Methods and apparatus for DNA storage coding and decoding and rules thereof

Abstract

The invention discloses a method and a device for DNA storage coding and decoding and rules thereof. The method comprises the following steps: performing single-molecule sequencing on a reference sequence to obtain actual sequencing data of single-molecule sequencing; comparing the actual sequencing data with reference data of the reference sequence, counting the frequency of sequencing errors of each sequence fragment with the length of k in the actual sequencing data, and calculating the proportion of the sequencing errors of each sequence fragment with the length of k in the actual sequencing data, namely the error rate; and taking the sequence fragments of which the error rates exceed a threshold value as limiting conditions to be eliminated. According to the method provided by the invention, the DNA storage coding and decoding steps are simplified, and the complexity of data processing is reduced through the time sequence of the threshold elimination step.

Description

Method and device for DNA storage encoding and decoding and its rules Technical Field

[0001] The present invention belongs to the field of biotechnology, and more specifically, discloses a method and device for DNA storage encoding and decoding and its rules. Background Art

[0002] With the development of modern technology, the amount of data in the world is increasing exponentially. The ever-increasing amount of data places higher and higher demands on storage technology. Traditional storage technologies, such as magnetic tape and optical disc storage, are increasingly unable to meet current data needs due to their low storage density and limited time. In recent years, the development of DNA storage technology has provided a new way to solve these problems. Compared with traditional storage media, DNA as a medium for information storage has the advantages of long storage time (up to thousands of years, more than a hundred times that of existing magnetic tape and optical disc media), high storage density (reaching ~10 9 Gb / mm 3 , which is more than 10 million times that of tape and CD media) and has high storage security.

[0003] The general workflow for DNA data storage involves: first, converting binary information into a DNA sequence through a specific information conversion process; second, physically storing the corresponding DNA sequence information by chemically or enzymatically synthesizing DNA molecules; and finally, sequencing the DNA molecules and reading the information using the aforementioned information conversion process. As the starting and end point of the DNA storage workflow, the information conversion process plays a crucial role in efficiently converting binary information into DNA sequences.

[0004] As a major drawback of DNA data storage, the process from raw sequencing data to information interpretation (i.e., the final step mentioned above) is very time-consuming and involves many steps. After obtaining the raw sequencing data, the data information is usually obtained through the following five data processing steps: (1) Clustering: The preliminarily filtered DNA sequences are classified based on a specific clustering method; (2) Alignment: Multiple DNA sequences in each class are aligned to calculate the base with the highest probability at each position in the sequence. A sequence composed of bases with the highest probability at different positions is called a high-confidence DNA sequence; (3) Decoding: The obtained high-confidence DNA sequence is decoded to obtain the corresponding bit sequence, and the DNA sequence that failed to be decoded is discarded; (4) Error correction: The decoded bit sequence is corrected based on the error correction region (including RS code or other error correction code) in its sequence, and the bit sequence that cannot be corrected is removed. After the error correction is completed, the error correction region is removed; (5) Assembly: The error-corrected bit sequence is sorted based on the index region of the bit sequence, and the data information is finally obtained.

[0005] Since the above five steps are sequential, they can only be performed step by step, but this will obviously cause a delay in the DNA storage retrieval process, thereby amplifying the defects of DNA data storage. In addition, single-molecule sequencing has been increasingly favored by the DNA data storage industry. This type of technology has the advantage of fast sequencing speed, but its high error rate is one of the important bottlenecks that limit its widespread application. The high error rate will lead to the inapplicability of clustering, alignment, and error correction processes in traditional retrieval processes. In order to merge decoding and error correction into one step, William H. Press, an academician of the U.S. Academy of Sciences, proposed in 2020 that HEDGES uses hash functions and greedy exhaustive searches to correct most index and replacement errors (Press WH, Hawkins JA, Jones Jr SK, et al. HEDGES error-correcting code for DNA storage corrects indels and allows sequence constraints [J]. Proceedings of the National Academy of Sciences, 2020, 117 (31): 18489-18496.). This method can correct errors in single-molecule sequencing with a certain probability. However, the problem caused by the principle of its method (i.e., greedy exhaustive search) is that the speed is slow (about 1 bit / second).

[0006] In summary, it is necessary to develop a retrieval process or mechanism that can interpret the raw sequencing data in real time for high error rates (for single-molecule sequencing).

[0007] Summary of the Invention

[0008] The present invention aims at least in part to solve the problem of DNA sequence repair in DNA data storage under high error rates caused by single-molecule sequencing, and to achieve end-to-end fast retrieval to achieve a throughput of GB-level decoding per line.

[0009] According to a first aspect of the present invention, the present invention provides a DNA storage encoding and decoding method for single molecule sequencing, the method comprising:

[0010] 1) Set the sliding window (k,n) for DNA storage encoding and decoding rules, where k represents the length of the sliding window and n represents the length of the base character for each slide, where n and k are positive integers and k ≥ n;

[0011] 2) Based on the length k of the sliding window, a full set of sequences is obtained, where the full set of sequences is the set of all possible random combinations of bases at each base position within the length of the sliding window;

[0012] 3) Performing single-molecule sequencing on the reference sequence to obtain actual sequencing data of single-molecule sequencing;

[0013] 4) Comparing the actual sequencing data with the reference sequence data, counting the frequency of sequencing errors for each sequence fragment of length k in the actual sequencing data, and calculating the proportion of sequencing errors for each sequence fragment of length k in the actual sequencing data, i.e., the error rate;

[0014] 5) Eliminate sequence segments whose error rates exceed a threshold as a restriction condition, and select a qualified sequence set whose error rates do not exceed the threshold from the entire set of sequences;

[0015] 6) Connecting the sequences in the above qualified sequence set through a directed graph, where each node in the directed graph represents each sequence;

[0016] 7) Obtaining an algorithm diagram, wherein the algorithm diagram includes DNA storage encoding and decoding rules.

[0017] In a preferred embodiment, the single molecule sequencing includes Nanopore single molecule sequencing and / or other related single molecule sequencing.

[0018] In a preferred embodiment, after step 6), the method includes:

[0019] i) deleting nodes in the directed graph whose out-degree number is less than a set out-degree limit;

[0020] ii) deleting redundant out-degrees of each node in the directed graph, wherein the redundant out-degrees are out-degrees exceeding a set out-degree limit.

[0021] In a preferred embodiment, the deleting of redundant out-degrees of each node in the directed graph includes: if the total number of out-degrees of the node exceeds a set out-degree limit, outputting the bases of the node in reverse order, and deleting the out-degrees pointing to the corresponding bases in sequence according to the order of the bases output in reverse order.

