjacent holes, while edge holes (blue) have three adjacent holes. This arrangement allows for more holes per unit area, increasing chip throughput. However, due to row misalignment, it cannot be directly mapped to a regular matrix; null values ​​need to be filled at the misaligned positions during matrix construction. A triangular array arranges holes according to the vertices of an equilateral triangle, with each hole surrounded by six adjacent holes (in a close arrangement), and adjacent rows are staggered by a certain distance. Triangular arrays have a certain geometric duality with honeycomb arrays and also require null value filling during matrix construction. A staggered array aligns holes in the row direction, but adjacent rows are staggered by a certain distance, forming a staggered structure similar to a brick wall. This arrangement is between rectangular and honeycomb arrays; the stagger distance can be half a hole spacing or other proportions. Similarly, null values ​​need to be filled during matrix construction to maintain matrix regularity. Arc-shaped or ring-shaped arrays refer to the arrangement of apertures along concentric circles or arcs. For example, some chips designed for specific applications distribute apertures in polar coordinates. When mapping this arrangement to a matrix, the radial direction is usually used as the matrix rows, and the angular direction as the matrix columns. For different angular positions on the same radius, if there is no actual aperture, a null value is set. Sparse or irregular distribution arrays refer to the apertures on the chip distributed in a specific pattern, with blank areas, but still maintaining a certain arrangement regularity. For example, some chips reserve blank areas in their design for positioning or as a reference; these areas need to be filled with null values ​​in the corresponding positions in the matrix. It can be seen that in the embodiments of this application, whether it is a regular row and column array or a special arrangement such as a honeycomb or ring, it can be converted into a unified data format through the corresponding matrix construction method, realizing the wide applicability and universality of this cross-aperture recognition method.

[0076] In some embodiments of this application, the process of translating a first matrix along at least one target direction to obtain at least one second matrix may include the following steps:

[0077] S301. Based on the pore array structure of the sequencing chip, determine all adjacent directions between pores.

[0078] S302. From all adjacent directions, select a subset that satisfies the first condition as the target direction set. The first condition indicates that for any direction in the subset, its opposite direction is not in the subset.

[0079] S303. Perform a global translation operation on the first matrix along each direction in the target direction set to obtain the second matrix corresponding to the target direction.

[0080] In step S301, based on the pore array structure of the sequencing chip, the outgoing directions of all pore positions are determined. The outgoing direction refers to the direction from each pore position to its adjacent pore positions, reflecting the adjacency relationship between pore positions. Taking a honeycomb array as an example, such as... Figure 4 As shown, each hole has six adjacent holes, with corresponding outgoing directions of left, right, top-left, bottom-left, top-right, and bottom-right. Since the hole array structure typically exhibits central symmetry, the comparison results obtained by translating in opposite directions are numerically equivalent. Taking the left-right direction as an example, regardless of whether the matrix is ​​translated to the left or right, the final matrix region for comparison is the part overlapping the original matrix after translation, and the calculation results are completely consistent. Therefore, by selecting one direction from each pair of opposite directions as the target direction for translation, the cross-hole information for all adjacent directions can be obtained, avoiding redundant calculations. Based on this, a subset satisfying the condition of non-opposite directions is selected from all outgoing directions as the target direction set; that is, for any direction in the target direction set, its opposite direction is not in this set. For example, for a honeycomb array, the right, top-right, and bottom-right directions can be selected from the six outgoing directions as the target directions, while their opposite directions, left, bottom-left, and top-left, are not calculated repeatedly.

[0081] For rectangular arrays, the outgoing direction can be flexibly chosen based on the definition of adjacency relationships: if considering four-neighbor adjacency relationships, the outgoing direction includes four directions: up, down, left, and right; if considering eight-neighbor adjacency relationships (i.e., including diagonal directions), the outgoing direction includes eight directions: up, down, left, right, upper left, upper right, lower left, and lower right. For other arrangements with regular geometric structures, such as triangular arrays and circular arrays, the corresponding set of outgoing directions can be determined based on their geometric characteristics.

[0082] In step S302, after determining all adjacent directions, this step selects a subset as the target direction set for the actual translation operation. The main criterion for selection is that the subset cannot contain a pair of opposite directions simultaneously. Opposite directions refer to a pair of directions that are geometrically opposite, such as "left" and "right", "up" and "down", "upper left" and "lower right", "upper right" and "lower left", etc. The reason for excluding opposite directions from appearing simultaneously in the target direction set is that the comparison results obtained by translating along opposite directions are equivalent.

[0083] by Figure 5For example, comparing the second matrix obtained by shifting the first matrix to the upper right (red area) with the second matrix obtained by shifting the first matrix to the lower left (green area) reveals that the adjacency relationship is essentially the same pair of adjacent aperture positions, just from different perspectives. Therefore, by selecting only one direction for translation and comparison, the cross-aperture information in that pair of adjacent directions can be obtained without repeated calculations. Based on this principle, a subset can be selected from all adjacent directions, where the opposite direction of each direction is not included. The size of this subset is usually half the number of all adjacent directions, depending on the symmetry of the aperture array structure.

[0084] As an optional implementation, the selection of the target direction set can be based on direction numbering. For example, all adjacent directions can be numbered sequentially, and then the directions with odd numbers can be selected as the target direction set. Alternatively, a specific combination of directions can be selected based on the actual application scenario. As another optional implementation, certain directions can be preferentially selected as target directions based on the chip's process characteristics or historical data. For example, if a certain type of chip is more prone to cross-vias in a specific direction, that direction can be prioritized for evaluation, while other directions are used for auxiliary evaluation. As yet another optional implementation, for some asymmetric aperture array structures, it may not be possible to simply select half the number of target directions. In this case, all or some directions can be selected for evaluation according to actual needs. The selection principle of this embodiment still applies, namely, avoiding the selection of mutually opposite directions to reduce redundant calculations.

[0085] In step S303, after determining the target direction set, a global translation operation is performed on the first matrix along each direction in the set. The translation operation is implemented as follows: the first matrix is ​​sliced ​​and padded according to the specified direction. Taking a rightward translation as an example, all rows and elements from column 2 to column n of the first matrix are taken as the translation body. Then, a column of null value markers (such as nan) is added to the right of the last column of the matrix, forming a second matrix with the same dimensions as the first matrix. Through this operation, the element at position (i,j) in the second matrix corresponds to the element at position (i,j-1) in the first matrix, which is the sequencing sequence of the adjacent well on the left side of the original matrix. Similarly, for a leftward translation, columns 1 to (n-1) are taken, and a column of null values ​​is added to the left of the first column; for an upward translation, rows 2 to m are taken, and a row of null values ​​is added above the first row; for a downward translation, rows 1 to (m-1) are taken, and a row of null values ​​is added below the last row. For diagonal translations, slicing and padded operations are performed simultaneously in both row and column directions. Perform a translation operation once for each target direction to obtain a corresponding second matrix. Therefore, if the set of target directions contains k directions, the translation will result in k second matrices, which are used to evaluate the sequence similarity between adjacent holes in k different directions.

