DNA compression method for recursively constructing dictionary based on suffix tree

Abstract

The invention discloses a DNA compression method for constructing a dictionary based on suffix tree recursion, which is applied to the crossing field of electronic information and synthetic biology and comprises the following steps: acquiring an original DNA sequence; constructing a suffix tree based on the original DNA sequence, and traversing the suffix tree to obtain a repetitive sequence in the original DNA sequence; constructing a prefix table according to the repetitive sequence; selecting an optimal entry from entries containing biological characteristics and entries not containing biological characteristics in the prefix table through entry definition screening; vocabulary entries are removed from the original DNA sequence, the vocabulary entries and position information are obtained during removal, and a new DNA sequence after removal is obtained; judging whether the new DNA sequence meets a preset cycle end condition or not; if so, ending the circulation; and if not, reconstructing the suffix tree by using the new DNA sequence.

Description

Technical Field

[0001] The present invention relates to the intersection of electronic information and synthetic biology, and in particular to a DNA compression method based on recursive dictionary construction using a suffix tree. Background Art

[0002] With the rapid development of bioinformatics, a huge amount of DNA sequence data has been generated. The rapid accumulation of bioinformatics data is unprecedented in the history of human scientific research. In order to better process and utilize this data, DNA sequence compression algorithms have become increasingly important.

[0003] DNA sequence data is an extremely important type of biological data. DNA is a long double-helix polymer used in cells to store biological genetic instructions for a long time. Its basic skeleton is composed of phosphate sugar substances and four bases: adenine, thymine, guanine and cytosine. DNA research is widely used in many important fields such as biology, medicine, and genetics. DNA sequence data is increasing rapidly, and the storage and use costs will eventually increase to an unbearable scale. Analyzing and managing these massive amounts of data has become a problem that must be solved at present. Data compression is the direction to solve this problem. Effective compression algorithms can reduce the storage and transmission costs of DNA sequences and improve transmission efficiency, thereby promoting the research and development of various biological information and academic exchanges. Compression technology is generally divided into two categories, lossless compression and lossy compression. In order to avoid the loss of biological information and ensure the reliability of original data, DNA sequence compression generally uses lossless compression algorithms. The modes used for lossless compression in DNA sequences are usually divided into two types, namely statistical models and dictionary models. The compression methods based on statistical models and dictionary models are different and have their own advantages. They both have good effects on DNA sequences with obvious patterns, but the compression effect is poor for DNA sequences without obvious patterns.

[0004] In order to overcome these defects, the present application proposes a DNA compression method based on recursive dictionary construction using a suffix tree. Summary of the invention

[0005] The purpose of this application is to provide a DNA compression method based on recursive dictionary construction of a suffix tree, aiming to solve the above-mentioned problems.

[0006] To achieve the above objectives, this application provides the following technical solutions:

[0007] The present application provides a DNA compression method based on recursive dictionary construction using a suffix tree, comprising:

[0008] Obtaining the original DNA sequence;

[0009] Constructing a suffix tree based on the original DNA sequence, and traversing the suffix tree to obtain repeated sequences in the original DNA sequence;

[0010] Constructing a prefix table according to the repeated sequence;

[0011] In the prefix table, the optimal term is selected from the terms containing biological characteristics and the terms containing non-biological characteristics by screening the term definitions;

[0012] Deleting entries from the original DNA sequence, acquiring the entry and position information during deletion, and obtaining a new DNA sequence after deletion;

[0013] Determine whether the new DNA sequence meets the preset loop end condition; if so, end the loop; if not, reconstruct the suffix tree of the new DNA sequence.

[0014] Furthermore, in the step of constructing a suffix tree based on the original DNA sequence, traversing the suffix tree, and obtaining a repeated sequence in the original DNA sequence, the following steps are specifically included:

[0015] For each position in the original DNA sequence, a substring from the position to the end of the sequence is inserted into the suffix tree as a suffix;

[0016] Starting from the root node, create nodes based on the characters of the current suffix;

[0017] If the current character does not exist in the node's child nodes, create a new child node and connect it to the current node;

[0018] Store the starting position of the current suffix in the traversed node.