[0022] In a preferred embodiment, a sliding window (k, n) is set to perform collaborative sampling of sequencing data error rate and sequence characteristics using the observation scale of the sliding window for actual sequencing data; and the error ratio information of sequencing data is statistically ranked based on the principle of higher error rate, higher priority.

[0023] In a preferred embodiment, the error rate ranges from 2% to 4.5%, and preferably, the error rate is 3%.

[0024] According to a second aspect of the present invention, the present invention provides an apparatus for generating DNA storage encoding and decoding rules, comprising:

[0025] A sliding window setting unit is used to set a sliding window (k, n) for DNA storage encoding and decoding rules, where k represents the length of the sliding window and n represents the length of the base character for each sliding, where n and k are positive integers and k≥n;

[0026] a constraint determination unit for performing single-molecule sequencing on a reference sequence to obtain actual sequencing data of the single-molecule sequencing; comparing the actual sequencing data with the reference data of the reference sequence, counting the frequency of sequencing errors occurring in the actual sequencing data for each sequence fragment of length k, and calculating the proportion of sequencing errors occurring in the actual sequencing data for each sequence fragment of length k, i.e., the error rate; and using sequence fragments with an error rate exceeding a threshold as constraint conditions;

[0027] A qualified sequence screening unit is configured to obtain a complete set of sequences based on the length k of the sliding window, wherein the complete set of sequences is a set of all possible base sequences formed by random combinations of all bases at each base position within the length of the sliding window, and to screen out qualified sequences from the complete set of sequences that meet the above-mentioned constraints using a constraint condition;

[0028] a directed graph connection unit, configured to connect the sequences in the qualified sequence set via a directed graph, wherein each node in the directed graph represents each sequence;

[0029] The algorithm chart obtaining unit is used to obtain an algorithm chart, which includes DNA storage encoding and decoding rules.

[0030] In a preferred embodiment, the device for generating DNA storage encoding and decoding rules further comprises:

[0031] The out-degree violation deletion unit is used to delete the nodes in the directed graph whose out-degree number is less than the set out-degree limit;

[0032] The redundant out-degree deletion unit is used to delete the redundant out-degree of each node in the directed graph, where the redundant out-degree is the out-degree that exceeds the set out-degree limit.

[0033] According to a third aspect of the present invention, the present invention provides a DNA storage encoding method, comprising:

[0034] 1) obtaining the DNA storage encoding and decoding rules generated by the method of the first aspect, and setting an initial node, which is defined as the current node;

[0035] 2) Obtain the binary sequence to be encoded and slice it to generate binary slices. Convert the binary value corresponding to the slice into an out-degree node or multiple layers of out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid fragment. The binary slice and the corresponding nucleic acid fragment form a pair of mapping relationships.

[0036] 3) According to the DNA storage encoding and decoding rules, the binary slice is input, the nucleic acid fragment mapped to the out-degree node or multiple out-degree nodes is output, and the out-degree node is updated as the current node. The binary slices are input and the nucleic acid fragments are output in a loop according to the binary slice sequence until all the binary slices are input;

[0037] 4) Connect the above nucleic acid fragments in order according to the output order and output the complete DNA sequence.

[0038] In a preferred embodiment, the above method slices the above binary sequence to be encoded according to a length of 2n-1, where n represents the base character length of each sliding of the sliding window.

[0039] In a preferred embodiment, the method further comprises: synthesizing the DNA sequence and then preserving it in an in vitro medium or living cells.

[0040] According to a fourth aspect of the present invention, the present invention provides a DNA storage encoding device, comprising:

[0041] A coding and decoding rule acquisition unit, configured to acquire the DNA storage coding and decoding rule generated by the method of the first aspect, and set an initial node, defining the initial node as the current node;

[0042] A binary sequence slicing and conversion unit is configured to obtain a binary sequence to be encoded and slice it to generate binary slices, and convert the binary values ​​corresponding to the slices into out-degree nodes or multiple layers of out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid fragment, and the binary slices and the corresponding nucleic acid fragments form a mapping relationship.

[0043] A nucleic acid fragment output unit is configured to input the binary slices according to the DNA storage encoding and decoding rules, output the nucleic acid fragments mapped to the out-degree node or multiple out-degree nodes, update the out-degree node as the current node, and continuously input binary slices and output nucleic acid fragments in a cyclic manner according to the binary slice sequence until all the binary slices are input;

[0044] The nucleic acid fragment connection unit is used to connect the above nucleic acid fragments in sequence according to the output order and output a complete DNA sequence.

[0045] In a preferred embodiment, the binary sequence to be encoded is sliced ​​according to a length of 2n-1, where n represents the length of base characters of each sliding of the sliding window.

[0046] According to a fifth aspect of the present invention, the present invention provides a DNA storage decoding method, comprising:

[0047] 1) obtaining the DNA storage encoding and decoding rules generated by the method of the first aspect, and setting an initial node, which is defined as the current node;

[0048] 2) Obtain the single-molecule sequencing DNA sequence to be decoded and slice it to generate nucleic acid slices. Based on the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slices, find the out-degree node or multiple layers of out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid information. The nucleic acid slice and the corresponding binary value or binary slice form a mapping relationship.

[0049] 3) Based on the current node and the out-degree node or multiple out-degree nodes, obtain the binary values ​​or binary slices between the nodes according to the mapping relationship, update the out-degree node to the current node, and continuously input nucleic acid slices and output binary values ​​or binary slices in a cyclic manner according to the sequence of nucleic acid slices until all nucleic acid slices are input;

[0050] 4) Connect the above binary values ​​or binary slices in order according to the output order and output the complete binary sequence.

[0051] In a preferred embodiment, step 3) includes an error correction step, which is a path-based probabilistic error correction strategy, including:

[0052] (1) An error was found during the decoding process:

[0053] Errors are found based on constraints, i.e. the current node cannot complete the transfer to the next node or cannot meet the requirements of a valid path;

[0054] (2) Check the location where the decoding error occurred

[0055] By reverse order, we mean checking backward from the location where the error was found to a location that is the length of the observation window away from the error location;

[0056] (3) Obtaining candidate repair sequences

[0057] After finding the position where the decoding error occurred, a local exhaustive reverse search is performed to determine the error type and try the following three types of adjustments to repair it: replacing the nucleotide at the current position with another nucleotide, inserting a nucleotide between the current position and the previous position, and deleting the nucleotide at the current position; thus, candidate repair sequences are obtained;

[0058] (4) Obtaining a set of repaired DNA sequences

[0059] The candidate repaired sequences are tested by the Varshamov-Tenengolts error correction code path, and candidate repaired sequences that meet the restriction conditions and the detection sequence requirements are collected; and a set of repaired DNA sequences that meet the restriction conditions and the detection sequence requirements are output.