[0086] like Figure 4 As shown, for a cellular array, if the target direction set includes three directions (upper right, right, and lower right, or other three directions are selected based on symmetry), then the array is translated along these three directions to obtain three second matrices. Each matrix is ​​used to evaluate the adjacency relationship in one direction.

[0087] Through the above steps, this embodiment achieves the rational selection of the target direction and the efficient execution of translation operations. Utilizing the symmetry of the aperture array structure, while ensuring a comprehensive evaluation of the cross-aperture situation in all directions, the number of translation operations is reduced to half that of all adjacent directions, thus lowering the computational load.

[0088] In one embodiment of this application, the aperture array structure is a honeycomb array structure, with six directions in all adjacent directions and three directions in the target direction set. For example... Figure 4 As shown, a honeycomb array is a common chip aperture arrangement method, characterized by a hexagonal, tightly packed arrangement of apertures, with each aperture (excluding edge apertures) surrounded by six adjacent apertures. Based on geometric position, these six adjacent apertures are located in the following six directions: upper left, upper right, left, right, lower left, and lower right. These six directions constitute the complete set of adjacent directions for the honeycomb array. From a geometric symmetry perspective, these six directions form three pairs of opposite directions: upper left and lower right opposite, left and right opposite, and lower left and upper right opposite.

[0089] For a cellular array, a subset containing three directions can be selected from all six adjacent directions as the target direction set. This subset satisfies the following condition: for any direction in the subset, its opposite direction is not in the subset. As an optional implementation, the upper right, right, and lower right directions can be selected as the target direction set. Figure 4 As shown, after selecting these three directions, their opposite directions, namely bottom left, left, and top left, are not in the target direction set. Translating along the top right, right, and bottom right directions respectively yields three second matrices, used to evaluate the sequence similarity between a hole and its adjacent holes in the top left, left, and bottom left directions. Alternatively, the top left, left, and bottom left directions can also be selected as the target direction set. Both selection methods cover the adjacency relationships of all six directions, only the evaluation perspective differs. Yet another optional implementation method can select other combinations of three directions that meet the conditions, such as top left, right, and bottom left, as long as the selected three directions are not opposite to each other. The specific combination chosen depends on actual application requirements, data storage habits, or the convenience of subsequent processing.

[0090] Taking the selection of the upper right, right, and lower right directions as an example, the specific implementation of the translation operation is as follows:

[0091] Shifting along the upper right: The entire first matrix is ​​shifted one unit to the upper right. After the shift, the element at position (i,j) in the second matrix corresponds to the element at position (i-1,j-1) in the first matrix (assuming row numbers increase downwards and column numbers increase to the right), which is the sequencing sequence of the well adjacent to the lower left in the original matrix. Shifting along the right: The entire first matrix is ​​shifted one unit to the right. After the shift, the element at position (i,j) in the second matrix corresponds to the element at position (i,j-1) in the first matrix, which is the sequencing sequence of the well adjacent to the left in the original matrix. Shifting along the lower right: The entire first matrix is ​​shifted one unit to the lower right. After the shift, the element at position (i,j) in the second matrix corresponds to the element at position (i+1,j-1) in the first matrix, which is the sequencing sequence of the well adjacent to the upper left in the original matrix.

[0092] By performing translation operations in these three directions, three second matrices can be obtained, which are used to evaluate the sequence similarity between a hole position and its adjacent holes to its lower left, left, and upper left, respectively. The sequence similarity between a hole position and its adjacent holes to its lower right, right, and upper right can be obtained indirectly through symmetry and does not need to be calculated separately.

[0093] When performing translation operations, holes located at the edges of the matrix may encounter boundary issues. For example, when translating to the right, the holes in the leftmost column will move out of the matrix boundary after translation, and a valid sequence of adjacent holes cannot be obtained at the corresponding position in the second matrix. For these edge holes, they can be ignored during similarity calculation, or their similarity can be set to a default value (such as 0). The specific handling method can be determined according to actual needs. By taking into account the six-neighborhood structure of the cellular array and utilizing its geometric symmetry to select three non-opposite directions as the target direction set, the number of translation operations is reduced by half while ensuring that all six directions of cross-hole situations can be fully evaluated, significantly reducing computational complexity. At the same time, the edge hole handling mechanism ensures the integrity and practicality of the method.

[0094] In some embodiments of this application, the process of comparing each second matrix with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction includes the following steps:

[0095] S401. For each target well in the first matrix, obtain its corresponding first sequencing sequence in the first matrix.

[0096] S402. Determine the second matrix element that has the same matrix position coordinates as the target well in the first matrix, and obtain the sequencing sequence of the adjacent well corresponding to the element as the second sequencing sequence.

[0097] S403. Compare the first sequencing sequence with the second sequencing sequence, and count the number of bases of the same type at the same position in the two sequencing sequences. This count is used as the sequence similarity of the target pore in the corresponding target direction.

[0098] In step S401, all well positions in the first matrix are traversed, and each well position is taken as a target well position to be evaluated. For each target well position, the sequencing sequence stored at that position is obtained from the first matrix according to its position coordinates (i,j), and is denoted as the first sequencing sequence. The first sequencing sequence is the raw sequence directly read from that well position on the sequencing chip, usually represented as a string, composed of four base characters: A, T, C, and G. The sequence length depends on the number of sequencing rounds or the read length. For example, if 100 sequencing rounds are performed, the length of each sequencing sequence is 100 bases. For positions in the first matrix filled with empty values ​​(i.e., positions where there are no actual well positions), these positions do not participate in the subsequent similarity calculation and can be skipped directly.

