Methods for predicting possible off-targets in a gene editing process

Abstract

The present application relates to a method for predicting possible off-targets in a gene editing process (e.g., a genome editing process) using a gene editing system.

Description

[Technical field]

[0001] The present application relates to a method for predicting possible off-targets in a gene editing process. The gene editing process may be, for example, a genomic DNA editing process using a CRISPR / Cas gene editing system. [Background technology]

[0002] Since 2005, investigational new drug (IND) applications have been submitted for various gene editors (e.g., zinc finger nuclease-based, TALEN-based, and CRISPR nuclease-based gene editors) (see reference [Mullard, Asher. "Gene-editing pipeline takes off." Nature Reviews Drug Discovery 19.6(2020):367-373.]). Unlike other drugs, such as chemicals or antibodies, which are generally associated with reversible side effects, the action of genome editing drugs is permanent. That is, it is particularly important to identify off-target sites throughout the entire genome for genome editing drugs, since the action that frequently occurs at undesired positions in the genome editing process (i.e., off-target action) poses important safety concerns. To ascertain information about possible off-target action in the genome editing process, many researchers have developed various methods to predict off-target action in the entire genome through various approaches.

[0003] However, currently developed off-target prediction tools (systems) have several limitations. For example, cell-based methods have problems such as missing true off-target sites, while in vitro and in silico methods have problems such as showing an excessive number of false positive data points. [Disclosure] [Technical issues]

[0004] In the genome editing process using gene editing tools (e.g., CRISPR / Cas gene editing system), there may be off-target problems. Such off-target effects cause strong side effects. The embodiment of the present application provides a method for predicting off-targets that occur in genome editing process. [Technical solution]

[0005] The present application provides a method for predicting off-targets that will occur in a gene editing process using a gene editing system.

[0006] One embodiment of the present application is a method for identifying information about off-targets generated in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) preparing a starting composition comprising a Cas protein, a guide RNA, and a cell; (ii) obtaining the composition to be analyzed by physically disrupting the cells, wherein, by physically disrupting the cells, genomic DNA is contacted with the Cas / gRNA complex formed by the Cas protein and the guide RNA, whereby the genomic DNA is cleaved at one or more cleavage sites; and (iii) obtaining information about the cleavage site through analysis of the composition to be analyzed; The present invention provides a method comprising:

[0007] In one particular embodiment, the physically disrupting the cells may comprise passing the cells through a filter having pores, the average diameter of the pores in the filter being smaller than the size of the cells.

[0008] In one particular embodiment, the force that causes the cells to pass through the filter can be pressure.

[0009] In one particular embodiment, the average diameter of the pores of the filter may be between 5 and 15 μm.

[0010] In one particular embodiment, the physically disrupting the first cells may be accomplished through the use of an extruder equipped with a filter having holes.

[0011] In one particular embodiment, the average diameter of the pores of the filter contained in the extruder may be smaller than the size of the cells.

[0012] In one particular embodiment, the average diameter of the pores of the filter may be between 5 and 15 μm.

[0013] In one particular embodiment, the information about the cleavage site is the location on the genomic DNA of each of said one or more cleavage sites; a cleavage score for each of said one or more cleavage sites; and Number of cleavage sites may include one or more of:

[0014] In one particular embodiment, the method further comprises: (iv) identifying information about potential off-targets from the information about the cleavage sites obtained from (iii); may further be provided.

[0015] In one particular embodiment, the information about off-target candidates comprises: the location on genomic DNA of each of one or more off-target candidates; an off-target prediction score for each of one or more off-target candidates; and Number of predicted off-target candidates may include one or more of:

[0016] In one particular embodiment, the step of analyzing the composition to be analyzed may comprise analyzing the cleaved genomic DNA contained in the composition to be analyzed through sequencing.

[0017] In one particular embodiment, the step of analyzing the composition to be analyzed may comprise analyzing the cleaved genomic DNA contained in the composition to be analyzed via a PCR-based method.

[0018] In a particular embodiment, the physical disruption of the cells may disrupt the membrane structures of the cells, including the cell membrane, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA derived from the cells.

[0019] In a particular embodiment, the physical disruption of the cells may disrupt membranous structures, including the nuclear membrane, of the cells, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA derived from the cells.

[0020] In one particular embodiment, the method further comprises: identifying a predetermined CRISPR / Cas genome editing system, wherein the step of identifying the predetermined CRISPR / Cas genome editing system is performed prior to (i). may further be provided.

[0021] In one particular embodiment, wherein the predetermined CRISPR / Cas genome editing system comprises the use of a predetermined guide RNA having a predetermined guide sequence, wherein the predetermined guide sequence and the guide sequence of the guide RNA may be the same.

[0022] In a specific embodiment, wherein the predetermined CRISPR / Cas genome editing system comprises the use of a predetermined cell, wherein the predetermined cell and the cell may be the same.

[0023] In a particular embodiment, the composition to be analyzed may comprise cleaved genomic DNA, where the genomic DNA from physically disrupted cells has been cleaved by the Cas / gRNA complex.

[0024] In one particular embodiment, the starting composition comprises a concentration of the Cas protein of 4000 nM or more and 6000 nM or less.

[0025] In one particular embodiment, the concentration of the guide RNA included in the starting composition may be 4000 nM or more and 6000 nM or less.

[0026] In a particular embodiment, the starting composition may have a concentration of the Cas / gRNA complexes of 4000 nM or more and 6000 nM or less.

[0027] In one particular embodiment, the concentration of the cells in the starting composition is 1×10 7 It may be in cells / mL.

[0028] In one particular embodiment, obtaining the composition to be analyzed comprises: The method may further comprise the step of incubating the composition obtained through said disrupting said cells.

[0029] In one particular embodiment, obtaining the composition to be analyzed comprises: The method may further comprise the step of removing RNA from the composition obtained through said disrupting said cells.

[0030] In one particular embodiment, obtaining the composition to be analyzed may further comprise purifying DNA from the composition obtained through the disruption of the cells.

[0031] One embodiment of the present application is a method for identifying information about off-targets generated in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) charging a starting composition comprising a Cas protein, a guide RNA, and cells into a first vessel of an extruder; (ii) using the extruder (a) applying pressure to the first vessel to move components of the starting composition from a first vessel of the extruder to a second vessel of the extruder; wherein the components of the starting composition are forced by the applied pressure through a filter having holes located between the first vessel and the second vessel of the extruder, thereby resulting in a mixture contained in the second vessel; Here, the cells, which are components larger in size than the diameter of the pores of the filter, are broken by the applied pressure and pass through the pores of the filter, wherein an environment is created in which the genomic DNA can come into contact with the Cas protein and guide RNA through physical disruption of the cells; This brings the genomic DNA into contact with the Cas / gRNA complex, Thereby, contacting the genomic DNA with a Cas / gRNA complex, This causes the genomic DNA to be cleaved at one or more cleavage sites. obtaining the composition to be analyzed; and (iii) analyzing the composition to obtain information about the cleavage site. The present invention provides a method comprising:

[0032] In one particular embodiment, the pressure applied to the first container is generated through the process of pushing a piston designed to apply the pressure to the first container in a direction toward the first container and the filter.

[0033] One embodiment of the present application is a method for identifying information about off-targets that arise in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) charging a starting composition comprising a Cas protein, a guide RNA, and cells into a first vessel of an extruder; (ii) using the extruder (a) applying pressure to the first vessel to move components of the starting composition from a first vessel of the extruder to a second vessel of the extruder; wherein the components of the starting composition are forced by the applied pressure through a filter having holes located between the first vessel and the second vessel of the extruder, thereby resulting in the mixture being contained in the second vessel; (b) applying pressure to the second vessel to transfer components of the mixture contained in the second vessel from the second vessel to the first vessel; wherein the components of the mixture contained in the second container are passed through the filter having a hole located between the first container and the second container by the applied pressure, and thus transferred from the second container to the first container, whereby the mixture transferred from the second container by passing through the filter by the pressure is received in the first container; and (c) repeatedly performing (a) and (b) a predetermined number of times; wherein the predetermined number of times is counted in increments of 0.5, where 0.5 indicates execution of a single process of (a) or (b); Here, the cells, which are components larger in size than the diameter of the pores of the filter, are broken by the applied pressure and pass through the pores of the filter, wherein an environment is created in which the genomic DNA can come into contact with the Cas protein and the guide RNA through physical disruption of the cells; This brings the genomic DNA into contact with the Cas / gRNA complex, This causes the genomic DNA to be cleaved at one or more cleavage sites. obtaining the composition to be analyzed; and (iii) analyzing the composition to obtain information about the cleavage site. The present invention provides a method comprising:

[0034] In one particular embodiment, the pressure applied to the first container is generated through a process of pushing a piston designed to apply the pressure to the first container in a direction toward the first container and the filter, and the pressure applied to the second container is generated through a process of pushing a piston designed to apply the pressure to the second container in a direction toward the second container and the filter. [Advantageous Effects]

[0035] The present application provides a method for predicting off-targets that may occur in a gene (e.g., genome) editing process. The present application provides a method for identifying candidates for off-targets that may occur in a genome editing process. The present application provides a method for predicting off-targets that can be performed more simply. The present application's method for predicting off-targets has the advantages of in vitro-based off-target prediction methods and cell-based off-target prediction methods. The present application's off-target prediction method shows a lower false positive rate. The present application's off-target prediction method shows a lower missed rate. That is, when the present application's off-target prediction method is used, it is possible to easily and accurately predict off-targets that may occur in a genome editing process. [Brief description of the drawings]

[0036] [Figure 1] Three categories of off-target prediction methods are shown: cell-based, in vitro, and in silico.

[0037] [Diagram 2] FIG. 1 is a schematic diagram showing a method for predicting off-targets provided by one embodiment of the present application.

[0038] [Diagram 3] Comparison results for off-target candidates predicted through different off-target prediction methods (Digenome-seq, Extru-seq, GUIDE-seq, and in silico) are shown. Comparison experiments for off-target prediction systems were performed using sgRNAs targeting human PCSK9 and human albumin, respectively.

[0039] [Figure 4] Comparison results for off-target candidates predicted through different off-target prediction methods (Digenome-seq, Extru-seq, GUIDE-seq, and in silico) are shown. Comparison experiments for the off-target prediction systems were performed using sgRNAs targeting mouse PCSK9 and mouse albumin, respectively.

[0040] [Diagram 5] Figure 1 shows the validation rates of the top off-target sites predicted by in silico methods, GUIDE-seq, Digenome-seq, and Extru-seq. Results are shown for sgRNAs targeting human PCSK9, human albumin targeting sgRNAs targeting mouse PCSK9, and mouse albumin targeting sgRNAs.

[0041] [Figure 6] Comparison results for off-target candidates predicted through different off-target prediction methods (Digenome-seq, Extru-seq, GUIDE-seq, and DIG-seq) are shown. Comparison experiments for off-target prediction systems were performed using sgRNAs targeting FANCF and VEGFA, respectively.

[0042] [Figure 7] Comparison results for off-target candidates predicted through different off-target prediction methods (Digenome-seq, Extru-seq, GUIDE-seq, and DIG-seq) are shown. A comparison experiment for the off-target prediction systems was performed using sgRNA targeting HBB.

[0043] [Figure 8] Figure 1 shows the validation rates of the top off-target sites predicted by DIG-seq, GUIDE-seq, Digenome-seq, and Extru-seq. Results are shown for sgRNAs targeting FANCF, sgRNAs targeting VEGFA, and sgRNAs targeting HBB.

[0044] [Figure 9] The comparative results of different off-target prediction methods are shown, analyzed through the intersection of Venn diagrams (Figures 3 and 4, and Figures 6 and 7), respectively.

[0045] [Figure 10] Validation results and a comparison of off-target results predicted by GUIDE-seq and Extru-seq are shown. Figure 10 (a) shows results associated with sgRNAs targeting human PCSK9. Figure 10 (b) shows results associated with sgRNAs targeting human albumin.

[0046] [Figure 11] Validation results and a comparison of off-target results predicted by GUIDE-seq and Extru-seq are shown. (c) of FIG. 11 shows results associated with sgRNAs targeting mouse PCSK9. (d) of FIG. 11 shows results associated with sgRNAs targeting mouse albumin.

[0047] [Figure 12] Validation results and a comparison of off-target results predicted by GUIDE-seq and Extru-seq are shown. (e) of FIG. 12 shows results associated with sgRNAs targeting human FANCF. (f) of FIG. 12 shows results associated with sgRNAs targeting human VEGFA.

[0048] [Figure 13] Figure 13(g) shows the validation results and a comparison of the results for off-targets predicted by GUIDE-seq and Extru-seq. Figure 13(g) shows the results related to sgRNAs targeting human HBB.

[0049] [Figure 14] The missed target rates for the GUIDE-seq and Extru-seq off-target prediction methods are shown, calculated based on the validation results.

[0050] [Figure 15] The distribution of the number of off-target mismatches missed by GUIDE-seq confirmed based on the validation results is shown.

[0051] [Figure 16] The ROC curves for each off-target prediction method are shown. Figure 16 (a) shows the results associated with sgRNA targeting human PCSK9. Figure 16 (b) shows the results associated with sgRNA targeting human albumin.

[0052] [Figure 17] The ROC curves for each off-target prediction method are shown. (c) of Figure 17 shows the results associated with sgRNA targeting mouse PCSK9. (d) of Figure 17 shows the results associated with sgRNA targeting mouse albumin.

[0053] [Figure 18] The ROC curves for each off-target prediction method are shown. Figure 18(e) shows the results associated with sgRNAs targeting human FANCF. Figure 18(f) shows the results associated with sgRNAs targeting human VEGFA. Figure 18(g) shows the results associated with sgRNAs targeting human HBB.

[0054] [Figure 19] The AUCs calculated using the ROC curve data from Figures 16 to 18 are shown. The AUCs were calculated for each of GUIDE-seq, Digenome-seq, Extru-seq, CROP, CFD, and DIG-seq.

[0055] [Figure 20] 1 shows the results and experimental conditions of experiments performed to determine the optimized conditions for the average pore size of the filter, the Cas9 RNP concentration of the mixture, and the number of cells in Extru-seq. [Figure 21] 1 shows the results and experimental conditions of experiments performed to determine the optimized conditions for the average pore size of the filter, the Cas9 RNP concentration of the mixture, and the number of cells in Extru-seq.

[0056] [Figure 22] Figure 22 shows the cleavage rates for on- and off-target sites recognized by sgRNAs targeting human PCSK9 sites as measured by quantitative PCR (qPCR). [Figure 23] Figure 23 shows the cleavage rates for on- and off-target sites recognized by sgRNAs targeting human PCSK9 sites as measured by quantitative PCR (qPCR).

[0057] [Figure 24] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Diagram 25] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Figure 26] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Figure 27] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Figure 28] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Figure 29] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns. [Diagram 30] Shown is Extru-seq WGS data analyzed using IGV to identify cleavage patterns.

[0058] [Diagram 31] Cleavage rates of the seven on-target sites for each target obtained through qPCR and manual calculations based on IGV analysis of WGS data are shown.

[0059] [Diagram 32] 1 shows the results of dip-sequencing on the non-cleaved population, which was performed to confirm the extent of NHEJ generation after the extrusion process of Extru-seq.

[0060] [Diagram 33] The results of the cleavage rate by SCR7 treatment, which was performed to confirm the extent of NHEJ generation after the Extru-seq extrusion process, are shown.

[0061] [Diagram 34]Figure 34 shows the sequence read results for GUIDE-seq obtained from HEK293T using sgRNA targeting PCSK9. [Diagram 35] Figure 35 shows the sequence read results for GUIDE-seq for off-target candidates predicted through GUIDE-seq. Figure 35 shows the sequence read results for GUIDE-seq obtained from HEK293T using sgRNA targeting PCSK9. [Diagram 36] Figure 36 shows sequence read results for off-target candidates predicted through GUIDE-seq. Figure 36 shows sequence read results for GUIDE-seq obtained from HEK293T using sgRNA targeting albumin. [Figure 37] Figure 37 shows sequence read results for off-target candidates predicted through GUIDE-seq. Figure 37 shows sequence read results for GUIDE-seq obtained from HEK293T using sgRNA targeting albumin. [Figure 38] Figure 38 shows the sequence read results for GUIDE-seq for off-target candidates predicted through GUIDE-seq. Figure 38 shows the sequence read results for GUIDE-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Figure 39] Figure 39 shows sequence read results for off-target candidates predicted through GUIDE-seq. Figure 39 shows sequence read results for GUIDE-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Diagram 40] Figure 40 shows sequence read results for off-target candidates predicted through GUIDE-seq. Figure 40 shows sequence read results for GUIDE-seq obtained from NIH-3T3 using sgRNA targeting albumin. [Diagram 41] Figure 41 shows sequence read results for off-target candidates predicted through GUIDE-seq. Figure 41 shows sequence read results for GUIDE-seq obtained from NIH-3T3 using sgRNA targeting albumin.

[0062] [Diagram 42] Figure 41 shows the Manhattan plot results of off-target candidates predicted through Digenome-seq. The Y-axis shows the DNA cleavage score. Figure 42 shows the Manhattan plot results of Digenome-seq obtained from HEK293T using sgRNA targeting PCSK9. [Diagram 43] Figure 43 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 43 shows Manhattan plot results of Digenome-seq obtained from HEK293T using sgRNA targeting PCSK9. [Diagram 44] Figure 43 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 44 shows Manhattan plot results of Digenome-seq obtained from HEK293T using sgRNA targeting albumin. [Diagram 45] Figure 45 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 45 shows Manhattan plot results of Digenome-seq obtained from HEK293T using sgRNA targeting albumin. [Diagram 46]Figure 46 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 46 shows Manhattan plot results of Digenome-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Figure 47] Figure 47 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 47 shows Manhattan plot results of Digenome-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Figure 48] Figure 48 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 48 shows Manhattan plot results of Digenome-seq obtained from NIH-3T3 using sgRNA targeting albumin. [Figure 49] Figure 49 shows Manhattan plot results of off-target candidates predicted through Digenome-seq. Y-axis shows DNA cleavage score. Figure 49 shows Manhattan plot results of Digenome-seq obtained from NIH-3T3 using sgRNA targeting albumin.

[0063] [Figure 50] Figure 50 shows Manhattan plot results of off-target candidates predicted through Extru-seq. Y-axis shows DNA cleavage score. Figure 50 shows Manhattan plot results of Extru-seq obtained from HEK293T using sgRNA targeting PCSK9. [Figure 51]Figure 51 shows Manhattan plot results of off-target candidates predicted through Extru-seq. Y-axis shows DNA cleavage score. Figure 51 shows Manhattan plot results of Extru-seq obtained from HEK293T using sgRNA targeting PCSK9. [Figure 52] Figure 51 shows the Manhattan plot results of off-target candidates predicted through Extru-seq. The Y-axis shows the DNA cleavage score. Figure 52 shows the Manhattan plot results of Extru-seq obtained from HEK293T using sgRNA targeting albumin. [Figure 53] Figure 53 shows Manhattan plot results of off-target candidates predicted through Extru-seq. Y-axis shows DNA cleavage score. Figure 53 shows Manhattan plot results of Extru-seq obtained from HEK293T using sgRNA targeting albumin. [Figure 54] Figure 53 shows the Manhattan plot results of off-target candidates predicted through Extru-seq. The Y-axis shows the DNA cleavage score. Figure 54 shows the Manhattan plot results of Extru-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Figure 55] Figure 55 shows Manhattan plot results of off-target candidates predicted through Extru-seq. Y-axis shows DNA cleavage score. Figure 55 shows Manhattan plot results of Extru-seq obtained from NIH-3T3 using sgRNA targeting PCSK9. [Figure 56] Figure 55 shows Manhattan plot results of off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 56 shows Manhattan plot results of Extru-seq obtained from NIH-3T3 using sgRNA targeting albumin. [Figure 57]Figure 57 shows Manhattan plot results of off-target candidates predicted through Extru-seq. Y-axis shows DNA cleavage score. Figure 57 shows Manhattan plot results of Extru-seq obtained from NIH-3T3 using sgRNA targeting albumin.

[0064] [Figure 58] The results are shown in relation to the score based on the number of mismatches between on-target and off-target sequences predicted using GUIDE-seq (a score calculated from sequence read count results).

[0065] [Figure 59] Results are shown in relation to a score based on the number of mismatches between on-target and off-target (Manhattan plot cleavage score) predicted using Digenome-seq.

[0066] [Figure 60] The results are shown in relation to a score based on the number of mismatches between on-target and off-target (CROP score) predicted using an in silico system.

[0067] [Figure 61] The results are shown in relation to a score based on the number of mismatches between on-target and off-target (CFD score) predicted using an in silico system.

[0068] [Figure 62] Results are shown in relation to a score based on the number of mismatches between on-target and off-target (Manhattan plot cleavage score) predicted using Extru-seq.

[0069] [Figure 63] 1 shows results regarding indel formation frequency with subretinal and systemic injections. [Figure 64] 1 shows results regarding indel formation frequency with subretinal and systemic injections.

[0070] [Figure 65] Figure 65 shows sequence read results for GUIDE-seq obtained from HeLa cells using sgRNA targeting FANCF. [Figure 66] Figure 66 shows the sequence read results for the GUIDE-seq obtained from HeLa cells using sgRNA targeting VEGFA. Figure 67 shows the sequence read results for GUIDE-seq obtained from HeLa cells using sgRNA targeting HBB. [Figure 67] Figure 67 shows sequence read results for GUIDE-seq obtained from HeLa cells using sgRNA targeting HBB.

[0071] [Figure 68] Figure 68 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 69 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting FANCF. [Figure 69] Figure 69 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 69 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting FANCF. [Figure 70]Figure 70 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 70 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting VEGFA. [Figure 71] Figure 71 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 71 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting VEGFA. [Figure 72] Figure 72 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 72 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting HBB. [Figure 73] Figure 73 shows Manhattan plot results for off-target candidates predicted through Extru-seq. The Y-axis shows DNA cleavage score. Figure 73 shows Manhattan plot results of Extru-seq obtained from HeLa cells using sgRNA targeting HBB.

[0072] [Figure 74] Figure 73 shows a Venn diagram comparing Extru-seq results obtained from MSCs and HEK293T cells. Figure 74 shows results related to sgRNAs targeting human PCSK9. [Figure 75] Figure 75 shows a Venn diagram comparing Extru-seq results obtained from MSCs and HEK293T cells. Figure 75 shows results related to sgRNA targeting human albumin.

[0073] [Figure 76]Shown are p-values ​​obtained by normalized rank sum test for each pair of off-target prediction methods for sgRNAs targeting PCSK9 and albumin in MSCs and HEK293T cells.

[0074] [Figure 77] Results are shown for off-target sites that were manually validated using IGV. [Figure 78] Results are shown for off-target sites that were manually validated using IGV. [Figure 79] Results are shown for off-target sites that were manually validated using IGV. [Figure 80] Results are shown for off-target sites that were manually validated using IGV. [Figure 81] Results are shown for off-target sites that were manually validated using IGV. [Figure 82] Results are shown for off-target sites that were manually validated using IGV. [Figure 83] Results are shown for off-target sites that were manually validated using IGV. [Figure 84] Results are shown for off-target sites that were manually validated using IGV. [Figure 85] Results are shown for off-target sites that were manually validated using IGV. [Figure 86] Results are shown for off-target sites that were manually validated using IGV. [Figure 87] Results are shown for off-target sites that were manually validated using IGV. [Figure 88] Results are shown for off-target sites that were manually validated using IGV. [Figure 89]Results are shown for off-target sites that were manually validated using IGV. [Figure 90] Results are shown for off-target sites that were manually validated using IGV. [Figure 91] Results are shown for off-target sites that were manually validated using IGV. [Figure 92] Results are shown for off-target sites that were manually validated using IGV. [Figure 93] Results are shown for off-target sites that were manually validated using IGV. [Figure 94] Results are shown for off-target sites that were manually validated using IGV. [Figure 95] Results are shown for off-target sites that were manually validated using IGV. [Figure 96] Results are shown for off-target sites that were manually validated using IGV. [Figure 97] Results are shown for off-target sites that were manually validated using IGV. [Figure 98] Results are shown for off-target sites that were manually validated using IGV. [Figure 99] Results are shown for off-target sites that were manually validated using IGV. [Figure 100] Results are shown for off-target sites that were manually validated using IGV. [Figure 101] Results are shown for off-target sites that were manually validated using IGV. [Figure 102] Results are shown for off-target sites that were manually validated using IGV. [Figure 103] Results are shown for off-target sites that were manually validated using IGV. [Figure 104] Results are shown for off-target sites that were manually validated using IGV. [Figure 105] Results are shown for off-target sites that were manually validated using IGV. [Fig. 106] Results are shown for off-target sites that were manually validated using IGV. [Figure 107] Results are shown for off-target sites that were manually validated using IGV. [Figure 108] Results are shown for off-target sites that were manually validated using IGV. [Fig. 109] Results are shown for off-target sites that were manually validated using IGV. [Figure 110] Results are shown for off-target sites that were manually validated using IGV. [Figure 111] Results are shown for off-target sites that were manually validated using IGV. [Figure 112] Results are shown for off-target sites that were manually validated using IGV. [Figure 113] Results are shown for off-target sites that were manually validated using IGV. [Fig. 114] Results are shown for off-target sites that were manually validated using IGV. [Fig. 115] Results are shown for off-target sites that were manually validated using IGV. [Fig. 116] Results are shown for off-target sites that were manually validated using IGV.

[0075] [Fig. 117] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Fig. 118] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 119] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 120] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 121] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 122] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 123] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Figure 124] Results are shown for false positive off-target candidates that were manually excluded using IGV from Digenome-seq and Extru-seq WGS data. [Fig. 125] The results show the false positive off-target candidates manually removed using IGV from Digenome-seq and Extru-seq WGS data.

