Method, system and device for rapid calculation of off-target editing efficiency of the whole genome

Abstract

The present invention provides a method, system and device for quickly calculating the editing efficiency of off-target sites in the whole genome. The present invention improves the calculation efficiency of the editing efficiency of off-target sites in the whole genome, reduces resource consumption, and improves the accuracy of the editing efficiency calculation results by optimizing the off-target site interval file, introducing strict quality control, avoiding unnecessary re-alignment steps, and utilizing multi-threading and random access characteristics.

Description

Technical Field

[0001] The present invention belongs to the field of bioinformatics, and specifically relates to a method, system and device for quickly calculating the editing efficiency of off-target sites in the whole genome. Background Art

[0002] As a technology that can modify the genetic material of species at the genome level, gene editing, especially the CRISPR-Cas9 system, is increasingly widely used in scientific research, medicine and other industries. However, this technology is also accompanied by potential off-target effects. Off-target effects may lead to loss of gene function or genetic mutations, posing a challenge to the safety and accuracy of gene editing. Therefore, it is necessary to evaluate the off-target effects in order to minimize them by adjusting the experimental methods. One of the most important evaluation indicators of off-target effects is the editing efficiency of off-target sites.

[0003] In the existing technology, for off-target sites with a limited number (for example, a few to dozens), the amplicon strategy combined with the second-generation sequencing technology can be used for analysis, and then the sequencing products can be analyzed for editing efficiency using conventional editing efficiency analysis software, such as CRISPRESSO2. However, for edited samples with a large number of off-target sites (for example, tens of thousands of off-target sites predicted by software), it is not suitable to use the amplicon second-generation sequencing method to evaluate the editing efficiency of off-target sites. At this time, whole genome or whole exome sequencing is often used to obtain randomly amplified sequences, and a huge number of off-target sites are evaluated from these sequences. However, there is currently no mature software, only some processes that call third-party software in series, the most classic of which is the WGS mode of CRISPRESSO2.

[0004] CRISPRESSO2 is a software for analyzing the effects of CRISPR gene editing. It can process high-throughput sequencing data and display editing efficiency, specificity, base distribution, and insertion / deletion events. The WGS mode (Whole Genome Sequencing) of CRISPRESSO2 is used to analyze editing events within the genome. It can support the analysis of pooled samples, input whole genome sequencing data before and after CRISPR editing, and then detect and analyze editing events across the genome. This mode allows users to obtain detailed statistical information on editing efficiency and specificity by comparing samples before and after editing. However, when it comes to sequencing data with many off-target sites, the above process is often computationally inefficient, consumes too many resources, lacks better quality control options, and the accuracy of the results is also poor.

[0005] In WGS mode, CRISPRESSO2 performs calculations through the following steps: using alignment tools such as Bowtie2 to align sequencing data with the reference genome; identifying mutations, insertions, and deletions at target sites and potential off-target sites through the alignment results; finally, screening and confirming variants based on the computational analysis model, and calculating mutation rates, insertion and deletion rates, etc.

[0006] Specifically, the WGS mode of the existing CRISPRESSO2 actually uses an amplicon analysis strategy for each off-target site, that is, extracting a fixed-length read near each off-target site, then re-aligning the extracted reads, and then calculating the editing efficiency. Therefore, this mode still has the following problems in practical applications.

[0007] 1. The number of extracted reads is too small. The WGS mode of CRISPRESSO2 requires that the cleavage site of the editing enzyme is located on one side of the gRNA, and the gRNA interval is used as the center to extend the fixed length on both sides to extract the reads. Since the length is a pre-set value, if the length value is set too long, the editing efficiency of the off-target site cannot be accurately calculated because the number of reads that can cover this length interval is too small, and even the program is abnormally interrupted and the calculation of other sites cannot be completed; if the length value is set too short, the extracted ultra-short read subsequence cannot be used for accurate re-alignment (read subsequence re-alignment refers to the process of selecting part of the sequence for re-sequence alignment in multiple sequence alignment, which is usually used to optimize the existing alignment results), and the results are often unreliable and easy to introduce false positive mutations. Therefore, it is necessary to manually debug repeatedly for a long time to obtain a suitable length value in order to at least prevent the program from being interrupted, but in any case, the length of the read segment extracted will be significantly shorter than the original read segment length in the sequencing library, so that the extracted read segment loses the overall sequence information of the original read segment.