[0099] In step S402, for the same target well location, the sequencing sequence at its corresponding position in the second matrix is ​​obtained. Specifically, based on the position coordinates (i,j) of the target well location in the first matrix, the element with the same coordinates (i,j) in the second matrix is ​​located, and the sequencing sequence stored in this element is the second sequencing sequence. It should be noted that since the second matrix is ​​obtained by translating the first matrix, the element at position (i,j) in the second matrix actually corresponds to the sequencing sequence of the adjacent well location in the first matrix in the opposite direction of the translation. Figure 4 Taking the upper-right translation as an example, in the second matrix obtained after the translation, the element at position (i,j) corresponds to the element at position (i-1,j-1) in the first matrix, which is the sequencing sequence of the well adjacent to the target well position (i,j) to the lower left. Therefore, through this correspondence, the second sequencing sequence is essentially the sequencing sequence of the well adjacent to the target well position in the target direction. For positions in the second matrix filled with empty values ​​(e.g., where the edge region cannot correspond to a valid well position after translation), the second sequencing sequence corresponding to that position is empty and can be ignored or specially processed in subsequent similarity calculations.

[0100] Step S403, after obtaining the first and second sequencing sequences, performs a position-by-position alignment of these two sequences, counting the number of bases with the same type at the same position. The specific alignment method is as follows: For each position k (k=1,2,...,L, where L is the sequence length), compare the base at position k of the first sequencing sequence with the base at position k of the second sequencing sequence. If both sequences have the same base (A, T, C, G) at that position, the position is considered a match, and the match count is incremented by 1; if the bases are different, or either sequence has a null value or invalid data at that position, the position is not a match. After traversing all positions, the total number of matches is the similarity between the two sequences. This similarity is an integer between 0 and L; a larger value indicates greater similarity between the two sequences.

[0101] like Figure 6 As shown, taking a sequencing sequence of length 10 as an example, the first sequencing sequence is "ACTTACGTAG" and the second sequencing sequence is "CCTTAGGTAG". Position-by-position alignment reveals that positions 1 to 6 are inconsistent, while the other positions are consistent. Therefore, the number of identical bases is 8, meaning the sequence similarity is expressed as a similarity score, which can be represented as 8. In practical applications, sequence lengths are typically hundreds or even thousands of bases, and the similarity calculation result is the number of matching bases within that length range.

[0102] For the second matrix corresponding to each target direction, repeat the above steps to obtain the sequence similarity of each hole position in that direction. If the set of target directions contains multiple directions, each hole position will obtain multiple similarity values, corresponding to adjacent holes in different directions. For example... Figure 7 As shown, for the score matrix (i.e., similarity matrix) corresponding to the lower left direction, each element in the matrix represents the sequence similarity between the corresponding hole and its adjacent hole to the lower left. Through this score matrix, the similarity distribution of each hole on the chip in a specific direction can be visually observed.

[0103] In practical sequencing applications, the sequence lengths obtained from each well may be the same (e.g., using the same number of sequencing epochs) or slightly different due to variations in the sequencing process. For inconsistent sequence lengths, the following approaches can be taken: As an optional implementation, sequence segments of the same length can be uniformly truncated for comparison, such as taking the first N bases of all sequences or the minimum effective length. Another optional implementation is to calculate only the common effective overlap between the two sequences during alignment, expressing the similarity as the proportion of matches to the overlap length, or directly using the number of matches as the similarity score, while adjusting the judgment threshold based on the actual overlap length in subsequent determinations.

[0104] For positions in the first or second matrix that are filled with empty values, these positions are not included in the similarity calculation because there are no valid sequencing sequences. In practice, when traversing the matrix, you can first check whether the current position is empty. If it is empty, skip it, or set its similarity to a default value (such as 0 or -1) to indicate that it is invalid.

[0105] In this embodiment, by acquiring the sequencing sequences at the same positions in the first and second matrices and performing position-by-position alignment, the precise calculation of the sequence similarity between each well position and its adjacent well positions in a specific direction is achieved. This alignment method based on matrix position mapping fully utilizes the correspondence established by translation operations, enabling efficient and accurate acquisition of the similarity distribution of all well positions across the entire chip, providing a reliable data foundation for subsequent cross-well identification.

[0106] In some embodiments of this application, the process of identifying cross-pore behavior based on sequence similarity may include:

[0107] S501. Based on sequence similarity, count the number of holes with cross-hole behavior in each target direction.

[0108] S502. Based on the number of pores spanning each target direction, determine the pore recognition result of the sequencing chip.

[0109] In step S501, for each target direction, based on the sequence similarity of each hole position in that direction, it is identified which holes exhibit cross-hole behavior and the number is counted. Specifically, for the similarity matrix corresponding to the target direction d (e.g., ... Figure 7 (As shown in the scoring matrix), traverse each valid hole location in the matrix and obtain the sequence similarity value S of that hole location in direction d. d(i,j) Then, based on the similarity value, it is determined whether the hole location exhibits cross-hole behavior in direction d. The determination is based on comparing the similarity with a preset judgment condition. When the similarity meets the judgment condition, it is considered that the hole location exhibits cross-hole behavior in that direction.

[0110] For each target direction, count the number of holes identified as exhibiting cross-hole behavior, denoted as N. d For example, for the three target directions of a cellular array, N can be obtained. 右上 N 右 N 右下 Three statistical values. For example... Figure 8As shown, taking the lower left direction as an example, each element in the scoring matrix represents the sequence similarity of the corresponding hole position. By comparing each similarity with the judgment criteria, the statistical result of the number of points crossing holes in that direction can be obtained. It should be noted that the specific judgment criteria used to determine the behavior of crossing holes in this step (such as threshold setting methods, comparison rules, etc.) can be flexibly determined according to actual application needs; this embodiment does not impose specific limitations on this. Subsequent embodiments will provide several optional judgment criteria and their specific implementation methods.

[0111] Then, step S502 is executed. After obtaining the number of vias in each target direction, this step performs a comprehensive analysis of these numbers to obtain an identification result that can characterize the overall via coverage of the chip. Specifically, the number of vias obtained from each target direction can be integrated to form a comprehensive index. The integration method can be flexibly selected according to actual application needs. As an optional implementation method, the number of vias in each direction can be directly output as a multi-dimensional identification result to analyze the distribution characteristics of via coverage in different directions. For example, if the number of vias in a certain direction is significantly higher than in other directions, it may indicate that the process parameters in that direction need to be optimized. As another optional implementation method, the number of vias in each direction can be summed to obtain a total score, which serves as a quantitative index of the overall via coverage of the chip. The summation formula is as follows:

[0112] S total = N d1 + N d2 + ... + N dk

[0113] Among them, S total N represents the total number of cross-hole positions, where d1, d2, ..., dk represent the first target direction, the second target direction, ..., the kth target direction, respectively, and k is the total number of target directions. d1 This indicates the number of cross-hole positions in the first target direction; the others are similar.