[0076] [Terminology definition] nucleic acid The term "nucleic acid" as used herein means a partial region or the entire molecule of a molecule consisting of DNA (double-stranded or single-stranded), RNA (double-stranded or single-stranded), or a hybrid of DNA and RNA (double-stranded or single-stranded). Nucleic acid is used to mean a collection of nucleotides (a partial region or the entire molecule), but is not limited thereto. The term "nucleic acid" or "nucleic acid region" may be used to refer to a partial region of a molecule. The term "nucleic acid" or "nucleic acid area" may be used to refer to the entire molecule. The term "nucleic acid" should be interpreted appropriately depending on the context, and the contents of each context, including the explanation of the term "nucleic acid", are helpful in understanding the meaning of the term "nucleic acid". Furthermore, the above terms include all meanings recognized by those skilled in the art and may be interpreted appropriately depending on the context.

[0077] "Connected" or "Connected" The term "linked" or "connected" as used herein refers to two or more elements present in a single conceptualizable structure being directly or indirectly (e.g., via a different element such as a linker) connected, and it is not intended that other additional elements cannot be present between these two or more elements. For example, a statement such as "element B connected to element A" is intended to include both cases where one or more other elements are intervening between element A and element B (i.e., element A is connected to element B via one or more third elements), and cases where no other elements are intervening between element A and element B (i.e., element A and element B are directly connected), and should not be construed as being limiting.

[0078] Sequence identity The term "sequence identity" as used herein is a term used in relation to the degree of similarity between two or more nucleotide sequences. For example, the term "sequence identity" is used with a reference sequence and a term referring to a ratio (e.g., percentage). For example, the term "sequence identity" can be used to describe a sequence that is similar or substantially the same as a reference nucleotide. When "a sequence having 90% or higher sequence identity with sequence A" is described, the reference sequence used is sequence A. For example, the percentage of sequence identity can be calculated by aligning the sequence subjected to the measurement of the percentage of sequence identity with the reference sequence, and the percentage of sequence identity may be calculated by including all mismatches, deletions, and insertions at one or more nucleotides. The method of calculating and / or determining the percentage of sequence identity is not otherwise limited and may be calculated and / or determined through a rational method or algorithm that can be used by a person skilled in the art.

[0079] Representation of amino acid sequences Unless otherwise stated, the amino acid sequences herein are written in the N-terminal to C-terminal direction using one-letter or three-letter codes for amino acids. For example, RNVP represents a peptide in which arginine, asparagine, valine, and proline are connected in the N-terminal to C-terminal direction in order. In another example, Thr-Leu-Lys represents a peptide in which threonine, leucine, and lysine are connected in the N-terminal to C-terminal direction in order. Amino acids that cannot be represented by one-letter codes are represented using different letters and are explained in supplementary terms.

[0080] The abbreviations for each amino acid are as follows: alanine (Ala, A); arginine (Arg, R); asparagine (Asn, N); aspartic acid (Asp, D); cysteine ​​(Cys, C); glutamic acid (Glu, E); glutamine (Gln, Q); glycine (Gly, G); histidine (His, H); isoleucine (Ile, I); leucine (Leu, L); lysine (Lys K); methionine (Met, M); phenylalanine (Phe, F); proline (Pro, P); serine (Ser, S); threonine (Thr, T); tryptophan (Trp, W); tyrosine (Tyr, Y); and valine (Val, V).

[0081] Representation of Nucleic Acid Sequences The symbols A, T, C, G, and U used herein are to be interpreted as understood by those skilled in the art. Depending on the context and technique, they may be interpreted as bases, nucleosides, or nucleotides on DNA or RNA, as appropriate. For example, when referring to bases, each symbol may be interpreted as adenine (A), thymine (T), cytosine (C), guanine (G), or uracil (U); when referring to nucleosides, each symbol may be interpreted as adenosine (A), thymidine (T), cytidine (C), guanosine (G), or uridine (U); when referring to nucleotides in a sequence, each symbol should be interpreted as indicating the nucleotides that comprise each nucleoside.

[0082] Target Array As used herein, the term "target sequence" refers to a specific sequence recognized by a guide RNA or a gene editing tool (e.g., a Cas / gRNA complex) to cleave a target gene or a target nucleic acid. The target sequence may be appropriately selected depending on the purpose. For example, the "target sequence" may refer to a sequence that is included in a target gene or a target nucleic acid sequence and is complementary to a spacer sequence included in a guide RNA. In another example, the "target sequence" may refer to a sequence that is included in a target gene or a target nucleic acid sequence and is complementary to a sequence that is complementary to a spacer sequence included in a guide RNA. Thus, the target sequence is used to refer to a sequence that is complementary to a spacer sequence included in a guide RNA and / or a sequence that is substantially the same as a spacer sequence of a guide RNA, and should not be interpreted as being limiting. In some embodiments, the target sequence may be disclosed as a sequence that does not include a PAM sequence. In some embodiments, the target sequence may be disclosed as a sequence that does not include a PAM sequence. The target sequence should be interpreted appropriately depending on the context. In general, the spacer sequence is determined by considering the sequence of the target gene or target nucleic acid and the PAM sequence recognized by the editing protein of the CRISPR / Cas system. The target sequence may refer to only the sequence of a specific strand that is complementary to the guide RNA of the CRISPR / Cas complex, only the sequence of a specific strand that is not complementary to the guide RNA, or the entire target duplex including a specific strand portion, which is interpreted appropriately according to the context. The definition of the term target sequence is provided to explain the strand in which the target sequence may exist, and is not intended to distinguish between on-target sequence and off-target sequence through the term target sequence. That is, in some embodiments, the intended target sequence may be referred to as the on-target sequence, and the unintended target sequence may be referred to as the off-target sequence. With respect to on-target and off-target, the term "target sequence" may be interpreted appropriately according to the content of the relevant paragraph.

[0083] Spacer Linking Chain The term "spacer binding strand" as used herein refers to a strand that contains a sequence complementary to a part or the entire sequence of the spacer region of a guide nucleic acid in a gene editing system (e.g., a CRISPR / Cas gene editing system) involving a guide nucleic acid (e.g., a guide RNA). A DNA molecule such as a genome generally has a double-stranded structure. In the double strand, a strand that has a sequence complementary to a part or the entire sequence of the spacer region of a guide nucleic acid, and thus forms a complementary bond with a part or the entire sequence of the spacer region, can be referred to as a spacer binding strand.

[0084] Spacer non-binding chain As used herein, the term "spacer-unbound strand" is used to refer to a strand other than the "spacer-bound strand," which is a strand that includes a sequence that forms a complementary bond with a portion or the entire sequence of the spacer region of a guide nucleic acid in a gene editing system (e.g., a CRISPR / Cas gene editing system) involving a guide nucleic acid (e.g., a guide RNA). A DNA molecule such as a genome generally has a double-stranded structure, and the term "spacer-unbound strand" can be used to refer to a strand other than the spacer-bound strand in the double strand.

[0085] functional equivalent The term "functional equivalent" or "equivalent" refers to a second biomolecule that is functionally equivalent to a first biomolecule, but not necessarily structurally equivalent. For example, a "Cas9 equivalent" refers to a protein that is the same or substantially the same function as Cas9, but not necessarily the same amino acid sequence. Throughout this application, when a particular protein is mentioned, the particular protein mentioned above is intended to encompass all of its functional equivalents. For example, when described as an "X protein," the term "X protein" can be interpreted to encompass the functional equivalent of the X protein. In this sense, a "functional equivalent" of an X protein includes any homolog, paralog, ortholog, fragment, natural, artificial, mutated, or synthetic version of protein X that ensures equivalent function. When described as a Cas protein, the term "Cas protein" can be interpreted to encompass the functional equivalent of the Cas protein.

[0086] Nuclear localization signal or sequence (NLS) The term "nuclear localization signal or sequence (NLS)" refers to an amino acid sequence that facilitates the introduction of a protein into a cell nucleus. For example, the introduction of a protein can be facilitated by nuclear transport. NLSs are well known in the art and are clear to those skilled in the art. For example, exemplary sequences of NLSs may be disclosed in PCT Application No. PCT / EP2000 / 011690 (Publication No. WO2021 / 038547), the contents of which regarding exemplary NLSs are incorporated herein by reference. In some embodiments, the NLS is PKKKRKV (SEQ ID NO: 01), KRPAATKKAGQAKKKK (SEQ ID NO: 02), PAAKRVKLD (SEQ ID NO: 03), RQRRNELKRSP (SEQ ID NO: 04), NQSSNFGPMKGGNFGGRSSGPYGGGGQYFAKPRNQGGY (SEQ ID NO: 05), RMRIZFKNKGKDTAELRRRRVEVSVELRKAKKDEQILKRRNV (SEQ ID NO: 06), VSRKRPRP (SEQ ID NO: 07), PPKKARED (SEQ ID NO: 08), PO The NLS may include, but is not limited to, amino acid sequences such as PKKKPL (SEQ ID NO: 09), SALIKKKKKMAP (SEQ ID NO: 10), DRLRR (SEQ ID NO: 11), PKQKKRK (SEQ ID NO: 12), RKLKKKIKKL (SEQ ID NO: 13), REKKKFLKRR (SEQ ID NO: 14), KRKGDEVDGVDEVAKKKSKK (SEQ ID NO: 15), RKCLQAGMNLEARKTKK (SEQ ID NO: 16), or MDSLLMNRRKFLYQFKNVRWAKGRRETYLC (SEQ ID NO: 17). The NLS may be optionally fused to a gene editing agent such as a Cas protein. The NLS fused to a protein may be used to facilitate the translocation of the attached protein to the desired location in the nucleus.

[0087] about The term "about" as used herein means a degree of approximation to a particular quantity and refers to an amount, level, value, number, frequency, percentage, dimension, size, volume, weight, or length that varies by approximately 30, 25, 20, 25, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1% from a reference amount, level, value, number, frequency, percentage, dimension, size, volume, weight, or length.

[0088] Orientation of the disclosed sequences Nucleotide sequences (e.g., DNA sequences, RNA sequences, or DNA / RNA hybrid sequences) disclosed herein should be understood as being disclosed in a 5' to 3' orientation, unless otherwise specified. Amino acid sequences disclosed herein should be understood as being disclosed in an N-terminal to C-terminal orientation, unless otherwise specified. For sequences disclosed in an orientation different from that described above, the orientation in the other direction is specified separately in the paragraph related to the corresponding sequence.

[0089] Overview of gene editing systems The present application relates to a method for predicting off-targets that may occur in a gene editing process using a gene editing system. Off-target prediction is used to encompass the prediction of off-target sites. Before describing the method for predicting off-targets provided by the present application, gene editing systems related to off-targets will be described. A gene editing system (e.g., a genome editing system) refers to a system used to achieve desired editing in a desired nucleic acid molecule (e.g., genomic DNA) through the use of gene editing tools such as editing proteins and guide nucleic acids. In many studies, gene editing systems are used to edit the genome of a cell, and the term "gene editing system" may be used interchangeably with genome editing system. However, the use of gene editing systems is not limited to genome editing. Furthermore, the term "gene editing system" may be used to refer to gene editing tools and may be interpreted appropriately according to the relevant context. Examples of known gene editing systems are zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and the CRISPR / Cas gene editing system (see Khan, Sikandar Hayat. "Genome-editing technologies: concept, pros, and cons of various genome-editing techniques and bioethical concerns for clinical application." Molecular Therapy-Nucleic Acids 16(2019):326-334.), the entire contents of which are incorporated herein by reference. In addition, there are base editing and prime editing systems developed based on the CRISPR / Cas gene editing system.

[0090] One of the characteristics of the off-target prediction method provided by the present application is that the membrane structure of the cell is disrupted through physical methods (e.g., using an extruder) to allow the components of the gene editing system (e.g., editing proteins and / or guide nucleic acids) to contact the genome. Therefore, the off-target prediction method of the present application can be applied to all of the above-mentioned gene editing systems.

[0091] As an example of a gene editing system, the CRISPR / Cas gene editing system, which is being actively researched to achieve genome editing goals, is described in detail below.

[0092] CRISPR / Cas gene editing system Overview of CRISPR / Cas gene editing systems The CRISPR / Cas gene editing system is used as a general term to refer to a gene editing system with editing proteins containing a Cas protein and a guide nucleic acid (e.g., a guide RNA) used to induce the desired editing of a gene (e.g., genomic DNA) at a desired location. The CRISPR / Cas gene system can be used as other terms that can be understood by those skilled in the art. For example, the CRISPR / Cas gene system can be referred to as CRISPR / Cas, a CRISPR / Cas system, a CRISRP system, or a Cas-based genome editing system, but the present application is not limited thereto. Furthermore, the CRISPR / Cas gene editing system is used to encompass all developed technologies based on CRISPR / Cas genes, such as base editing (see Gaudelli, Nicole M., et al. "Programmable base editing of A·T to G·C in genomic DNA without DNA cleavage." Nature 551.7681(2017):464-471.) and prime editing (see Anzalone, Andrew V., et al. "Search-and-replace genome editing without double-strand breaks or donor DNA." Nature 576.7785(2019):149-157.). The results of gene editing (e.g., genome editing) may include, but are not limited to, cleavage, indels, insertions, deletions, substitutions, base editing (e.g., can be achieved by base editing), and writing (e.g., can be achieved by prime editing). Below, we provide a detailed description of the CRISPR / Cas gene editing system, including its origins.

[0093] CRISPR The "CRISPR" section is intended to aid in the understanding of those of skill in the art, and the terms used in this section are not intended to limit the terms disclosed herein.

[0094] CRISPR is a family of DNA sequences (i.e., CRISPR clusters) in bacteria and archaea that represent snippets of previous infections brought about by viruses that have invaded prokaryotes. The snippets of DNA are used by prokaryotic cells to detect and destroy DNA from subsequent attacks by similar viruses, and together with an array of CRISPR-associated proteins (Cas proteins) and CRISPR-associated RNAs, effectively constitute a prokaryotic immune defense system. CRISPR clusters are transcribed and processed into CRISPR RNA (crRNA). Cas9 / crRNA / tracrRNA then endonucleolytically cleaves linear or circular dsDNA targets that are complementary to the RNA. In particular, target strands that are not complementary to the crRNA are first endonucleolytically cleaved and then 3'→5' exonucleolytically trimmed. DNA binding and cleavage typically require a protein and two RNAs. However, single guide RNA (sgRNA, or simply gRNA) has been developed, a single stranded RNA engineered to incorporate aspects of both crRNA and tracrRNA into a single RNA species. See, for example, Jinek, Martin, et al., "A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity." Science 337.6096(2012):816-821., the entire contents of which are incorporated herein by reference. Cas9 recognizes a short motif (PAM or protospacer adjacent motif) in the CRISPR repeat sequence to help distinguish self versus non-self.CRISPR biology, as well as Cas9 nuclease sequences and structures, are well known to those of skill in the art (see, e.g., Ferretti, Joseph J., et al., "Complete genome sequence of an M1 strain of Streptococcus pyogenes." Proceedings of the National Academy of Sciences 98.8(2001):4658-4663.; Deltcheva, Elitza, et al., "CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III." Nature 471.7340(2011):602-607.; and Jinek, Martin, et al., "A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity." science 337.6096(2012):816-821., the entire contents of each of which are incorporated herein by reference). Cas9 orthologues have been described in various species, including, but not limited to, Streptococcus pyogenes (S. pyogenes) and Streptococcus thermophilus (S. thermophilus). Additional suitable Cas9 nucleases and their sequences will be apparent to those of skill in the art based on this disclosure, including Cas9 sequences from organisms and loci disclosed in Chylinski, Krzysztof, Anais Le Rhun, and Emmanuelle Charpentier. "The tracrRNA and Cas9 families of type II CRISPR-Cas immunity systems." RNA biology 10.5 (2013): 726-737., the entire contents of which are incorporated herein by reference.

[0095] CRISPR / Cas gene editing system The CRISPR / Cas gene editing system, developed from the above-mentioned CRISPR, is a technology for editing a gene (e.g., a cell's genome) at a desired location using a Cas protein derived from the cell's CRISPR system and a guide nucleic acid that guides the Cas protein to a target region. For example, the Cas protein forms a Cas / gRNA complex together with a guide RNA (gRNA). The Cas / gRNA complex guides the guide RNA contained in the complex to a desired location. The Cas protein contained in the Cas / gRNA complex induces a double strand break (DSB) or a nick to a desired location. The CRISPR / Cas gene editing system can edit not only the cell's genome, but also DNA molecules that are not located on the genome. Since the discovery of CRISPR, as described above, with regard to the CRISPR / Cas genome editing system, single guide RNAs to which tracrRNA and crRNA are connected have been developed (see the reference [Jinek, Martin, et al. "A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity." science 337.6096(2012):816-821.], the entire contents of which are incorporated herein by reference), and various kinds and / or types of Cas proteins (Cas equivalents) such as cas9, cas12a(cpf1), cas12b(c2c1), cas12e(casX), cas12k(c2c5), cas14, cas14a, cas13a(c2c2), cas13b(c2c6), cas9 nickase, dead cas, etc. have been developed. Furthermore, based on the CRISPR / Cas gene editing system, base editing capable of realizing the purpose of base modification and prime editing capable of realizing the writing of desired edits have been developed. As described above, the CRISPR / Cas gene editing system can be used to include the conventional CRISPR / Cas gene editing system and the technology aimed at gene editing developed based on it.To understand the CRISPR / Cas gene editing system, reference may be made to document WO2018 / 231018 (PCT No.), the entire contents of which are incorporated herein by reference. To aid those skilled in the art in their understanding, the editing protein (e.g., Cas protein) that can be used in the CRISPR / Cas gene editing system is further described below.

[0096] CRISPR / Cas gene editing system 1 - Editing proteins Overview of editing proteins and editing proteins in CRISPR / Cas gene editing systems Editing protein (editor protein) may be used to refer to a protein that generates a DSB or nick in a desired region to achieve gene editing or helps induce editing. In general, a protein with nuclease activity that cleaves nucleic acid may be referred to as an editing protein. In the CRISPR / Cas gene editing system, editing protein may be used interchangeably with Cas protein. A representative example of a Cas protein is Cas9. The term "Cas protein" as used herein is used as a general term for gene editing proteins that can generate a DSB or nick in a target region or inactive Cas proteins used in the CRISPR / Cas gene editing system. Examples of Cas proteins include, but are not limited to, Cas9, Cas9 variants, Cas9 nickase (nCas9), dead Cas9, Cpf1 (type V CRISPR-Cas system), C2c1 (type V CRISPR-Cas system), C2c2 (type VI CRISPR-Cas system), and C2c3 (type V CRISPR-Cas system). Additional examples of Cas proteins are disclosed in Abudayyeh, Omar O., et al., "C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector." Science 353.6299(2016):aaf5573., the entire contents of which are incorporated herein by reference.In one embodiment, the Cas protein is selected from the group consisting of Streptococcus pyogenes, Streptococcus thermophilus, Streptococcus spp., Staphylococcus aureus, Campylobacter jejuni, Nocardiopsis dassonvillei, Streptomyces pristinaespiralis, Streptomyces viridochromogenes, Streptomyces viridochromogenes, Streptosporangium roseum, Streptosporangium purpurea, and Alicyclobacter spp. Bacillus acidocaldarius, Bacillus pseudomycoides, Bacillus serenityreducens, Exiguobacterium sibiricum, Lactobacillus delbrueckii, Lactobacillus salivarius, Microscilla marina, Burkholderiales bacterium, Polaromonas naphthalenivorans, Polaromonas spp., Crocosphaera watsonii, Cyanoseis spp., Microcystis aeruginosa, Synechococcus spp., Acetohalobium arabaticum, Ammonifec S. degensis, Caldicellulosylpter bessii, Candidatus desulfordis, Clostridium botulinum, Clostridium difficile, Finegoldia magna, Natlanaerobius thermophilus, Perotomacrum thermopropionicum, Acidithiobacillus caldus, Acidithiobacillus ferrooxidans, Allochromatium vinosum, Marinobacter spp., Nitrosococcus halophilus, Nitrosococcus watsonii, Pseudoalteres moniliformes, The Cas9 or Cpf1 may be derived from various microorganisms such as Romonas haloplanktis, Ctedonobacter racemifera, Methanohalobium ebestigatum, Anabaena variabilis, Nodularia spumigena, Nostoc spp., Arthrospira maxima, Arthrospira platensis, Arthrospira spp., Lyngbya spp., Microcoleus ctnoplastes, Oscillatoria spp., Petrotoga mobilis, Thermosipho africanus, and Acaryochloris marina. The Cas9 proteins used in the CRISPR / Cas9 gene editing system are shown below.

[0097] Cas9 protein In the CRISPR / Cas9 gene editing system, the protein with nuclease activity that cleaves nucleic acid is called Cas9 protein. Cas9 proteins correspond to class 2 (type II) in the classification of CRISPR / Cas systems, and include Cas9 proteins from Streptococcus pyogenes, Streptococcus thermophilus, Streptococcus spp., Streptomyces pristinaespiralis, Streptomyces viridochromogenes, Streptomyces viridochromogenes, Streptosporangium roseum, and Streptosporangium roseum. Additional Cas9 proteins and their sequences are disclosed in Chylinski, Krzysztof, Anais Le Rhun, and Emmanuelle Charpentier. "The tracrRNA and Cas9 families of type II CRISPR-Cas immunity systems." RNA biology 10.5 (2013): 726-737., the entire contents of which are incorporated herein by reference. For example, the DNA cleavage domain of Cas9 is known to contain two subdomains, such as the NHN nuclease subdomain and the RucC1 subdomain. The NHN subdomain cleaves the strand complementary to the gRNA, and the RuvC1 subdomain cleaves the strand that is not complementary. Inactivation of these subdomains can silence the nuclease activity of Cas9. For example, both mutants D10A and H840A completely inactivate the nuclease activity of S. pyogenes Cas9 (see Jinek, Martin, et al. "A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity." Science 337.6096(2012):816-821.). For example, mutant H840A provides the Cas9 nickase.

[0098] CRISPR / Cas Gene Editing System 2 - Guide Nucleic Acid Guide Nucleic Acid Overview In the CRISPR / Cas gene editing system, the Cas protein is associated with the guide nucleic acid (e.g., guide RNA) to form a Cas / guide nucleic acid complex (e.g., Cas / gRNA complex). The Cas / gRNA complex may be referred to as a ribonucleoprotein (RNP). The Cas / gRNA complex generates a double-strand break (DSB) or a nick in the target region that contains a sequence corresponding (e.g., complementary) to the spacer sequence of the guide RNA (gRNA), and the DSB or nick is induced by the Cas protein. The position where the DSB or nick is generated may be near the PAM sequence on the genome. The protospacer adjacent motif (PAM) on the genome and the spacer sequence of the gRNA are involved in Cas / gRNA targeting. The Cas protein (e.g., Cas9) guided to the target region by the PAM and the spacer sequence of the gRNA generates a DSB in the target region.

[0099] In the CRISPR / Cas gene editing system, an RNA that has the function of guiding a Cas protein to a target region in order to recognize a specific sequence contained in a target DNA molecule is called a guide RNA.

[0100] When the composition of guide RNA is divided by function, it can be mainly divided into 1) a scaffold sequence portion and 2) a guide domain including a guide sequence. The scaffold sequence portion is a portion that interacts with a Cas protein (e.g., a Cas9 protein) and can bind to the Cas protein to form a complex. In general, the scaffold sequence portion includes a tracrRNA and a crRNA repeat sequence portion, and the scaffold sequence is determined depending on which Cas9 protein is used. The guide sequence is a portion that can bind complementary to a nucleotide sequence portion of a specific length in a target nucleic acid (e.g., a target DNA molecule or a cell genome). The guide sequence can be artificially modified and is determined by the target nucleotide sequence of interest related to the desired gene editing.

[0101] In some embodiments, the guide RNA may be described as including a crRNA and a tracrRNA. The crRNA may include a spacer and a repeat sequence. The repeat sequence portion of the crRNA may interact with (e.g., bind complementary to) a portion of the tracrRNA. As described above, a single guide RNA (sgRNA) may be provided in which the crRNA and the tracrRNA are linked.

[0102] In one embodiment, the guide RNA may be provided as two strands. In one embodiment, the guide RNA may be provided as one strand. A sgRNA in which the tracrRNA and the crRNA are linked has been developed (see Jinek, Martin, et al. "A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity." Science 337.6096(2012):816-821.), the entire contents of which are incorporated herein by reference). In a particular embodiment, the guide RNA may be an sgRNA.

[0103] Guide domain and guide sequence of guide RNA Guide nucleic acid (e.g., guide RNA) may include a guide domain that includes a guide sequence. Guide sequence is used interchangeably with spacer sequence. Guide sequence is determined by the target nucleotide sequence of interest as a part that can be artificially designed. In some embodiments, guide sequence can be designed to target the sequence adjacent to the PAM sequence located on the DNA molecule for editing. As described above, the localization of Cas / gRNA complex to the target position is induced. The structure of guide nucleic acid can vary depending on CRISPR type. For example, guide RNA used in CRISPR / Cas9 gene editing system can have the structure of 5'-[guide domain]-[scaffold]-3'.

[0104] In one embodiment, the guide sequence may have a length of 5 to 40 nt. In one embodiment, the guide sequence included in the guide domain of the guide RNA may have a length of 10 to 30 nt. In one embodiment, the guide sequence may have a length of 15 to 25 nt. In one embodiment, the guide sequence may have a length of 18 to 22 nt. In one embodiment, the guide sequence may have a length of 20 nt. In one embodiment, the target sequence (including both the target sequence present in the spacer-bound strand and the target sequence present in the spacer-unbound strand) which is a sequence in the genome that forms a complementary bond with the guide sequence may have a length of 5 to 40 nt or 5 to 40 bp. In one embodiment, the target sequence (including both the target sequence present in the spacer-bound strand and the target sequence present in the spacer-unbound strand) which is a sequence in the genome that forms a complementary bond with the guide sequence may have a length of 10 to 30 nt or 10 to 30 bp. In one embodiment, the target sequence may have a length of 15 to 25 nt or 15 to 25 bp. In one embodiment, the target sequence may have a length of 18 to 22 nt or 18 to 22 bp. In one embodiment, the target sequence may have a length of 20 nt or 20 bp.