[0060] In a preferred embodiment, after (4), all sequences in the DNA sequence set are counted and sorted, and sequences whose frequencies reach a threshold are directly decoded as correct sequences.

[0061] In a preferred embodiment, the sequencing depth of the sequence is 20×, and the sequence frequency threshold is 4, that is, the sequence with a frequency of 4 is directly decoded as the correct sequence.

[0062] In a preferred embodiment, the above method slices the DNA sequence to be decoded according to the length n, where n represents the length of the base character of each sliding of the sliding window.

[0063] In a preferred embodiment, the DNA sequence to be decoded is generated by encoding using the method of the third aspect or the apparatus of the fourth aspect.

[0064] According to a sixth aspect of the present invention, the present invention provides a DNA storage and decoding device, comprising:

[0065] A coding and decoding rule acquisition unit, configured to acquire the DNA storage coding and decoding rule generated by the method of the first aspect, and set an initial node, defining the initial node as the current node;

[0066] A DNA slicing and conversion unit is configured to obtain a DNA sequence to be decoded and slice it to generate nucleic acid slices. Based on the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slices, the unit finds the out-degree node or multiple out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid information. The nucleic acid slices and the corresponding binary values ​​or binary slices form a mapping relationship.

[0067] a binary value output unit, configured to obtain binary values ​​or binary slices between nodes according to the mapping relationship between the current node and the out-degree node or multiple out-degree nodes, and update the out-degree node to the current node, and continuously and cyclically input nucleic acid slices and output binary values ​​or binary slices according to the sequence of nucleic acid slices until all nucleic acid slices are input;

[0068] The binary value connection unit is used to connect the above binary values ​​in sequence according to the output order and output a complete binary sequence.

[0069] In a preferred embodiment, the DNA sequence to be decoded is sliced ​​according to the length n, where n represents the length of the base character of each sliding of the sliding window.

[0070] According to a seventh aspect of the present invention, the present invention provides a computer-readable storage medium comprising a program, which can be executed by a processor to implement the method of the first aspect, the method of the third aspect, or the method of the fifth aspect.

[0071] The present invention designs a real-time repair mechanism that requires no preprocessing and no strict sequence (repair speed of up to 12,000 bits / second for a 2% editing error). The method simplifies the DNA storage encoding and decoding steps, eliminates the sequential nature of the steps through thresholding, and reduces the complexity of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the specific embodiments of the present invention, the drawings in the specific embodiments are briefly introduced below.

[0073] Figure 1 illustrates the path-based error correction strategy according to the present invention. (a) A GC-balanced directed graph of length 2 is created; using this directed graph, the binary message 001101 is encoded as the DNA sequence TCTGAC; based on this DNA sequence, a VT-based check string GCT is also generated. (b) The real-time error detection during decoding is shown. (c) The process of local exhaustive reverse search is illustrated; after 13 searches, two candidate strings are obtained. (d) The final check process using VT-based checking is shown; only one of the two candidates meets the correct check value, resulting in a unique solution.

[0074] FIG2A shows an existing end-to-end retrieval mechanism.

[0075] FIG2B shows an end-to-end retrieval mechanism according to the present invention, which reduces one step compared to the strategy in FIG2A and cancels the strict asynchronous requirement. For example, when the frequency is greater than 4, the decoding operation can be performed directly.

[0076] FIG3 is a flow chart of a method for generating DNA storage encoding and decoding rules according to an embodiment of the present invention.

[0077] FIG4 is a flow chart of a DNA storage encoding method according to an embodiment of the present invention.

[0078] FIG5 is a flow chart of a DNA storage decoding method according to an embodiment of the present invention.

[0079] FIG. 6 shows an exemplary image used in the embodiment. DETAILED DESCRIPTION

[0080] The present invention will be further described in detail below by way of specific embodiments in conjunction with the accompanying drawings. In the following embodiments, many details are described to facilitate a better understanding of the present invention. However, those skilled in the art will readily appreciate that some of the features may be omitted in different circumstances, or may be replaced by other materials or methods.

[0081] In addition, the features, operations, or characteristics described in the specification may be combined in any appropriate manner to form various embodiments. Furthermore, the steps or actions in the method description may be reordered or adjusted in a manner readily apparent to those skilled in the art. Therefore, the various sequences in the specification and drawings are provided solely for the purpose of clearly describing a particular embodiment and are not intended to be mandatory, unless otherwise specified.

[0082] The terms used in the present invention are explained as follows:

[0083] The error rate refers to the probability that a DNA sequence or fragment changes into another sequence or fragment.

[0084] Co-sampling refers to statistics based on the correct sequence and the aligned sequencing sequence.

[0085] VT error correction code, full name Varshamov-Tenengolts, corrects single insertion / deletion errors with asymptotically optimal redundancy.

[0086] Path, generation algorithm graph WO / 2021 / 243605 has paths connecting nodes.

[0087] Threshold: after repair, whether the number of occurrences of a sequence reaches a certain value.

[0088] Coding method refers to a mapping relationship between binary and base. Generally speaking, the traditional fixed rule coding method will perform multiple steps of optimization processing to finally obtain the final mapping relationship. In the present invention, the coding method is implemented by coding and decoding rules. The coding and decoding rules of the present invention are generated by the method of generating DNA storage coding and decoding rules of the present invention.

[0089] The generator, also referred to as the "method for generating DNA storage encoding and decoding rules" in the present invention, obtains a potential mapping relationship between binary and bases through graph theory based on different combinations, that is, obtains the encoding and decoding rules of the present invention.

[0090] Algorithm stability refers to the ability of the algorithm to consistently meet constraints for any given electronic file input and DNA sequence output. Typically, in the "arbitrary" case, flood-like inputs are used to observe the algorithm's stability under extreme conditions.

[0091] End-to-end is a data recovery mechanism that does not require clustering and comparison processes. From the original data input to the result output, from the input end to the output end, the intermediate mapping process is self-contained.

[0092] Time complexity. The time complexity of an algorithm is a function that qualitatively describes the algorithm's running time. It is a function of the length of the string representing the algorithm's input. Time complexity is often expressed using Big O notation, which excludes the function's lower-order terms and the leading coefficient. When expressed this way, the time complexity is said to be asymptotic, meaning it considers the case when the input value approaches infinity.