[0019] Furthermore, in the step of constructing a prefix table according to the repeated sequence, the following steps are specifically included:

[0020] Starting from the root node, the suffix tree is searched in depth first; for each character of the input string, the substring from the character to the end of the string is inserted into the suffix tree; for each suffix, a path is traversed or created starting from the root node until the node representing the last character of the current suffix is ​​reached;

[0021] During the traversal process, if the current character does not exist in the child nodes of the node, a new child node is created and connected to the current node; based on the newly created node, the node with the longest common prefix suffix is ​​selected, and a prefix table from the current node to the node is created.

[0022] Furthermore, the step of selecting the best entry from the entries containing biological characteristics and the entries containing non-biological characteristics by screening the entry definitions in the prefix table specifically includes the following steps:

[0023] It is defined that the number of occurrences of the term is greater than 1, and the number of characters in the term is greater than 1;

[0024] Divide the substrings that satisfy the terms into terms containing biological characteristics and terms not containing biological characteristics; select the maximum value of the term length and the number of occurrences among the terms containing biological characteristics, and select the maximum value of the term length and the number of occurrences among the terms not containing biological characteristics;

[0025] In each iteration a term is selected, with preference given to terms containing biological properties.

[0026] Furthermore, in the step of removing entries from the original DNA sequence, obtaining the entries and position information during the removal, and obtaining a new DNA sequence after the removal, the following steps are specifically included:

[0027] The obtained optimal entries are divided according to whether they contain biological characteristics, to obtain entry and position information; and the entry and position information are stored in the entry stack table.

[0028] Furthermore, after removing the entries from the original DNA sequence, the method further comprises:

[0029] When a new entry appears after the DNA sequence before and after the entry is removed, the entry is checked and removed in the DNA sequence in an incremental manner, specifically:

[0030] The characters of the original DNA sequence are added to the sequence to be tested in sequence, and the sequence to be tested is checked to see whether there is an entry to be removed; if so, the character added is the last character of the entry to be removed, which is removed from the original DNA sequence. When it is removed, the position information is obtained and stored in the entry stack table, and characters are continued to be added in sequence until the end of the DNA sequence.

[0031] This application provides a DNA compression method based on recursive dictionary construction of suffix tree, which has the following features:

[0032] Beneficial effects:

[0033] (1) This application uses a suffix tree data structure to perform global analysis on DNA sequences, realize global search for repeated sequences, avoid falling into local repeated sequences, and improve compression efficiency;

[0034] (2) By defining the concept of terms and performing secondary division based on whether they contain biological characteristics, the optimal terms are selected for elimination, which not only takes into account compression efficiency but also retains the biological information of DNA sequences;

[0035] (3) Efficient compression of DNA sequences is achieved by recursively constructing dictionaries and incrementally removing entries. At the same time, the original DNA sequence is restored by recording the entries and their position information, ensuring lossless compression. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a flow chart of a DNA compression method for recursively constructing a dictionary based on a suffix tree according to Example 1 of the present application;

[0037] Figure 2 This is a schematic diagram of an example suffix tree of Embodiment 1 of the present application;

[0038] Figure 3 A schematic diagram of an example prefix table of Embodiment 1 of the present application;

[0039] Figure 4 This is a schematic diagram of the entry table after the entry definition screening in Example 1 of the present application;

[0040] Figure 5 This is a schematic diagram of the entry table after being divided by presence or absence of biological characteristics in Example 1 of the present application;

[0041] Figure 6 A schematic diagram of selecting the current entry and pushing it into a stack for storage in Example 1 of the present application;

[0042] Figure 7 This is a schematic diagram of an abnormality occurring during removal in Example 1 of the present application;

[0043] Figure 8 A schematic diagram of incrementally removing entries according to Example 1 of the present application;

[0044] Fig. 9 This is a schematic diagram of the entry stack table of Example 1 of the present application. DETAILED DESCRIPTION

[0045] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] The following is an analysis of the DNA sequence data compression scheme in the prior art in combination with related technologies.

[0047] The BioCompress algorithm in the literature [1] was proposed by S. Grumbach and F. Tahi in 1993 as the first algorithm specifically for compressing DNA sequence data. The algorithm is based on the LZ77 dictionary compression concept and uses a sliding window to search for and compress repeated sequences, ultimately achieving compression of sequence-matching fragments. Subsequently, the authors of the algorithm proposed an improved BioCompress-2 algorithm in the literature [2]. The new algorithm uses Arithmetic encoding of order 2 (Arith-2) instead of 2-bit encoding only for the compression of the remaining non-repeating sequences.