[0105] CRISPR / Cas Gene Editing System 3 - Protospacer Adjacent Motif (PAM) Two conditions are required for a target gene or target nucleic acid to be cleaved by the CRISPR / Cas9 gene editing system.

[0106] First, there must be a specific length of base sequence (nucleotide sequence) that the Cas9 protein can recognize in the target gene or target nucleic acid. Here, the specific length of base sequence (nucleotide sequence) that the Cas9 protein recognizes is called a protospacer adjacent motif (PAM) sequence. The PAM sequence is a unique sequence determined by the Cas9 protein. Second, there must be a sequence that can bind complementarily to the spacer sequence contained in the guide RNA around the specific length of the PAM sequence. Here, the PAM sequence can be used to include a sequence present on the spacer non-binding strand and a sequence present on the spacer binding strand.

[0107] As described above, in the CRISPR / Cas gene editing system, the Cas / gRNA complex is guided to the target region by the PAM sequence on the target DNA molecule (e.g., the genome of a cell) and the guide sequence of the gRNA. In the target DNA molecule, the PAM sequence may be located on the non-binding strand of the guide sequence, which is not the strand to which the guide sequence of the guide RNA binds. The PAM sequence may be determined independently depending on the type of Cas protein used. In one embodiment, the PAM sequence may be any one selected from NGG (SEQ ID NO: 18); NNNNRYAC (SEQ ID NO: 19); NNAGAAW (SEQ ID NO: 20); NNNNGATT (SEQ ID NO: 21); NNGRR(T) (SEQ ID NO: 22); TTN (SEQ ID NO: 23); and NNNVRYAC (SEQ ID NO: 24) (shown in the 5'→3' direction). Each N may independently be A, T, C, or G. Each R may independently be A or G. Each Y may independently be C or T. Each W may be independently A or T. For example, when SpCas9 is used as the Cas protein, the PAM sequence may be NGG (SEQ ID NO: 18). For example, when Streptococcus thermophilus Cas9 (StCas9) is used as the Cas protein, the PAM sequence may be NNAGAAW (SEQ ID NO: 20). For example, when Neisseria meningitidis Cas9 (NmCas9) is used, the PAM sequence may be NNNNGATT (SEQ ID NO: 21). For example, when Campylobacter jejuni Cas9 (CjCas9) is used, the PAM may be NNNVRYAC (SEQ ID NO: 24). In one embodiment, the PAM sequence may be connected to the 3' end of the target sequence present on the non-spacer binding strand (wherein the target sequence present on the non-spacer binding strand refers to a sequence that does not bind to the guide RNA). In one embodiment, the PAM sequence may be located at the 3' end of the target sequence located on the non-spacer binding strand. The target sequence present on the non-spacer-binding strand refers to a sequence that does not bind to the guide sequence of the guide RNA. The target sequence present on the non-spacer-binding strand is complementary to the target sequence present on the spacer-binding strand.

[0108] The position where the DSB or nick is generated may be near the PAM sequence on the genome. In one embodiment, the position where the DSB or nick is generated may be -0 to -20 or +0 to +20 based on the 5' or 3' end of the PAM sequence present on the non-spacer binding strand. In one embodiment, the position where the DSB or nick is generated may be -1 to -5 or +1 to +5 of the PAM sequence on the non-spacer binding strand. For example, in a CRISPR / Cas gene editing system using SpCas9, SpCas9 cleaves between the third and fourth nucleotides located upstream of the PAM sequence.

[0109] Genome editing process using CRISPR / Cas gene editing system To aid in the understanding of those skilled in the art, the genome editing process using the CRISPR / Cas gene editing system is briefly described using the following example.

[0110] For example, an environment can be provided in which a DNA molecule for editing can contact the Cas / gRNA complex. The DNA molecule can be the DNA molecule that is the object of editing. For the purpose of genome editing in a cell, a Cas protein or a nucleic acid encoding the same and a guide RNA or a nucleic acid encoding the same are introduced into the cell, thereby generating an environment in which the Cas protein and the guide RNA can contact the genomic DNA of the cell. In an environment in which the Cas protein and the guide RNA can contact the genomic DNA of the cell, the Cas protein and the guide RNA can form a Cas / gRNA complex. Of course, the Cas protein and the gRNA can form a Cas / gRNA complex in an appropriate environment even in the absence of genomic DNA of the cell. The involvement of the guide sequence of the gRNA contained in the Cas / gRNA complex and the PAM sequence on the genome allows the Cas / gRNA complex to guide to a target region where a pre-designed target sequence is present. The Cas / gRNA complex guided to the target region generates a DSB (for example, in the case of Cas9) in the target region. Then, when the DNA where the DSB is generated (cleaved) is repaired in the DNA repair process, gene editing is performed at the target region or target position. There are two main pathways for repairing DSB generated on DNA, such as homology-directed repair (HDR) and non-homologous end joining (NHEJ). Among these, HDR, a natural DNA repair system, can be used to correct genomes in various organisms, including humans. HDR-mediated repair can be mainly used to insert a desired sequence into a target region or target site, or to induce specific point mutations, but the present application is not limited thereto. HDR-mediated repair can be performed by the DNA repair system HDR and a HDR template (e.g., a donor template that can be provided from outside the cell). NHEJ refers to the process of repairing DSB on DNA, and in contrast to HDR, it connects the cleaved ends without the HDR template. That is, the repair process does not require a HDR template.NHEJ may be the DNA repair mechanism that can be mainly selected to induce indels. Insertion / deletion may refer to a mutation in which some nucleotides are deleted, any nucleotides are inserted, and / or insertion and deletion are mixed in the nucleotide sequence of the nucleic acid before gene editing. The occurrence of some indels in the target gene may inactivate the corresponding gene. The DNA repair mechanisms HDR and NHEJ are disclosed in detail in Sander, Jeffry D., and J. Keith Joung. "CRISPR-Cas systems for editing, regulating and targeting genomes." Nature biotechnology 32.4(2014):347-355., the entire contents of which are incorporated herein by reference.

[0111] Up to this point, gene editing systems including CRISPR / Cas gene editing systems have been described in detail. This application relates to a method for predicting possible off-targets in gene editing (e.g., genome editing) processes using gene editing systems. Hereinafter, the off-targets that may occur in gene editing systems will be described in detail.

[0112] Off-target In the field of gene editing (e.g., genome editing), off-target refers to gene modification that occurs at an unintended location. Gene modification induced by off-target can be non-specific. Genome editing tools developed include CRISPR / Cas gene editing system, transcription activator-like effector nuclease (TALEN), meganuclease, and zinc finger nuclease. These genome editing tools or gene editing systems are designed to perform editing in a target region through a special mechanism that can bind to a predetermined sequence (e.g., a sequence in a target region). For example, in the CRISPR / Cas gene editing system, a guide RNA (gRNA) guides the movement of the Cas / gRNA complex to the intended target location. A PAM sequence in the genome can also be involved in the movement to the target location. However, the Cas / gRNA complex may still bind to a sequence in an unintended location that is not a sequence in the target region. In this way, when the Cas / gRNA complex binds to a sequence in an unintended location and a DSB is generated in the unintended location, an unintended gene modification occurs. Off-target effects lead to unintended gene modifications such as unintended point mutations, deletions, insertions, inversions, and translocations. It is known that the binding of genome editing tools to undesired regions results from partial but sufficient matching of the sequence in the undesired region to the target sequence. Without being bound by theory, Lin, Yanni, et al. "CRISPR / Cas9 systems have off-target activity with insertions or deletions between target DNA and guide RNA sequences." Nucleic acids research 42.11(2014):7473-7485., the entire contents of which are incorporated herein by reference, shows that the mechanisms of off-target binding can be grouped into base mismatch tolerance and bulge mismatch.For example, the off-target site may contain 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more mismatches with the guide RNA sequence, but the application is not limited thereto.

[0113] Off-target problems suggest the possibility of causing disruption in important coding regions leading to serious problems such as cancer. Furthermore, in the field of research, off-target problems may also lead to confusion in variables in biological research, suggesting the possibility of causing irreproducible results (see Eid, Ayman, and Magdy M. Mahfouz. "Genome editing: the road of CRISPR / Cas9 from bench to clinic." Experimental & Molecular Medicine 48.10(2016):e265-e265.), the entire contents of which are incorporated herein by reference).

[0114] The problem of off-target remains not only in the CRISPR / Cas gene editing system, but also in base editing and prime editing developed based on the CRISPR / Cas gene editing system. In this specification, off-target can be used as the opposite concept of on-target and can be used to refer to gene modification at an unintended position.

[0115] As described above, off-targets cause serious side effects in various aspects (e.g., side effects that are difficult to identify and / or irreversible side effects). Therefore, identification of off-targets that may occur when using a gene editing system (e.g., a genome editing system) is very important in the development and research of treatments. In order to identify true off-targets that occur in a designed gene editing system (e.g., a CRISPR / Cas9 gene editing system and a specific guide RNA), a lot of cost and time are required. For this reason, various methods that can identify off-target candidates, i.e., predict off-targets, have been researched and developed. However, the methods for predicting off-targets that may occur in gene editing processes (e.g., genome editing processes using gene editing systems) developed up to the filing date of this application still have various problems. Hereinafter, off-target prediction systems that have been researched and developed so far and their problems will be disclosed.

[0116] Known off-target prediction systems and their limitations Known off-target prediction systems As mentioned above, various methods have been developed to predict off-targets that may occur in genome editing using a gene editing system (e.g., a CRISPR / Cas gene editing system). Conventional off-target prediction or off-target candidate identification methods can be classified into three categories according to the mechanism of action (MOA) of the method, such as cell-based off-target prediction systems, in vitro off-target prediction systems, and in silico off-target prediction systems. Examples of prediction systems included in the above categories are as follows: - Cell-based systems: GUIDE-seq (see reference [Tsai, SQ, Zheng, Z., Nguyen, NT, Liebers, M., Topkar, VV, Thapar, V., ... & Joung, JK(2015). GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nature biotechnology, 33(2), 187-197.]), GUIDE-tag, DISCOVER-seq, BLISS, BLESS, integrase-deficient lentiviral vector-mediated DNA cleavage capture, HTGTS, ONE-seq, CReVIS-Seq, ITR-seq, and TAG-seq. - In vitro systems: Digenome-seq (see Kim, Daesik, et al. "Digenome-seq: genome-wide profiling of CRISPR-Cas9 off-target effects in human cells." Nature methods 12.3(2015):237-243.), DIG-seq (see Kim, Daesik, and Jin-Soo Kim. "DIG-seq: a genome-wide CRISPR off-target profiling method using chromatin DNA." Genome research 28.12(2018):1894-1900.), SITE-Seq, CIRCLE-Seq, and CHANGE-seq. - In silico systems: Cas-OFFinder (see reference [Bae, Sangsu, Jeongbin Park, and Jin-Soo Kim. "Cas-OFFinder: a fast and versatile algorithm that searches for potential an off-target effect of Cas9 RNA-guided endonucleases." Bioinformatics 30.10(2014):1473-1475.]), CHOPCHOP, CRISPOR, etc.

[0117] As mentioned above, various methods have been developed to predict off-targets that may occur in the genome editing process using the CRISPR / Cas genome editing system (or tool), but the prediction methods or tools developed so far still have limitations. The limitations of each category are shown below.

[0118] Cell-based class problems For example, cell-based methods have presented problems with potentially missing true off-target locations and reducing the efficiency of predictive methods in more clinically relevant cell types (e.g., cells used clinically) (see Wienert, Beeke, et al. "Unbiased detection of CRISPR an off-target effect in vivo using DISCOVER-Seq." Science 364.6437(2019):286-289; and Shapiro, Jenny, et al. "Increasing CRISPR efficiency and measuring its specificity in HSPCs using a clinically relevant system." Molecular Therapy-Methods & Clinical Development 17(2020):1097-1107.).

[0119] Furthermore, the inventors of the present application have further confirmed through their own experiments that problems may occur in cell-based prediction methods (e.g., GUIDE-seq). For example, the inventors of the present application have confirmed that cell-based prediction methods have a higher miss rate, and through experiments, have confirmed that off-target prediction results may vary depending on cell type (see the experimental examples of the present application).

[0120] In vitro and in silico class issues In vitro off-target prediction methods and in silico off-target prediction methods have the problem of providing an excessively large number of false positive data points and not reflecting the intracellular environment, which may be cell-specific, such as chromatin structure and epigenetic modifications (see the reference [Kim, Daesik, and Jin-Soo Kim. "DIG-seq: a genome-wide CRISPR off-target profiling method using chromatin DNA." Genome research 28.12(2018):1894-1900.]).

[0121] Furthermore, the inventors of the present application have further identified problems that may occur in in vitro-based prediction methods (e.g., Digenome-seq, DIG-seq) and in silico-based prediction methods in their own experiments. For example, the inventors of the present application have confirmed a higher false positive rate of in vitro-based prediction methods (see the experimental examples of the present application).

[0122] Off-target prediction system used in the IND process for gene therapy Since each method has its own advantages and disadvantages, various prediction methods were used in combination to determine or confirm the off-targets of the CRISPR / Cas genome editing system. In a recent study conducted at Intellia, GUIDE-seq, SITE-Seq, and Cas-OFFinder were used to identify potential off-target locations (NTLA-2001, see Gillmore, Julian D., et al. "CRISPR-Cas9 in vivo gene editing for transthyretin amyloidosis." New England Journal of Medicine 385.6 (2021): 493-502.). EDITAS Medicine used GUIDE-seq, Digenome-seq, and Cas-OFFinder off-target prediction tools for the candidate therapeutic EDIT-101 (see Maeder ML, Stefanidakis M, Wilson CJ, Baral R, Barrera LA, Bounoutas GS, Bumcrot D, Chao H, Ciulla DM, DaSilva JA et al: Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10. Nat Med 2019, 25(2):229-233.).

[0123] However, the use of a combination of various prediction methods requires a lot of effort and cost, and is difficult to use in a general group. In addition, the use of various off-target prediction methods may not guarantee that more off-target candidates can be detected. For example, for NTLA-2001, the seven valid off-target positions identified by SITE-Seq used herein include all of the valid off-target positions discovered by GUIDE-seq (three valid off-targets were discovered) and Cas-OFFinder (three valid off-targets were discovered). In this case, the output of one in vitro method is the same as the output of the combination of the three methods. For NTLA-2001, SITE-Seq identified 475 off-target candidates, of which 468 candidates were identified as false positives. In clinical trials, it is difficult to verify all 475 off-target candidates for each patient or organ cell.

[0124] Therefore, for various reasons including the above-mentioned aspects, it is necessary to develop a focused off-target prediction method. The inventors of the present application have developed an effective off-target prediction method that is more accurate and has a lower false positive rate. The inventors of the present application have confirmed the excellent performance of the new off-target prediction method of the present application by comparing the performance of the conventional method and the newly developed off-target prediction method of the present application in detail using various and multiple test methods. In particular, the inventors of the present application have confirmed that the off-target prediction method of the present application has better performance than the in vitro off-target prediction method by comparing the in vitro off-target prediction method and the off-target prediction method of the present application. Furthermore, the inventors of the present application have confirmed that the off-target prediction method of the present application shows better performance than the cell-based off-target prediction method by comparing the cell-based off-target prediction method and the off-target prediction method of the present application. Furthermore, the inventors of the present application have comprehensively confirmed through various and multiple tests that the off-target prediction method of the present application shows better performance than other off-target prediction methods (see the experimental examples of the present application).

[0125] The off-target prediction method provided by the present application has all the advantages of cell-based prediction method and in vitro prediction method. With respect to the advantages of cell-based prediction method, the off-target prediction method provided by the present application can provide an environment in which Cas / gRNA complex can contact with genomic DNA in which chromatin structure and epigenetic modifications are maintained. With respect to the advantages of in vitro-based prediction method, the off-target prediction method provided by the present application can prevent DNA repair mechanism from accumulating cut rate, thereby preventing true off-target from being overlooked. The off-target prediction system (method or tool) provided by the present application will be described in detail below.

[0126] Overview of the off-target prediction system provided by the present application Overview of the off-target prediction system provided by the present application The present application provides a method for predicting off-targets that may occur in a gene editing process. The present application provides a method for predicting off-targets that may occur in a genome editing process. In one embodiment, the gene editing process may be performed using a CRISPR / Cas gene editing system. In one embodiment, the genome editing process may be performed using a CRISPR / Cas gene editing system. In one embodiment of the present application, a method for predicting off-targets that may occur in a genome editing process using a CRISPR / Cas gene editing system is provided. Off-target encompasses the concept of off-target sites. For example, an off-target site or position may be described as an off-target. In the present application, off-target prediction may mean identifying an off-target candidate. Off-target prediction may mean identifying the position of an off-target candidate. Off-target prediction may mean identifying an off-target candidate. The descriptions of "off-target", "off-target prediction" and "off-target candidate" used herein should not be construed as limiting.

[0127] The novel off-target prediction system provided by the present application physically disrupts a cell, thereby contacting the genomic DNA of the cell and a gene editing protein (e.g., a Cas protein such as a Cas9 protein) and a gRNA, or a Cas / gRNA complex (e.g., a Cas9 / gRNA complex). By physically disrupting the cell, the cell membrane and / or nuclear membrane of the cell can be disrupted to create an environment in which the Cas / gRNA complex can contact the genome.

[0128] In one embodiment, physical disruption of cells can be performed using a filter with holes of appropriate size to have less impact on genomic DNA of cells. For example, an extruder with a filter with holes of appropriate size can be used to provide an environment in which genomic DNA of cells can contact with Cas / gRNA complexes. Pressure can be applied to the area containing cells, which can cause cells to pass through holes with a diameter smaller than the size of the cells, and the cells can be destroyed while passing through the holes. If the size of the holes is adjusted appropriately, the genomic DNA or the structure of genomic DNA in the cells (e.g., the structure according to epigenetic characteristics such as chromatin structure) may not be destroyed or altered in the cell disruption process.

[0129] Depending on the properties of the off-target system of the present application, an environment may be created in which the Cas protein and gRNA (or Cas / gRNA complex) can approach or contact the genomic DNA of the cell, while at the same time maintaining the genomic DNA more intact until it is cleaved by the Cas / gRNA complex.

[0130] The off-target prediction system (or method) provided by the present application may be referred to as Extru-seq.

[0131] The off-target prediction system provided by the present application can be mainly divided into two processes, including obtaining a composition to be analyzed (target composition) and analyzing the composition to be analyzed.

[0132] Here, obtaining the composition to be analyzed may be carried out by a process that includes providing an environment in which the genomic DNA can come into contact with the Cas / gRNA complex.

[0133] Analysis of the composition to be analyzed may be carried out by a process that involves analyzing the DNA (e.g., cleaved or uncleaved DNA) contained in the composition to be analyzed.

[0134] Below, we disclose a method for providing an environment in which genomic DNA can come into contact with a Cas / gRNA complex.

[0135] Methods for Providing an Environment in Which Genomic DNA Can Contact the Cas / gRNA Complex As described above, the method provided by the present application for predicting possible off-targets in the gene editing process can provide an environment in which genomic DNA can contact the Cas / gRNA complex.

[0136] In order to provide an environment in which the genomic DNA present in the cell (e.g., nucleus) can contact the Cas / gRNA complex, a process of disrupting the cell by physical methods (e.g., using physical force) may be carried out. Through cell disruption, the membrane structure of the cell, such as the cell membrane and / or the cell nuclear membrane, may be disrupted, or a space may be generated in the membrane structure such that the Cas protein and gRNA, or the Cas / gRNA complex, can access the genomic DNA. For example, the nuclear membrane of the cell may be disrupted to expose the genomic DNA to the Cas protein and gRNA. In another example, the cell membrane may be disrupted to provide an environment in which the Cas protein can contact the gRNA, and the Cas protein and gRNA (or the Cas / gRNA complex) may pass through the nuclear membrane of the cell to access the genomic DNA. In some embodiments, the Cas protein may be fused or linked to an NLS (i.e., an NLS-linked Cas protein may be provided), and the NLS fused or linked to the Cas protein may help the Cas protein (or the Cas / gRNA complex) to pass through the nuclear membrane of the cell. In one embodiment, cell destruction can cause destruction of the membrane structure of the cell.In one embodiment, cell destruction can cause destruction of the cell membrane.In one embodiment, cell destruction can cause destruction of the nuclear membrane of the cell.

[0137] In one embodiment, the above-mentioned cell destruction can be achieved through passing the cells through a porous structure with holes. The porous structure can be a filter or a membrane with holes. For example, the cell destruction can be performed by passing the cells through a filter with holes. For example, the cell destruction can be performed by passing the cells through holes with a diameter smaller than the size of the cells. For example, the cell destruction can be performed by passing the cells through holes with a diameter smaller than the size of the cell nucleus. Here, the driving force that allows the cells to pass through the filter can be pressure. In particular, pressure is applied to the part where the cells are located, and the applied pressure forces the cells to pass through holes smaller than the cell size. Here, the cells can be destroyed while passing through the holes smaller than the cell size. In one embodiment, the cell destruction can be performed by an extrusion process.

[0138] In one embodiment, cell disruption and contacting of the genomic DNA and the Cas / gRNA complex may be achieved by use of an extruder. Extruders and the use of extruders are described in detail later in this disclosure.

[0139] Although the above descriptions have been exemplified through Cas proteins and gRNAs or Cas / gRNA complexes, they may be fully applied to editing proteins used in gene editing systems other than the CRISPR / Cas gene editing system.

[0140] When genomic DNA contacts with Cas / gRNA complex, cleavage occurs at on-target and off-target positions on genomic DNA. Here, cleavage can be achieved by DSB or nick induced by Cas / gRNA complex (especially Cas protein). Since DNA repair mechanism of cell can be destroyed in cell destruction process, cleaved DNA cannot be repaired. By analyzing cleaved or uncleaved DNA, the position where off-target is likely to occur can be analyzed. That is, off-target (or off-target site) can be predicted, or off-target candidate (or candidate off-target site) can be identified.

[0141] Advantages of the method for predicting off-targets disclosed in the present application The inventors of the present application have extensively tested the method of predicting off-targets provided by the present application. By comparing the method of predicting off-targets of the present application with another off-target prediction method, it has been confirmed that the off-target prediction method of the present application performs better than another off-target prediction method (see the experimental examples of the present application). The off-target prediction method of the present application exhibits several advantages that other off-target prediction methods do not have. The off-target prediction method of the present application can have both the advantages of cell-based off-target prediction methods and the advantages of in vitro off-target prediction methods.

[0142] The off-target prediction method of the present application may have a lower false positive rate than the in vitro off-target prediction method. For example, the in vitro off-target prediction method has a high false positive rate because it is difficult to reflect epigenetic features such as chromatin structure and epigenetic modifications. Detecting a site that is not a true off-target as an off-target candidate may be represented by a false positive result. A high false positive rate may be associated with a low validation rate. Furthermore, in conventional in vitro off-target prediction methods, epigenetic features that may be cell-specific are difficult to reflect in in vitro off-target prediction results. However, in the off-target prediction method of the present application, since cells are physically destroyed without using chemical additives to maintain the structure of genomic DNA, the cell-specific environment may be partially maintained, resulting in a lower false positive rate. The off-target prediction method of the present application may show a high validation rate. Furthermore, epigenetic features may be reflected in the off-target prediction results.

[0143] The off-target prediction method of the present application may show a lower miss rate than the cell-based off-target prediction method. The miss rate may mean that the true off-target is missed. For example, false negative results such as when the true off-target site is not detected as an off-target candidate will increase the miss rate. For example, in the process of the cell-based prediction method, the involvement of DNA repair mechanisms may be unavoidable, and the repair cleavage sites repaired by DNA repair mechanisms will prevent the identification of true off-targets or off-target candidates. However, in the off-target prediction method of the present application, DNA repair mechanisms may not be involved because cells may be destroyed.

[0144] The off-target prediction method of the present application can be applied without limiting cell type. For example, cell-based prediction methods may be difficult to perform on some cells and difficult to apply to cells used in actual clinical practice. If off-target prediction is performed based on cells that are not related to the cells used in actual clinical practice, inaccurate results may be obtained. For example, epigenetic features vary by cell type, so the use of different cell types may lead to inaccurate results. However, off-target prediction methods have no or fewer restrictions on cell types.

[0145] Furthermore, the off-target prediction methods of the present application can be performed more conveniently and at lower cost than cell-based or in vitro off-target prediction methods.

[0146] The off-target prediction method of the present application involves physically destroying cells, so the above-mentioned advantages can occur. The inventors of the present application have tested and validated the off-target prediction method of the present application through numerous and various types of experiments. The advantages of the off-target prediction method of the present application have been confirmed through the experimental examples of the present application.

[0147] In one embodiment, the validation rate calculated based on the top 10 off-target candidates of the off-target candidates identified by the off-target prediction method of the present application may be 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%, or may be higher, but the present application is not limited thereto. In one embodiment, the validation rate calculated based on the top 10 off-target candidates of the off-target candidates identified by the off-target prediction method of the present application may be in a range formed by two values ​​selected from the above values, but the present application is not limited thereto. The validation rate may be affected by the type of gene editing tool and the type of cell used in the off-target prediction system.

[0148] In one embodiment, the missed rate of the off-target prediction method of the present application may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, or 40% or less, but the present invention is not limited thereto. In one embodiment, the missed rate of the off-target prediction method of the present application may be in a range formed by two values ​​selected from the above values, but the present application is not limited thereto. The missed rate may be affected by the type of gene editing tool and the type of cell used in the off-target prediction system.