[0093] To address the sequence constraints of different sequencing or synthesis instruments, the present invention proposes a constraint-based optimal codec generator, specifically a method for generating DNA storage codec rules. This generator (or method) addresses the problem that existing fixed rules cannot completely avoid extreme GC sequences or special motifs. Special motifs in this context refer to sequences that are difficult to analyze using fixed rules.

[0094] Furthermore, the encoding method generated by this generator does not require a screening process, thus eliminating the risk of not being able to accept all inputs. Furthermore, the encoding and decoding time complexity of this generator is O(n). Compared to most encoding and decoding methods that require extensive optimization, this encoding and decoding method is significantly faster, making it more efficient for future large-scale DNA storage transcoding.

[0095] The following describes in detail the technical components of the present invention. It should be understood that these descriptions are exemplary and that those skilled in the art can make many modifications based on the technical content of the present invention.

[0096] During single-molecule sequencing, each time a base is read, the base is added to the end of the obtained sequence, and the sequence is corrected by a path-based probabilistic error correction strategy. The correction speed is much faster than the sequencing speed, so the path-based probabilistic error correction strategy can be used for real-time decoding. Combined with the frequency sorting mechanism mentioned in the present invention, the purpose of real-time decoding can be ultimately achieved. In the present invention, the path-based probabilistic error correction strategy can be seen in Figure 1 and (Zhang H, Lan Z, Zhang W, et al. SPIDER-WEB enables stable, repairable, and encryptible algorithms under arbitrary local biochemical constraints in DNA-based storage[J]. 2022. DOI: 10.48550 / arXiv.2204.02855.; https: / / arxiv.org / abs / 2204.02855, the paper is incorporated herein by reference in its entirety). Figure 1 shows a path-based error correction strategy according to the present invention. (a) A GC-balanced directed graph of length 2 is created; using this directed graph, the binary message 001101 is encoded into the DNA sequence TCTGAC; based on this DNA sequence, a VT-based check string GCT is also generated. (b) This shows real-time error detection during decoding. (c) This illustrates the process of a local exhaustive reverse search; after 13 searches, two candidate strings are obtained. (d) This shows the final check using VT-based checking; only one of the two candidates matches the correct check value, resulting in a unique solution. Therefore, the path-based error correction strategy can include: (1) first discovering the error based on the constraint conditions, that is, the current node cannot complete the transfer to the next node (or cannot meet the requirements of the valid path); (2) for the node where the error occurs, in reverse order, check the previous position from the position where the error occurs and perform path correction until the position is within the observation window length from the error position; (3) path correction includes guessing the error type (one of the editing errors) and trying to adjust it to the following three types for repair: replacing the nucleotide at its current position with another nucleotide; inserting a nucleotide between the current position and the previous position; and deleting the nucleotide at the current position; (4) passing the repaired sequence through the path detection mechanism of the Varshamov-Tenengolts error correction code to collect the sequences that meet the detection requirements; (5) outputting a set of repaired DKA sequences that meet the constraint conditions and detection sequence requirements.

[0097] In the present invention, the frequency-based end-to-end retrieval mechanism can be seen in Figures 2A and B, where Figure 2A shows the existing end-to-end retrieval mechanism and Figure 2B shows the end-to-end retrieval mechanism according to the present invention. Assuming that there are m types of synthesized DNA sequences, (1) the DNA sequences obtained by sequencing are passed through the path-based probabilistic error correction strategy one by one; (2) the set of repaired DNA sequences that meet the restriction conditions and detection sequence requirements are counted; (3) they are sorted according to the frequency priority principle; (4) the first m sequences are taken as the final recovery sequence and decoded. The frequency-based end-to-end retrieval mechanism of the present invention optimizes the original five steps into four steps (i.e., repair (DNA level) - counting - decoding - assembly). Based on the results of path error correction, the correct sequence is retrieved through frequency sorting without going through the clustering and alignment process. At the same time, the strategy of the present invention cancels the strict asynchronous requirement. For example, when the frequency is greater than 4, the decoding operation can be performed directly. These steps can eliminate necessary sequentiality by threshold value, such as when the sequencing depth is 20, as long as the frequency of a certain sequence is greater than 4, it can be directly regarded as the correct sequence, complete decoding and subsequent assembly process. On the other hand, compared with the global search method, the probability error correction strategy based on path of the present invention is a local search, and the number of searches of its nodes is reduced by an average of thousands of times (see Table 1), and in theory its time-consuming also can be reduced by thousands of times. Table 1 shows the vertex access times of local and global searches under different error rates. The DNA sequence length is 200nt (that is, the minimum node access number is 200nt). The purpose of statistical search vertex number rather than actual time consumption is to eliminate the uncertain influence of implementation strategy differences and programming language on the time-consuming operation.

[0098] Table 1:

[0099] According to the first aspect of the present invention, the present invention provides a DNA storage encoding and decoding method for single-molecule sequencing. As shown in Figure 3, the exemplary DNA storage encoding and decoding method includes: 1) setting a sliding window (k, n) of the DNA storage encoding and decoding rule, where k represents the length of the sliding window, n represents the length of the base character of each slide, where n and k are positive integers, k ≥ n; 2) based on the length k of the sliding window, obtaining a full set sequence, where the full set sequence is a set of all base sequences formed by random combinations of all possible bases at each base position within the length of the sliding window; 3) performing single-molecule sequencing on the reference sequence to obtain actual sequencing data of the single-molecule sequencing; 4) comparing the actual sequencing data with the reference sequence 5) comparing the sequence fragments with a length of k with reference data, counting the frequency of sequencing errors in the actual sequencing data for each sequence fragment with a length of k, and calculating the proportion of sequencing errors in the actual sequencing data for each sequence fragment with a length of k, i.e., the error rate; 5) removing sequence fragments with an error rate exceeding a threshold as a restriction condition, and screening out a set of qualified sequences in the above complete set of sequences with an error rate not exceeding the threshold; 6) connecting the sequences in the above qualified sequence set via a directed graph, where each node in the directed graph represents each sequence; 7) obtaining an algorithm graph, which includes DNA storage encoding and decoding rules.