[0048] The DNACompress algorithm proposed in the literature [3] is based on biological sequence comparison and uses a compression algorithm based on the LZ concept. The algorithm first uses the biological software PatternHunter (a general, fast-running and high-matching precision approximate sequence search software) to assist in the search for precise repeated fragments; secondly, an algorithm similar to BioCompress-2 is used to compress repeated fragments using the LZ concept, and the remaining non-repeating fragments are compressed using Arith-2. Compared with previous compression algorithms, the compression rate of this algorithm is not much improved, but the compression time of the algorithm is significantly reduced. It is one of the most commonly used sequence compression algorithms.

[0049] The Cfact algorithm proposed in the literature [4] is a typical lossless compression algorithm based on a statistical model. The execution of this algorithm is divided into two stages. The first is the parsing stage: Since it is based on statistical ideas, in order to ensure the selection of the most important repetitions, the parsing stage uses a suffix tree data structure to obtain the repeated fragments in the DNA sequence. The repeated sequences are considered from large to small in length; the second is the encoding stage: the encoding stage scans the DNA sequence from left to right, encodes all non-repeating sequences and the first occurrence of the repeated sequence, and each character is represented by two bits, but no compression is performed. The second occurrence of the repeated sequence is replaced with a pointer (position, length) pointing to the first occurrence, and subsequent occurrences are replaced with a repeated index identifier. The Cfact algorithm first searches for redundancy in the entire sequence and uses the conceptual information of the distribution of bases in the DNA sequence to optimize compression.

[0050] [1]S.Grumbach and F.Tahi, "Compression of DNAsequences," [Proceedings]DCC`93: Data Compression Conference, Snowbird, UT, USA, 1993, pp.340-350, doi:10.1109 / DCC.1993.253115.

[0051] [2]Grumbach S, Tahi FA new challenge for compression algorithms: Genetic sequences. Information Processing and Management, 1994, 30(6): 875-886

[0052] [3]Chen X, et al.DNACompress: Fast and effective DNA sequencecompression[J].Bioinformatics, 2002,18(12):1696-1698

[0053] [4] E. Rivals, J..-P. Delahaye, M. Dauchet and O. Delgrange, "Aguaranteedcompression scheme for repetitive DNA sequences," Proceedings of DataCompression Conference-DCC'96, Snowbird, UT, USA, 1996, pp.453-, doi:10.1109 / DCC.1996.488385.

[0054] Based on the existing methods mentioned above, they have the following disadvantages:

[0055] 1) The algorithms proposed in references [1] and [2] are both based on a dictionary model to achieve compression. The specific operation is to use a sliding window to find repeated sequences in the sequence. Both algorithms only utilize the local redundancy of the data and cannot detect many overlapping repeated fragments in the DNA sequence. The compression effect is relatively poor.

[0056] 2) For example, the algorithm proposed in the literature [3] uses a method similar to the BioCompress-2 in the literature [2] to process repeated fragments. Although the compression time is significantly improved, the compression rate is not much improved.

[0057] 3) As mentioned in the algorithm in [4], this algorithm needs to scan the sequence twice. The first scan is to build a suffix tree. The second scan is based on the idea of ​​the LZ algorithm to find the longest repeated segment in the suffix tree. Then, different encoding methods are used for the first, second, and multiple referenced repeated segments to achieve compression. Although this method searches for repeated sequences globally in the first pass, it will affect the compression effect of the second scan encoding for DNA sequences without obvious patterns, ultimately resulting in poor results.

[0058] The main working method of compression technology is to find repeated patterns and perform compact encoding. On the one hand, the above-mentioned classic dictionary compression algorithm cannot compress repeated DNA sequences and is prone to fall into local repeated sequences; on the other hand, the above-mentioned Cfact algorithm based on statistical models will affect the compression encoding effect for DNA sequences without obvious patterns. Therefore, this application combines the advantages of dictionary-based models and statistical models, avoids the above-mentioned shortcomings, and under the inspiration of the Cfact algorithm, it not only utilizes the probability information of base distribution in DNA sequences, but also takes into account biological characteristics, and proposes a DNA compression method based on recursive dictionary construction based on suffix trees.