[0149] In one embodiment, a receiver operating characteristic (ROC) curve for the off-target prediction method of the present application may be plotted. In one embodiment, the area under receiver operating characteristic curve (AUC) for the off-target prediction method of the present application may be calculated. ROC curves and AUC are powerful tools that can indicate the diagnostic ability of a binary classifier system. ROC curves can be roughly plotted by matching true positive rate (TPR) and false positive rate (FPR), or by matching sensitivity and specificity. For example, ROC curves can be plotted with TPR on the y-axis and FPR on the x-axis. For example, ROC curves can be plotted with sensitivity on the y-axis and specificity on the x-axis. The closer the AUC is to 1 (i.e., the larger the area of ​​the AUC), the higher the performance model of the system. In one embodiment, the AUC for the off-target prediction method of the present application may be calculated, and the AUC is approximately 0.4, 0.42, 0.44, 0.46, 0.48, 0.5, 0.52, 0.54, 0.56, 0.58, 0.6, 0.62, 0.64, 0.66, 0.68, 0.7, 0.72, 0.74, 0.75, 0.76, 0.77 , 0.78, 0.79, 0.8, 0.81, 0.82, 0.83, 0.84, 0.85, 0.86, 0.87, 0.88, 0.89, 0.9, 0.91, 0.92, 0.93, 0.94, 0.95, 0.96, 0.97, 0.98, 0.99, or more, or 1, but the application is not limited thereto. In one embodiment, the AUC calculated for the off-target prediction method of the application can be in the range of two values ​​selected from the above values, but the application is not limited thereto. The AUC can be influenced by the gene editing tool and cell type used in the off-target prediction system.

[0150] As described above, the off-target prediction method of the present application comprises obtaining a composition to be analyzed and analyzing the composition to be analyzed. It is clear to those skilled in the art that in addition to the above two processes, additional processes may be included. Hereinafter, the obtaining of the composition to be analyzed will be described in detail.

[0151] Acquisition of the composition to be analyzed The off-target prediction method of the present application may comprise a process of obtaining a composition to be analyzed. Here, the composition to be analyzed may refer to a composition containing cleaved DNA and / or uncleaved DNA. The off-target prediction method of the present application may be realized through a method comprising obtaining a composition to be analyzed and analyzing the composition to be analyzed (e.g., analyzing cleaved DNA contained in the composition to be analyzed). In order to obtain a composition to be analyzed containing cleaved DNA, the genomic DNA must be contacted with a Cas protein and a gRNA (or a Cas / gRNA complex). The Cas / gRNA complex cleaves the on-target and / or off-target site by contacting the genomic DNA. In order to bring the Cas / gRNA complex into contact with the genomic DNA, the cell may be destroyed. That is, an environment in which the genomic DNA can come into contact with the Cas / gRNA complex may be provided by cell destruction (e.g., destruction of the membrane structure of the cell).

[0152] One of the main processes for obtaining the composition to be analyzed is to physically destroy the cells. In one embodiment, the cells can be destroyed by physical methods. In one embodiment, the cells can be physically destroyed. In one embodiment, the cells can be destroyed by physical forces. In one embodiment, the composition to be analyzed can be obtained through destroying the cells from the starting composition. In one particular embodiment, the starting composition can include cells. In one particular embodiment, the starting composition can include cells and gene editing tools (e.g., Cas proteins and gRNAs).

[0153] The physical destruction of cells, which is one of the main features of the off-target prediction method of the present application, will be disclosed in detail below.

[0154] Physical disruption of cells Overview of physical cell disruption As mentioned above, the off-target prediction method of one embodiment of the present application may include physically destroying cells. Here, it is important to note that chemical additives that may cause damage to genomic DNA or genomic DNA structure (e.g., chromatin structure, etc.) are not used for the primary purpose of destroying cells. In one embodiment, the physical destruction of cells may be performed by forcing the cells to pass through a porous structure with pores smaller than the cell size. In one embodiment, the porous structure may be a filter with pores. The process of physically destroying cells, which is performed by forcing the cells to pass through a porous structure with pores smaller than the cell size, is described in detail below.

[0155] Cell disruption using filters with holes and pressure In one embodiment, physical disruption of the cells may be carried out by a method comprising passing the cells through a filter with pores smaller than the size of the cells, where the force for passing the cells through the filter may be pressure.

[0156] For example, pressure may be applied to the first container in which the first composition containing cells is located. Here, the applied pressure breaks the cells while passing them through a filter with pores smaller than the cell size. That is, the driving force for passing the cells through the filter may be pressure. When pressure is applied to the first container or the first composition, the mixed solution and the contained components (e.g., cells) in the first container may be released from the first container through the pores of the filter. In this process, the cells may be broken by the pores smaller than the cell size.

[0157] In one embodiment, the cell membrane may be disrupted by pores smaller than the size of the cell.In one embodiment, the cell membrane and the nuclear membrane may be disrupted by pores smaller than the size of the cell.

[0158] When referring to a plurality of cells, some or all of the cells may be destroyed in the process of passing through a hole smaller than the cell size, some of the cells may not be destroyed by passing through a hole larger than the cell size, or may not be destroyed by passing through a hole smaller than the cell size.

[0159] In one embodiment, the first composition located in the first container may further include a tool (e.g., a gene editing tool) used in the gene editing system. For example, the first composition located in the first container may further include a Cas protein and a gRNA. When pressure is applied to the first container, the cell is disrupted, and the components of the disrupted cell may move to the second container, and the Cas protein and the gRNA may move to the second container through the holes. In the second container located on the opposite side of the first container based on the filter, the Cas protein and the gRNA (or the Cas / gRNA complex) of the cell and the genomic DNA may come into contact. Here, the contact between the Cas / gRNA and the genomic DNA may be performed in a newly generated vesicle (e.g., a liposome) and / or in an extravesicular environment that is not inside the vesicle, but is not otherwise limited.

[0160] In another embodiment, the first composition located in the first container may contain cells, and the second container located on the opposite side of the first container based on the filter may contain tools used in the gene editing system. For example, when pressure is applied to the first container, the cells are broken, and the components of the broken cells move to the second container where the gene editing tools are present. Thus, in the second container, the gene editing tools (e.g., Cas proteins and gRNAs) can contact the DNA molecules derived from the cells.

[0161] In the second vessel, when contact between the gene editing tool and the DNA molecule is achieved, an environment is created in which the DNA molecule (e.g., genomic DNA) can be cleaved by the gene editing tool.

[0162] Filter with holes As mentioned above, cell disruption can be achieved by passing the cells through a porous structure with pores.

[0163] In one embodiment, the porous structure may be a filter with holes. In one embodiment, the filter may be a filter made of any one of polycarbonate, cellulose, mixed cellulose ester membrane, glass, polyethersulfone, nylon, polytetrafluoroethylene (PTFE), and PVDF, or a combination thereof, but the present application is not limited thereto, and may be a filter conventionally used in the biological and / or chemical fields. In a particular embodiment, the filter may be a polycarbonate membrane filter, but is not otherwise limited thereto.

[0164] In one embodiment, the filter may have pores with a diameter smaller than the cell size. In one embodiment, the filter may have pores with a diameter smaller than the average size of the cells. In one embodiment, the filter may have pores with a diameter smaller than the nucleus size of the cells. In one embodiment, the filter may have pores with a size smaller than the average size of the cell nucleus. The filter may be suitably designed depending on the type of cells. In one embodiment, the average diameter of the pores contained in the filter may be smaller than the cell size (e.g., cell diameter). In one embodiment, the average diameter of the pores contained in the filter may be smaller than the size of the cell nucleus (e.g., diameter of the cell nucleus).

[0165] In one embodiment, the filters are sized to approximately 0.1 μm, 0.2 μm, 0.3 μm, 0.4 μm, 0.5 μm, 0.6 μm, 0.7 μm, 0.8 μm, 0.9 μm, 1 μm, 1.5 μm, 2 μm, 2.5 μm, 3 μm, 3.5 μm, 4 μm, 4.54 μm, 5 μm, 5.5 μm, 6 μm, 6.5 μm, 7 μm, 7.5 μm, 8 μm, 8.5 μm, 9 μm, 9.5μm, 10μm, 11μm, 12μm, 13μm, 14μm, 15μm, 16μm, 17μm, 18μm, 19μm, 20μm, 21μm, 22μm, 23μm, 24μm, 25μm, 26μm, 27μm, 28μm, 29μm, 30μm, 31μm, 32μm, 33μm, 34μm, 35μm, 36μm, 37μm, 38μm, 39μm, 40μm, 4 1μm, 42μm, 43μm, 44μm, 45μm, 46μm, 47μm, 48μm, 49μm, 50μm, 51μm, 52μm, 53μm, 54μm, 55μm, 56μm, 5 7μm, 58μm, 59μm, 60μm, 61μm, 62μm, 63μm, 64μm, 65μm, 66μm, 67μm, 68μm, 69μm, 70μm, 71μm, 72μm, 73 In one embodiment, the filter may have pores with a diameter of any one of the following values: 74 μm, 75 μm, 76 μm, 77 μm, 78 μm, 79 μm, 80 μm, 81 μm, 82 μm, 83 μm, 84 μm, 85 μm, 86 μm, 87 μm, 88 μm, 89 μm, 90 μm, 91 μm, 92 μm, 93 μm, 94 μm, 95 μm, 96 μm, 97 μm, 98 μm, 99 μm, and 100 μm. In one embodiment, the filter may have pores with a diameter of or less than any one of the foregoing values.

[0166] In one embodiment, the average diameter of the pores contained in the filter is approximately 0.1 μm, 0.2 μm, 0.3 μm, 0.4 μm, 0.5 μm, 0.6 μm, 0.7 μm, 0.8 μm, 0.9 μm, 1 μm, 1.5 μm, 2 μm, 2.5 μm, 3 μm, 3.5 μm, 4 μm, 4.54 μm, 5 μm, 5.5 μm, 6 μm, 6.5 μm, 7 μm, 7.5 μm, 8 μm, 8.5 μm, 9μm, 9.5μm, 10μm, 11μm, 12μm, 13μm, 14μm, 15μm, 16μm, 17μm, 18μm, 19μm, 20μm, 21μm, 22μm, 23μm, 24μm, 25μm, 26μm, 27μm, 28μm, 29μm, 30μm, 31μm, 32μm, 33μm, 34μm, 35μm, 36μm, 37μm, 38μm, 39μm, 40μm, 41μm, 4 2μm, 43μm, 44μm, 45μm, 46μm, 47μm, 48μm, 49μm, 50μm, 51μm, 52μm, 53μm, 54μm, 55μm, 56μm, 57μm, 58μm, 5 9μm, 60μm, 61μm, 62μm, 63μm, 64μm, 65μm, 66μm, 67μm, 68μm, 69μm, 70μm, 71μm, 72μm, 73μm, 74μm, 75μm, 76 In one particular embodiment, the average diameter of the pores in the filter may be approximately 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, 10 μm, 11 μm, 12 μm, 13 μm, 14 μm, or 15 μm. In one particular embodiment, the filter may contain pores having an average diameter of 5 μm, 6 μm, 7 μm, 8 μm, 9 μm, 10 μm, 11 μm, 12 μm, 13 μm, 14 μm, or 15 μm, or smaller.

[0167] In one embodiment, the average diameter of the pores in the filter is 0.1 μm, 0.2 μm, 0.3 μm, 0.4 μm, 0.5 μm, 0.6 μm, 0.7 μm, 0.8 μm, 0.9 μm, 1 μm, 1.5 μm, 2 μm, 2.5 μm, 3 μm, 3.5 μm, 4 μm, 4.54 μm, 5 μm, 5.5 μm, 6 μm, 6.5 μm, 7 μm, 7.5 μm, 8 μm, 8.5 μm , 9μm, 9.5μm, 10μm, 11μm, 12μm, 13μm, 14μm, 15μm, 16μm, 17μm, 18μm, 19μm, 20μm, 21μm, 22μm, 23μm, 24μm, 25μm, 26μm, 27μm, 28μm, 29μm, 30μm, 31μm, 32μm, 33μm, 34μm, 35μm, 36μm, 37μm, 38μm, 39μm, 40 μm, 41μm, 42μm, 43μm, 44μm, 45μm, 46μm, 47μm, 48μm, 49μm, 50μm, 51μm, 52μm, 53μm, 54μm, 55μm, 56μ m, 57μm, 58μm, 59μm, 60μm, 61μm, 62μm, 63μm, 64μm, 65μm, 66μm, 67μm, 68μm, 69μm, 70μm, 71μm, 72μm, It may be in the range of two values ​​selected from 73 μm, 74 μm, 75 μm, 76 μm, 77 μm, 78 μm, 79 μm, 80 μm, 81 μm, 82 μm, 83 μm, 84 μm, 85 μm, 86 μm, 87 μm, 88 μm, 89 μm, 90 μm, 91 μm, 92 μm, 93 μm, 94 μm, 95 μm, 96 μm, 97 μm, 98 μm, 99 μm, and 100 μm.

[0168] In some embodiments, one or more filters may be used to achieve physical disruption of cells. For example, one filter may be used, e.g., a first filter having pores with a first average diameter. In another example, multiple filters may be used. For example, a first filter having pores with a first average diameter may be used, and a second filter having pores with a second average diameter (i.e., a filter having a pore profile different from the pore profile of the first filter) may be used. The type and number of filters that can be used to achieve physical disruption of cells are not otherwise limited.

[0169] pressure As mentioned above, in order for a cell to pass through a pore that is smaller than the cell size, a force must be applied to the area where the cell is located.

[0170] The force that makes the cells pass through the holes (e.g., break through the holes) can be pressure. That is, when pressure is applied to the area where the cells are located (e.g., the container containing the cells), the cells can break through the holes having a size smaller than the cell size. Here, the pressure can be applied by various methods and is not otherwise limited.

[0171] In one embodiment, applying pressure may be performed by a human. For example, pressure may be applied by pushing a piston designed to apply pressure to a vessel containing cells. In one embodiment, applying pressure may be achieved by a machine or device. For example, pressure may be applied by pushing a piston designed to apply pressure to a vessel containing cells through a machine. In another example, applying pressure may be achieved by centrifugation. In one embodiment, the pressure may be centrifugal force or osmotic pressure. The magnitude or strength of the applied force (e.g., pressure) is not otherwise limited. For example, a minimal force or pressure may be applied to force cells through the pores and / or filters, or a greater force or pressure may be applied.

[0172] In one embodiment, cell disruption may be achieved using an extruder. The method of disrupting cells using an extruder is described in detail below.

[0173] How to use an extruder to disrupt cells Extruder Overview In this specification, an extruder may refer to a tool or machine that includes a container and a porous structure with holes, and is designed to pass a composition filled in the container through the porous structure with holes by a force applied to the container. One example of an extruder used in the bio and chemical fields is the Avanti Mini-Extruder. The Mini-Extruder includes two containers contained in two syringes, respectively, and a porous filter (or membrane) located between the two containers. The inventors of the present application discovered that such an extruder structure is suitable for physically disrupting cells, and used the extruder to disrupt cells. The above description of the Avanti Mini-Extruder is an example to assist the understanding of those skilled in the art, and the extruder disclosed in this specification is not limited to the Mini-Extruder. The extruder disclosed in this specification may be recognized to include a tool or machine that includes at least one container and a porous structure (filter or membrane) and can achieve cell disruption.

[0174] In one embodiment, the term "extrusion" can be recognized to include a series of processes in which components located in a container are forced through a porous structure (filter or membrane) by pressure. For example, a process in which pressure is applied to a first container in which a composition containing cells is located, and the cells are forced through a filter while being broken by the pressure is an example of extrusion. In another example, a process in which Cas protein and / or gRNA are forced through a filter from a first container by pressure to a region other than the first container (e.g., a second container) is an example of extrusion.

[0175] In one embodiment, the extruder may be a unidirectional extruder designed to pass the filter once, but the present application is not limited thereto. In one embodiment, the extruder may be a bidirectional extruder designed to pass the filter multiple times, but the present application is not limited thereto. The bidirectional extruder may include at least two containers and a filter located between the two containers. For example, the Avanti Mini-Extruder described above may be a bidirectional extruder. In one embodiment, when a bidirectional extruder is used, the cell destruction rate may be increased by passing the cells through the filter multiple times, but the present application is not limited thereto.

[0176] Below, we will explain in more detail how the above-mentioned extruder can be used to create an environment in which gene editing tools and DNA molecules are accessible, including a cell disruption process.

[0177] Cell disruption through an extruder The off-target prediction method of the present application may include the use of an extruder. For example, an extruder including a first vessel, a second vessel, and a filter may be used. Here, the filter may be located between the first vessel and the second vessel. An example of the use of an extruder including a first vessel, a second vessel, and a filter is disclosed below.

[0178] In one embodiment, a starting composition including cells, Cas protein, and gRNA may be loaded into a first container. Pressure may be applied to the first container in which the starting composition is located. For example, applying pressure may be performed by pushing a piston connected to the first container designed to apply pressure to the first container. That is, pressure may be applied to the first container by pushing a piston connected to the first container in the direction of the first container and the filter. In one embodiment, pressure may be applied to move components of the starting composition (including cells, Cas protein, and gRNA) to a second container.

[0179] When pressure is applied to the first container, the components contained in the starting composition can move through a filter with holes into the second container. Here, cells larger than the size of the holes can be destroyed while passing through the filter. As described above, cell destruction can be the destruction of the cell membrane, or the destruction of the cell membrane and the nuclear membrane. Finally, a mixture containing the destroyed cells, the Cas protein, and the components obtained from the gRNA can be contained in the second container. In this process, the Cas / gRNA complex contacts with DNA (e.g., genomic DNA), which is one of the components obtained from the destroyed cells. Furthermore, the mixture in the second container may or may not contain unbroken cells.

[0180] Optionally, pressure may then be applied to the second container in which the mixture is located to move the components of the mixture (through a filter) into the first container. Thus, the mixture may fit into the first container. Optionally, pressure may then be applied to the first container in which the mixture is located to move the components of the mixture (through a filter) into the second container. Similarly, an extruder may be used such that the components of the composition or the components derived from the composition loaded into the extruder pass through its filter several times. Passing through the filter several times may increase the rate of cell destruction and / or increase the rate of contact between the Cas / gRNA complex and the genomic DNA.

[0181] In one embodiment, the extrusion may be performed n times. In one embodiment, the filter pass may be performed n times, where n may be an integer, where n may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, In one embodiment, n may be 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, or 100, but the invention is not limited thereto. In one embodiment, n may be any one or less of the aforementioned values. In one embodiment, n may be any one or more of the aforementioned values. In one embodiment, n may be within a range determined by any two selected from the aforementioned values.

[0182] The composition to be analyzed may be obtained through a process that includes the use of an extruder. In one embodiment, the extrusion process may be followed by any one or more of incubation, RNA removal, and DNA purification.

[0183] In one embodiment, after the extrusion process, an incubation process may be carried out to accumulate the cleavage rate. That is, after the cell disruption process, a process of incubating the composition containing the components of the disrupted cells may be further carried out. For example, the incubation time may be approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 26, 28, 30, 32, 36, 38, 40, 42, 44, 46, or 48 hours, or longer than the above values, but is not otherwise limited. For example, after the extrusion process is completed, through incubation (e.g., incubation at 37°C) and RNA removal, the composition to be analyzed may finally be obtained. Here, the composition to be analyzed is the composition to be used in the subsequent analysis process. Here, for example, the DNA contained in the composition to be analyzed may be suitable for analysis (e.g., sequencing). For example, the composition to be analyzed may include cleaved DNA. The composition to be analyzed may include uncleaved DNA in addition to cleaved DNA. The process of forming cleaved DNA has been fully described above. For example, the Cas / gRNA complex may contact DNA (e.g., genomic DNA), and the DNA may be cleaved via DSB or nick induced by the Cas / gRNA complex.

[0184] Analysis of the composition to be analyzed Composition to be analyzed As described above, the composition to be analyzed may be obtained by a method comprising contacting a gene editing tool (e.g., Cas / gRNA complex) and genomic DNA according to one embodiment of the present application. In one embodiment, the composition to be analyzed may include cleaved genomic DNA. In one embodiment, the composition to be analyzed may include one or more cleaved DNAs (e.g., double-stranded DNA or single-stranded DNA). In one embodiment, the cleaved genomic DNA may include one or more cleavage at one or more cleavage sites. For example, as described above, the one or more cleavage may occur via a DSB or a nick induced by the Cas / gRNA complex contacting the genomic DNA. In one embodiment, the cleavage site may involve an off-target site or an on-target site. In one embodiment, the cleavage site may be an off-target site or an on-target site. That is, cleavage may occur via a DSB or a nick induced (or generated) at an off-target site or an on-target site by the Cas / gRNA complex contacting the genomic DNA.

[0185] In one embodiment, the composition to be analyzed can take advantage of the in vitro off-target prediction system. For example, the cleaved genomic DNA contained in the composition to be analyzed may not be repaired genomic DNA, because some or all of the DNA repair mechanisms have been inactivated.

[0186] In one embodiment, the composition to be analyzed can take advantage of the cell-based off-target prediction system. For example, the cleaved genomic DNA contained in the composition to be analyzed can reflect the cell's unique epigenetic signature.

[0187] After the composition to be analyzed is obtained, it can be analyzed to obtain information about the cleavage of genomic DNA.Therefore, information about the off-target candidates that may be generated with the use of gene editing system can be obtained.The information about the off-target candidates can be applied to predict off-target.That is, the off-target that is likely to be generated when using gene editing system (e.g., CRISPR / Cas gene editing system) can be predicted.

[0188] Disclosed below are methods for obtaining information about potential off-targets by analyzing a composition of interest.

[0189] Summary of the analysis of the composition being analyzed As described above, the method of the present application comprises analyzing the composition to be analyzed, which contains the cleaved genomic DNA obtained above. Information about the cleavage of the genomic DNA (e.g., information about one or more cleavage positions and / or cleavage scores at one or more cleavage positions) can be obtained by analyzing the composition to be analyzed. Based on the information about the cleavage of the genomic DNA, information about off-target candidates (e.g., information about one or more off-targets and / or scores for one or more off-targets) can be obtained.

[0190] Analysis of the composition of interest and obtaining information on genomic DNA cleavage In one embodiment, information about the cleavage of genomic DNA can be obtained by analyzing DNA (e.g., cleaved and / or uncleaved genomic DNA) contained in the composition to be analyzed. In one embodiment, information about the cleavage of genomic DNA can be obtained by analyzing cleaved DNA contained in the composition to be analyzed. In one embodiment, information about the cleavage of genomic DNA can be obtained by analyzing one or more cleavage sites. Here, the analysis method capable of identifying the cleavage site of cleaved DNA is not particularly limited. For example, any analysis method capable of identifying the cleavage site of cleaved DNA can be sufficiently used in the off-target prediction method of the present application.

[0191] In one embodiment, DNA analysis can be performed through DNA analysis methods well known to those skilled in the art. In one embodiment, DNA analysis can be performed by any one or more selected from PCR-based assays (see Cameron, Peter, et al. "Mapping the genomic landscape of CRISPR-Cas9 cleavage." Nature methods 14.6(2017):600-606.) and sequencing (see Metzker, Michael L. "Sequencing technologies-the next generation." Nature reviews genetics 11.1(2010):31-46.; and Kumar, Kishore R., Mark J. Cowley, and Ryan L. Davis. "Next-generation sequencing and emerging technologies." Seminars in thrombosis and hemostasis. Vol. 45. No. 07. Thieme Medical Publishers, 2019.) (e.g., DNA sequencing).

[0192] For example, the sequencing may be any one or more of the sequencing methods referred to as whole genome sequencing (WGS), deep sequencing, high throughput sequencing (HTS), de novo sequencing, second generation sequencing, next generation sequencing, third generation sequencing, deep sequencing, shotgun sequencing, long read sequencing, and short read sequencing, but is not otherwise limited to.

[0193] In one embodiment, the sequencing depth of the sequencing method used in the analysis of the composition to be analyzed can be approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000x.In one embodiment, the sequencing depth can be in the range of two values ​​selected from the above values.In one embodiment, the sequencing depth can be lower or higher than the above values. In a particular embodiment, the sequencing depth of the sequencing used in the analysis may be approximately 10 to 40x. The sequencing depth is not otherwise limited, and any sequencing depth that can confirm the cleavage site on the DNA is sufficient.

[0194] Information about DNA cleavage In one embodiment, by analyzing the analyte composition, information about DNA cleavage (eg, information about cleavage sites in genomic DNA) can be obtained.

[0195] In one embodiment, the information about DNA cleavage may include information about one or more cleavage sites, where the cleavage sites may be generated by a gene editing tool.

[0196] In one embodiment, the information about DNA cleavage may include information about the location of one or more cleavage sites on genomic DNA. For example, the information about DNA cleavage may include location information about each cleavage site on genomic DNA among all cleavage sites on cleaved DNA contained in the composition to be analyzed. For example, the information about DNA cleavage may include location information about each cleavage site on genomic DNA among one or more cleavage sites on cleaved DNA contained in the composition to be analyzed. That is, location information about all cleavage sites or location information about some cleavage sites may be obtained through analysis. The location information obtained thereby may be associated with off-target candidates and / or on-target sites. For example, by comparing the location information about the identified cleavage site and the predetermined on-target site, it may be determined whether the cleavage site is associated with an off-target candidate or on-target.