[0114] As another optional implementation, the proportion of vias in each direction to the total number of valid vias in that direction can be calculated to obtain the via occurrence rate. Then, a weighted average of the occurrence rates in each direction can be performed to obtain the overall via occurrence rate index. The weighting coefficient can be determined based on the importance of each direction or the density of via distribution. As yet another optional implementation, the number of vias in each direction can be compared with a preset threshold to output a pass / fail judgment for the chip. For example, if the number of vias in any direction exceeds a certain upper limit, or the total score exceeds a certain upper limit, the chip is determined to have a serious via problem. The via identification result can be expressed in numerical form (e.g., total score, occurrence rate), grade (e.g., A, B, C), or status indicator (e.g., pass / fail), which can be determined according to the application scenario and user needs.

[0115] In this embodiment, by counting the number of vias in each target direction and combining these numbers to determine the overall via identification result of the chip, the massive amount of via similarity data is transformed into an intuitive and quantitative evaluation index. This index can reflect the distribution characteristics of via phenomena in different directions and characterize the degree of vias in the chip as a whole, providing a reliable decision-making basis for chip process optimization, quality control and data analysis.

[0116] In some embodiments of this application, the process of counting the number of holes spanning holes in each target direction based on sequence similarity may include the following steps:

[0117] S601. Compare the sequence similarity determined for each hole location in each target direction with a preset cross-hole determination threshold.

[0118] S602. If the sequence similarity is greater than or equal to the cross-hole determination threshold, it is determined that there is a cross-hole at the hole location in the corresponding target direction.

[0119] S603. Count the number of holes identified as having cross-holes in each target direction.

[0120] In step S601, for each target direction d, the similarity matrix is ​​traversed to obtain the sequence similarity value S of each valid hole position (i,j) in the matrix. d(i,j) Simultaneously, a pre-set threshold T for determining cross-holes is obtained. This threshold is a numerical value used to distinguish between random similarity between normally adjacent hole positions and high similarity caused by cross-holes. Then, each similarity value S is... d(i,j) Compare with the threshold T. The comparison result has two possibilities: S d(i,j) ≥ T or S d(i,j) < T.

[0121] In step S602, based on the comparison results, for those satisfying S... d(i,j) For pores with a sequence similarity ≥ T, it is determined that the pore exhibits cross-pore behavior in direction d. This means that the sequence similarity between this pore and its adjacent pores in direction d has reached a level higher than randomness, which is highly likely due to the migration of template molecules from one pore to an adjacent pore. For pores satisfying S d(i,j) If the hole position is less than T, it is determined that the hole position does not have cross-hole behavior in direction d, and the sequence similarity between it and its adjacent hole positions is within the normal random range.

[0122] After determining all hole positions in the target direction, step S603 is executed to perform statistics for each target direction, calculating the total number of hole positions determined to have cross-holes in that direction. The specific statistical method is as follows:

[0123] For the target direction d, count all cases that satisfy S d(i,j) The number of hole positions (i,j) ≥ T is denoted as N. d This statistic reflects the number of holes that occur across the borehole in direction d. For example... Figure 8 As shown, taking the lower left direction as an example, each element in the score matrix represents the sequence similarity of the corresponding hole. After comparing each similarity with a threshold, holes with similarity reaching or exceeding the threshold are determined to have cross-holes. Counting these numbers yields the number of cross-hole points in that direction. For multiple target directions, N is obtained respectively. d1 N d2 ... N dk Multiple statistical values ​​are collected for use in subsequent steps.

[0124] The threshold for determining cross-well overlap can be set based on empirical values. For example, for a 100bp sequencing sequence, a threshold of 80 can be set, meaning that when two sequences have more than 80 identical bases, cross-well overlap is considered to have occurred. This method is simple, intuitive, and easy to implement, suitable for applications with relatively fixed sequence lengths. Alternatively, the threshold can be calculated based on the sequence length and a preset threshold ratio: T = p × L, where L is the sequence length and p is the preset threshold ratio. This method is adaptive to the sequence length and suitable for scenarios with different sequencing read lengths. A statistical distribution method can also be used to determine the threshold. For example, the similarity between all wells on the chip and their adjacent wells can be statistically analyzed to obtain a similarity distribution curve. Then, a value far from the main distribution can be selected as the threshold, such as the 95th or 99th percentile. This method is based on actual data distribution and can adaptively identify abnormally high similarity cases. Furthermore, the threshold can be dynamically adjusted by combining multiple factors. For example, considering that the aperture positions in different regions may have different background similarity levels, different thresholds can be set for different regions; or considering the systematic differences between different batches of chips, the thresholds can be calibrated between batches.

[0125] Furthermore, in some embodiments of this application, the method for determining the cross-pore determination threshold includes: calculating the cross-pore determination threshold based on the length of the sequencing sequence and a preset cross-pore determination ratio; wherein the cross-pore determination ratio is greater than the expected ratio of base matching between two sequencing sequences under random conditions.

[0126] In a sequencing chip, the sequencing sequence at each well consists of four bases: A, T, C, and G. Under normal circumstances, sequences from two different wells originate from different template molecules, and their base compositions should be independent. Therefore, the probability of two random sequences having identical bases at the same position is 1 / 4 (assuming a uniform distribution of the four bases), and this probability constitutes the expected matching ratio in a random case. When cross-well matching occurs, sequences in adjacent wells may originate from the amplification of the same template molecule, leading to a similarity between the two sequences that is much higher than the random expectation. Therefore, by setting a judgment ratio higher than the random expectation, the abnormally high similarity caused by cross-well matching can be effectively distinguished from normal random similarity.

[0127] In this embodiment, the cross-pore determination threshold T is calculated based on the length L of the sequencing sequence and the preset cross-pore determination ratio p, using the formula T = p × L; where L is the length of the sequencing sequence, for example, L = 100 when the number of sequencing rounds is 100; p is the preset cross-pore determination ratio, a value between 0 and 1, and p is greater than the expected base matching ratio of two sequencing sequences under random conditions. The expected base matching ratio under random conditions is 1 / 4 = 0.25, therefore the value of p should be greater than 0.25, and is usually set above 0.7 to ensure the reliability of the determination.