[0100] In the present invention, error-prone sequences are eliminated based on error sorting and a DNA storage encoding and decoding method is generated. The DNA storage encoding and decoding method refers to WO / 2021 / 243605, and the patent disclosure text is incorporated herein by reference in its entirety. Specifically, the sequence fragments that need to be eliminated so that the error rate is controlled to be less than or equal to 3% are calculated and used for DNA storage encoding and decoding. Referring to WO / 2021 / 243605, the DNA storage encoding and decoding process is as follows: The encoding process is a graph of the loading algorithm, and an initial node is selected. The binary sequence is converted to a decimal number, the decimal number is divided by the out-degree node corresponding to the current node, the subsequent DNA fragment information of the output out-degree node is recorded, the out-degree node is updated to the current node, and the cycle is continued until no decimal number is equal to 0. The DNA fragments are connected in sequence and the complete DNA sequence is output. Since the node paths accessed by encoding and decoding are the same, the decoding process is a mirror process of the encoding process.

[0101] In the present invention, the steps of extracting and sorting single-molecule sequencing error features can be: (1) setting a sliding window (k, n) of DNA storage encoding and decoding rules, where k represents the length of the sliding window, and n represents the length of the base character of each slide, where n and k are positive integers, and k≥n; (2) based on the length k of the sliding window, obtaining a full set sequence, where the above-mentioned full set sequence is a set of all base sequences formed by random combinations of all possible bases at each base position within the length of the sliding window; (3) performing single-molecule sequencing on the reference sequence to obtain actual sequencing data of the single-molecule sequencing; comparing the actual sequencing data with the reference data of the reference sequence, counting the frequency of sequencing errors in each sequence fragment of length k in the actual sequencing data, and calculating the proportion of sequencing errors in each sequence fragment of length k in the actual sequencing data, i.e., the error rate; (4) statistically sorting the error rates of the sequencing data errors, and eliminating the sequence fragments with error rates exceeding the threshold on the principle of higher error rates and higher priority, thereby screening out a qualified sequence set with error rates not exceeding the threshold in the above-mentioned full set sequence.

[0102] The method of the present invention can be used to model single-molecule sequencing technology, including but not limited to ONT single-molecule sequencing, Nanopore single-molecule sequencing, etc. The disadvantage of single-molecule sequencing technology is its high error rate, which may cause problems when applied to DNA storage technology. The present invention compares the sequencing data with the reference sequence, screens out error-prone sequence fragments for DNA storage encoding, and improves the reliability of DNA storage technology. In the present invention, the reference sequence is used to proofread the erroneous sequencing sequences in the single-molecule sequencing data. Therefore, the reference sequence and the DNA sample sequence of the single-molecule sequencing are as consistent as possible, preferably the same. The reference sequence is a sequence with a known sequence, which can be a genomic sequence, and is used to test the single-molecule sequencing system, so that the frequency of sequencing errors in each sequence fragment of length k in the actual sequencing data is calculated, and the proportion of sequencing errors in each sequence fragment of length k in the actual sequencing data is calculated. The reference sequence can come from a public genomic sequence or be obtained by sequencing using sequencing technology.

[0103] In a preferred embodiment, after step 6), the method includes:

[0104] i) deleting nodes in the directed graph whose out-degree number is less than a set out-degree limit;

[0105] ii) deleting redundant out-degrees of each node in the directed graph, wherein the redundant out-degrees are out-degrees exceeding a set out-degree limit.

[0106] In a preferred embodiment, the deleting of redundant out-degrees of each node in the directed graph includes: if the total number of out-degrees of the node exceeds a set out-degree limit, outputting the bases of the node in reverse order, and deleting the out-degrees pointing to the corresponding bases in sequence according to the order of the bases output in reverse order.

[0107] In a preferred embodiment, a sliding window (k, n) is set to perform collaborative sampling of sequencing data error rate and sequence characteristics using the observation scale of the sliding window for actual sequencing data; and the error ratio information of sequencing data is statistically ranked based on the principle of higher error rate, higher priority.

[0108] In a preferred embodiment, the error rate ranges from 2% to 4.5%, and preferably, the error rate is 3%.

[0109] According to a second aspect of the present invention, the present invention provides an apparatus for generating DNA storage encoding and decoding rules, comprising:

[0110] A sliding window setting unit is used to set a sliding window (k, n) for DNA storage encoding and decoding rules, where k represents the length of the sliding window and n represents the length of the base character for each sliding, where n and k are positive integers and k≥n;

[0111] a constraint determination unit for performing single-molecule sequencing on a reference sequence to obtain actual sequencing data of the single-molecule sequencing; comparing the actual sequencing data with the reference data of the reference sequence, counting the frequency of sequencing errors occurring in the actual sequencing data for each sequence fragment of length k, and calculating the proportion of sequencing errors occurring in the actual sequencing data for each sequence fragment of length k, i.e., the error rate; and using sequence fragments with an error rate exceeding a threshold as constraint conditions;

[0112] A qualified sequence screening unit is configured to obtain a complete set of sequences based on the length k of the sliding window, wherein the complete set of sequences is a set of all possible base sequences formed by random combinations of all bases at each base position within the length of the sliding window, and to screen out qualified sequences from the complete set of sequences that meet the above-mentioned constraints using a constraint condition;

[0113] a directed graph connection unit, configured to connect the sequences in the qualified sequence set via a directed graph, wherein each node in the directed graph represents each sequence;

[0114] The algorithm chart obtaining unit is used to obtain an algorithm chart, which includes DNA storage encoding and decoding rules.

[0115] In a preferred embodiment, the device for generating DNA storage encoding and decoding rules further comprises:

[0116] The out-degree violation deletion unit is used to delete the nodes in the directed graph whose out-degree number is less than the set out-degree limit;

[0117] The redundant out-degree deletion unit is used to delete the redundant out-degree of each node in the directed graph, where the redundant out-degree is the out-degree that exceeds the set out-degree limit.

[0118] According to a third aspect of the present invention, a DNA storage encoding method is provided. As shown in FIG4 , an exemplary DNA storage encoding method includes: 1) obtaining the DNA storage encoding and decoding rules generated by the method of the first aspect, and setting an initial node, which is defined as the current node; 2) obtaining a binary sequence to be encoded and slicing it to generate binary slices, converting the binary values ​​corresponding to the slices into out-degree nodes or multiple out-degree nodes connected to the current node, wherein each out-degree node describes a nucleic acid fragment, and the binary slices and the corresponding nucleic acid fragments form a pair of mapping relationships; 3) according to the DNA storage encoding and decoding rules, inputting the binary slices, outputting the nucleic acid fragments mapped by the out-degree nodes or multiple out-degree nodes, and updating the out-degree nodes as the current node, and continuously inputting binary slices and outputting nucleic acid fragments in a cyclic manner according to the binary slice sequence until all the binary slices are input; 4) sequentially connecting the nucleic acid fragments in the output order and outputting the complete DNA sequence.