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0060] Example 1

[0061] See also Figure 1 , is a flow chart of a DNA compression method for recursively constructing a dictionary based on a suffix tree according to Example 1 of the present application; the steps include:

[0062] S1: Obtain the original DNA sequence.

[0063] S2: constructing a suffix tree based on the original DNA sequence, traversing the suffix tree, and obtaining repeated sequences in the original DNA sequence.

[0064] In this embodiment, taking the sequence "ATCGCGATCGC" as an example, except for the root node, the path from the root node to each child node is the prefix of the node, and the prefix is ​​a substring in the original sequence. The path from each child node to the leaf node is the suffix of the node, and the number of paths to the leaf node means the number of times the prefix of the node appears in the original sequence. When constructing the suffix tree, a $ sign is generally added to the end of the sequence to indicate the end, so there will be an extra $ node from the root node, and the last sign of each leaf node is also $; the numbers in the nodes only represent the order during construction.

[0065] See also Figure 2 , which is a schematic diagram of an example suffix tree of Example 1 of the present application. The node numbered 5 has a prefix of "CG". There are three paths from node 5 to a leaf node, so the number of times "CG" appears in the original DNA sequence is 3.

[0066] S3: Construct a prefix table according to the repeated sequence.

[0067] In this embodiment, a suffix tree constructed from a DNA sequence, except for the root node, the path from each node to the leaf node is a substring of the DNA sequence, and if the node is regarded as a subtree, the number of leaf nodes in the subtree is the number of occurrences of the substring from the root node to the node, that is, the frequency. Therefore, a depth-first traversal post-order operation is performed on the tree, and the substring from the root node to each node and the number of occurrences are returned to form a prefix table of the suffix tree, which is composed of prefixes and the number of occurrences.

[0068] The suffix tree is searched in depth first starting from the root node; for each character of the input string, the substring from the character to the end of the string is inserted into the suffix tree; for each suffix, a path is traversed or created starting from the root node until the node representing the last character of the current suffix is ​​reached. During the traversal process, if the current character does not exist in the child nodes of the node, a new child node is created and connected to the current node; based on the newly created node, the node with the longest common prefix suffix is ​​selected, and a prefix table is created from the current node to the node.

[0069] S4: Select the best entry from the entries containing biological characteristics and the entries not containing biological characteristics by screening the entry definitions in the prefix table.

[0070] In this embodiment, the concept of "entry" is defined: the length of the string must be greater than 1; the number of times the string appears must be greater than 1. Based on the definition of the entry, an optimal substring is selected as the entry and directly removed from the original DNA sequence. The substrings that meet the concept of the entry are divided into two categories: those containing biological characteristics and those containing non-biological characteristics. The selection strategy is: first select Max (entry length * number of occurrences) from the entries containing biological characteristics, and then select Max (entry length * number of occurrences) from the entries containing non-biological characteristics. Only one entry is selected in each iteration.

[0071] See also Figure 3 , is a schematic diagram of the example prefix table of Example 1 of the present application. There are prefixes in the prefix table that do not meet the above definition and should not be selected as entry objects. Here is the following explanation: For the first item, if the acceptance string is 1, then the prefix table will appear A, T, C, G single bases as entries. The number of times a single base appears must be huge, which will affect the selection strategy, so a single base is not accepted as an entry. Figure 3 As shown in the figure, the prefix with 1 occurrences is almost highly overlapped with the original sequence. If it is accepted as an entry, the substring of the original sequence minus the first character will almost certainly be selected and removed. Such compression is meaningless and deviates from the essence of compression technology. Therefore, the prefix with 1 occurrences is not accepted into the selectable range. Through the definition of the entry, some inappropriate prefixes can be preliminarily screened out to obtain the entry table after the first screening.

[0072] After the entry definition screening, the selected entries are divided into two categories according to whether they contain biological characteristics. Here, it is assumed that "TCGC" contains biological characteristics. Figure 4 By performing secondary division, we can obtain Figure 5 In the selection process, the entries containing biological characteristics are selected first, and then the optimal entry is selected by Max (number of characters * times). Since the number of characters * times of the entry "ATCGC" is greater than the number of characters * times of "TCGC", the sequence "ATCGC" is finally selected as the entry in this cycle and the entry is pushed onto the stack, as shown in Figure 6 The third step in each cycle is to select the best entry to be eliminated and push the entry into the stack to facilitate the subsequent recovery of the original DNA sequence.