[0197] In one embodiment, the information about DNA cleavage may include a cleavage score for one or more cleavage sites. For example, the information about DNA cleavage may include a cleavage score for each cleavage site among all cleavage sites on the cleaved DNA contained in the composition to be analyzed. For example, the information about DNA cleavage may include a cleavage score for each cleavage site among all cleavage sites on the cleaved DNA of the composition to be analyzed. That is, cleavage scores for all cleavage sites or scores for some cleavage sites may be obtained through analysis. In one embodiment, the cleavage score may be calculated from the sequence read. In one embodiment, the cleavage score may be calculated from the results of the Manhattan plot. The calculation mechanism for the cleavage score is not limited otherwise and may be suitably selected depending on which analysis method is used. In one embodiment, a cleavage ranking may be calculated based on the cleavage score. For example, a cleavage site showing a high cleavage score may be ranked higher. For example, a cleavage site showing the highest cleavage score may be ranked first. In one embodiment, the cleavage score may be related to the cleavage rate of the corresponding cleavage site, and the information obtained about the cleavage score may be related to the scores of off-target candidates and / or on-target candidates.

[0198] In one embodiment, the information about DNA cleavage may include information about the number of cleavage sites generated. For example, the total number of cleavage sites may be calculated. For example, in one calculation of the number of cleavage sites, overlapping sites may be counted as 1. In another example, in another calculation of the number of cleavage sites, overlapping sites may be counted as multiples. For example, if 5 DNAs show cleavage at cleavage site x, the result may be counted as 1 or 5 as appropriate. From the information about the number of cleavage sites, the total number of off-target candidates that may be generated by the use of the gene editing system may be ascertained.

[0199] In one embodiment, information about DNA cleavage obtained by analyzing the analyte composition includes, but is not limited to, the location of the cleavage site or sites on the genomic DNA; a cleavage score for one or more cleavage sites; and The number of cleavages that occurred may include any one or more of:

[0200] In one embodiment, the process of obtaining information about DNA cleavage by analyzing the composition to be analyzed may further include additional processes for obtaining information about DNA cleavage. For example, the process may further include processes such as information (or data) processing and / or normalization of the obtained information (or data). For example, the process may further include a process of comparing the obtained cleavage information with information about the on-target previously determined. The process of obtaining cleavage information may further include additional processes such as those described above, but is not otherwise limited.

[0201] In one embodiment, the information about DNA cleavage may further include other information that can be obtained through analysis of the composition being analyzed (e.g., DNA sequencing), but is not otherwise limited to such.

[0202] Obtaining information about off-target In one embodiment, information about off-target candidates can be obtained from the obtained cleavage information.Those skilled in the art in the field related to the present application can easily obtain information about off-targets based on cleavage information, and therefore the present disclosure does not limit the process of the off-target prediction system of the present application.Those skilled in the art in the field related to the present application can obtain information about off-targets by using cleavage information (e.g., information about DNA cleavage) obtained by analyzing the composition to be analyzed through a suitable process or without additional process.

[0203] In one embodiment, the off-target prediction method of the present application may include a process of identifying information about off-target candidates from obtained information about cleavage.

[0204] In one embodiment, the information about the off-target candidates may include information about the positions of one or more off-target candidates on genomic DNA (e.g., information about candidate off-target sites). For example, the information about the positions of the off-target candidates may include information about each position (position on genomic DNA) of all the off-target candidates. For example, the information about the positions of the off-target candidates may include information about each position of one or more off-target candidates. That is, position information about all the candidate off-target sites may be obtained, or position information about one or more candidate off-target sites but not about all the candidate off-target sites may be obtained. Among the off-target candidates, there may be true off-targets (e.g., true off-targets generated from the use of a gene editing system). The information about the positions of the off-target candidates may be obtained based on the above-mentioned cleavage information (e.g., position information about one or more cleavage sites).

[0205] In one embodiment, the information about the off-target candidates may include off-target scores (e.g., off-target prediction scores) for one or more off-target candidates. For example, the information about the off-target candidates may include off-target scores for each off-target candidate for all off-target candidates. For example, the information about the off-target candidates may include off-target scores for each off-target candidate for one or more off-target candidates. That is, off-target scores for all candidate off-target sites may be obtained, or off-target scores for one or more candidate off-target sites may be obtained but not for all candidate off-target sites. The information about the off-target scores for the off-target candidates may be obtained based on the cleavage information (e.g., scores for one or more cleavage sites) described above. In one embodiment, a ranking of the off-target candidates may be calculated based on the obtained off-target scores. For example, off-target candidates (e.g., candidate off-target sites) having higher off-target scores may be ranked higher. For example, the off-target candidate having the highest off-target score may be ranked first. For example, a high off-target score of a candidate off-target may be associated with a true off-target, but is not otherwise limited.

[0206] In one embodiment, the information about off-target candidates may include information about the number of off-target candidates. For example, the total number of off-target candidates may be calculated. For example, in calculating the number of off-target candidates, overlapping sites may be counted as 1. In another example, in calculating the number of off-target candidates, overlapping sites may be counted as multiples. For example, if 5 candidate off-target sites x are found, they may be counted as 1 or 5. From the information about the number of off-target candidates, the total number of off-target candidates that can be generated by using the gene editing system can be identified. That is, the total number of predicted off-targets can be identified.

[0207] In one embodiment, information about off-target candidates includes, but is not limited to, a location on genomic DNA for each of one or more off-target candidates; an off-target score for each of one or more off-target candidates; and Number of predicted off-target candidates may include any one or more of:

[0208] In one embodiment, the process of obtaining information about off-target candidates may further include additional processes for obtaining information about off-target candidates. For example, processes such as information (or data) processing and / or normalization of obtained information (or data) may further be included. For example, a process of comparing the obtained information about off-target candidates with information about previously determined on-targets may further be included. As described above, the process of obtaining information about off-target candidates may further include additional processes, but is not otherwise limited.

[0209] In one embodiment, the information about potential off-targets may further include additional information that aids in the prediction of off-targets likely to be generated by using the gene editing system, but is not otherwise limited.

[0210] Relationship with predicted gene editing systems The off-target prediction system of the present application may be related to a gene editing system to be predicted. Here, the gene editing system to be predicted may refer to a gene editing system determined to be used for research or treatment, but is not otherwise limited thereto. That is, the gene editing system to be predicted may refer to a gene editing system (or gene editing process) for which off-target needs to be predicted.

[0211] For example, when a specific cell is used in the gene editing system to be predicted, the specific cell can also be used in the method of predicting off-target of the present application.For example, when a guide RNA having a specific guide sequence is used in the gene editing system to be predicted, the guide RNA having the same guide sequence can also be used in the method of predicting off-target of the present application.

[0212] In this aspect, the method for predicting off-target according to one embodiment of the present application may further comprise a process of confirming the gene editing system to be predicted. The gene editing system to be predicted may be referred to as a predetermined gene editing system. The predetermined gene editing (e.g., genome editing) system may include one or more of a cell to be subjected to gene editing (genome editing) and a predetermined gene editing tool. The predetermined gene editing tool may include, for example, multiple types of guide RNA, guide sequence, and gene editing protein (e.g., Cas protein).

[0213] In one embodiment, the method for predicting off-targets of the present application may further comprise confirming or designing a predetermined gene editing system. The predetermined gene editing system may be confirmed, and thereby a component suitable for use in the off-target prediction system may be designed. Here, the process of confirming the predetermined gene editing system may be performed before the composition to be analyzed is obtained. Hereinafter, an example of confirmation of a predetermined gene editing system (to be predicted) is shown based on a CRISPR / Cas gene editing system. The exemplary explanation based on the CRISPR / Cas gene editing system does not limit the aspects of the off-target prediction system of the present application, and it is suggested that the exemplary explanation is fully applicable to different gene editing systems in the same or the same context as the following explanation.

[0214] In one embodiment, the method for predicting off-targets of the present application may include a predetermined CRISPR / Cas gene editing system, where the confirmation of the predetermined CRISPR / Cas gene editing system may include confirmation of any one or more of information on a predetermined cell (i.e., a cell for editing to be used in the predicted CRISPR / Cas-based gene editing), a predetermined type of Cas protein (i.e., a type of Cas protein to be used in the predicted CRISPR / Cas-based gene editing), and a predetermined guide RNA (a guide RNA sequence or a guide sequence).

[0215] In a particular embodiment, validating the predetermined CRISPR / Cas gene editing system may include validating a predetermined cell. In a particular embodiment, the same cell as the predetermined cell may be used in the off-target prediction system of the present application. As a result, the cell's inherent characteristics may be reflected in the results of the off-target prediction system. The cell undergoing genome editing is not otherwise limited. In one embodiment, the predetermined cell may be an animal cell or a plant cell. In one embodiment, the predetermined cell may be, but is not otherwise limited to, a human cell or a cell derived from a non-human animal (e.g., mouse, rat, dog, cat, cow, pig, horse, and sheep). In a particular embodiment, the predetermined cell may be a human cell.

[0216] In a specific embodiment, validating the predetermined CRISPR / Cas gene editing system may include validating the predetermined Cas protein. In a specific embodiment, the same Cas protein as the predetermined Cas protein may be used in the off-target prediction system of the present application. As a result, the properties that may be influenced by the Cas protein may be reflected in the result of the off-target prediction system. In one embodiment, it may be a gene editing system that uses SpCas9.

[0217] In a specific embodiment, validating the predefined CRISPR / Cas gene editing system may include validating the predefined guide RNA. In a specific embodiment, the predefined guide RNA may be used in the off-target prediction system of the present application. As a result, the properties that may be influenced by the guide RNA may be reflected in the results of the off-target prediction system.

[0218] In a specific embodiment, validating the pre-defined CRISPR / Cas gene editing system may include validating the pre-defined guide sequence. In a specific embodiment, the off-target prediction system of the present application may use a guide RNA having the same guide sequence as the pre-defined guide sequence. As a result, the properties influenced by the guide sequence may be reflected in the results of the off-target prediction system.

[0219] In a specific embodiment, the off-target prediction system of the present application may use any one or more selected from a cell identical to a predetermined cell, a Cas protein identical to a predetermined Cas protein, and a guide RNA having a guide sequence identical to a predetermined guide sequence.

[0220] The above description does not necessarily limit that the same components as those used in the predetermined gene editing system must be used in the off-target prediction system, and the off-target prediction system of the present application may be suitably selected according to the purpose of the off-target prediction system used by a person skilled in the art. For example, a type of Cas protein different from the predetermined Cas protein (e.g., a Cas protein known to have similar properties) may be used in the off-target prediction system. In another example, a type of cell different from the predetermined cell (e.g., a cell known to have similar properties) may be used in the off-target prediction system. In yet another example, a type of guide RNA different from the predetermined guide RNA (e.g., a guide RNA modified to be more effectively applied to the off-target prediction system) may be used in the off-target prediction system.

[0221] Can be used with different off-target prediction systems In one embodiment, the off-target prediction system of the present application may be used together with different off-target prediction systems. For example, the off-target prediction system of the present application may be used together with any one or more selected from an in silico-based off-target prediction system, an in vitro-based off-target prediction system, and a cell-based off-target prediction system. For example, the off-target prediction system of the present application may be used together with any one or more selected from Cas-OFFinder, CHOPCHOP, CRISPOR, Digenome-seq, DIG-seq, SITE-seq, CIRCLE-seq, CHANGE-seq, GUIDE-seq, GUIDE-tag, DISCOVER-seq, BLISS, BLESS, integrase-deficient lentiviral vector-mediated DNA cleavage capture, HTGTS, ONE-seq, CReVIS-Seq, ITR-seq, and TAG-seq. In order to more effectively discover true off-target sites, an off-target prediction system different from the off-target prediction system of the present application may be used, and the different off-target prediction system may be an off-target prediction system developed before the filing date of the present application or an off-target prediction system developed after the filing date of the present application, but is not otherwise limited thereto.

[0222] Starting Compositions and Components That May Be Included in the Starting Compositions Summary of starting composition As described above, according to one embodiment of the present application, the composition to be analyzed can be obtained by disrupting cells, and information about the cleavage of genomic DNA can be obtained by analyzing the composition to be analyzed.

[0223] In one embodiment, the composition to be analyzed can be obtained from the starting composition containing cells through cell destruction. In one embodiment, the starting composition can further contain gene editing tools (e.g., Cas protein and guide RNA) in addition to cells. Below, the conditions regarding the components that can be included in the starting composition of the off-target prediction method of one embodiment of the present application are disclosed.

[0224] Cells and cell concentration In one embodiment, the starting composition may include cells. In one embodiment, the concentration of cells included in the starting composition is approximately 1x10 5 cells / mL, 2x10 5 cells / mL, 3x10 5 cells / mL, 4x10 5 cells / mL, 5x10 5 cells / mL, 6x10 5 cells / mL, 7x10 5 cells / mL, 8x10 5 cells / mL, 9x10 5 cells / mL, 1x10 6 cells / mL, 2x10 6 cells / mL, 3x10 6 cells / mL, 4x10 6 cells / mL, 5x10 6 cells / mL, 6x10 6 cells / mL, 7x10 6 cells / mL, 8x10 6 cells / mL, 9x10 6 cells / mL, 1x10 7 cells / mL, 2x10 7 cells / mL, 3x10 7 cells / mL, 4x10 7 cells / mL, 5x10 7 cells / mL, 6x10 7 cells / mL, 7x10 7 cells / mL, 8x10 7 cells / mL, 9x10 7 cells / mL, 1x10 8 cells / mL, 2x10 8 cells / mL, 3x10 8 cells / mL, 4x108 cells / mL, 5x10 8 cells / mL, 6x10 8 cells / mL, 7x10 8 cells / mL, 8x10 8 cells / mL, 9x10 8 cells / mL, 1x10 9 cells / mL, 2x10 9 cells / mL, 3x10 9 cells / mL, 4x10 9 cells / mL, 5x10 9 cells / mL, 6x10 9 cells / mL, 7x10 9 cells / mL, 8x10 9 cells / mL, 9x10 9 cells / mL, 1x10 10 cells / mL, 2x10 10 cells / mL, 3x10 10 cells / mL, 4x10 10 cells / mL, 5x10 10 cells / mL, 6x10 10 cells / mL, 7x10 10 cells / mL, 8x10 10 cells / mL, or 9x10 10 The concentration of cells in the starting composition may be, but is not otherwise limited to, cells / mL. In one embodiment, the concentration of cells in the starting composition may be in a range between two values ​​selected from the values ​​above. In one embodiment, the concentration of cells in the starting composition may be higher or lower than any one value selected from the values ​​above. In one particular embodiment, the concentration of cells in the starting composition is approximately 1x10 6 cells / mL, 2x10 6 cells / mL, 3x10 6 cells / mL, 4x10 6 cells / mL, 5x10 6 cells / mL, 6x10 6 cells / mL, 7x10 6 cells / mL, 8x10 6 cells / mL, 9x10 6 cells / mL, 1x10 7 cells / mL, 2x10 7 cells / mL, 3x107 cells / mL, 4x10 7 cells / mL, 5x10 7 cells / mL, 6x10 7 cells / mL, 7x10 7 cells / mL, 8x10 7 cells / mL, 9x10 7 cells / mL or 1x10 8 It may be in cells / mL.

[0225] The cells that can be used in the off-target prediction system of the present application are not otherwise limited. In one embodiment, the cells can be animal cells or plant cells. In one embodiment, the cells can be human cells or cells derived from non-human animals (e.g., mice, rats, dogs, cats, cows, pigs, horses, and sheep), but are not otherwise limited. In a particular embodiment, the cells can be human cells.

[0226] Gene editing tools and concentration of editing tools In one embodiment, the starting composition may comprise a gene editing tool, hi one embodiment, the starting composition may comprise a Cas protein and a gRNA.

[0227] In one embodiment, the starting composition includes a Cas protein at a concentration of approximately 10 nM, 20 nM, 30 nM, 40 nM, 50 nM, 60 nM, 70 nM, 80 nM, 90 nM, 100 nM, 200 nM, 300 nM, 400 nM, 500 nM, 600 nM, 700 nM, 800 nM, 900 nM, 1000 nM (1 μM), 2000 nM, 3000 The concentration of the Cas protein in the starting composition may be, but is not otherwise limited to, 4000nM, 5000nM, 6000nM, 7000nM, 8000nM, 9000nM, 10000nM (10 μM), 20000nM, 30000nM, 40000nM, 50000nM, 60000nM, 70000nM, 80000nM, 90000nM, or 100000nM (100 μM). In one embodiment, the concentration of the Cas protein in the starting composition may be in a range between two values ​​selected from the aforementioned values. In one embodiment, the concentration of the Cas protein in the starting composition may be higher or lower than any one value selected from the aforementioned values. In a particular embodiment, the starting composition may comprise a Cas protein at a concentration of approximately 1000 nM (1 μM), 2000 nM, 3000 nM, 4000 nM, 5000 nM, 6000 nM, 7000 nM, 8000 nM, 9000 nM, or 10000 nM (10 μM).

[0228] In one embodiment, the starting composition contains a guide RNA at a concentration of approximately 10 nM, 20 nM, 30 nM, 40 nM, 50 nM, 60 nM, 70 nM, 80 nM, 90 nM, 100 nM, 200 nM, 300 nM, 400 nM, 500 nM, 600 nM, 700 nM, 800 nM, 900 nM, 1000 nM (1 μM), 2000 nM, 3000 nM, The concentration of the guide RNA in the starting composition may be, but is not otherwise limited to, M, 4000nM, 5000nM, 6000nM, 7000nM, 8000nM, 9000nM, 10000nM (10 μM), 20000nM, 30000nM, 40000nM, 50000nM, 60000nM, 70000nM, 80000nM, 90000nM, or 100000nM (100 μM). In one embodiment, the concentration of the guide RNA in the starting composition may be in a range between two values ​​selected from the above values. In one embodiment, the concentration of the guide RNA in the starting composition may be higher or lower than any one value selected from the above values. In a particular embodiment, the concentration of guide RNA included in the starting composition may be approximately 1000 nM (1 μM), 2000 nM, 3000 nM, 4000 nM, 5000 nM, 6000 nM, 7000 nM, 8000 nM, 9000 nM, or 10000 nM (10 μM).

[0229] In one embodiment, the starting composition may include ribonucleoproteins (RNPs) (e.g., Cas / gRNA complexes). In this case, since the Cas protein and gRNA are present in the form of RNPs in the starting composition, the off-target prediction method of the present application may further include mixing the guide RNA and Cas protein and pre-incubating the mixture. That is, before providing the starting composition, the method may further include a process of incubating the mixture containing the guide RNA and Cas protein. For example, RNPs (Cas / gRNA complexes) may be obtained from the incubated mixture containing the guide RNA and Cas protein, and the obtained RNPs may be mixed with cells to obtain the starting composition. In one embodiment, the starting composition includes an RNP (e.g., Cas / gRNA complex) at a concentration of approximately 10 nM, 20 nM, 30 nM, 40 nM, 50 nM, 60 nM, 70 nM, 80 nM, 90 nM, 100 nM, 200 nM, 300 nM, 400 nM, 500 nM, 600 nM, 700 nM, 800 nM, 900 nM, 1000 nM (1 μM), 20 ... The RNP concentration may be, but is not limited to, M, 3000nM, 4000nM, 5000nM, 6000nM, 7000nM, 8000nM, 9000nM, 10000nM (10 μM), 20000nM, 30000nM, 40000nM, 50000nM, 60000nM, 70000nM, 80000nM, 90000nM, or 100000nM (100 μM). In one embodiment, the RNP concentration in the starting composition may be in a range between two values ​​selected from the above values. In one embodiment, the RNP concentration in the starting composition may be higher or lower than any one value selected from the above values. In a particular embodiment, the concentration of RNP included in the starting composition may be approximately 1000 nM (1 μM), 2000 nM, 3000 nM, 4000 nM, 5000 nM, 6000 nM, 7000 nM, 8000 nM, 9000 nM, or 10000 nM (10 μM).

[0230] Exemplary embodiment of the off-target prediction system of the present application (1) Below, exemplary embodiments (non-limiting embodiments) of the off-target prediction system of the present application are disclosed. The following embodiments may feature the mechanism of the off-target prediction system of the present application. Some or all of the following embodiments may include one or all of the embodiments disclosed in the embodiment featuring the use of an extruder described below.

[0231] One embodiment of the present application provides a method for predicting off-targets that may occur in a gene editing (e.g., genome editing) process. One embodiment of the present application provides a method for confirming off-target candidates that may occur in a genome editing process. One embodiment of the present application provides a method for predicting off-targets of a CRISPR / Cas gene editing system. One embodiment of the present application provides a method for confirming off-target candidates that may be generated in a gene editing process using a CRISPR / Cas gene editing system. The description of the method for predicting off-targets that may be generated in a genome editing process or the method for confirming information about off-targets may be used without limitation.

[0232] One embodiment of the present application is (i) obtaining a composition to be analyzed that contains cleaved genomic DNA; and (ii) obtaining information about the cleavage sites by analyzing the composition to be analyzed; The present invention provides a method for predicting off-targets that may occur in a gene editing process, comprising:

[0233] In one particular embodiment, the cleaved genomic DNA contained in the composition to be analyzed may be cleaved genomic DNA formed by cleaving genomic DNA of a cell that has been physically disrupted by a gene editing system.

[0234] In a particular embodiment, the cleaved genomic DNA may have a cell-specific epigenetic signature.

[0235] In a particular embodiment, the cleaved genomic DNA may not be repaired genomic DNA.

[0236] One embodiment of the present application is (i) preparing a starting composition comprising a gene editing tool and a first cell; (ii) obtaining the composition to be analyzed by physically disrupting the first cell, where the physical disruption of the first cell leads to the creation of an environment in which the genomic DNA in the cell can come into contact with the gene editing tool, whereby the genomic DNA and the gene editing tool come into contact, whereby the genomic DNA is cleaved at one or more cleavage sites; and (iii) obtaining information about the cleavage sites by analyzing the composition to be analyzed; The present invention provides a method for predicting off-targets that may occur in a gene editing (e.g., genome editing) process, comprising:

[0237] In one particular embodiment, the method for predicting off-targets comprises: (iv) identifying information about potential off-targets from the information about the cleavage sites obtained in (iii); may further be provided.

[0238] One embodiment of the present application is (i) preparing a starting composition comprising a first Cas protein, a first guide RNA, and a first cell, wherein the Cas protein and the guide RNA are capable of forming a Cas / gRNA complex; (ii) obtaining the composition to be analyzed by physically disrupting the first cell, whereby an environment is created in which the genomic DNA can contact the Cas / gRNA complex through the physical disruption of the first cell, whereby the genomic DNA and the Cas / gRNA complex come into contact, where the genomic DNA is cleaved at one or more cleavage sites; and (iii) obtaining information about the cleavage sites by analyzing the composition to be analyzed; The present invention provides a method for predicting off-targets that may occur in a CRISPR / Cas gene editing system, comprising:

[0239] In a particular embodiment, the method for predicting off-targets may further comprise the step of (iv) confirming information about potential off-targets from the information about cleavage sites obtained in (iii).

[0240] In a particular embodiment, the information about the cleavage sites may include one or more of the location of the cleavage site(s) on the genomic DNA, a cleavage score for the cleavage site(s), and the number of cleavage sites.

[0241] In a particular embodiment, the location of the one or more cleavage sites on the genomic DNA may be the location of each of the one or more cleavage sites on the genomic DNA.

[0242] In a particular embodiment, the cleavage score for the one or more cleavage sites may be the cleavage score for each of the one or more cleavage sites.

[0243] In a particular embodiment, the number of cleavage sites may be the total number of cleavage sites.

[0244] In a particular embodiment, the information about the off-target candidates may include one or more of: a location of the one or more off-target candidates on the genomic DNA; an off-target prediction score for the one or more off-target candidates; and a number of predicted off-target candidates.

[0245] In a particular embodiment, the location of the one or more off-target candidates on the genomic DNA may be a location on each of the one or more off-target candidates on the genomic DNA.

[0246] In a particular embodiment, the off-target prediction score for one or more off-target candidates may be the off-target prediction score for each of the one or more off-target candidates.

[0247] In a particular embodiment, the number of off-target candidates may be the total number of predicted off-target candidates.

[0248] In a particular embodiment, in (ii), the membrane structure comprising the cell membrane of the first cell may be disrupted by physically disrupting the first cell, thereby providing an environment in which the Cas / gRNA complex can contact the genomic DNA.

[0249] In a particular embodiment, in (ii), the membrane structure including the nuclear membrane of the first cell can be disrupted by physically disrupting the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA.

[0250] In one particular embodiment, the physical disruption of the first cells may include passing the first cells through a filter with pores.

[0251] In one particular embodiment, the force that causes the first cell to pass through a pore smaller than the size of the first cell can be pressure.

[0252] In one particular embodiment, the average pore size of the filter may be smaller than the size of the first cell.

[0253] In one particular embodiment, the filter may have pores with a diameter smaller than the size of the first cells.

[0254] In one particular embodiment, the average pore size of the filter may be between 5 and 15 μm.

[0255] In one particular embodiment, the average pore size of the filter may be approximately 8 μm.

[0256] In one particular embodiment, the filter may have pores with a diameter of 5 to 15 μm.

[0257] In one particular embodiment, the physical disruption of the first cells may be carried out by use of an extruder.

[0258] In one particular embodiment, the extruder may include a filter with pores, and the filter may have pores with a diameter smaller than the first cell size.

[0259] In one particular embodiment, the extruder may be equipped with a filter with holes, and the average pore size of the filter may be between 5 and 15 μm.

[0260] In a particular embodiment, the exposed genomic DNA upon physical disruption of the first cell may maintain the cell-specific epigenetic characteristics (eg, chromatin structural characteristics) of the first cell.

[0261] In one particular embodiment, the information about the off-target candidates may be information reflecting the intrinsic epigenetic signature of the first cell.

[0262] In a particular embodiment, the environment in which the genomic DNA and the Cas / gRNA complex can come into contact can be an environment in which DNA repair mechanisms are inactivated.

[0263] In one particular embodiment, cell disruption inactivates the cell's DNA repair mechanisms, so that the cleaved DNA cannot be repaired.

[0264] Certain embodiments may provide a method of predicting off-targets, further comprising validating the predetermined CRISPR / Cas gene editing system subjected to the off-target prediction, where the predetermined CRISPR / Cas gene editing system may include any one or more of the use of a predetermined cell, the use of a predetermined Cas protein, and the use of a predetermined guide RNA, where the validation of the predetermined CRISPR / Cas gene editing system may be performed prior to (i).