[0119] In a preferred embodiment, the above method slices the above binary sequence to be encoded according to a length of 2n-1, where n represents the base character length of each sliding of the sliding window.

[0120] In a preferred embodiment, the method further comprises: synthesizing the DNA sequence and then preserving it in an in vitro medium or living cells.

[0121] According to a fourth aspect of the present invention, the present invention provides a DNA storage encoding device, comprising:

[0122] A coding and decoding rule acquisition unit, configured to acquire the DNA storage coding and decoding rule generated by the method of the first aspect, and set an initial node, defining the initial node as the current node;

[0123] A binary sequence slicing and conversion unit is configured to obtain a binary sequence to be encoded and slice it to generate binary slices, and convert the binary values ​​corresponding to the slices into out-degree nodes or multiple layers of out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid fragment, and the binary slices and the corresponding nucleic acid fragments form a mapping relationship.

[0124] A nucleic acid fragment output unit is configured to input the binary slices according to the DNA storage encoding and decoding rules, output the nucleic acid fragments mapped to the out-degree node or multiple out-degree nodes, update the out-degree node as the current node, and continuously input binary slices and output nucleic acid fragments in a cyclic manner according to the binary slice sequence until all the binary slices are input;

[0125] The nucleic acid fragment connection unit is used to connect the above nucleic acid fragments in sequence according to the output order and output a complete DNA sequence.

[0126] In a preferred embodiment, the binary sequence to be encoded is sliced ​​according to a length of 2n-1, where n represents the length of base characters of each sliding of the sliding window.

[0127] According to a fifth aspect of the present invention, a DNA storage decoding method is provided. As shown in FIG5 , the exemplary DNA storage decoding method includes: 1) obtaining the DNA storage encoding and decoding rules generated by the method of the first aspect, and setting an initial node, which is defined as the current node; 2) obtaining a single-molecule sequencing DNA sequence to be decoded and slicing it to generate nucleic acid slices. Based on the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slices, an out-degree node or multiple out-degree nodes connected to the current node are found, wherein each out-degree node describes a nucleic acid information, and the nucleic acid slices and the corresponding binary values ​​or binary slices form a pair of mapping relationships; 3) based on the current node and the out-degree node or multiple out-degree nodes, binary values ​​or binary slices between nodes are obtained according to the mapping relationship, and the out-degree node is updated as the current node. Nucleic acid slices are continuously input and output binary values ​​or binary slices according to the nucleic acid slice sequence until all nucleic acid slices are input; 4) the binary values ​​or binary slices are sequentially connected in the output order and a complete binary sequence is output.

[0128] In a preferred embodiment, 3) includes an error correction step, which is a path-based probabilistic error correction strategy, including: (1) discovering an error during the decoding process: discovering an error based on a constraint condition, that is, the current node cannot complete the transfer to the next node, or cannot meet the requirements of a valid path; (2) checking the position where the decoding error occurred: in reverse order, that is, checking backward from the position where the error was found to the position of the observation window length away from the error position; (3) obtaining a candidate repair sequence: after searching for the position where the decoding error occurred, through local exhaustive reverse search, determine the error type and try the following three adjustment types for repair: replace the nucleotide at its current position with another nucleotide, insert a nucleotide between the current position and the previous position, and delete the nucleotide at the current position; obtain a candidate repair sequence; (4) obtain a set of repaired DNA sequences: pass the candidate repair sequence through the path detection of the Varshamov-Tenengolts error correction code, collect candidate repair sequences that meet the constraint condition and the detection sequence requirements; output a set of repaired DNA sequences that meet the constraint condition and the detection sequence requirements.

[0129] In a preferred embodiment, after (4), all sequences in the DNA sequence set are counted and sorted, and sequences whose frequencies reach a threshold are directly decoded as correct sequences.

[0130] In a preferred embodiment, the sequencing depth of the sequence is 20×, and the sequence frequency threshold is 4, that is, the sequence with a frequency of 4 is directly decoded as the correct sequence.

[0131] In a preferred embodiment, the above method slices the DNA sequence to be decoded according to the length n, where n represents the length of the base character of each sliding of the sliding window.

[0132] In a preferred embodiment, the DNA sequence to be decoded is generated by encoding using the method of the third aspect or the apparatus of the fourth aspect.

[0133] According to a sixth aspect of the present invention, the present invention provides a DNA storage and decoding device, comprising:

[0134] A coding and decoding rule acquisition unit, configured to acquire the DNA storage coding and decoding rule generated by the method of the first aspect, and set an initial node, defining the initial node as the current node;

[0135] A DNA slicing and conversion unit is configured to obtain a DNA sequence to be decoded and slice it to generate nucleic acid slices. Based on the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slices, the unit finds the out-degree node or multiple out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid information. The nucleic acid slices and the corresponding binary values ​​or binary slices form a mapping relationship.

[0136] a binary value output unit, configured to obtain binary values ​​or binary slices between nodes according to the mapping relationship between the current node and the out-degree node or multiple out-degree nodes, and update the out-degree node to the current node, and continuously and cyclically input nucleic acid slices and output binary values ​​or binary slices according to the sequence of nucleic acid slices until all nucleic acid slices are input;

[0137] The binary value connection unit is used to connect the above binary values ​​in sequence according to the output order and output a complete binary sequence.

[0138] In a preferred embodiment, the DNA sequence to be decoded is sliced ​​according to the length n, where n represents the length of the base character of each sliding of the sliding window.

[0139] According to a seventh aspect of the present invention, the present invention provides a computer-readable storage medium comprising a program, which can be executed by a processor to implement the method of the first aspect, the method of the third aspect, or the method of the fifth aspect.

[0140] Example

[0141] In this example, the sliding window (k, n) has a sliding window length k of 4, a base character length n of 1, and the example image is the Mona Lisa (see Figure 6). The image size is 2283400 bytes (approximately 2.17 MB), with dimensions of 709 pixels (width) × 1073 pixels (height), and a bit depth of 24.