[0128] For example, if the sequencing sequence length L = 100 bp and the preset cross-pore detection ratio p = 0.8, then the cross-pore detection threshold is calculated as follows: T = 0.8 × 100 = 80; that is, when two 100 bp sequences have more than 80 identical bases at the same position, cross-pore behavior is considered to exist. As another example, if the sequencing sequence length L = 150 bp and the preset cross-pore detection ratio p = 0.75, then the cross-pore detection threshold is: T = 0.75 × 150 = 112.5; since the similarity is an integer, it can be rounded down to 112 or rounded up to 113, depending on the actual needs. It is generally recommended to use rounding down to maintain the strictness of the detection.

[0129] It should be noted that,

[0130] In this embodiment, the value of the cross-hole determination ratio p needs to satisfy the following condition: p > p random , where p random This represents the expected proportion of base matches between two sequencing sequences under random conditions. For the case of a uniform distribution of the four bases, p random =1 / 4 = 0.25. However, in practical applications, due to factors such as potentially uneven base distribution and sequencing errors, p randomThe actual value of p may deviate slightly, so the value of p is usually much greater than 0.25 to ensure the reliability of the judgment. As a preferred implementation, the value of p is between 0.7 and 0.9. This range can effectively distinguish abnormally high similarity caused by cross-pores without causing missed judgments due to an excessively high threshold. For example, in a preferred embodiment of this application, the cross-pore judgment ratio is not less than 80%, that is, p ≥ 0.8. That is, when the number of identical bases at the same position in two sequencing sequences reaches 80% or more of the sequence length, cross-pore behavior is judged to exist. For example, for a sequence with a length of 100 bp, a similarity of 80 or more is judged as cross-pore; for a sequence with a length of 200 bp, a similarity of 160 or more is judged as cross-pore.

[0131] The selection of 80% as the threshold for determining cross-pore similarity stems from the probability distribution analysis of the similarity between two sequences under random conditions. According to the binomial distribution principle, the number X of identical bases at the same position in two random sequences follows a binomial distribution:

[0132] X ~ B(n, p)

[0133] Where n is the sequence length (number of trials), and p is the probability of success in each trial (the probability of having the same base). In the case of a uniform distribution of the four bases, p = 1 / 4 = 0.25.

[0134] like Figure 10 As shown, the horizontal axis represents the number of identical bases, and the vertical axis represents the probability density. The graph clearly shows that the distribution of the number of identical bases in a random case is concentrated around 25, with a rapid decline at the tail. When the number of identical bases reaches 40, the probability is close to 0; when it reaches 80, the probability is negligible. This probability distribution provides a clear statistical basis for choosing 80% as the judgment ratio in this embodiment. Taking a sequence length of n=100 as an example, a probability distribution graph for the number of identical bases is plotted. From... Figure 10 As can be seen from this: Under random conditions, the expected number of identical bases in two sequences is n×p = 100×0.25 = 25; the probability distribution is concentrated around the expected value, and the probability of having more than 40 identical bases is already very small; the probability of having more than 50 identical bases is close to 0. Specifically, this conclusion can be quantified by calculating the cumulative probability of the binomial distribution.

[0135] .

[0136] For n=100 and p=0.25, the calculation yields:

[0137] P(X ≥ 40) ≈ 0.0006, meaning the probability of having more than 40 identical bases is only 0.06%.

[0138] P(X ≥ 50) ≈ 1.2×10^(-7), which is almost 0;

[0139] P(X ≥ 80) is even more negligible and can be ignored.

[0140] This probabilistic analysis indicates that, under completely random conditions, two 100bp sequences generally will not have more than 40 identical bases, and the possibility of having more than 80 is almost nonexistent. Therefore, when the observed similarity reaches 80% or more, it can be concluded that these two sequences did not originate from independent random templates, but are very likely derived from the amplification of the same template molecule, i.e., transpore amplification occurred.

[0141] In this embodiment, 80% is a preferred value. However, in practical applications, the cross-hole determination ratio can be adjusted according to specific needs: if it is desired to improve detection sensitivity and reduce false negatives, the determination ratio can be appropriately reduced, for example, set to 75% or 70%. This can capture more suspected cross-hole cases, but may introduce a small number of false positives. If it is desired to improve the accuracy of determination and reduce false positives, the determination ratio can be appropriately increased, for example, set to 85% or 90%. This can ensure that the cases determined to be cross-holes have a very high confidence level, but may miss some cross-hole events with slightly lower similarity.

[0142] In some embodiments of this application, the cross-pore identification result of the sequencing chip is determined based on the number of cross-pore positions in each target direction, including: summing the statistically obtained number of cross-pore positions in each target direction to obtain a total cross-pore evaluation score, and determining the total cross-pore evaluation score as the cross-pore identification result.

[0143] like Figure 9 As shown, for the three target directions of the cellular array, the number of cross-hole points in the lower left direction, the left-hand direction, and the upper left direction (or the three directions selected in practice) are summed to obtain the cumulative number of cross-hole points, which is the total score for cross-hole evaluation.

[0144] For example, suppose a sequencing chip uses a honeycomb array, and the target direction set includes three directions: upper right, right, and lower right. After cross-hole determination and statistics, the number of cross-hole positions in each direction is as follows: 156 cross-hole positions in the upper right direction, 142 cross-hole positions in the right direction, and 138 cross-hole positions in the lower right direction. The total cross-hole evaluation score is calculated as: 156 + 142 + 138 = 436. This total score of 436 is the cross-hole identification result output by this embodiment, reflecting the quantitative value of the overall cross-hole degree of the chip.

[0145] The total score for via cross-via evaluation is an absolute value, and its magnitude is related to factors such as the total number of vias on the chip, the via array structure, and the number of target directions. Generally speaking, the higher the total score, the more vias cross-via occur on the chip, the more serious the via cross-via problem, and the worse the amplification effect; the lower the total score, the fewer via cross-via phenomena on the chip, and the better the amplification effect.

[0146] Furthermore, this method also includes: comparing the total score of the transpore evaluation with a quality threshold, and outputting a quality level judgment of the sequencing chip based on the comparison results.

[0147] Obtain the pre-set quality threshold. A quality threshold is a critical value used to divide different quality levels, and can be determined based on actual application needs, historical data statistics, or industry standards. There can be one or more quality thresholds. When there is only one quality threshold, it can be used to distinguish between acceptable and unacceptable levels; when there are multiple quality thresholds, they can be used to divide multiple quality levels, such as excellent, good, average, and poor. For example, for a certain type of sequencing chip, based on statistical analysis of historical production data and actual application effects, the following quality thresholds can be set: Acceptable threshold: DupScore ≤ 500 is acceptable, DupScore > 500 is unacceptable.