[0265] In a particular embodiment, the guide sequence of the first guide RNA may have the same sequence as the guide sequence of the predetermined guide RNA.

[0266] In a particular embodiment, the predetermined CRISPR / Cas gene editing system may include the use of a predetermined cell, and the first cell and the predetermined cell may be the same.

[0267] In a particular embodiment, in (iii), analyzing the composition to be analyzed may comprise analyzing DNA contained in the composition to be analyzed through sequencing.

[0268] In a particular embodiment, in (iii), analyzing the composition to be analyzed may comprise analyzing the cleaved genomic DNA contained in the composition to be analyzed through sequencing.

[0269] In a particular embodiment, in (iii), the analysis of the analyte composition may comprise analyzing the DNA contained in the analyte composition through a PCR-based analytical method.

[0270] In a particular embodiment, in (iii), the analysis of the composition to be analyzed may comprise analyzing the cleaved genomic DNA contained in the composition to be analyzed via a PCR-based analytical method.

[0271] In one particular embodiment, the concentration of Cas protein included in the starting composition may be approximately 5000 nM.

[0272] In one particular embodiment, the concentration of the first cells in the starting composition is approximately 1×10 7 It may be in cells / mL.

[0273] In one particular embodiment, obtaining the composition to be analyzed may further comprise incubating the obtained composition through cell disruption.

[0274] In a particular embodiment, obtaining the composition to be analyzed may further comprise incubating the composition comprising the disrupted cellular component, the Cas protein, and the guide RNA.

[0275] In a particular embodiment, obtaining the composition to be analyzed may further comprise removing RNA from the obtained composition by cell disruption.

[0276] In a particular embodiment, obtaining the composition to be analyzed may further comprise removing the RNA component of the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA.

[0277] In a particular embodiment, obtaining the composition to be analyzed may further comprise purifying DNA from the composition obtained by cell disruption.

[0278] In a particular embodiment, obtaining the composition to be analyzed may further comprise purifying DNA from the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA.

[0279] In a specific embodiment, the method for predicting off-targets of the present application may be used in combination with one or more other off-target prediction methods, where the other off-target prediction methods may be any one or more selected from Cas-OFFinder, CHOPCHOP, CRISPOR, Digenome-seq, DIG-seq, SITE-Seq, CIRCLE-Seq, CHANGE-seq, GUIDE-seq, GUIDE-tag, DISCOVER-seq, BLISS, BLESS, integrase-deficient lentiviral vector-mediated DNA cleavage capture, HTGTS, ONE-seq, CReVIS-Seq, ITR-seq, and TAG-seq.

[0280] Below, we describe an exemplary embodiment of the off-target prediction system of the present application that features the use of an extruder.

[0281] In one embodiment, a method for predicting possible off-targets in a CRISPR / Cas gene editing system may be provided, the method comprising: (i) charging a starting composition comprising a first editing protein, a first guide RNA, and a first cell into a first vessel of an extruder; (ii) using an extruder (a) applying pressure to a first vessel to move components of an initial composition from a first vessel of an extruder to a second vessel of the extruder; wherein the components of the starting composition are transferred by the applied pressure from the first vessel through a filter with holes located between the first vessel of the extruder and the second vessel of the extruder to the second vessel, such that the mixture is contained in the second vessel; Here, the first cells, which are larger than the diameter of the filter pores, are destroyed by the applied pressure and at the same time pass through the filter pores, wherein, through physical disruption of the first cell, an environment is created in which the genomic DNA in the cell can come into contact with the Cas / gRNA complex; This brings the genomic DNA into contact with the Cas / gRNA complex, resulting in cleavage of the genomic DNA at one or more cleavage sites to obtain the composition to be analyzed; and (iii) analyzing the composition to be analyzed to obtain information about the cleavage site(s). Equipped with.

[0282] In one particular embodiment, the off-target prediction method comprises: (iv) confirming information about off-target candidates from the information about cleavage sites obtained from (iii) to predict off-targets that may occur in the CRISPR / Cas gene editing system; may further be provided.

[0283] In one particular embodiment, the extrusion process in (ii) comprises (a) applying pressure to a first vessel to move components of an initial composition from a first vessel of an extruder to a second vessel of the extruder; wherein the components of the starting composition are transferred by the applied pressure from the first vessel through a filter with holes located between the first vessel of the extruder and the second vessel of the extruder to the second vessel, so that the mixture is contained in the second vessel; (b) applying pressure to the second container to transfer components of the mixture contained in the second container from the second container to the first container. wherein the components of the mixture contained in the second container are transferred from the second container to the first container through a filter with holes located between the first container and the second container under applied pressure, and thus the mixture transferred from the second container through the filter to the first container under pressure is contained in the first container; and (c) repeating the processes of (a) and (b) a predetermined number of times; where the predetermined number of times is counted in increments of 0.5, with 0.5 indicating execution of a single process of (a) or (b); Here, the first cells, which are components larger than the pore size of the filter, are destroyed by the applied pressure and at the same time pass through the pores of the filter, Here, through physical disruption of the first cell, an environment is created in which the genomic DNA in the cell can come into contact with the Cas / gRNA complex; This brings the genomic DNA into contact with the Cas / gRNA complex, This results in cleavage of the genomic DNA at one or more cleavage sites. may be provided.

[0284] In one particular embodiment, the pressure applied to the first container may be generated through the process of pushing a piston designed to apply pressure to the first container in a direction toward the first container and the filter.

[0285] In one particular embodiment, the pressure applied to the first container may be generated through a process of pushing a piston designed to apply pressure to the first container in the direction of the first container and the filter, and the pressure applied to the second container may be generated through a process of pushing a piston designed to apply pressure to the second container in the direction of the second container and the filter.

[0286] In a particular embodiment, the information about the cleavage sites may include one or more of a position on the genomic DNA for the one or more cleavage sites, a cleavage score for the one or more cleavage sites, and a number of cleavage sites.

[0287] In a particular embodiment, the information about the off-target candidates may include one or more of a location on the genomic DNA for one or more off-target candidates, an off-target prediction score for one or more off-target candidates, and a number of predicted off-target candidates.

[0288] In a particular embodiment, the membrane structure including the cell membrane of the first cell can be disrupted by physically disrupting the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell.

[0289] In a particular embodiment, in (ii), the physical disruption of the first cell may lead to the disruption of membranous structures, including the nuclear membrane, of the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell.

[0290] In one particular embodiment, the filter may have pores with a diameter smaller than the first cell size.

[0291] In one particular embodiment, the average diameter of the pores of the filter may be between 5 and 15 μm.

[0292] In one particular embodiment, the average diameter of the pores of the filter may be 8 μm.

[0293] In one particular embodiment, the predetermined number may be between four and seven.

[0294] In one particular embodiment, the predetermined number may be 5.5.

[0295] In a particular embodiment, in (ii), the genomic DNA exposed through the physical disruption of the first cell may maintain the unique epigenetic characteristics of the first cell.

[0296] In a particular embodiment, the information about the cleavage site obtained in (iii) may be information reflecting an intrinsic epigenetic signature of the first cell.

[0297] In a particular embodiment, the information about the off-target candidates obtained in (iv) may be information reflecting an intrinsic epigenetic signature of the first cell.

[0298] In one particular embodiment, the cell's DNA repair mechanisms may be disrupted due to the destruction of the cell, and thus the cleaved DNA cannot be repaired.

[0299] In one particular embodiment, the off-target prediction method comprises: Identifying a CRISPR / Cas gene editing system that is subject to off-target predictions, wherein the predicted CRISPR / Cas gene editing system includes the use of a predicted Cas protein and the use of a predicted guide RNA. may further be provided.

[0300] In one particular embodiment, the off-target prediction method comprises: identifying a predicted off-target CRISPR / Cas gene editing system, wherein the predicted CRISPR / Cas gene editing system includes use of a predicted Cas protein and use of a predicted guide RNA, and wherein the identification of the predicted CRISPR / Cas gene editing system is performed prior to (i). may further be provided.

[0301] In a particular embodiment, the guide sequence of the first guide RNA may have the same sequence as the guide sequence of the guide RNA to be predicted.

[0302] In a particular embodiment, the predicted CRISPR / Cas gene editing system may include the use of a predicted cell, where the first cell and the predicted cell may be the same.

[0303] In a particular embodiment, the analysis of the analyte composition in (iii) may comprise analyzing the cleaved genomic DNA contained in the analyte composition through sequencing.

[0304] In a particular embodiment, the analysis of the analyte composition in (iii) may comprise analyzing the cleaved genomic DNA contained in the analyte composition through sequencing.

[0305] In a particular embodiment, the analysis of the analyte composition in (iii) may comprise analyzing DNA contained in the analyte composition through a PCR-based analytical method.

[0306] In a particular embodiment, the analysis of the analyte composition in (iii) may comprise analyzing the cleaved genomic DNA contained in the analyte composition via a PCR-based analytical method.

[0307] In one particular embodiment, the concentration of Cas protein included in the starting composition may be 5000 nM.

[0308] In one particular embodiment, the concentration of the first cells in the starting composition is 1×10 7 It may be in cells / mL.

[0309] In a particular embodiment, a process of incubation of the composition comprising the disrupted cellular components, the Cas protein and the guide RNA may be further carried out to obtain the composition to be analyzed.

[0310] In one particular embodiment, a process of incubation of the composition obtained through cell disruption may be further carried out to obtain the composition to be analyzed.

[0311] In a particular embodiment, a process of removal of the RNA component of the composition comprising the disrupted cellular components, the Cas protein and the guide RNA may be further carried out to obtain the composition to be analyzed.

[0312] In one particular embodiment, a process of removal of RNA from the composition obtained by cell disruption may be further carried out in order to obtain the composition to be analyzed.

[0313] In a particular embodiment, a process of DNA purification of the composition comprising the disrupted cellular components, the Cas protein and the guide RNA may be further carried out to obtain the composition to be analyzed.

[0314] In one particular embodiment, a process of DNA purification from the composition obtained by cell disruption may be further carried out in order to obtain the composition to be analyzed.

[0315] In a particular embodiment, the method of predicting off-targets of the present application may be used in combination with one or more different off-target prediction methods, where the different off-target prediction methods may be any one or more selected from Cas-OFFinder, CHOPCHOP, CRISPOR, Digenome-seq, DIG-seq, SITE-Seq, CIRCLE-Seq, CHANGE-seq, GUIDE-seq, GUIDE-tag, DISCOVER-seq, BLISS, BLESS, integrase-deficient lentiviral vector-mediated DNA cleavage capture, HTGTS, ONE-seq, CReVIS-Seq, ITR-seq, and TAG-seq.

[0316] Exemplary embodiment (2) of the off-target prediction system of the present application Hereinafter, exemplary embodiments (non-limiting embodiments) will be disclosed through explanation forms different from the above-mentioned "exemplary embodiment (1) of the off-target prediction system of the present application." Exemplary embodiments featuring mechanisms A01. (i) preparing a starting composition comprising a first Cas protein, a first guide RNA, and a first cell, wherein the Cas protein and the first guide RNA are capable of forming a Cas / gRNA complex; (ii) obtaining the composition to be analyzed through physical disruption of the first cell, where the physical disruption of the first cell creates an environment in which the genomic DNA can come into contact with the Cas / gRNA complex, whereby the genomic DNA and the Cas / gRNA complex come into contact, whereby the genomic DNA is cleaved at one or more cleavage sites; (iii) obtaining information about the cleavage sites by analyzing the composition of interest; and (iv) identifying information about off-target candidates from the information about the cleavage sites obtained from (iii) and predicting off-targets that will occur in the CRISPR / Cas gene editing system; A method for predicting off-targets that may occur in a genome editing process using a CRISPR / Cas gene editing system.

[0317] A02. Information on the cleavage site is given below. the location on the genomic DNA of one or more cleavage sites; a cleavage score for one or more cleavage sites, and Number of cleavage sites The method of A01, comprising one or more of:

[0318] A03. For information on potential off-target targets, the location on genomic DNA of one or more off-target candidates; an off-target prediction score for one or more off-target candidates; and Number of predicted off-target candidates Any one of the methods of A01 to A02, including one or more of:

[0319] A04 In (ii), the membrane structure including the cell membrane of the first cell is disrupted by physically disrupting the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell. One of the methods from A01 to A03.

[0320] A05. In (ii), the membrane structure including the nuclear membrane of the first cell is disrupted by physically disrupting the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell. Any one of methods A01 to A04.

[0321] A06. Any one of the methods of A01 to A05, wherein physically disrupting the first cells includes passing the first cells through a filter having pores having a size smaller than a size of the first cells, and the first cells are disrupted while passing through the pores having a size smaller than the first cell size.

[0322] A07. Any one of the methods of A01 to A06, wherein physically disrupting the first cells comprises passing a composition comprising the first cells or a composition comprising cellular components derived from the disrupted cells through a filter having pores that are half or smaller in size than the size of the first cells, and the first cells are disrupted while passing through the pores having a size smaller than the size of the first cells.

[0323] A08. The method of any one of A06 and A07, wherein the filter has pores with a diameter smaller than the size of the first cell.

[0324] A09. The method of any one of A06 to A08, wherein the average pore size of the filter is smaller than the first cell size.

[0325] A10. The method of any one of A06 to A09, wherein the average pore size of the filter is from 5 to 15 μm.

[0326] A11. The method of any one of A06 to A10, wherein the average pore size of the filter is approximately 8 μm.

[0327] A12. The method of any one of A01 to A05, wherein physically disrupting the first cells is accomplished by use of an extruder.

[0328] A13. The method of any one of A01 to A05 and A12, wherein physically disrupting the first cells is accomplished by use of an extruder, and a filter included in the extruder has pores with a diameter smaller than the first cell size.

[0329] A14. The method of any one of A01 to A05, A12 and A13, wherein physically disrupting the first cells is accomplished by use of an extruder, and the average pore size of a filter contained in the extruder is from 5 to 15 μm.

[0330] A15. Any one of the methods of A01 to A14, wherein in (ii), the genomic DNA exposed by physically disrupting the first cell maintains the unique epigenetic characteristics of the first cell.

[0331] A16. Any one of the methods of A01 to A15, wherein the information about the cleavage site obtained in (iii) is information reflecting an intrinsic epigenetic signature of the first cell.

[0332] A17. Any one of the methods of A01 to A16, wherein in (iv), the information about the off-target candidate is information reflecting a unique epigenetic signature of the first cell.

[0333] A18. The method of any one of A01 to A17, wherein the cell's DNA repair mechanisms are disrupted by cell disruption and thus the cleaved DNA is not repaired.

[0334] A19. Identifying a CRISPR / Cas gene editing system that is subject to off-target predictions, wherein the predicted CRISPR / Cas gene editing system includes the use of a predicted Cas protein and the use of a predicted guide RNA. The method of any one of A01 to A18, further comprising:

[0335] A20. validating the off-target predicted CRISPR / Cas gene editing system, wherein the predicted CRISPR / Cas gene editing system includes use of the predicted Cas protein and use of the predicted guide RNA, and the validation of the predicted CRISPR / Cas gene editing system is performed prior to (i); The method of any one of A01 to A18, further comprising:

[0336] A21. The method of any one of A19 to A20, wherein the guide sequence of the first guide RNA has the same sequence as the guide sequence of the guide RNA undergoing prediction.

[0337] A22. The method of any one of A19 to A21, wherein the CRISPR / Cas gene editing system for receiving prediction comprises use of a cell for receiving prediction, and wherein the first cell and the cell for receiving prediction are the same.

[0338] A23. Any one of the methods of A01 to A22, wherein in (iii) analyzing the analyte composition comprises analyzing DNA contained in the analyte composition through sequencing.

[0339] A24. Any one of the methods of A01 to A23, wherein in (iii) the step of analyzing the analyte composition comprises a step of analyzing the cleaved genomic DNA contained in the analyte composition via sequencing.

[0340] A25. The method of any one of A01 to A22, wherein in (iii) the step of analyzing the analyte composition comprises analyzing DNA contained in the analyte composition via a PCR-based analytical method.

[0341] A26. Any one of the methods A01 to A22 and A25, wherein in (iii) the step of analyzing the composition to be analyzed comprises a step of analyzing genomic DNA contained in the composition to be analyzed via a PCR-based analytical method.

[0342] A27. The method of any one of A01 to A26, wherein the concentration of the Cas protein included in the starting composition is 5000 nM.

[0343] A28. The concentration of the first cells in the starting composition is 1X10 7 Any one of the methods A01 to A27, wherein the cell mass per mL is cells / mL.

[0344] A29. The composition to be analyzed is obtained by Incubating a composition comprising the disrupted cellular components, the Cas protein, and the guide RNA. The method of any one of A01 to A28, further comprising:

[0345] A30. The composition to be analyzed is obtained by Removing the RNA component of the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA. The method of any one of A01 to A29, further comprising:

[0346] A31. The composition to be analyzed is obtained by Purifying DNA from a composition containing disrupted cellular components, Cas proteins, and guide RNAs. The method of any one of A01 to A30, further comprising:

[0347] Exemplary embodiments featuring the use of an extruder B01. (i) loading a starting composition comprising a first Cas protein, a first guide RNA, and a first cell into a first container; (ii) using an extruder (a) applying pressure to a first vessel to move components of an initial composition from a first vessel of an extruder to a second vessel of the extruder; wherein the components of the starting composition are transferred from the first vessel to the second vessel by the applied pressure through a filter with holes located between the first vessel of the extruder and the second vessel of the extruder, so that the mixture is contained in the second vessel; Here, the first cells, which are components larger than the diameter of the filter pores, are broken by the applied pressure and pass through the filter pores, Here, an environment is created in which the genomic DNA can be contacted with the Cas protein and the guide RNA through physical disruption of the first cell; This brings the genomic DNA into contact with the Cas / gRNA complex, This results in cleavage of the genomic DNA at one or more cleavage sites. to obtain the composition to be analyzed; and (iii) analyzing the composition to obtain information about the cleavage site; and (iv) (iii) identifying information about off-target candidates from the information about the cleavage site obtained in (iii) and predicting off-targets that will occur in the CRISPR / Cas gene editing system. A method for predicting off-targets that may occur in a genome editing process using a CRISPR / Cas gene editing system.

[0348] B02. The extrusion process in (ii) (a) applying pressure to a first vessel to move components of an initial composition from a first vessel of an extruder to a second vessel of the extruder; wherein the components of the starting composition are transferred from the first vessel to the second vessel by the applied pressure through a filter with holes located between the first vessel of the extruder and the second vessel of the extruder, thus resulting in the mixture being contained in the second vessel; (b) applying pressure to the second container to transfer components of the mixture contained in the second container from the second container to the first container. wherein the components of the mixture contained in the second container are transferred from the second container to the first container by the applied pressure through a filter with holes located between the first container and the second container, and thus the mixture transferred from the second container to the first container by passing through the filter by the pressure is received in the first container; and (c) repeating the processes of (a) and (b) a predetermined number of times; wherein the predetermined number of times is counted in increments of 0.5, where 0.5 indicates execution of a single process of (a) or (b); Here, first cells larger than the diameter of the filter pores pass through the filter pores while being broken by the applied pressure, Here, through physical disruption of the first cell, an environment is created in which the genomic DNA in the cell can come into contact with the Cas / gRNA complex; This brings the genomic DNA into contact with the Cas / gRNA complex, This results in cleavage of the genomic DNA at one or more cleavage sites. The method of B01, comprising the process of

[0349] B03. The method of any one of B01 to B02, wherein the pressure applied to the first container may be generated through a process of pushing a piston designed to apply pressure to the first container in the direction of the first container and the filter.

[0350] B04. The method of B02, wherein the pressure applied to the first container may be generated through a process of pushing a piston designed to apply pressure to the first container in the direction of the first container and the filter, and the pressure applied to the second container may be generated through a process of pushing a piston designed to apply pressure to the second container in the direction of the second container and the filter.

[0351] B05. Information on the cleavage site is given below. the location on the genomic DNA of one or more cleavage sites; a cleavage score for one or more cleavage sites, and Number of cleavage sites Any one of the methods of B01 to B04, including one or more of:

[0352] B06. For information on potential off-target targets, the location on genomic DNA of one or more off-target candidates; an off-target prediction score for one or more off-target candidates; and Number of predicted off-target candidates The method of any one of B01 to B05, comprising one or more of:

[0353] B07. Any one of the methods of B01 to B06, wherein in (ii), a membrane structure including a cell membrane is disrupted through physical disruption of the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell.

[0354] B08. Any one of the methods of B01 to B07, wherein in (ii), a membrane structure including a nuclear membrane of the first cell is disrupted through physical disruption of the first cell, thereby preparing an environment in which the Cas / gRNA complex can contact the genomic DNA of the first cell.

[0355] B09. The method of any one of B01 to B08, wherein the filter has pores with a diameter smaller than the first cell size.

[0356] B10. The method of any one of B01 to B09, wherein the average pore size of the filter is from 5 to 15 μm.

[0357] B11. The method of any one of B01 to B10, wherein the average pore size of the filter is 8 μm.

[0358] B12. Any one of the methods B02 to B10, wherein the predetermined number of times is 4 to 7.

[0359] B13. Any one of the methods B02 to B12, wherein the predetermined number of times is 5.5.

[0360] B14. Any one of the methods of B01 to B13, wherein in (ii), the genomic DNA exposed by the physical disruption of the first cell maintains the unique epigenetic characteristics of the first cell.

[0361] B15. Any one of the methods of B01 to B14, wherein the information about the cleavage site obtained in (iii) is information reflecting an intrinsic epigenetic signature of the first cell.

[0362] B16. Any one of the methods of B01 to B15, wherein the information about the off-target candidates obtained in (iv) is information reflecting a unique epigenetic signature of the first cell.

[0363] B17. The method of any one of B01 to B16, wherein due to the cell destruction, the cell's DNA repair mechanisms are destroyed, whereby the cleaved DNA is not repaired.

[0364] B18. The method is: Identifying a CRISPR / Cas gene editing system that is subject to off-target predictions, wherein the predicted CRISPR / Cas gene editing system includes the use of a predicted Cas protein and the use of a predicted guide RNA. The method of any one of B01 to B17, further comprising:

[0365] B19. The method further comprises: validating a predicted off-target CRISPR / Cas gene editing system, wherein the predicted CRISPR / Cas gene editing system comprises use of a predicted Cas protein and use of a predicted guide RNA, and wherein validating the predicted CRISPR / Cas gene editing system is performed prior to (i); The method of any one of B01 to B17, further comprising:

[0366] B20. The method of any one of B18 and B19, wherein the guide sequence of the first guide RNA has the same sequence as the guide sequence of the guide RNA undergoing prediction.

[0367] B21. The method of any one of B18 to B20, wherein the CRISPR / Cas gene editing system for receiving prediction comprises use of a cell for receiving prediction, and wherein the first cell and the cell for receiving prediction are the same.

[0368] B22. (iii) analyzing the composition to be analyzed includes analyzing the cleaved genomic DNA contained in the composition to be analyzed through sequencing; One of the methods from B01 to B21.

[0369] B23. (iii) analyzing the analyte composition includes analyzing the cleaved genomic DNA contained in the analyte composition through sequencing; One of the methods from B01 to B22.

[0370] B24. (iii) analyzing the analyte composition includes analyzing DNA contained in the analyte composition through a PCR-based analytical method; One of the methods from B01 to B21.

[0371] B25. (iii) Analyzing the composition to be analyzed in (iii) comprises analyzing the cleaved genomic DNA contained in the composition to be analyzed through a PCR-based analytical method. One of the methods from B01 to B21.

[0372] B26. The method of any one of B01 to B25, wherein the concentration of the Cas protein included in the starting composition is 5000 nM.

[0373] B27. The concentration of the first cells in the starting composition is 1X10 7 Any one of the methods B01 to B26, wherein the nucleus is cells / mL.

[0374] B28. The method includes the steps of: Incubating the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA. The method of any one of B01 to B27, further comprising the process of:

[0375] B29. The method includes the steps of: Removing the RNA component of the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA. The method of any one of B01 to B28, further comprising the process of:

[0376] B30. The method includes the steps of: Purifying the DNA of the composition comprising the disrupted cellular components, the Cas protein, and the guide RNA. The method of any one of B01 to B29, further comprising the process of:

[0377] Anticipated Uses of the Off-Target Prediction System of the Present Application Hereinafter, the expected use example of the off-target prediction system of the present application (the assumed situation where a person skilled in the art uses the off-target prediction system of the present application) will be described without being limited thereto. The off-target prediction system of the present application (for example, Extru-seq) is an off-target prediction system characterized by physical destruction of cells, and is a more efficient and more accurate off-target prediction system that has the advantages of conventional in vitro-based off-target prediction systems and the advantages of in vivo-based off-target prediction systems. Therefore, using the above-mentioned characteristics of the off-target prediction system, all of the methods of confirming off-target candidates or predicting off-targets that are performed to realize the purpose of confirming off-targets that may possibly occur in the gene editing process are included as one embodiment of the use or application of the off-target prediction method of the present application, but the following examples do not limit the scope of the present application.

[0378] For example, the off-target prediction method (or system) of the present application may be used by engineers or researchers who use CRISPR / Cas gene editing systems to edit cell genomes.