[0008] 2. Lack of quality control of the raw reads in the sequencing library, such as checking the alignment quality value of the reads, checking whether the read alignment is abnormal, etc.

[0009] 3. The output result files include too many temporary files. For tens of thousands of off-target sites, tens of GB of temporary files are often generated, taking up too much disk space.

[0010] 4. The run time is too long. For tens of thousands of off-target sites, running the WGS mode of CRISPRESSO2 on the whole genome often takes more than a dozen or even dozens of hours.

[0011] 5. Excessive memory consumption. The memory consumption of CRISPRESSO2's WGS mode is about 5G, and there is room for optimization. Summary of the invention

[0012] In response to the above problems, the present invention proposes a method, system and device for quickly calculating the editing efficiency of off-target sites in the whole genome, aiming to improve the computing efficiency, reduce resource consumption and improve the accuracy of the results by optimizing the off-target site interval file, introducing strict quality control, avoiding unnecessary re-alignment steps, and utilizing multi-threading and random access characteristics.

[0013] The technical solution adopted by the present invention is: a method for quickly calculating the editing efficiency of off-target sites in the whole genome, including: obtaining a comparison file of the whole genome sequencing results of the editing sample and an off-target site interval file; optimizing the off-target interval in the off-target site interval file according to the cutting site of the editing enzyme to obtain an optimized off-target interval; wherein the optimized off-target interval is based on the cutting site as the midpoint; according to the optimized off-target interval, extracting the comparison information of the target read segment and the target read segment from the comparison file of the whole genome sequencing results of the editing sample; wherein at least part of the sequence of the target read segment overlaps with the optimized off-target interval, and the target read segment contains a cutting site; according to the comparison information of the target read segment, quality control is performed on the target read segment to obtain the retained read segment and the comparison information of the retained read segment; according to the comparison information of the retained read segment, the editing efficiency of the off-target sites in the whole genome is calculated, and the result is output.

[0014] Preferably, the method for quality control of the target read segment includes: if the target read segment is an unqualified read segment, it is discarded, otherwise it is retained and used as a retained read segment; the unqualified read segment is any one of a repeated read segment, a non-primary alignment read segment, a low alignment quality read segment, and an end-clipped read segment.

[0015] Preferably, the non-primary alignment reads are any one of secondary alignment reads and subsidiary alignment reads.

[0016] Preferably, the alignment quality of the low alignment quality read segment is lower than a preset alignment quality threshold, and the alignment quality threshold is preferably Q20 to Q30.

[0017] Preferably, the shearing length of the terminal sheared read segment is higher than a preset shearing threshold, and the shearing threshold is preferably 15 bp to 20 bp.

[0018] Preferably, the output results include: editing efficiency, sequencing depth, number of edited reads, number of unedited reads, and number of discarded reads.

[0019] Preferably, the method for extracting target reads and target read alignment information from the edited sample whole genome sequencing result alignment file comprises: randomly accessing the edited sample whole genome sequencing result alignment file, and extracting target reads and target read alignment information therefrom in multiple threads.

[0020] The present invention also provides a system for rapidly calculating the editing efficiency of off-target sites in the whole genome, comprising:

[0021] An input module is used to obtain the comparison file of the whole genome sequencing results of the edited sample and the off-target site interval file;

[0022] An off-target interval optimization module is used to optimize the off-target interval in the off-target site interval file according to the cutting site of the editing enzyme to obtain an optimized off-target interval; wherein the optimized off-target interval takes the cutting site as the midpoint;

[0023] A target read segment extraction module is used to extract the target read segment and the alignment information of the target read segment from the comparison file of the whole genome sequencing result of the edited sample according to the optimized off-target interval; wherein at least a part of the sequence of the target read segment overlaps with the optimized off-target interval, and the target read segment contains a cleavage site;

[0024] A target read segment quality control module is used to perform quality control on the target read segment according to the comparison information of the target read segment, and obtain the reserved read segment and the comparison information of the reserved read segment;

[0025] The editing efficiency calculation module is used to calculate the editing efficiency of off-target sites in the whole genome based on the alignment information of the retained read segments and output the results.

[0026] Preferably, in the target read quality control module, the method for performing quality control on the target read includes: if the target read is an unqualified read, it is discarded, otherwise it is retained and used as a retained read; the unqualified read includes at least one of a repeated read, a non-primary alignment read, a low alignment quality read, and an end-clipped read.