[0142] 1) Collect error rates of different sequences through feature extraction and sorting strategies:

[0143] (1) Nanopore single-molecule sequencing of Escherichia coli,

[0144] Obtain actual sequencing data of single-molecule sequencing. Obtain sequencing data of Escherichia coli using Nanopore single-molecule sequencing technology;

[0145] (2) comparing the actual sequencing data with the reference data of the reference sequence;

[0146] (3) intercepting the correct sequence fragments according to the sliding window and sequencing the corresponding sequence fragments;

[0147] (4) Generate 256 different DNA fragments based on the sliding window length;

[0148] (5) Count the number of correct sequence fragments a and the number of incorrect sequence fragments b, and calculate the error rate = b / (a+b);

[0149] (6) sorting each obtained fragment from high to low according to the error rate to obtain the sorting result for the sliding window;

[0150] (7) Calculate the segments with an error rate greater than or equal to 3% (a total of 44, see Table 2, in which the high error rate sequences are statistically analyzed by k-mer) and select them into the set of restriction conditions. Table 2:

[0151] 2) By eliminating the above segments with an error rate greater than or equal to 3%, the corresponding encoding method diagram is generated:

[0152] (1) The fragments with an error rate greater than or equal to 3% were removed from the 256 DNA fragments, and a qualified sequence set with an error rate not exceeding the threshold was screened out from the complete set of sequences.

[0153] (2) The sequences in the above qualified sequence set are processed by the method described in patent application WO / 2021 / 243605 to complete the creation of the encoding method diagram.

[0154] 3) Encode the example image using the encoding method diagram:

[0155] (1) Set the sequence length to 200 nt and the check sequence length to 5 nt;

[0156] (2) 60,891 DNA sequences were obtained through sample sequencing.

[0157] 4) For each of the 60,891 encoded DNA sequences, a 2% random error rate (including substitutions, insertions, and deletions) was set, and the sequencing depth was set to 20×, completing real-time decoding:

[0158] (1) Perform a path-based error correction strategy on each DNA sequence to obtain the corresponding sequence solution set. Combine the solution sets of all sequences and record the frequency of each sequence solution.

[0159] (2) Through the frequency-based end-to-end detection mechanism, the repaired sequences with a frequency greater than 4 are preferentially decoded;

[0160] (3) The program runs on an Intel(R) Core(TM) i7-4710MQ CPU (single core) for approximately 1562 seconds, achieving an overall repair rate (the number of correct bit sequences to all bit sequences) of approximately 98.18% (59783 / 60891).

[0161] (4) The repaired DNA sequence decodes the sequence information according to the coding rules to obtain a binary value, restores the binary information, and interprets it as an example image.

Claims

1. A DNA storage encoding and decoding method, comprising: 1) Set a sliding window (k, n) for storing DNA encoding and decoding rules, where k represents the length of the sliding window, and n represents the length of the base character for each sliding, where n and k are positive integers, and k ≥ n; 2) Based on the length k of the sliding window, a full set sequence is obtained, wherein the full set sequence is a set of all base sequences formed by random combinations of all possible bases at each base position within the length of the sliding window; 3) Sequencing the reference sequence to obtain actual sequencing data; 4) comparing the actual sequencing data with the reference data of the reference sequence, counting the frequency of sequencing errors in each sequence fragment of length k in the actual sequencing data, and calculating the proportion of sequencing errors in each sequence fragment of length k in the actual sequencing data, i.e., the error rate; 5) Eliminate sequence segments whose error rates exceed a threshold as a restriction condition, and select a qualified sequence set whose error rates do not exceed the threshold from the above-mentioned full set of sequences; 6) connecting the sequences in the above qualified sequence set by means of a directed graph, wherein each node in the above directed graph represents each sequence; 7) Obtaining an algorithm diagram, wherein the algorithm diagram includes DNA storage encoding and decoding rules.

2. The method according to claim 1, characterized in that The sequencing is single molecule sequencing.

3. The method according to claim 2, characterized in that The single molecule sequencing includes ONT single molecule sequencing and / or Nanopore single molecule sequencing.

4. The method according to claim 1, characterized in that: After step 6) includes: i) deleting nodes in the directed graph whose out-degree number is less than a set out-degree number limit; ii) deleting redundant out-degrees of each node in the directed graph, wherein the redundant out-degrees are out-degrees exceeding a set out-degree number limit.

5. According to the method of claim 4, the step of deleting the redundant out-degree of each node in the directed graph comprises: If the total out-degree number of the node exceeds the set out-degree number limit, the bases of the node are output in reverse order, and the out-degrees pointing to the corresponding bases are deleted in sequence according to the order of the bases output in reverse order.

6. The method according to any one of claims 1 to 5, characterized in that: A sliding window (k, n) is set to perform collaborative sampling of sequencing data error rate and sequence features using the observation scale of the sliding window for the actual sequencing data; the error ratio information of sequencing data is statistically sorted based on the principle of higher error rate, higher priority.

7. The method according to any one of claims 1 to 6, characterized in that: The error rate ranges from 2% to 4.5%.

8. The method according to any one of claims 1 to 7, characterized in that: The error rate is 3%.

9. A device for generating DNA storage encoding and decoding rules, characterized in that: include: A sliding window setting unit, used to set a sliding window (k, n) for storing DNA encoding and decoding rules, where k represents the length of the sliding window, and n represents the length of the base character of each sliding, where n and k are positive integers, and k ≥ n; A restriction condition determination unit, used for performing single-molecule sequencing on a reference sequence to obtain actual sequencing data of the single-molecule sequencing; Compare the actual sequencing data with the reference data of the reference sequence, count the frequency of sequencing errors in each sequence fragment of length k in the actual sequencing data, and calculate the proportion of sequencing errors in each sequence fragment of length k in the actual sequencing data, i.e., the error rate; use the sequence fragments with an error rate exceeding a threshold as restriction conditions; A qualified sequence screening unit is used to obtain a full set of sequences based on the length k of the sliding window, wherein the full set of sequences is a set of all base sequences formed by random combinations of all possible bases at each base position within the length of the sliding window, and to screen out a set of qualified sequences that meet the above restriction conditions from the full set of sequences using restriction conditions; A directed graph connection unit, used to connect the sequences in the above qualified sequence set in a directed graph, wherein each node in the above directed graph represents each sequence; The algorithm chart obtaining unit is used to obtain an algorithm chart, wherein the algorithm chart includes DNA storage encoding and decoding rules.

10. The device according to claim 9, characterized in that The device for generating DNA storage encoding and decoding rules also includes: The out-degree violation deletion unit is used to delete the nodes whose out-degree number is less than the set out-degree limit in the above directed graph; The redundant out-degree deletion unit is used to delete the redundant out-degree of each node in the directed graph, wherein the redundant out-degree is the out-degree exceeding the set out-degree number limit.