[0379] For example, a researcher selects a gene editing system to be used in cellular genome editing. For example, a researcher selects a CRISPR / Cas gene editing system as a gene editing system to be used in cellular genome editing. In addition, a researcher may select a cell that is the main purpose of genome editing. In the process of selecting a gene editing system to be used for cellular genome editing, an in silico-based off-target prediction method may be used to design a suitable guide sequence. Here, the researcher will develop a treatment involving the use of a gene editing system. In the development of the treatment, information about the off-target of the selected gene editing system (particularly, the guide RNA) must be confirmed. Based on the selected gene editing system, the details of the off-target prediction method of the present application are designed to be fit for purpose. By carrying out the off-target prediction method of the present application, information about off-target candidates that may possibly occur in the use of the selected gene editing system was confirmed. Then, using the confirmed information about the off-target candidates, information about the off-targets that are problematic in the use of the selected gene editing system was confirmed. In particular, by verifying the candidate off-target sites identified by the off-target prediction system of the present application, the true off-target was finally confirmed. In this process, known off-target prediction methods (in silico, in vitro, and cell-based off-target prediction methods) can be used in combination to discover true off-target sites.

[0380] In another example, the off-target prediction system of the present application may be used in the selection process of a gene editing system (particularly, a guide sequence of a guide RNA). A researcher generates a guide RNA library containing various types of guide RNAs. The off-target prediction method is performed on a gene editing system containing one or more guide RNAs contained in the guide RNA library. Then, based on the results of the off-target prediction method of the present application, a gene editing system to be used in therapeutic development or research is selected. In this process, known off-target prediction methods (in silico, in vitro, and cell-based off-target prediction methods) can be used in combination to find true off-target sites.

[0381] As described above, the off-target prediction system of the present application can be used in various situations, and the manner of use of the off-target prediction system is not limited to the above examples. [Example]

[0382] The present invention provided by the present application will be described in more detail below with reference to experimental examples or examples. These experimental examples are provided merely to illustrate the contents disclosed by the present application, and it is clear to those skilled in the art that the scope of the present invention disclosed in the present specification should not be construed as being limited to the following experimental examples.

[0383] Experimental Example Experimental Method Experimental method 1. Design of promiscuous sgRNAs Candidate target sequences containing protospacer adjacent motifs (NGG PAM) located in mouse genome (mm10) PCSK9 and albumin genes were extracted by Cas-Designer (see references [Park, Jeongbin, Sangsu Bae, and Jin-Soo Kim. "Cas-Designer: a web-based tool for choice of CRISPR-Cas9 target sites." Bioinformatics 31.24(2015):4014-4016.]). The extracted sequences were aligned to the human genome (hg19). Among the extracted sequences, sequences with 1 or more targets with 0 mismatches when the extracted sequences were aligned to the human genome were selected. The selected candidates were analyzed by Cas-OFFinder (see reference [Bae, Sangsu, Jeongbin Park, and Jin-Soo Kim. "Cas-OFFinder: a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases." Bioinformatics 30.10(2014):1473-1475.]). Candidates with various sets of related sequences containing various numbers of mismatches (0 to 5 mismatches per site) widely distributed throughout the human and mouse genomes were selected as targets. Information on the target and guide sequences of the promiscuous sgRNAs used in subsequent experiments is as follows:

[0384] Target sequence of single-stranded guide RNA (sgRNA) targeting mouse PCSK9 (excluding NGG PAM): AGGTGGGAAACTGAGGCTT (SEQ ID NO: 25)

[0385] Target sequence for sgRNA targeting mouse albumin (excluding NGG PAM): ACATGCATATGTATGTGTG (SEQ ID NO: 26)

[0386] As explained below in the experimental results, these sgRNAs perfectly matched target sequences present on the human genome (however, the PCSK9 and albumin loci on the human genome are not targets). Although the target and guide sequences targeted loci other than PCSK9 or albumin on the human genome, for convenience they were indicated as human PCSK9 targeting sgRNA (sgRNA targeting human PCSK9) and human albumin targeting sgRNA (sgRNA targeting human albumin).

[0387] That is, the target sequence of the promiscuous sgRNA, referred to as the human PCSK9 targeting sgRNA, is the same as the target sequence of the mouse PCSK9 targeting sgRNA (a sgRNA that targets mouse PCSK9). The target sequence of the human PCSK9 targeting sgRNA (excluding the NGG PAM) is as follows: AGGTGGGGAAACTGAGGCTT (SEQ ID NO: 25).

[0388] The target sequence of the promiscuous sgRNA, designated human albumin targeting sgRNA, is the same as the target sequence of mouse albumin targeting sgRNA (sgRNA targeting mouse albumin). The target sequence of the human albumin targeting sgRNA (excluding the NGG PAM) is as follows: ACATGCATATGTATGTGTG (SEQ ID NO: 26).

[0389] Experimental method 2. Construction of plasmids for expression of sgRNA and Cas9 The Streptococcus pyogenes Cas9 sequence (see reference [Cho, Seung Woo, et al. "Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease." Nature biotechnology 31.3(2013):230-232.]) and the designed promiscuous sgRNA sequences (albumin-targeting sgRNA and PCSK9-targeting sgRNA sequences) were cloned into the backbone of the AAV plasmid used in the previous study (see reference [Kim, Eunji, et al. "In vivo genome editing with a small Cas9 orthologue derived from Campylobacter jejuni." Nature communications 8.1(2017):1-12.]) to generate Cas9 (pAAV-Cas9) and sgRNA (pAAV-albumin and pAAV-PCSK9) expression vectors. Cas9 expression was carried out under the control of the CMV promoter, and sgRNA expression was carried out under the control of the U6 promoter. Guide sequences targeting the FANCF, VEGFA, and HBB genes were cloned in the pRG2 vector (Addgene #104174).

[0390] Experimental method 3. GUIDE-seq Human HEK293T cells (ATCC, Cat# CRL-3216) and mouse NIH-3T3 cells (ATCC, Cat# CRL-1658) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin under conditions of 5% CO2 and 37 °C. HEK293T and NIH3T3 cells were subcultured every 72 hours to maintain 80% confluency. For GUIDE-seq, 2x10 5HEK293T cells were transfected with sgRNA expression plasmid (500 ng, pAAV-Albumin or pAAV-PCSK9), Cas9 expression plasmid (500 ng, p3s-Cas9HC; Addgene plasmid #43945), and 5 pmol dsODN using lipofectamine 2000. 5 NIH-3T3 cells were transfected with sgRNA expression plasmid (250 ng, pAAV-Albumin or pAAV-PCSK), Cas9 expression plasmid (500 ng, p3s-Cas9HC; Addgene plasmid #43945), and 100 pmol dsODN using the Amaxa P3 electroporation kit (V4XP-3032; program EN-158). Transfected cells were transferred to 24-well plates containing pre-cultured DMEM (1 mL / well) at 37°C. After 72 hours, genomic DNA was isolated using a QIAamp DNA mini kit (Qiagen).

[0391] Human HeLa cells (ATCC, Cat#CCL-2) were maintained in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin at 5% CO2 and 37°C. HeLa cells were subcultured every 72 hours to maintain 80% confluency. For GUIDE-seq, 2x10 5 HeLa cells were transfected with sgRNA expression plasmids (500 ng, pRG2-FANCF, pRG2-VEGFA, or pRG2-HBB), Cas9 expression plasmids (500 ng, p3s-Cas9HC; Addgene plasmid #43945), and 25 pmol dsODN using an Amaxa 4D-nucleofector (V4XC-1024; program CN-114). Transfected cells were transferred to 24-well plates (1 mL / well) containing DMEM pre-incubated at 37°C. After 72 hours, genomic DNA was isolated using a QIAamp DNA Mini Kit (Qiagen).

[0392] The purified DNA was fragmented at 1000 nm using a Covaris system (duty factor: 10%, PIP: 50, cycles per burst: 200, time: 50 sec, temperature: 20° C.) and purified using Ampure XP beads (A63881). Sequencing libraries were generated from the DNA using the NEBNext® Ultra™ II DNA Library Prep Kit for Illumina (E7546L) according to the manufacturer's protocol. Regions of the library containing dsODN sequences were then amplified using primers specific for the dsODN and sequenced using MiSeq (Illumina, TruSeq HT kit). Other procedures were the same as those described in previous studies (see references [Tsai, Shengdar Q., et al. "GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases." Nature biotechnology 33.2 (2015): 187-197.]). For data analysis, GUIDE-seq (1.0.2; https: / / pypi.org / project / guide-seq / ), compatible with Python 3, was used.

[0393] Experimental method 4. Construction of plasmids for sgRNA transcription and in vitro transcription reactions To improve the yield and accuracy of sgRNA transcription, the inventors of this application modified a previously described method (see Kim, Daesik, Beum-Chang Kang, and Jin-Soo Kim. "Identifying genome-wide off-target sites of CRISPR RNA-guided nucleases and deaminases with Digenome-seq." Nature Protocols 16.2 (2021): 1170-1192.). Briefly, the sgRNA template was generated by annealing two complementary oligonucleotides followed by PCR amplification. Through a second PCR, BamHI, BsaI, and KpnI restriction sites were attached to the ends of the sgRNA template. The tailed sgRNA template was inserted into a pUC19 plasmid digested with BamHI and Kpnl. The sgRNA-encoding plasmid was linearized by Bsal, resulting in the generation of appropriate sgRNA end sequences. The linearized plasmid was incubated with 7.5 U / μl T7 RNA polymerase (NEB, M0251L) for 8 hours at 37° C. in a reaction buffer (NEB, B9012S) containing 14 mM MgCl2 (NEB, B0510A), 10 mM DTT (Sigma, 43816), 0.02 U / μl yeast inorganic pyrophosphatase (NEB, M2403L), 1 U / μl mouse RNase inhibitor (NEB, M0314L), 4 mM ATP (NEB, N0451AA), 4 mM GTP (NEB, N0452AA), 4 mM UTP (NEB, N0453AA), and 4 mM CTP (NEB, N0454AA). Yeast inorganic phosphatase was included to enhance sgRNA synthesis. After the reaction, the mixture was mixed and incubated with DNase I to remove the DNA template; then, the transcribed sgRNA was purified using a PCR purification kit (Favorgen, #FAGCK001-1).

[0394] Experimental method 5. Digenome-seq Genomic DNA was purified from HEK293T cells (ATCC, Cat#CRL-3216) and NIH-3T3 cells (ATCC, Cat#CRL-1658) using DNeasy Blood & Tissue Kit (Qiagen). Two types of genomic DNA (10 μg) were incubated (8 h, 37° C.) in 1 mL reaction solution containing NEB3 buffer [100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, 100 μg / mL bovine serum albumin (BSA), pH 7.9] together with Cas9 protein (10 μg) and albumin or PCSK9 targeting sgRNA (10 μg each). To degrade sgRNA, the digested genomic DNA was treated with RNase A (50 μg / mL, Qiagen) for 10 min and purified using DNeasy Blood & Tissue Kit (Qiagen).

[0395] Genomic DNA (1 μg) was fragmented to 300 bp range using Covaris system (Life Technologies) and blunt-ended using End Repair Mix (Thermo Fischer). Fragmented DNA was ligated to adapters to generate libraries, which were then applied to WGS using HiSeq X Ten Sequencer (Illumina) in Macrogen. WGS was performed at a sequencing depth of 30 to 40x. DNA cleavage sites were identified using the Digenome1.0 program (see references [Park, Jeongbin, et al. "Digenome-seq web tool for profiling CRISPR specificity." Nature methods 14.6 (2017): 548-549.]).

[0396] Experimental Method 6. In silico prediction of off-target sites Using Cas-OFFinder, we obtained hg19 genome-wide candidate off-target sites with fewer than 7 mismatches with the selected sgRNA. Based on a previous paper (Liu, Qiaoyue, et al. "Deep learning improves the ability of sgRNA off-target propensity prediction." BMC bioinformatics 21.1 (2020): 1-15.), we calculated the CROP score (a heuristic score indicating whether a candidate off-target site will be edited) using the CROP prediction model and optimization parameters (https: / / github.com / vaprilyanto / crop). The "crisprScore" R package was used to calculate the CFD score (a percent activity value provided in a matrix of penalties based on each possible type of mismatch at each position in the guide RNA sequence) (see reference [Doench, John G., et al. "Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9." Nature biotechnology 34.2 (2016): 184-191.]). For two calculations, GX19 (GACATGCATATGTATGTGTG (SEQ ID NO: 27) for albumin and GAGGTGGGAAACTGAGGCTT (SEQ ID NO: 28) for PCSK9 sgRNA sequence and X20 target sequence were used.

[0397] Experimental method 7. Extru-seq To prepare Extru-seq, the transcribed sgRNA was refolded in 1X NEBuffer 3.1 reaction buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, 100 μg / mL BSA, pH 7.9). After heating the sgRNA for 2 min at 98°C, the temperature was reduced at a rate of 0.1°C / sec until it reached 20°C. To reduce reaction inhibition by high concentration of glycerol, the Cas9 buffer (10 mM Tris-HCl, 0.15 M NaCl, 50% glycerol, pH 7.4) was exchanged with elution buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, pH 8.0). Buffer exchange was performed using a 10K Amicon® Ultra-15 centrifugal filter (Millipore).

[0398] Each type of HEK293T (ATCC, Cat# CRL-3216), NIH-3T3 (ATCC, Cat# CRL-1658), and HeLa (ATCC, Cat# CCL-2) cells were harvested with 0.25% trypsin-EDTA, and human bone marrow mesenchymal stem cells (BM-MSCs; Lonza, Cat# PT-2501) were harvested with 0.05% trypsin-EDTA. Harvested cells were resuspended in Dulbecco's phosphate-buffered saline (PBS). Buffer-exchanged Cas9 (800 mg) and refolded sgRNA (530 μg) were pre-incubated at room temperature for 10 min to form RNP complexes (for multiplex Extru-seq, buffer-exchanged Cas9 (800 mg) and 5 different refolded sgRNAs (106 μg each) were used). 1x10 7100 nM of cells were mixed with 5000 nM RNP complex in 1 mL 1X NEBuffer 3.1 reaction buffer (100 mM NaCl, 50 mM Tris-HCl, 10 mM MgCl2, 100 μg / mL BSA, pH 7.9). To perform Extru-seq in the presence of SCR7, SCR7 pyrazine (Sigma, SML1546) (1 μM) was added. After gentle pipetting, the suspended cells were extruded 11 times through 8 μm pore size polycarbonate membrane filters (Whatman) using a mini extruder (Avanti Polar Lipids). The extruded samples were incubated at 37 °C for 16 h. To remove sgRNA and RNA, RNase A (2 mg / mL) was added, and then genomic DNA was purified from the extruded samples using the FavorPrep Blood Genomic DNA Extraction mini kit (Favorgen, #FAGCK001-2). Whole genome sequencing (WGS) was performed at a sequencing depth of 30 to 40x. WGS was performed at Macrogen using Nova-seq equipment according to the manufacturer's standard protocol. DNA cleavage sites were identified using the Digenome-seq standalone program (http: / / www.rgenome.net / digenome-js / standalone). Analysis filtering options were minimum depth 10, minimum score 0.05, and minimum ratio 0.01; other options were default. In addition, the inventors of this application, as developers of new tools, checked all sites identified by Extru-seq using Integrative Genomic Viewer (IGV). Some loci appeared to be false positive candidates (i.e., non-cleavage sites by IGV (see Figure 117 to Figure 125)). False positives (non-cleavage sites) identified using IGV were treated as negatives and excluded from the analysis. Figures 117 to 125 show false positive off-target sites that were manually excluded from Digenome-seq and Extru-seq WGS data and visualized using IGV.These false positives were observed in Digenome-seq. The associated bam file is available in the NCBI Bioproject (https: / / www.ncbi.nlm.nih.gov / bioproject / ) under the accession number PRJNA796642.

[0399] Referring to Figures 117 through 119, the sequence ACtTGtgTgTGTgTGTGgGGGG (SEQ ID NO:49) is disclosed. In these figures, mismatches are indicated by lower case letters and bulges (if present) are indicated by dashes. In these figures, the PAM sequence is underlined.

[0400] Referring to Figures 120 to 122, there is disclosed the sequence AtATATATATaTATaTaTaTGGAG (SEQ ID NO:50) (bulge-related markers omitted). In these figures, mismatches are indicated by lower case letters and bulges (if present) are indicated by dashes. In these figures, the PAM sequence is underlined.

[0401] Referring to Figures 123 to 125, there is disclosed the sequence TAgATATATATGaATGgGTaGAG (SEQ ID NO:51) (bulge-related markers omitted). In these figures, mismatches are indicated by lower case letters and bulges (if present) are indicated by dashes. In these figures, the PAM sequence is underlined.

[0402] Experimental Method 8. Assignment of Digenome-seq and Extru-seq non-target results to Cas-OFFinder results Unlike GUIDE-seq and Cas-OFFinder, the standalone Digenome-seq program does not have an sgRNA:off-target alignment function that provides information about the number of mismatches and bulge type (DNA or RNA) between the guide and off-target sites. The web version of the Digenome-seq analysis tool (http: / / www.rgenome.net / digenome-js / #!) has an optional alignment function with an alignment score that does not provide any information about the number of mismatches or the type of bulge. Instead, the inventors of the present application used Cas-OFFinder to identify off-target sites with up to 7 mismatches and 2 bulges relative to the target sequence. The positions of the off-target candidates identified by Digenome-seq and Extru-seq were compared to those identified by Cas-OFFinder. Information about mismatches and bulge types obtained from Cas-OFFinder could be assigned to the loci identified by Digenome-seq and Extru-seq.

[0403] Experimental Method 9. Validation of candidate off-target sites using human cell lines Human HEK293T and HeLa cells were maintained in DMEM supplemented with 10% FBS (ATCC, CRL-3216) and 1% penicillin-streptomycin at 37°C in the presence of 5% CO. To determine the indel frequency at candidate off-target sites, 2x10 5 HEK293T cells and 8x10 4Each of the HeLa cells was transfected with sgRNA expression plasmids (500 ng, pAAV-Albumin, pAAV-PCSK9, pRG2-HBB, pRG2-FANCF, or pRG2-VEGFA) and Cas9 expression plasmids (500 ng, pAAV-Cas9 or p3s-Cas9HC; Addgene plasmid #43945) using lipofectamine 2000 (vendor, amount). After incubating the cells for 3 days at 37 °C, genomic DNA was prepared using the FavorPrep Blood Genomic DNA Extraction Mini Kit (Favorgen, #FAGCK001-2). Deep sequencing was then used to analyze the target sites and potential off-target sites. Deep sequencing libraries were generated by PCR. Each sample was labeled using TruSeq HT dual index primers. Paired-end sequencing was performed on the pooled libraries using MiSeq (Illumina). In particular, multiple targets were combined by PCR performed using different indexed primers and then subjected to deep sequencing analysis.

[0404] Deep sequencing data is available under accession number PRJNA796642 at NCBI Bioproject (https: / / www.ncbi.nlm.nih.gov / bioproject / ). To determine whether a target is validated or false, the inventors of the present application used the following criteria used in EDITAS Medicine (see reference [Maeder, Morgan L., et al. "Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10." Nature medicine 25.2 (2019): 229-233.]). First, to be validated, the indels of the sample must be higher than 0.1%. Second, the treated / control ratio must be higher than 2. The validation results for off-target candidates through deep sequencing are detailed in Table 1. The mismatches of the validated off-target candidates with the target are shown in lowercase. For human PCSK9 in Table 1, SEQ ID NOs: 74 to 132 were assigned to off-target sequences 1 to 59 disclosed in human PCSK9 in the order of their disclosure. For human albumin in Table 1, SEQ ID NOs: 133 to 174 were assigned to off-target sequences 1 to 42 disclosed in human albumin in the order of their disclosure. For mouse PCSK9 in Table 1, SEQ ID NOs: 175 to 211 were assigned to off-target sequences 1 to 37 disclosed in mouse PCSK9 in the order of their disclosure. For mouse albumin in Table 1, SEQ ID NOs: 212 to 249 were assigned to off-target sequences 1 to 38 disclosed in mouse albumin in the order of their disclosure. For human HBB in Table 1, SEQ ID NOs: 250 to 293 were assigned to off-target sequences 1 to 44 disclosed in human HBB in the order of their disclosure.For human VEGFA in Table 1, SEQ ID NOs:294 to 343 were assigned to off-target sequences 1 to 50 disclosed in human VEGFA in the order of their disclosure. For human FANCF in Table 1, SEQ ID NOs:344 to 383 were assigned to off-target sequences 1 to 40 disclosed in human FANCF in the order of their disclosure. Table 1 is shown below.

[0405] Table 1. Targeted deep sequencing results for off-target validation. [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4] [Table 1-5] [Table 1-6] [Table 1-7] [Table 1-8] [Table 1-9] [Table 1-10] [Table 1-11] [Table 1-12] [Table 1-13] [Table 1-14] [Table 1-15] [Table 1-16] [Table 1-17] [Table 1-18] [Table 1-19] [Table 1-20] [Table 1-21]

[0406] Experimental method 10. AAV generation AAV8 carrying the desired clone sequences (pAAV-PCSK9, pAAV-albumin, and pAAV-Cas9) were produced on a large scale by VigeneBioscience (10 13 The generated AAV was distributed and stored at -70°C until use.

[0407] Experimental Methods 11. Animal Studies All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at Yonsei University College of Medicine (IACUC number 2019-0215). C57BL / 6 mice were maintained on a 12:12-h light:dark cycle.

[0408] Experimental Method 12. AAV Injection Each of the two types of AAV8 carrying pAAV-Cas9 and one of the two pAAV-sgRNAs (pAAV-PCSK9 or pAAV-Albumin) was delivered into C57BL / 6 mice through systemic (intravenous) and subretinal injections. Both injections were performed at a 1:1 GC (pAAV-Cas9:pAAV-sgRNA) ratio. The dose for intravenous injection was 2.5x10 11 GC / animal, and the dose for subretinal injection was 1.5x10 10 It was GC / eye.

[0409] For systemic injections, 200 μL of AVV8 diluted in PBS was injected into 7- to 9-week-old male mice via tail vein injection. The dose was 2.5×10 11 It was GC AAV8.

[0410] For subretinal injections, male mice aged 7 to 9 weeks were selected. Under general anesthesia, one pupil per mouse was dilated with eye drops containing tropicamide and phenylfurine. During the experiment, the body temperature of the mice was maintained at 37 °C using a heating pad. A small incision was made 1 mm away from the corneal limbus with a 1 / 2 30 G needle. A Hamilton syringe with a 33 G blunt needle, filled with 2 μL solution containing the AAV8 mixture, was inserted through the incision to the point of resistance (subretinal area). To avoid unnecessary tissue damage, the volume was injected carefully and gently, waiting 20 to 30 seconds to allow even diffusion, and then the syringe was slowly removed. Subsequently, antibiotic ointment was applied to the surface of the eye. For different injection methods and different sgRNAs, four mice were used.

[0411] Experimental Method 13. DNA preparation from harvested organs and tissues Organs and tissues were removed 3 months and 2 weeks after injection. At the end of the experiment, animals were euthanized by cardiac puncture under isoflurane anesthesia. Organs including eyes, liver, spleen, lungs, kidneys, muscles, brain, and testes were excised, snap frozen in liquid nitrogen, and stored at -70°C until further analysis.

[0412] For subretinal injection, neural retina and retinal pigment epithelium (RPE) were isolated and prepared. The cornea, iris, lens, and vitreous were removed from the enucleated eyes. The remaining ocular tissue was incubated in hyaluronidase solution for 45 min (37°C, 5% CO2). It was then incubated in cold PBS for 30 min to inactivate hyaluronidase activity. The ocular tissue was then transferred to fresh PBS and the neural retina was gently separated from the retina / RPE / choroid / sclera complex. The remaining retina / RPE / choroid / sclera complex was incubated in trypsin solution at 37°C in 5% CO2 for 45 min and gently shaken until the RPE sheet was completely detached. All isolated RPE sheets and RPE cells were collected. Genomic DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen, Cat. No. 69506) according to the manufacturer's instructions.

[0413] Experimental Method 14. Targeted Deep Sequencing Genomic DNA of mouse retinal pigment epithelial (RPE) cells was amplified using the REPLI-g single cell kit (Qiagen) according to the manufacturer's protocol.

[0414] Target sites and potential off-target sites were analyzed through targeted deep sequencing. Deep sequencing libraries were generated by PCR. TruSeq HT dual index primers were used to label each sample. Paired-end sequencing was performed on the pooled libraries using MiSeq (Illumina). In particular, multiple targets were combined by PCR performed with different indexed primers, and then targeted deep sequencing analysis was performed.

[0415] Experimental Method 15. Statistical Analysis Scores / sequence read counts were min-max normalized. In each group, the maximum value was normalized to 1 and the minimum value was normalized to 0. To test whether the median scores of two different groups were the same or not, a Wilcoxon rank sum test was performed on the samples falling into each intersection of the Venn diagram. The results of a two-tailed unpaired Mann-Whitney test, calculated by Prism (version 9.4.1), are shown.

[0416] Nucleic acid sequences of sgRNA and amino acid sequences of SpCas9 used in the experiments The following shows the sequences of sgRNAs and their related sequences used herein, and the SpCas9 sequence.As mentioned above, the human PCSK9 targeting sgRNA actually targets a different locus on the human genome other than PCSK9, but for convenience it is referred to as the human PCSK9 targeting sgRNA.The human albumin targeting sgRNA actually targets a different locus on the human genome other than albumin, but for convenience it is referred to as the human albumin targeting sgRNA.