[0027] The present invention also provides a device for rapidly calculating the editing efficiency of off-target sites in the whole genome, comprising a processor and a memory; the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; the memory is used to store a program, and the program is at least used to execute the method for rapidly calculating the editing efficiency of off-target sites in the whole genome.

[0028] Beneficial effects of the present invention:

[0029] 1. The number of extracted reads increases. Compared with the prior art method of extracting reads, the present invention optimizes the off-target site interval and takes the cutting site of the editing enzyme as the midpoint of each interval. The optimized off-target interval evenly covers the sequences on both sides of the cutting site, avoiding the coverage preference problem caused by the gRNA interval as the center. According to the optimized off-target site interval, more target reads that overlap with the optimized off-target interval and contain the cutting site and their alignment information can be extracted from the comparison file of the whole genome sequencing results of the editing sample, thereby effectively improving the accuracy of the editing efficiency calculation results by increasing the number of extracted reads.

[0030] 2. Quality control of reads. Compared with the prior art, the method provided by the present invention performs quality control on the extracted target reads used for subsequent calculation of editing efficiency, discarding reads of non-primary alignments that may cause counting distortion, reads with poor alignment quality values, and reads with suspected structural variation signals at both ends of the alignment process (i.e., reads with long shears at the ends), thereby avoiding unnecessary re-alignment steps and improving the accuracy of the editing efficiency calculation results.

[0031] 3. Improved computational efficiency. The prior art requires the extraction of subsequences after truncating the reads, and then the subsequences are used for realignment, which is a complicated process. The present invention directly uses the original reads extracted from the alignment file of the whole genome sequencing results of the edited sample as the target reads for subsequent calculations of the editing efficiency, which greatly improves the computational efficiency and eliminates a large amount of intermediate file overhead. In addition, the target reads themselves have alignment information of the full length. Compared with the re-alignment using subsequences, the alignment information of the present invention is more accurate, which avoids the introduction of false positive mutations during the calculation process. It also avoids problems such as program crashes and errors caused by fewer reads or abnormal re-alignment, and inability to perform calculations.

[0032] 4. Comprehensive output results. The method provided by the present invention outputs a detailed result report for each off-target interval, including the sequencing depth of each site, the number of edited reads, the number of unedited reads, the editing efficiency, the number of discarded reads, etc., and outputs it as an integrated table that can be viewed directly, without the need for multiple subsequent extraction processes like CRISPRESSO2 to obtain effective results.

[0033] 5. Improved analysis speed and reduced resource consumption. A batch of WGS samples were compared and tested using the method provided by the present invention and the WGS mode of CRISPRESSO2. The results showed that the analysis time of CRISPRESSO2 was about 9 hours, while this software only needed 10 seconds, which was about 3000 times faster; at the same time, the memory consumption of CRISPRESSO2 was about 5G, while this software only needed about 0.2G at most, which was 25 times lower; in terms of disk usage, CRISPRESSO2 required about 20G of disk space for each sample result, while this software only needed about 0.003G for each sample, which was about 6000 times lower. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a flow chart of a method for rapidly calculating genome-wide off-target site editing efficiency provided in an embodiment of the present invention.

[0035] Figure 2 A reference schematic diagram of a target read segment extraction method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following describes the embodiments of the present invention by specific specific embodiments, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied by other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or is formed by a column.

[0037] The method of the embodiment of this specification is executed by a computer device, which may be a terminal device or a server with computing capabilities. The terminal device may be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, desktop computers, etc.

[0038] The computer device may be configured with sequencing data analysis software. The method of the embodiment of this specification may be configured in a whole genome off-target site editing efficiency calculation tool or calculation module, and the above tool or module is integrated in the sequencing data analysis software. After obtaining the whole genome sequencing result comparison file of the edited sample and the off-target site interval file, the above files are analyzed using a computer device, and the relevant whole genome off-target site editing efficiency calculation processing is performed. For details, please refer to a method for quickly calculating the editing efficiency of whole genome off-target sites provided in the embodiment of the present invention ( Figure 1 ). This method mainly improves the calculation efficiency and accuracy of the results by optimizing the off-target site interval and introducing quality control, while greatly reducing the consumption of computing resources. The method specifically includes the following steps:

[0039] Obtain the whole genome sequencing result comparison file and off-target site interval file of the edited sample;

[0040] According to the cutting site of the editing enzyme, the off-target interval in the off-target site interval file is optimized to obtain an optimized off-target interval; wherein the optimized off-target interval takes the cutting site as the midpoint;

[0041] Extracting target reads and alignment information of target reads from the alignment file of whole genome sequencing results of the edited sample according to the optimized off-target interval; wherein at least a portion of the sequence of the target read overlaps with the optimized off-target interval, and the target read contains a cleavage site;

[0042] According to the alignment information of the target read segment, quality control is performed on the target read segment to obtain the retained read segment and alignment information of the retained read segment;

[0043] Based on the alignment information of the retained reads, the editing efficiency of off-target sites in the whole genome is calculated and the results are output.

[0044] First, this example optimizes the off-target interval in the off-target site interval file. Figure 2 In practical applications, to calculate whether a certain off-target interval has been cut, the focus is on comparing a certain window near the cutting site, generally the 5bp sequence on both sides of the cutting site. Therefore, in this embodiment, the cutting site of the editing enzyme is used as the midpoint of each off-target interval, so that the optimized off-target interval can evenly cover the sequences on both sides of the cutting site, which not only avoids the coverage preference problem caused by the gRNA interval as the center, but also can extract more target reads that overlap with the optimized off-target interval and contain the cutting site from the comparison file of the whole genome sequencing results of the editing sample according to the optimized off-target interval. For example, Figure 2In the comparison file of the whole genome sequencing result of the edited sample, there are reads read1 to read4, and only reads read3 and read4 overlap with the optimized off-target interval and contain a cutting site, so reads read3 and read4 are extracted as target reads, and their corresponding comparison information is extracted at the same time. In addition, compared with the method of extracting subsequences and re-aligning using subsequences in the prior art, this embodiment directly uses the original reads extracted from the comparison file of the whole genome sequencing result of the edited sample as the target read for subsequent calculation of the editing efficiency. Since the comparison file of the whole genome sequencing result of the edited sample already has the comparison information of the reads, the target reads also have the original comparison information, and there is no need to use the subsequences for re-alignment. This method not only greatly saves calculation time, greatly improves calculation efficiency, saves a lot of intermediate file overhead, but also avoids the generation of false positive mutations, significantly improves the accuracy of the comparison, and also avoids the problem that the program crashes and reports errors when there are fewer reads or abnormal re-alignment. It can be understood that the longer the length of the read, the higher the specificity of the comparison. For example, an ultra-short subsequence ACG with only 3 bases may have tens of thousands of matches on the genome during the realignment process; however, the target read segment extracted in this embodiment is still the original read segment of the full length (generally 150bp), so there will not be too many matches on the genome, so it has better alignment specificity and more accurate alignment information, thereby improving the accuracy of subsequent editing efficiency calculation results.

[0045] Then, according to the alignment information of the target read, the target read is quality controlled to obtain the reserved read and the alignment information of the reserved read. In a specific embodiment, the method for quality control of the target read includes: if the target read is an unqualified read, it is discarded, otherwise it is retained and used as a reserved read; the unqualified read is any one of a repeated read, a non-primary alignment read, a low alignment quality read, and an end clipping read.

[0046] The repeated reads refer to repeated sequences generated during sequencing due to PCR amplification and other reasons. For example, a read segment appears multiple times in the genome, which may be caused by PCR amplification, and such a read segment may affect the accurate calculation of editing efficiency.

[0047] The non-primary alignment reads generally refer to reads that are not at the best alignment position, such as any of the secondary alignment reads and the subsidiary alignment reads. For example, a read may find similar sequences at multiple positions in the genome, but it has only one best alignment position, and the alignment reads at other positions may be inaccurate. Such non-primary alignment reads generally cause distortion in the read counts, and are therefore discarded as unqualified reads.