[0148] Alternatively, multiple quality level thresholds can be set: Excellent: DupScore ≤ 300; Good: 300 < DupScore ≤ 500; Average: 500 < DupScore ≤ 800; Poor: DupScore > 800. The calculated total score for the cross-hole evaluation is compared with these quality thresholds to determine which threshold range the total score falls into. Based on the comparison results, the corresponding quality level judgment is output. The quality level can be represented by textual descriptions (e.g., "Pass," "Fail"), level symbols (e.g., "Grade A," "Grade B," "Grade C"), color codes (e.g., "Green," "Yellow," "Red"), or any other form that is easy to understand and use.

[0149] In the embodiments of this application, the quality threshold can be determined in various ways. Several optional determination methods are provided below:

[0150] As an optional implementation, the threshold can be determined based on statistical data of the total cross-pore evaluation score of historical batches of chips. For example, the total cross-pore evaluation score of all qualified chips produced over a period of time can be collected, and its mean and standard deviation can be calculated. The mean plus a certain multiple of the standard deviation can be used as the pass threshold. This method, based on actual production data, can reflect the normal fluctuation range under the current process level. Another optional implementation can be determined based on the expected performance indicators of the chip. For example, based on the chip design goals, the total cross-pore evaluation score can be set to not exceed a certain proportion of the total number of pores on the chip, and the value corresponding to this proportion can be used as the pass threshold. Yet another optional implementation can be determined based on the needs of downstream applications. For example, some high-precision sequencing applications are particularly sensitive to cross-pore issues, and a stricter quality threshold can be set; while for some applications where throughput requirements are higher than accuracy, the quality threshold can be appropriately relaxed. Yet another optional implementation can be determined through experimental verification. For example, a series of chips with different cross-pore degrees can be prepared, and their performance can be verified through actual sequencing. The maximum total cross-pore evaluation score that meets the performance requirements can be found, and this value can be used as the pass threshold. In practical applications, the quality threshold is not static. As technology levels improve and application requirements change, quality thresholds can be dynamically adjusted.

[0151] In some embodiments of this application, the method further includes: generating process adjustment suggestions for sequencing chips based on the cross-pore identification results; wherein the process parameters include at least one of the following: pore spacing, pore size, biochemical reaction conditions, and algorithm.

[0152] Correspondingly, the results of cross-well evaluation can provide quantitative guidance for the synergistic optimization of chip design and biochemical reaction systems. In terms of chip structure design, by analyzing the correlation between cross-well degree and parameters such as pore spacing and pore size, the physical structure design of the chip can be optimized, effectively suppressing cross-well diffusion of template molecules while ensuring throughput. In terms of biochemical reaction system optimization, cross-well identification results can be used to evaluate the impact of different amplification methods, enzyme combinations, and reaction conditions on cross-well phenomena, guiding improvements to the amplification system. For example, adjusting the number of amplification cycles, optimizing polymerase concentration, or improving the amplification program can reduce the likelihood of template molecules migrating to adjacent wells. Through this joint optimization of chip structure and biochemical reaction conditions, the amplification effect and signal recognition accuracy of the sequencing process can be comprehensively improved.

[0153] After obtaining the cross-pore identification results of the sequencing chip, these results can be used not only for quality assessment but also as feedback information to guide the optimization of the chip fabrication process. When the cross-pore identification results show that the chip has a serious cross-pore problem, it is necessary to adjust the relevant process parameters to reduce the cross-pore occurrence rate of subsequent batches of chips. In this embodiment, based on the cross-pore identification results, process parameter adjustment suggestions for the sequencing chip are generated. Specifically, the cross-pore identification results (such as the total cross-pore evaluation score, the number of cross-pores in each direction, etc.) are matched with preset process parameter adjustment rules, and corresponding adjustment suggestions are generated based on the matching results. The process parameter adjustment rules can be pre-established based on theoretical analysis, experimental data, or experience. For example, a mapping relationship between the total cross-pore evaluation score and the adjustment amount of each process parameter can be established. When the total cross-pore evaluation score exceeds a certain threshold, it is recommended to adjust the corresponding parameter by a specific amount.

[0154] According to this embodiment, the process parameters include at least one of the following:

[0155] Hole Spacing: When the cross-hole recognition results show a high cross-hole incidence rate, it is recommended to increase the hole spacing to reduce the possibility of template molecules migrating from adjacent pores. When the cross-hole incidence rate is low and it is desired to increase chip throughput, it is recommended to appropriately reduce the hole spacing. Hole Size: Hole size affects the spatial confinement and molecular diffusion of the amplification reaction. Based on the cross-hole recognition results, it is recommended to adjust the diameter or depth of the holes. For example, if the number of cross-holes in a specific direction is significantly high, it may indicate an anisotropy problem in the hole shape design in that direction. Biochemical Reaction Conditions: These include amplification temperature, reaction time, reagent concentration, etc. If the cross-hole recognition results indicate a high overall cross-hole degree, it is recommended to reduce the number of amplification cycles, adjust the annealing temperature, or reduce the polymerase concentration to reduce the possibility of amplification products diffusing outward. Algorithm Parameters: These include cross-hole determination threshold, quality grading threshold, etc. Based on application feedback from the cross-hole recognition results, it is recommended to adjust these algorithm parameters to better adapt to the characteristics of different batches of chips or the needs of different application scenarios.

[0156] In this embodiment, the generated process parameter adjustment suggestions can be output in the form of text descriptions, numerical adjustment amounts, or visual charts, providing them to chip design engineers, process engineers, or production managers as a decision-making reference. In actual production, the adjustment suggestions can be integrated into the automated production control system to achieve automatic optimization and adjustment of process parameters.

[0157] In some embodiments of this application, the method further includes: comparing and analyzing the cross-pore identification results with the cross-pore identification results of historical batches of chips; and monitoring the stability of the sequencing chip manufacturing process based on the comparison and analysis results.