[0073] S5: removing entries from the original DNA sequence, acquiring the entry and position information during the removal, and obtaining a new DNA sequence after the removal.

[0074] In this embodiment, the obtained optimal terms are divided according to whether they contain biological characteristics, and the terms and position information are obtained; and the terms and position information are stored in the term stack table. When a new term appears after the DNA sequence is merged before and after the term is removed, the term is checked and removed in the DNA sequence in an incremental manner, specifically: the characters of the original DNA sequence are added to the sequence to be tested in sequence, and the sequence to be tested is checked to see whether there is a term to be removed; if there is, the added character is the last character of the term to be removed, and it is removed from the original DNA sequence, and the position information is obtained when it is removed, and stored in the term stack table, and characters are continued to be added in sequence until the end of the DNA sequence.

[0075] See also Figure 8 , is a schematic diagram of incremental elimination of terms in Example 1 of the present application. "ATCGC" was selected as the term to be eliminated in this step. It should be noted that after a term is eliminated, the merging of the front and back DNA sequences may cause new entries of the term to appear. In order to avoid omission of the eliminated terms, an incremental method is used to check and eliminate the terms in the DNA sequence.

[0076] First, each character of the DNA sequence is added to the test sequence in turn, and the test sequence is checked to see if there is an entry to be removed. If so, it must be that the character just added is exactly the last character of the entry to be removed, such as Figure 8 As shown in (a), characters are added in sequence. After the fifth character "C" is added, there is a term to be eliminated in the current sequence to be detected. After eliminating the term, the new term to be eliminated that appears due to the connection between the previous and next sequences will not be missed. The reason is as follows: the term to be eliminated must appear at the end of the sequence to be detected. After it is eliminated, if the subsequent characters added are combined with the previous ones to appear again, the step of adding characters one by one for detection is also required. Therefore, this method will not miss the new term to be eliminated that appears due to the connection between the previous and next sequences, and this method can also be improved when recording positions. When the term appears for the first time, it is necessary to record the starting position of the occurrence, such as Figure 8 As shown in the stack on the right side of (a), 0 represents the first position of the word to be removed in the sequence to be detected. Its position is recorded and connected to the stack element in the form of a linked list. After completing the above recording and removal operations, continue to add characters to the sequence to be detected. For the words to be removed that appear later, only the increment of the position of the previous word to be removed is recorded. This further reduces the amount of data that needs to be recorded, such as Figure 8 As shown in (b), the second occurrence position is 1, and the increment between the first occurrence position is "1"-"0"="1", so 1 is added to the linked list, and the current operation is continued after the entry is removed, such as Figure 8(c) When all characters are added to the sequence to be tested, all entries have been removed, and the sequence to be tested is now a new DNA sequence.

[0077] S6: Determine whether the new DNA sequence meets the preset loop end condition; if so, end the loop; if not, reconstruct the suffix tree of the new DNA sequence.

[0078] In this embodiment, if the new DNA sequence does not meet the loop end condition, the first step is re-executed on the new DNA sequence until the end condition is met, and finally a compressed DNA remainder sequence and an entry stack table storing the entry and position information are output.

[0079] Specifically, it also includes a recovery operation: assuming that the original DNA sequence is "ATCGCGATCGC", after the above compression operation, the remaining DNA sequence is "G", which is now restored.

[0080] The first step is to pop the top element of the current stack by popping the stack to obtain the entry and position information.

[0081] In the second step, the entry information currently obtained is "ATCGC". The position information is obtained from the head of the linked list. Since the first position information is the real position, and the subsequent ones are increments from the previous real position, the first position information is "0", and the second position information is "0" + "1" = "1", at position 1; if there is a third position information of "3", then the third position is "1" + "3" = "4", at position 4, and so on. After scanning the entire linked list, all the real position information can be obtained. Scan Figure 8 The first position information of the stack table is "0", and the second position information is "1". The recovery operation starts from the end of the stack table. The position information of the end of the table is "1", so the entry is inserted into the position 1 of the remaining DNA sequence "G" to obtain the "GATCGC" sequence; the second position information is "0", so similarly, the entry is inserted into the position 0 of the "GATCGC" sequence to obtain the "ATCGCGATCGC" sequence, until the entries at all positions are restored.

[0082] The third step is to loop the above two steps until the stack is empty.