[0048] The alignment quality of the low alignment quality read is lower than the preset alignment quality threshold. The reads in the sequencing library are matched and aligned with the reference sequence to obtain the alignment information of the reads. Due to the existence of sequencing errors and the presence of many complex sequences (such as repeated regions) in the reference genome, the alignment between the reads and the reference genome is not absolutely correct, so the alignment needs to have a certain amount of error tolerance to find the source of the sequence itself. However, if there are many sequencing errors or many variations caused by editing, the alignment quality value will decrease. On the other hand, if the same read is aligned to several similar regions of the genome, these alignments are often unreliable, which will also cause the alignment quality value to decrease. Therefore, in some embodiments, the alignment quality threshold is generally set in a range where the probability of unreliable alignment is low, such as Q20 to Q30. Among them, Q20 represents a 1% probability of alignment error, and Q30 represents a 1 / 1000 probability of alignment error.

[0049] The shearing length of the terminal sheared read is higher than the preset shearing threshold. During the alignment process, if a read cannot be continuously matched to the same region, the read will be sheared and aligned. For example, part of a read matches to region A, and the other part matches to region B, and region A and region B may be located at non-adjacent and distant positions on the same chromosome or on different chromosomes, and the length of the sequence aligned to region A and region B is long (generally longer than 20bp), then the read may have structural variation or sequencing errors, and it is generally not caused by editing, so such terminal sheared reads should be discarded. Therefore, in some embodiments, the shearing threshold is generally set at 15bp to 20bp.

[0050] Finally, based on the alignment information of the retained reads, the editing efficiency of off-target sites in the whole genome is calculated and the results are output. In a specific embodiment, the output results include: editing efficiency, sequencing depth, number of edited reads, number of unedited reads, and number of discarded reads. The above output results are integrated into the same table and reported, which can be viewed directly without the need for multiple subsequent extraction processes like CRISPRESSO2 to obtain effective results. The edited reads have editing enzyme cleavage sites, and insertion / deletion mutations are generated after the action of the editing enzyme. The unedited reads have editing enzyme cleavage sites, but no insertion / deletion mutations are generated after the action of the editing enzyme. The sum of the number of edited reads and the number of unedited reads is the number of reads with the editing enzyme cleavage site.

[0051] In a specific embodiment, the target reads and the comparison information of the target reads are extracted from the comparison file of the whole genome sequencing results of the edited sample in a multi-threaded manner through random access. This embodiment utilizes the characteristics of multi-threading and random access of the input comparison file to parallelize the massive off-target sites, and can quickly generate and output the results. The report is in the form of an information table including the sequencing depth, number of edited reads, number of normal reads, editing efficiency and other output results of each site, which can be viewed directly without the need for multiple subsequent extraction processes like CRISPRESSO2 to obtain effective results.

[0052] In a specific application example, a batch of WGS samples were compared and tested using the method for rapid calculation of whole-genome off-target site editing efficiency provided by one or more of the above embodiments and the WGS mode of CRISPRESSO2. The analysis time of CRISPRESSO2 required about 9 hours, while the present method only required 10 seconds, which was 3000 times faster. At the same time, the memory consumption of CRISPRESSO2 was about 5 GB, while the present method only required a maximum of about 0.2 GB, reducing the memory consumption by 25 times. In terms of disk usage, each sample result of CRISPRESSO2 required about 20 GB of disk space, while each sample result of the present method only required about 0.003 GB, which was about 6000 times less.

[0053] In a specific embodiment, a system for rapid calculation of genome-wide off-target site editing efficiency is also provided, including:

[0054] An input module is used to obtain the comparison file of the whole genome sequencing results of the edited sample and the off-target site interval file;

[0055] An off-target interval optimization module is used to optimize the off-target interval in the off-target site interval file according to the cutting site of the editing enzyme to obtain an optimized off-target interval; wherein the optimized off-target interval takes the cutting site as the midpoint;

[0056] A target read segment extraction module is used to extract the target read segment and the alignment information of the target read segment from the comparison file of the whole genome sequencing result of the edited sample according to the optimized off-target interval; wherein at least a part of the sequence of the target read segment overlaps with the optimized off-target interval, and the target read segment contains a cleavage site;

[0057] A target read segment quality control module is used to perform quality control on the target read segment according to the comparison information of the target read segment, and obtain the reserved read segment and the comparison information of the reserved read segment;

[0058] The editing efficiency calculation module is used to calculate the editing efficiency of off-target sites in the whole genome based on the alignment information of the retained read segments and output the results.

[0059] The system is implemented as a C++ software that only requires one command line run to complete the calculation of the editing efficiency of a large number of off-target sites, without the need for support from third-party software and various libraries.