[0158] In the mass production of sequencing chips, continuous monitoring of the manufacturing process stability is necessary to promptly identify potential anomalies. Cross-pore identification results, as a sensitive process quality indicator, can be used for this monitoring purpose. First, a database of cross-pore identification results from historical batches of chips can be established. For each batch of chips produced, its cross-pore identification results are recorded, including the total cross-pore evaluation score, the number of cross-pores in each direction, and the cross-pore occurrence rate, along with corresponding production time, process parameters, and production environment information. The accumulation of historical batch data provides a benchmark and reference for subsequent comparative analysis. The cross-pore identification results of the current batch of chips are then compared with those of historical batches. For example, a statistical distribution comparison can be used, comparing the results of the current batch with the statistical distribution of historical batches. The mean μ and standard deviation σ of the total cross-pore evaluation scores of historical batches are calculated, and the result x of the current batch is compared with the range of μ ± kσ. If x exceeds this range (e.g., k=3, i.e., exceeding the 3σ range), the current batch is considered to have a significant anomaly, triggering an alert. Trend analysis can also be used for comparative analysis, such as performing trend analysis on the cross-hole identification results of consecutive batches to identify whether there is systematic drift. If multiple batches show a monotonous upward or downward trend, even if each individual batch may be within the normal range, the overall trend still suggests that the production process may have changed slowly.

[0159] Based on the results of comparative analysis, the stability of the sequencing chip manufacturing process is monitored. For example, it can periodically generate process stability reports, including comparison charts between the current batch and historical batches, trend analysis results, etc. It can also automatically trigger early warning notifications when significant anomalies are detected, reminding relevant personnel to pay attention and investigate the cause. Furthermore, it can issue trend alerts when continuous trend changes are detected, suggesting preventative adjustments to relevant process parameters. Through this historical batch comparison and continuous monitoring mechanism, abnormal fluctuations in the manufacturing process can be detected in a timely manner, ensuring the stability and consistency of product quality.

[0160] This application also provides a perforation recognition device, see [link to relevant documentation]. Figure 12 The device includes:

[0161] The acquisition unit 11 is used to obtain a first matrix corresponding to each well position in the sequencing chip, wherein the elements of the first matrix correspond to the well positions of the sequencing chip, and the value of the element represents the sequencing sequence of the corresponding well position.

[0162] Translation unit 12 is used to translate the first matrix along at least one target direction to obtain at least one second matrix; wherein the translation is used to make the elements at each position in the second matrix correspond to the sequencing sequences of the adjacent pores at that position in the first matrix in the target direction;

[0163] Comparison unit 13 is used to compare each of the second matrices with the first matrix to determine the sequence similarity between each hole position and its adjacent hole positions in the corresponding target direction;

[0164] The identification unit 14 is used to identify cross-hole behavior based on the sequence similarity.

[0165] In one possible implementation, the acquisition unit includes:

[0166] The framework construction subunit is used to construct an initial matrix framework based on the actual well arrangement characteristics of the sequencing chip, wherein the initial matrix framework has a position that corresponds one-to-one with the actual well position;

[0167] The filling subunit is used to fill the sequencing sequences of each well of the sequencing chip into the corresponding positions of the initial matrix frame to obtain the first matrix.

[0168] In one possible implementation, the framework building subunit is specifically used for:

[0169] In the case where the actual hole arrangement features are the first type of arrangement that is directly mapped to a matrix, the matrix frame with the same row and column dimensions is determined as the initial matrix frame.

[0170] When the actual hole arrangement is a second type of arrangement that cannot be directly mapped to a matrix but has a regular geometric structure, an initial matrix architecture capable of accommodating all holes is constructed, wherein the initial matrix architecture sets null value markers at positions where no actual holes exist.

[0171] Furthermore, the first type of arrangement includes a rectangular array or a square array; the second type of arrangement includes a honeycomb array or a triangular array.

[0172] In one possible implementation, the translation unit includes:

[0173] The orientation determination subunit is used to determine all adjacent orientations between pores based on the pore array structure of the sequencing chip;

[0174] The direction selection subunit is used to select a subset of the all adjacent directions that satisfy a first condition as the target direction set, wherein the first condition indicates that for any direction in the subset, its opposite direction is not in the subset.

[0175] The translation execution subunit is used to perform a global translation operation on the first matrix along each direction in the target direction set to obtain a second matrix corresponding to the target direction.

[0176] In one possible implementation, for a cellular array structure, the number of all adjacent directions is six, and the number of directions in the target direction set selected by the direction selection sub-unit is three.

[0177] In one possible implementation, the comparison unit includes:

[0178] The first sequence acquisition subunit is used to acquire the first sequencing sequence corresponding to each target pore in the first matrix.

[0179] The second sequence acquisition subunit is used to determine the second matrix element that has the same matrix position coordinates as the target pore in the first matrix, and to acquire the sequencing sequence of the adjacent pore corresponding to the element as the second sequencing sequence.

[0180] The similarity calculation subunit is used to compare the first sequencing sequence with the second sequencing sequence, and count the number of bases of the same type at the same position in the two sequencing sequences, which is used as the sequence similarity of the target pore in the corresponding target direction.

[0181] In one possible implementation, the identification unit includes:

[0182] The statistical subunit is used to count the number of holes exhibiting the cross-hole behavior in each of the target directions based on the sequence similarity.

[0183] The result determination subunit is used to determine the pore recognition result of the sequencing chip based on the number of pores that span the pore in each of the target directions.

[0184] In one possible implementation, the statistical subunit is specifically configured as follows:

[0185] The sequence similarity of each hole location in each target direction is compared with a preset cross-hole determination threshold.

[0186] If the sequence similarity is greater than or equal to the cross-hole determination threshold, then it is determined that there is a cross-hole at the hole location in the corresponding target direction;

[0187] Count the number of holes that are determined to exist across holes in each target direction.

[0188] In one possible implementation, the cross-pore determination threshold is calculated based on the length of the sequencing sequence and a preset cross-pore determination ratio; wherein the cross-pore determination ratio is greater than the expected base matching ratio of two sequencing sequences under random conditions.

[0189] Preferably, the cross-hole determination ratio is not less than 80%.

[0190] In one possible implementation, the result determination subunit is specifically used to: sum the number of holes with cross-holes in each of the target directions obtained by statistics to obtain a total cross-hole evaluation score, and determine the total cross-hole evaluation score as the cross-hole identification result.

[0191] In one possible implementation, the identification unit further includes:

[0192] The quality judgment subunit is used to compare the total score of the transpore evaluation with the quality threshold, and output the quality level judgment of the sequencing chip based on the comparison result.

[0193] In one possible implementation, the identification unit further includes:

[0194] The process suggestion subunit is used to generate process parameter adjustment suggestions for the sequencing chip based on the cross-pore identification results; wherein the process parameters include at least one of the following: pore spacing, pore size, biochemical reaction conditions, and algorithm.