[0083] In summary, Example 1 of the present application combines the suffix tree data structure, recursively constructs a dictionary, realizes global search for repeated sequences; and defines the concept of entries, selects the best entry for elimination, and records the entry and position information at the same time, and finally obtains a remaining DNA sequence and an entry stack table, thereby realizing efficient compression of the DNA sequence. The present application not only takes into account the advantages of global search for repeated sequences, but also combines a reasonable entry selection strategy, and solves the storage and use cost problems caused by the increasing surge in DNA sequence data.

[0084] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0085] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

[0086] Although the embodiments of the present application have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the appended claims and their equivalents.

[0087] Of course, the present invention may have many other implementations. Based on this implementation, other implementations obtained by ordinary technicians in this field without any creative work are all within the scope of protection of the present invention.

Claims

1. A DNA compression method based on recursive dictionary construction using a suffix tree, characterized in that: include: Obtaining the original DNA sequence; Constructing a suffix tree based on the original DNA sequence, and traversing the suffix tree to obtain repeated sequences in the original DNA sequence; Constructing a prefix table according to the repeated sequence; In the prefix table, the optimal term is selected from the terms containing biological characteristics and the terms containing non-biological characteristics by screening the term definitions; Deleting entries from the original DNA sequence, acquiring the entry and position information during deletion, and obtaining a new DNA sequence after deletion; Determine whether the new DNA sequence meets the preset loop end condition; if so, end the loop; if not, reconstruct the suffix tree of the new DNA sequence.

2. A DNA compression method based on recursive dictionary construction using suffix trees according to claim 1, characterized in that: The step of constructing a suffix tree based on the original DNA sequence, traversing the suffix tree, and obtaining a repeated sequence in the original DNA sequence specifically includes the following steps: For each position in the original DNA sequence, a substring from the position to the end of the sequence is inserted into the suffix tree as a suffix; Starting from the root node, create nodes based on the characters of the current suffix; If the current character does not exist in the node's child nodes, create a new child node and connect it to the current node; Store the starting position of the current suffix in the traversed node.

3. A DNA compression method based on recursive dictionary construction using suffix tree according to claim 1, characterized in that: The step of constructing a prefix table according to the repeated sequence specifically includes the following steps: Starting from the root node, the suffix tree is searched in depth first; for each character of the input string, the substring from the character to the end of the string is inserted into the suffix tree; For each suffix, traverse or create a path starting from the root node until you reach the node representing the last character of the current suffix; During the traversal process, if the current character does not exist in the child nodes of the node, a new child node is created and connected to the current node; based on the newly created node, the node with the longest common prefix suffix is ​​selected, and a prefix table from the current node to the node is created.

4. The DNA compression method based on recursive dictionary construction of suffix tree according to claim 1, characterized in that: The step of selecting the best entry from the entries containing biological characteristics and the entries containing non-biological characteristics by screening the entry definitions in the prefix table specifically includes the following steps: It is defined that the number of occurrences of the term is greater than 1, and the number of characters in the term is greater than 1; Divide the substrings that satisfy the terms into terms containing biological characteristics and terms not containing biological characteristics; select the maximum value of the term length and the number of occurrences among the terms containing biological characteristics, and select the maximum value of the term length and the number of occurrences among the terms not containing biological characteristics; In each iteration a term is selected, with preference given to terms containing biological properties.

5. The DNA compression method based on recursive dictionary construction of suffix tree according to claim 1, characterized in that: The step of removing entries from the original DNA sequence, acquiring the entries and position information during the removal, and obtaining a new DNA sequence after the removal specifically includes the following steps: The obtained optimal entries are divided according to whether they contain biological characteristics, to obtain entry and position information; and the entry and position information are stored in the entry stack table.

6. The DNA compression method based on recursive dictionary construction of suffix tree according to claim 1, characterized in that: After removing the term from the original DNA sequence, it also includes: When a new entry appears after the DNA sequence before and after the entry is removed, the entry is checked and removed in the DNA sequence in an incremental manner, specifically: The characters of the original DNA sequence are added to the sequence to be tested in sequence, and the sequence to be tested is checked to see whether there is an entry to be removed; if so, the character added is the last character of the entry to be removed, which is removed from the original DNA sequence. When it is removed, the position information is obtained and stored in the entry stack table, and characters are continued to be added in sequence until the end of the DNA sequence.

Images