[0060] In a specific embodiment, in the target read segment quality control module, the method for performing quality control on the target read segment includes: if the target read segment is an unqualified read segment, it is discarded, otherwise it is retained and used as a retained read segment; the unqualified read segment includes at least one of a repeated read segment, a non-primary alignment read segment, a low alignment quality read segment, and an end-clipped read segment.

[0061] In a specific embodiment, a device for rapidly calculating the editing efficiency of off-target sites in the whole genome is also provided, comprising a processor and a memory; the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; the memory is used to store a program, and the program is at least used to execute the method for rapidly calculating the editing efficiency of off-target sites in the whole genome.

[0062] The embodiments described above are merely descriptions of preferred implementation modes of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope of the present invention.

Claims

1. A method for rapidly calculating the editing efficiency of off-target sites in the whole genome, characterized in that: include: Obtain the whole genome sequencing result comparison file and off-target site interval file of the edited sample; According to the cutting site of the editing enzyme, the off-target interval in the off-target site interval file is optimized to obtain an optimized off-target interval; wherein the optimized off-target interval takes the cutting site as the midpoint; Extracting target reads and alignment information of target reads from the alignment file of whole genome sequencing results of the edited sample according to the optimized off-target interval; wherein at least a portion of the sequence of the target read overlaps with the optimized off-target interval, and the target read contains a cleavage site; According to the alignment information of the target read segment, quality control is performed on the target read segment to obtain the retained read segment and the alignment information of the retained read segment; Based on the alignment information of the retained reads, the editing efficiency of off-target sites in the whole genome is calculated and the results are output.

2. The method according to claim 1, characterized in that The method for performing quality control on the target read segment comprises: If the target read is a non-qualified read, it is discarded, otherwise it is retained and used as a retained read; The unqualified reads are any one of duplicate reads, non-primary alignment reads, low alignment quality reads, and end clipping reads.

3. The method according to claim 2, characterized in that The non-primary alignment read segment is any one of a secondary alignment read segment and an auxiliary alignment read segment.

4. The method according to claim 2, characterized in that The alignment quality of the low alignment quality read segment is lower than a preset alignment quality threshold.

5. The method according to claim 2, characterized in that The shearing length of the terminal sheared read segment is higher than a preset shearing threshold.

6. The method according to claim 1, wherein the output result comprises: Editing efficiency, sequencing depth, number of edited reads, number of unedited reads, number of discarded reads.

7. The method according to claim 1, wherein the method of extracting the target read segment and the comparison information of the target read segment from the comparison file of the whole genome sequencing result of the edited sample comprises: The whole genome sequencing result comparison file of the sample is randomly accessed and edited, and the target read segment and the comparison information of the target read segment are extracted from it in multiple threads.

8. A rapid calculation system for off-target editing efficiency of the whole genome, characterized by: include: An input module is used to obtain the comparison file of the whole genome sequencing results of the edited sample and the off-target site interval file; An off-target interval optimization module is used to optimize the off-target interval in the off-target site interval file according to the cutting site of the editing enzyme to obtain an optimized off-target interval; wherein the optimized off-target interval takes the cutting site as the midpoint; A target read segment extraction module is used to extract the target read segment and the alignment information of the target read segment from the comparison file of the whole genome sequencing result of the edited sample according to the optimized off-target interval; wherein at least a part of the sequence of the target read segment overlaps with the optimized off-target interval, and the target read segment contains a cleavage site; A target read segment quality control module is used to perform quality control on the target read segment according to the comparison information of the target read segment, and obtain the reserved read segment and the comparison information of the reserved read segment; The editing efficiency calculation module is used to calculate the editing efficiency of off-target sites in the whole genome based on the alignment information of the retained read segments and output the results.

9. The system according to claim 8, characterized in that In the target read segment quality control module, a method for performing quality control on the target read segment includes: If the target read is a non-qualified read, it is discarded, otherwise it is retained and used as a retained read; The unqualified reads include at least one of duplicate reads, non-primary alignment reads, low alignment quality reads, and end clipping reads.

10. A rapid calculation device for off-target editing efficiency of the whole genome, characterized in that: It comprises a processor and a memory; the processor and the memory are connected via a communication bus; wherein the processor is used to call and execute a program stored in the memory; the memory is used to store a program, and the program is at least used to execute the method for rapidly calculating the editing efficiency of off-target sites in the whole genome as described in any one of claims 1 to 7.

Images