[0195] In one possible implementation, the identification unit further includes:

[0196] The historical comparison subunit is used to compare and analyze the cross-hole identification results with the cross-hole identification results of historical batches of chips;

[0197] The monitoring subunit is used to monitor the stability of the manufacturing process of the sequencing chip based on the results of the comparison analysis.

[0198] It should be noted that the specific implementation of each unit and subunit in this embodiment can be referred to the corresponding content above, and will not be described in detail here.

[0199] This application embodiment also provides a storage medium, characterized in that the storage medium stores a computer program, which, when executed by a processor, implements the cross-hole recognition method as described above.

[0200] In another embodiment of this application, an electronic device is also provided, comprising:

[0201] Memory, used to store computer programs;

[0202] A processor is configured to execute a computer program stored in the memory, which, when executed by the processor, implements the trans-hole recognition method as described above.

[0203] It should be noted that the specific implementation of the processor in this embodiment can be referred to the corresponding content above, and will not be described in detail here.

[0204] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0205] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0206] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0207] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying perforated holes, characterized in that, include: A first matrix corresponding to each well position in the sequencing chip is obtained, wherein the elements of the first matrix correspond to the well positions of the sequencing chip, and the value of the element represents the sequencing sequence of the corresponding well position; The first matrix is ​​translated along at least one target direction to obtain at least one second matrix; wherein the translation is such that each element in the second matrix corresponds to the sequencing sequence of the adjacent pore in the first matrix along the target direction at that position; Each second matrix is ​​compared with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction; Based on the sequence similarity, cross-pore behavior is identified.

2. The method according to claim 1, characterized in that, The first matrix corresponding to each well position in the sequencing chip is obtained, including: Based on the actual well arrangement characteristics of the sequencing chip, an initial matrix framework is constructed, wherein the initial matrix framework has a position that corresponds one-to-one with the actual well position; The sequencing sequences of each well of the sequencing chip are filled into the corresponding positions of the initial matrix frame to obtain the first matrix; Optionally, the initial matrix framework constructed based on the actual well arrangement characteristics of the sequencing chip includes: In the case where the actual hole arrangement features are the first type of arrangement that is directly mapped to a matrix, the matrix frame with the same row and column dimensions is determined as the initial matrix frame. In the case where the actual hole arrangement is a second type of arrangement that cannot be directly mapped to a matrix but has a regular geometric structure, an initial matrix architecture that can accommodate all hole positions is constructed, wherein the initial matrix architecture sets null value markers at positions where no actual hole positions exist. Optionally, the first type of arrangement includes a rectangular array or a square array; the second type of arrangement includes a honeycomb array or a triangular array.

3. The method according to claim 1 or 2, characterized in that, The step of translating the first matrix along at least one target direction to obtain at least one second matrix includes: Based on the pore array structure of the sequencing chip, all adjacent directions between pores are determined; From all the adjacent directions, a subset that satisfies the first condition is selected as the target direction set, wherein the first condition indicates that for any direction in the subset, its opposite direction is not in the subset; Perform a global translation operation on the first matrix along each of the target direction sets to obtain a second matrix corresponding to the target direction; Optionally, the aperture array structure is a honeycomb array structure, the number of all adjacent directions is six, and the number of directions in the target direction set is three.

4. The method according to any one of claims 1-3, characterized in that, The step of comparing each of the second matrices with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction includes: For each target well in the first matrix, obtain its corresponding first sequencing sequence in the first matrix; Determine a second matrix element that has the same matrix position coordinates as the target pore in the first matrix, and obtain the sequencing sequence of the adjacent pore corresponding to this element as the second sequencing sequence; The first sequencing sequence and the second sequencing sequence are compared, and the number of bases with the same type at the same position in the two sequencing sequences is counted as the sequence similarity of the target pore in the corresponding target direction.

5. The method according to any one of claims 1-4, characterized in that, The identification of cross-pore behavior based on the sequence similarity includes: Based on the sequence similarity, the number of holes exhibiting the cross-hole behavior in each of the target directions is counted; The pore recognition result of the sequencing chip is determined based on the number of pores spanning each target direction. Optionally, the step of counting the number of holes spanning holes in each of the target directions based on the sequence similarity includes: The sequence similarity of each hole location in each target direction is compared with a preset cross-hole determination threshold. If the sequence similarity is greater than or equal to the cross-hole determination threshold, then it is determined that there is a cross-hole at the hole location in the corresponding target direction; Count the number of holes that are determined to exist across holes in each target direction; Optionally, the method for determining the cross-hole determination threshold includes: The cross-pore determination threshold is calculated based on the length of the sequencing sequence and the preset cross-pore determination ratio. Wherein, the cross-pore determination ratio is greater than the expected ratio of base matching between two sequencing sequences under random conditions; Optionally, the cross-hole determination ratio is not less than 80%.

6. The method according to claim 5, characterized in that, The determination of the pore identification result of the sequencing chip based on the number of pores spanning each of the target directions includes: The total number of holes that cross the hole in each of the target directions obtained by statistics is summed to obtain the total score of the cross-hole evaluation, and the total score of the cross-hole evaluation is determined as the cross-hole identification result; Optionally, the method further includes: The total score of the transpore evaluation is compared with the quality threshold, and the quality level of the sequencing chip is determined based on the comparison result.

7. The method according to claim 5 or 6, characterized in that, The method further includes: Based on the cross-pore identification results, suggestions for adjusting the process parameters of the sequencing chip are generated; The process parameters include at least one of the following: pore spacing, pore size, biochemical reaction conditions, and algorithm; Optionally, the method further includes: The via identification results are compared and analyzed with the via identification results of historical batches of chips; Based on the results of the comparison analysis, the stability of the manufacturing process of the sequencing chip is monitored.

8. A perforation recognition device, characterized in that, include: The acquisition unit is used to obtain a first matrix corresponding to each well position in the sequencing chip, wherein the elements of the first matrix correspond to the well positions of the sequencing chip, and the value of the element represents the sequencing sequence of the corresponding well position. A translation unit is used to translate the first matrix along at least one target direction to obtain at least one second matrix; wherein the translation is used to make the elements at each position in the second matrix correspond to the sequencing sequences of the adjacent pores at that position in the first matrix in the target direction; A comparison unit is used to compare each of the second matrix with the first matrix to determine the sequence similarity between each hole position and its adjacent holes in the corresponding target direction; The identification unit is used to identify cross-pore behavior based on the sequence similarity.

9. A storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the trans-aperture identification method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing a computer program stored in the memory, wherein the computer program, when executed by the processor, implements the trans-hole recognition method as described in any one of claims 1 to 7.

Images