11. A DNA storage encoding method, characterized in that: include: 1) obtaining a DNA storage encoding and decoding rule generated by the method according to any one of claims 1 to 8, and setting an initial node, and setting the initial node as the current node; 2) Obtaining a binary sequence to be encoded and slicing it to generate binary slices, converting the binary value corresponding to the slice into an out-degree node or multiple out-degree nodes connected to the current node, wherein each out-degree node describes a nucleic acid fragment, and the above binary slice and the corresponding nucleic acid fragment form a pair of mapping relationships; 3) According to the DNA storage encoding and decoding rules, the binary slice is input, the nucleic acid fragment mapped to the out-degree node or multiple out-degree nodes is output, and the out-degree node is updated to the current node. The slicing sequence continuously cycles through the input of binary slices and the output of nucleic acid fragments until all the binary slices are input; 4) Connect the above nucleic acid fragments in order according to the output order and output the complete DNA sequence.

12. The method according to claim 11, characterized in that The above method slices the above binary sequence to be encoded according to the length of 2n-1, where n represents the base character length of each sliding of the sliding window.

13. The method according to claim 11 or 12, characterized in that: The method also includes: synthesizing the DNA sequence and then storing it in an in vitro medium or in a living cell.

14. A DNA storage encoding device, comprising: A coding and decoding rule acquisition unit, used to acquire the DNA storage coding and decoding rule generated by the method of any one of claims 1 to 8, and set an initial node, and set the initial node as the current node; A binary sequence slicing and conversion unit is used to obtain a binary sequence to be encoded and slice it to generate a binary slice, and convert the binary value corresponding to the slice into an out-degree node or multiple out-degree nodes connected to the current node, wherein each out-degree node describes a nucleic acid fragment, and the above binary slice and the corresponding nucleic acid fragment form a pair of mapping relationships; A nucleic acid fragment output unit is used to input the binary slices according to the DNA storage encoding and decoding rules, output the nucleic acid fragments mapped by the out-degree node or multiple out-degree nodes, and update the out-degree node as the current node, and continuously input binary slices and output nucleic acid fragments in a loop according to the binary slice sequence until all the binary slices are input; The nucleic acid fragment connection unit is used to connect the above nucleic acid fragments in sequence according to the output order and output a complete DNA sequence.

15. The device according to claim 14, characterized in that The above binary sequence to be encoded is sliced ​​according to a length of 2n-1, where n represents the length of the base character of each sliding of the sliding window.

16. A DNA storage decoding method, comprising: 1) obtaining the DNA storage encoding and decoding rules generated by the method of any one of claims 1 to 8, and setting an initial node, and setting the initial node as the current node; 2) Obtain the single-molecule sequencing DNA sequence to be decoded and slice it to generate nucleic acid slices. According to the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slices, find the out-degree node or multiple out-degree nodes connected to the current node, where each out-degree node describes a nucleic acid information, and the nucleic acid slice and the corresponding binary value or binary slice form a pair of mapping relationships; 3) According to the above current node and the above out-degree node or multi-layer out-degree node, obtain the node according to the above mapping relationship The binary value or binary slice between the points, and update the out-degree node as the current node, and continuously input nucleic acid slices and output binary values ​​or binary slices in a loop according to the sequence of nucleic acid slices, until all the nucleic acid slices are input; 4) Connect the above binary values ​​or binary slices in order according to the output order and output a complete binary sequence.

17. The method according to claim 16, characterized in that 3) includes an error correction step, which is a path-based probabilistic error correction strategy, including: (1) An error was found during the decoding process: Find errors based on constraints, i.e. the current node cannot complete the transfer to the next node or cannot meet the requirements of a valid path; (2) Check the location where the decoding error occurred By reverse order, it means checking backwards from the position where the error is found to the position that is the length of the observation window away from the error position; (3) Obtaining candidate repair sequences After searching for the position where the decoding error occurred, the error type is determined through local exhaustive reverse search and the following three types of adjustments are tried to repair it: replacing the nucleotide at its current position with another nucleotide, inserting a nucleotide between the current position and the previous position, and deleting the nucleotide at the current position; obtaining a candidate repair sequence; (4) Obtaining a set of repaired DNA sequences The candidate repaired sequences are tested by the Varshamov-Tenengolts error correction code path, and the candidate repaired sequences that meet the restriction conditions and the detection sequence requirements are collected; and a set of repaired DNA sequences that meet the restriction conditions and the detection sequence requirements are output.

18. The method according to claim 17, characterized in that After (4), all sequences in the DNA sequence set are counted and sorted, and the sequences whose frequencies reach the threshold are directly decoded as correct sequences.

19. The method according to claim 17 or 18, characterized in that The sequencing depth of the sequence was 20×, and the sequence frequency threshold was 4, that is, the sequence with a frequency of 4 was directly decoded as the correct sequence.

20. The method according to any one of claims 16 to 19, characterized in that: The DNA sequence to be decoded is sliced ​​according to the length of n, where n represents the length of the base characters of each sliding of the sliding window.

21. The method according to any one of claims 16 to 20, characterized in that: The DNA sequence to be decoded is generated by encoding using the method of any one of claims 11 to 14 or the device of claim 14 or 15.

22. A DNA storage decoding device, comprising: A coding and decoding rule acquisition unit, used to acquire the DNA storage code generated by the method according to any one of claims 1 to 8. Decode the rules and set the initial node, which is set as the current node; A DNA slicing and conversion unit is used to obtain a single-molecule sequencing DNA sequence to be decoded and slice it to generate a nucleic acid slice, and find an out-degree node or multiple out-degree nodes connected to the current node and the current node according to the DNA storage encoding and decoding rules and the nucleic acid information corresponding to the slice, wherein each out-degree node describes a nucleic acid information, and the nucleic acid slice and the corresponding binary value or binary slice form a pair of mapping relationships; A binary value output unit is used to obtain binary values ​​or binary slices between nodes according to the above-mentioned current node and the above-mentioned out-degree node or multi-layer out-degree node according to the above-mentioned mapping relationship, and update the above-mentioned out-degree node to the current node, and continuously and cyclically input nucleic acid slices and output binary values ​​or binary slices according to the above-mentioned nucleic acid slice sequence until all the above-mentioned nucleic acid slices are input; The binary value connection unit is used to connect the above binary values ​​in sequence according to the output order and output a complete binary sequence.

23. The device according to claim 22, characterized in that The DNA sequence to be decoded is sliced ​​according to the length of n, where n represents the length of the base characters of each sliding of the sliding window.

Images