[0417] Complete sequence of mouse PCSK9 targeting sgRNA GAGGUGGGAAACUGAGGCUUGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 29)

[0418] Guide sequence for mouse PCSK9 targeting sgRNA GAGGUGGGAAACUGAGGCUU (SEQ ID NO: 30)

[0419] Target sequence of mouse PCSK9 targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) AGGTGGGAAACTGAGGCTT (SEQ ID NO: 25)

[0420] Complete sequence of mouse albumin-targeting sgRNA GACAUGCAUGUAUGUGUGGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 31)

[0421] Guide sequence for mouse albumin targeting sgRNA GACAUGCAUGUAUGUGUG (SEQ ID NO: 32)

[0422] Target sequence of mouse albumin targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) ACATGCATATGTATGTGTG (SEQ ID NO: 26)

[0423] Complete sequence of human PCSK9 targeting sgRNA GAGGUGGGAAACUGAGGCUUGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 29)

[0424] Guide sequence for human PCSK9 targeting sgRNA GAGGUGGGAAACUGAGGCUU (SEQ ID NO: 30)

[0425] Target sequence of human PCSK9 targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) AGGTGGGAAACTGAGGCTT (SEQ ID NO: 25)

[0426] Complete sequence of human albumin targeting sgRNA GACAUGCAUGUAUGUGUGGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 31)

[0427] Guide sequence for human albumin targeting sgRNA GACAUGCAUGUAUGUGUG (SEQ ID NO: 32)

[0428] Target sequence of human albumin targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) ACATGCATATGTATGTGTG (SEQ ID NO: 26)

[0429] Full sequence of FANCF targeting sgRNA GGAAUCCCUUCUGCAGCACCGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 33)

[0430] Guide sequence of FANCF targeting sgRNA GGAAUCCCUUCUGCAGCACC (SEQ ID NO: 34)

[0431] Target sequence of FANCF targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) GAATCCCTTCTGCAGCACC (SEQ ID NO: 35)

[0432] Full sequence of VEGFA-targeting sgRNA GGGUGGGGGGAGUUUGCUCCGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 36)

[0433] Guide sequence of VEGFA-targeting sgRNA GGGUGGGGGGAGUUUGCUCC (SEQ ID NO: 37)

[0434] Target sequence of VEGFA-targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) GGTGGGGGGAGTTTGCTCC (SEQ ID NO: 38)

[0435] Full sequence of HBB-targeting sgRNA GUUGCCCCACAGGGCAGUAAGUUUUAGAGCUAGAAAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGAAAAAGUGGCACCGAGUCGGUGCUUUUUU (SEQ ID NO: 39)

[0436] Guide sequence of HBB-targeting sgRNA GUUGCCCCACAGGGCAGUAA (SEQ ID NO: 40)

[0437] Target sequence of HBB-targeting sgRNA (target sequence on the spatially unbound strand, excluding PAM) TTGCCCCACAGGGCAGTAA (SEQ ID NO: 41)

[0438] Amino acid sequence of SpCas9 IAKSEQEIGKATAKYFFYSNIMNFFKTEITLANGEIRKRPLIETNGETGEIVWDKGRDFATVRKVLSMPQVNIVKKTEVQTGGFSKESILPKRNSDKLIARKKDWDPKKYGGFDSPTVAYSVLVVAKVEKGKSKKLKSVKELLGITIMERSSFEKNPIDFLEAKGYKEVKKDLIIKLPKYSLFELENGRKRMLASAGELQKGNELALPSKYVNFLYLASHYEKLKGSPEDNEQKQLFVEQHKHYLDEIIEQISEFSKRVILADANLDKVLSAYNKHRDKPIREQAENIIHLFTLTNLGAPAAFKYFDTTIDRKRYTSTKEVLDATLIHQSITGLYETRIDLSQLGGD (SEQ ID NO: 42)

[0439] DNA sequence encoding SpCas9 TGAGCCAGCTGGGCGGCGAC (SEQ ID NO: 43)

[0440] result Result 1 Selection of off-target prediction methods to be used for comparison Genome-wide off-target prediction methods can be categorized into three groups according to their approaches: cell-based, in vitro, and in silico methods. Examples of approaches for the three groups are shown in Figure 1.

[0441] Different combinations of off-target prediction methods have been used in IND studies for genome editing therapeutics. Information on the off-target prediction methods used in IND studies for genome editing therapeutics is shown in Table 2.

[0442] Table 2. Off-target prediction methods used in gene editing drug and IND studies (TALEN: transcription activator-like effector; NA: not available; LCA10: Leber congenital amaurosis 10; MPS: Mucopolysaccharidosis. ) [Table 2]

[0443] To compare the performance of off-target prediction methods, the inventors of this application selected one method for each of the above-mentioned categories. For cell-based off-target prediction methods, GUIDE-seq was selected. For in silico off-target prediction methods, CAS-OFFinder was selected. GUIDE-seq and CAS-OFFinder were most frequently used to predict the off-targets of Cas9 therapeutics, including EDIT101 and NTLA-2001. For in vitro off-target prediction methods, Digenome-seq was selected. This is one of the most popular protocols in previous studies for comparison, since Digenome-seq was used in the EDIT101 study.

[0444] Result 2 Overview of Extru-seq and optimization of Extru-seq conditions The inventors of this application aimed to design a new method that combines the positive features of cell-based and in vitro methods. For this purpose, an off-target prediction method, Extru-seq, was developed, which features the use of physical force to lyse cells and mixing genomic DNA with Cas9 and sgRNA. A schematic diagram of this novel off-target prediction method, Extru-seq, is shown in Figure 2.

[0445] For example, live HEK294T or NIH-3T3 cells are mixed with the pre-incubated Cas9-sgRNA RNP complex. Using an extruder (see Goh, Wei Jiang, et al., "Bioinspired cell-derived nanovesicles versus exosomes as drug delivery systems: a cost-effective alternative." Scientific reports 7.1 (2017): 1-10.), the mixture is passed through a filter (e.g., filter paper) with a pore size smaller than the cell diameter. As it passes through a filter with a pore size smaller than the cell diameter, the cell (e.g., cell membrane) is disrupted, allowing Cas9 RNP to access genomic DNA. Figures 20 and 21 show the results of experiments carried out to determine the optimization conditions for the average filter pore size, the Cas9 RNP concentration of the mixture, and the number of cells in Extru-seq.

[0446] In particular, FIG. 20 shows the quality of genomic DNA incubated overnight with Cas9 RNP, analyzed through gel electrophoresis. Different numbers of NIH-3T3 cells and different pore sizes were tested. Here, "Con" represents control genomic DNA of sufficient quality for WGS analysis. "L" represents ladder DNA. FIG. 21 shows information for each of samples 1 to 9. For example, the condition of sample 8 (1X10 7 The electrophoretic results for the genomic DNA of the sample extruded under 1000 μl of ... cells / mL (cells / mL; 8 μm pore size) are shown in line 8 of FIG.

[0447] Figures 22 and 23 show the cleavage rates for on- and off-target sites recognized by sgRNAs targeting human PCSK9 sites as measured by quantitative PCR (qPCR). Figure 22 shows the cleavage rate results for on-target and off-target 2 sites for samples using human PCSK9 targeting sgRNA. Figure 23 shows the cleavage rate results for off-target 4 and off-target 7 sites for samples using human PCSK9 targeting sgRNA.

[0448] Through this experiment, the inventors of the present application determined the optimized conditions for Extru-seq for subsequent experiments as follows: pore size of 8 μM, Cas9 RNP concentration of 5000 nM, and 10 7 Under optimized conditions, the quality of genomic DNA incubated overnight at 37 °C with Cas9 RNP was confirmed to be high enough for constructing a whole genome sequencing (WGS) library.

[0449] Result 3. Measurement of NHEJ levels after the extrusion step The inventors of this application hypothesized that there is no DNA repair mechanism in the Extru-seq process to religate the genomic DNA cleaved by Cas9. In fact, when DNA was cleaved through the Extru-seq process and the cleavage rate of the target site was measured using quantitative PCR, an average rate of 70% was observed. The results on the cleavage rate of the target site are shown in Figures 24 to 31.

[0450] Figures 24 to 30 show WGS data from Extru-seq analyzed using IGV to identify cleavage patterns. Figure 24 shows results obtained using a human PCSK9 targeting sgRNA. Figure 25 shows results obtained using a human albumin targeting sgRNA. Figure 26 shows results obtained using a mouse PCSK9 targeting sgRNA. Figure 27 shows results obtained using a mouse albumin targeting sgRNA. Figure 28 shows results obtained using a human FANCF targeting sgRNA. Figure 29 shows results obtained using a human VEGFA targeting sgRNA. Figure 30 shows results obtained using a human HBB targeting sgRNA.

[0451] Figure 31 shows the cleavage rates of the seven on-target sites for each target, obtained through manual calculations based on IGV analysis of qPCR and WGS data. In Figure 31, the y-axis represents the cleavage rate.

[0452] The results shown in Figures 24 to 31 demonstrate the absence of DNA repair mechanisms such as NHEJ, indicating that Extru-seq can reflect positive characteristics in vitro.

[0453] In addition, the inventors of the present application analyzed the cleaved and uncleaved populations of the on-target site to investigate which NHEJ occurred or the extent to which NHEJ occurred after the extrusion process. First, considering that indel mutations accumulated in the uncleaved group when the NHEJ process was complete during the incubation period after the extrusion process, the uncleaved population in the Extru-seq sample was analyzed through deep sequencing. The deep sequencing results for the Extru-seq sample treated with Cas9 RNP complex were compared with the results for the control sample not treated with Cas9 RNP complex. As a result of the comparison, there was no significant difference between the two samples. The deep sequencing results of the uncleaved population are shown in Figure 32. In particular, Figure 32 shows the indel frequency measured by targeted deep sequencing for the uncleaved population of the Extru-seq sample related to Figures 24 to 30. The indel frequency was measured for the untreated sample and the Cas9 treated sample. In Figure 32, samples not treated with Cas9 are indicated as Cas9(-), and samples treated with Cas9 are indicated as Cas9(+). For t-test, unpaired Student's t-test was used. Error bars indicate standard deviation (n=3).

[0454] These results demonstrate that the level of NHEJ after the extrusion process is not significant.

[0455] Second, using a protocol from multiplex Digenome-seq (see Kim, Daesik, et al. "Genome-wide target specificities of CRISPR-Cas9 nucleases revealed by multiplex Digenome-seq." Genome research 26.3(2016):406-415.), the inventors of the present application performed multiplex Extru-seq to measure the change in cleavage rate at five different on-target sites in the presence or absence of SCR7 (chemical DNA ligase IV or NHEJ inhibitor; see Chu, Van Trung, et al. "Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells." Nature biotechnology 33.5(2015):543-548.). When NHEJ occurs, the cleavage rate increases due to the presence of SCR7, and this effect may also accumulate during the incubation step. However, the difference in the average cleavage rate at the 5 on-target sites in the presence or absence of SCR7 is not significant. The results related to the presence or absence of SCR7 are shown in Figure 33. In particular, Figure 33 shows the cleavage rate (%) measured using qPCR at the 5 target positions. The results were obtained by multiplex Extru-seq in the presence (+SCR7 in Figure 33) or absence (-SCR7 in Figure 33) of 1 μM SCR7. The horizontal lines in the graphs represent the average for the experiments (n=5). For t-tests, an unpaired Student's t-test was used.

[0456] These results further demonstrate that NHEJ has no significant effect on on-target cleavage rates.

[0457] The inventors of this application hypothesized that cellular components other than genomic DNA are not damaged, so that the cleavage pattern is similar to that of cell-based off-target prediction methods. This hypothesis was tested by comparing Extru-seq results with those of cell-based and in vitro-based methods (described below). Extru-seq was confirmed to be an off-target prediction method that can reflect the positive features of both cell-based off-target prediction methods (complete cellular components other than genomic DNA) and in vitro off-target prediction methods (absence of DNA repair mechanisms).

[0458] Result 4. Design and Use of Promiscuous Guide Sequences The second objective of this study was to develop a standard test that could effectively measure the performance metrics for each method. Previous studies used guide sequences predicted to recognize only a small number of off-target sites in the genome to compare other methods. As a result, only a small number of guide sequence-validated off-target loci were found, and therefore it was difficult to effectively compare different prediction methods using a statistically significant number of loci. In more recent papers (see Wienert, Beeke, et al. "Unbiased detection of CRISPR off-targets in vivo using DISCOVER-Seq." Science 364.6437(2019):286-289.; and Akcakaya, Pinar, et al. "In vivo CRISPR editing with no detectable genome-wide off-target mutations." Nature 561.7723(2018):416-419.), predicted promiscuous guide sequences were used to recognize a large number of off-target loci. The use of promiscuous guide sequences provided a powerful testbed for genome-wide off-target prediction methods. However, these promiscuous guide sequences were not used in this study. One of these involved a mouse guide sequence targeting PCSK9 that was not complementary to the sequence of human cells (as shown in previous studies). Another one targeting VEGFA lacked the off-target locus predicted as a single mismatch (see Table 3).

[0459] To overcome such limitations, the present study looked for two types of promiscuous guide sequences targeting PCSK9 and albumin in the mouse genome. The guide sequences also perfectly matched target sequences present in the human genome (however, in the human genome, PCSK9 and albumin loci are not targets). Even if the guide sequences target loci other than PCSK9 or albumin in the human genome, they are conveniently referred to as human PCSK9 and human albumin. The number of off-target sequences for these promiscuous guide sequences in each of the human and mouse genomes was calculated using Cas-OFFinder. It was confirmed that the selected guide sequences are associated with a large number of off-target sequences in both genomes. Information on the target sequences of the guide sequences used in the previous study and the promiscuous guide sequences used in this study, as well as the survey results on off-target sites, are shown in Table 3 below. In particular, Table 3 shows survey results on genome-wide off-target loci containing 0 to 6 mismatches. Off-target sites were predicted through Cas-OFFinder in genome hg19 (Table 3(a)) and genome mm10 (Table 3(b)). Referring to Table 3, the sequence GACCCCCTCCACCCCGCCTC (SEQ ID NO: 72) (VEGFA target sequence), the sequence AGCAGCAGCGCGGCAACAG (SEQ ID NO: 73) (PCSK9 target sequence, previous study), the sequence ACATGCATATGTATGTGTG (SEQ ID NO: 26) (Albumin-target sequence), and the sequence AGGTGGGGAAACTGAGGCTT (SEQ ID NO: 25) (PCSK9 target sequence) are shown.

[0460] Table 3. Genome-wide off-target site survey results following target or guide sequences [Table 3] [Table 4]

[0461] Result 5. Genome-wide off-target site prediction using GUIDE-seq, Digenome-seq, in silico methods, and Extru-seq Using the promiscuous sgRNA sequences targeting PCSK9 and albumin, respectively, in silico predictions based on GUIDE-seq, Digenome-seq, Extru-seq, and Cas-OFFinder were performed. Off-target prediction systems for each sgRNA sequence were performed on human cell lines (HEK293T) and mouse cell lines (NIH-3T3). The results are shown in Figures 34 to 62 and Figures 3 and 4.

[0462] In particular, Figures 34 and 35 show GUIDE-seq results obtained from HEK293T cells using PCSK9 targeting sgRNA. In Figures 34 and 35, the target sequence (including the PAM sequence) AGGTGGGGAAACTGAGGCTTNGG (SEQ ID NO: 44) is shown.

[0463] Figures 36 and 37 show GUIDE-seq results obtained from HEK293T cells using an albumin-targeting sgRNA. In Figures 36 and 37, the target sequence (including the PAM sequence) ACATGCATATGTATGTGTGNGG (SEQ ID NO: 45) is shown.

[0464] Figures 38 and 39 show GUIDE-seq results obtained from NIH-3T3 cells using PCSK9 targeting sgRNA. In Figures 38 and 39, the target sequence (including the PAM sequence) AGGTGGGGAAACTGAGGCTTNGG (SEQ ID NO: 44) is shown.

[0465] Figures 40 and 41 show GUIDE-seq results obtained from NIH-3T3 cells using an albumin-targeting sgRNA. In Figures 40 and 41, the target sequence (including the PAM sequence) ACATGCATATGTATGTGTGNGG (SEQ ID NO: 45) is shown.

[0466] Relatively low-ranking off-target loci were omitted from the corresponding plots by the GUIDE-seq analysis program, and the omitted loci were included in subsequent analyses.

[0467] Figures 42 and 43 show Manhattan plot results of Digenome-seq obtained from HEK293T cells using PCSK9 targeting sgRNA, where the y-axis represents DNA cleavage score.

[0468] Figures 44 and 45 show Manhattan plot results of Digenome-seq obtained from HEK293T cells using albumin-targeting sgRNA, where the y-axis represents DNA cleavage score.

[0469] Figures 46 and 47 show Manhattan plot results of Digenome-seq obtained from NIH-3T3 cells using PCSK9 targeting sgRNA, where the y-axis represents DNA cleavage score.

[0470] Figures 48 and 49 show Manhattan plot results of Digenome-seq obtained from NIH-3T3 cells using albumin-targeting sgRNA, where the y-axis represents DNA cleavage score.

[0471] Figures 50 and 51 show Manhattan plot results of Extru-seq obtained from HEK293T cells using PCSK9 targeting sgRNA, where the y-axis represents DNA cleavage score.

[0472] Figures 52 and 53 show Manhattan plot results of Extru-seq obtained from HEK293T cells using albumin-targeting sgRNA, where the y-axis represents DNA cleavage score.

[0473] Figures 54 and 55 show Manhattan plot results of Extru-seq obtained from NIH-3T3 cells using PCSK9 targeting sgRNA, where the y-axis represents DNA cleavage score.

[0474] Figures 56 and 57 show Manhattan plot results of Extru-seq obtained from NIH-3T3 cells using albumin-targeting sgRNA, where the y-axis represents DNA cleavage score.

[0475] The inventors of the present application predicted off-target sites (candidate off-target sites) for sgRNAs (sgRNA targeting human PCSK9, sgRNA targeting human albumin, sgRNA targeting mouse PCSK9, and sgRNA targeting mouse albumin) by GUIDE-seq, Digenome-seq, Extru-seq, and in silico methods, and the results were compared. Using Venn diagrams, the comparison results are shown in Figures 3 and 4. In particular, Figure 3 shows the comparison results for sgRNA targeting human PCSK9 and the comparison results for sgRNA targeting human albumin. Figure 4 shows the comparison results for sgRNA targeting mouse PCSK9 and the comparison results for sgRNA targeting mouse albumin. For the results of Figures 3 and 4, a human cell line (HEK293T) and a mouse cell line (NIH-3T3) were used.

[0476] Using the sequence read counts from GUIDE-seq and DNA cleavage scores from Digenome-seq and Extru-seq, it was possible to rank the candidate off-target loci. For in silico predictions based on Cas-OFFinder, there is no score that can be used for these rankings. Therefore, the inventors of the present application calculated prediction scores for candidate off-target sites for ranking using two different scripts of machine learning studies (see Liu, Qiaoyue, et al. "Deep learning improves the ability of sgRNA off-target propensity prediction." BMC bioinformatics 21.1 (2020): 1-15.; and Doench, John G., et al. "Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9." Nature biotechnology 34.2 (2016): 184-191.). The prediction scores are: CRISPR off-target predictor (CROP) score (a heuristic score indicating whether the candidate off-target site will be edited) and cutting frequency determination (CFD) score (a percent activity value provided to a penalty matrix based on each possible type of mismatch at each position in the guide RNA sequence). The distribution of sequence read counts, DNA cleavage, and in silico prediction scores for each candidate off-target locus was tabulated against the number of mismatches with the guide sequence. Tables showing the distribution of sequence read counts, DNA cleavage scores, and in silico prediction scores as a function of the number of mismatches with the guide sequence are shown in Figures 58 to 62.

[0477] In particular, Figure 58 shows results related to on-target and off-target scores (scores calculated from sequence read count results) as a function of the number of mismatches predicted using GUIDE-seq. The x-axis shows the number of mismatches, and the scores for each number of mismatches are shown. That is, the number of mismatches for on-target and off-target sites is shown on the x-axis. The scores converted from sequence read counts are shown on the y-axis.

[0478] Figure 59 shows the results related to on-target and off-target scores (Manhattan plot cleavage scores) according to the number of mismatches predicted using Digenome-seq. The number of mismatches is shown on the x-axis, and the score for each number of mismatches is shown. That is, the number of mismatches for on-target and off-target sites is shown on the x-axis. The cleavage score is shown on the y-axis.

[0479] Figure 60 shows the results related to the on-target and off-target scores (CROP scores) according to the number of mismatches predicted using the in silico system. The x-axis shows the number of mismatches and the score for each number of mismatches. That is, the number of mismatches for on-target and off-target sites is displayed on the x-axis. The CROP score is displayed on the y-axis.

[0480] Figure 61 shows the results related to on-target and off-target scores (CFD scores) according to the number of mismatches predicted using the in silico system. The x-axis shows the number of mismatches and the score for each number of mismatches. That is, the number of mismatches for on-target and off-target sites is displayed on the x-axis. The CFD score is displayed on the y-axis.

[0481] Figure 62 shows the results related to on-target and off-target scores (Manhattan plot cleavage scores) as a function of the number of mismatches predicted using Extru-seq. The x-axis shows the number of mismatches and the score for each number of mismatches. That is, the number of mismatches for on-target and off-target sites is displayed on the x-axis. The cleavage score is displayed on the y-axis.

[0482] It was expected that as the number of mismatches increased, the corresponding prediction score would decrease. Although GUIDE-seq and in silico predictions followed such a trend, outliers with high DNA cleavage scores were observed even when there were 4, 5, or 6 mismatches in the Digenome-seq results. Unlike the Digenome-seq results, when the DNA cleavage scores for off-target site candidates were calculated using the Extru-seq approach, high DNA cleavage scores for off-target candidates of sgRNAs with 4 or more mismatches were not observed. This result indicates that Extru-seq identified fewer false positives than Digenome-seq. The inventors of the present application confirmed such an idea by validating the off-target candidates with the top scores.

[0483] Result 6. Validation rate of GUIDE-seq and Extru-seq GUIDE-seq and Extru-seq showed high validation rates. Validation of predicted off-target loci was performed in human cell lines and mouse models. In human cell line experiments, plasmids encoding Cas9 protein and sgRNA were transfected into HEK293T cells. In mouse experiments, sequences encoding Cas9 protein and sgRNA were packaged into adeno-associated virus (AAV) serotype 8 (i.e., AAV8). These AAVs were then delivered to C57BL / 6 mice via systemic or subretinal injection. As only subretinal injections have shown high frequencies of on-target indel formation, model retinal pigment epithelial cells were used in validation experiments.

[0484] Results regarding indel formation frequency in subretinal and systemic injections are shown in Figures 63 and 64. In particular, the indel rate was calculated through analysis of genomic DNA obtained from organs of C57BL / 6 mice injected with two AAV8 vectors expressing Cas9 and sgRNAs targeting PCSK9 or albumin, respectively. In Figures 63 and 64, the results marked as iv represent the results for systemic injections, and the results marked as subretinal represent the results for subretinal injections. In Figures 63 and 64, error bars represent sem (n=3). NR represents neural retina, and RPE represents retinal pigment epithelial cells.

[0485] The top 10 candidates of each prediction method were tested using targeted deep sequencing. As a result of the testing, Extru-seq showed a 92.5% validation rate, and GUIDE-seq showed a 97.5% validation rate. However, Digenome-seq showed a 45% validation rate, and the in silico methods showed a 62.5% validation rate in CROP and a 67.5% validation rate in CFD. Extru-seq and GUIDE-seq showed much higher validation rates than Digenome-seq and the in silico methods. The results of the validation rates by each prediction method are shown in Figure 5 and Table 4.

[0486] In particular, Figure 5 shows the validation rates of the top off-target sites predicted by in silico methods, GUIDE-seq, Digenome-seq, and Extru-seq, showing the experimental results in human and mouse cells for promiscuous sgRNAs targeting PCSK9 and albumin (*P<0.05, ns, not significant in a two-tailed unpaired Mann-Whitney test).

[0487] Table 4 below shows the validation results obtained through targeted deep sequencing of the top 10 off-target sites predicted by each method (using sgRNA targeting human PCSK9, sgRNA targeting human albumin, sgRNA targeting mouse PCSK9, and sgRNA targeting mouse albumin). For validation, the indel frequency at the off-target site must be higher than 0.1% and the formula '(indel frequency at off-target locus) / (indel frequency in control without Cas9 treatment)>2' must be satisfied (see reference [Frangoul, Haydar, et al. "CRISPR-Cas9 gene editing for sickle cell disease and β-thalassemia." New England Journal of Medicine 384.3(2021):252-260.]). * indicates that the target was manually confirmed. Here, manually confirmed (off) target sites refer to sites that were determined to be negative using the Digenome software but were determined to be positive using the IGV software. As a result of confirmation using the IGV software, 1 additional site was identified in the human PCSK9 sample. 5 additional sites were identified in the human albumin sample. 3 additional sites were identified in the mouse PCSK9 sample. 3 additional sites were identified in the mouse albumin sample.

[0488] Table 4. Validation results for the top 10 candidate off-target sites predicted by each off-target prediction method (a) [Table 5] (b) [Table 6] (c) [Table 7] (d) [Table 8]

[0489] Information about manually identified targets is shown in Figures 77 to 116. In particular, WGS data visualized using IGV for manually validated off-target sites from Extru-seq is shown in Figures 77 to 116. For Figures 77 to 116, the sequences of the off-target sites are shown, with the sequences mismatched to the guide sequence shown in lower case. The PAM sequences are underlined.

[0490] With reference to Figures 77 and 78, the off-target site sequence AGGTGGGAAAACTGAGGCccAGG (SEQ ID NO:52) is disclosed.

[0491] With reference to Figures 79 and 80, the off-target site sequence tgATGCATATGTATGTGTGGaGG (SEQ ID NO:53) is disclosed.

[0492] With reference to Figures 81 and 82, the sequence of the off-target site AaATGCATATGTATGaGTGTGG (SEQ ID NO:54) is disclosed.

[0493] With reference to Figures 83 and 84, the off-target site sequence CATGCATATGcATGTGgGAGG (SEQ ID NO:55) is disclosed.

[0494] With reference to Figures 85 and 86, the off-target site sequence AgATGCATAgGTATGTGTGTGG (SEQ ID NO:56) is disclosed.

[0495] With reference to Figures 87 and 88, the off-target site sequence ACtTGCATATcTATGTGTGTGG (SEQ ID NO:57) is disclosed.

[0496] With reference to Figures 89 and 90, the off-target site sequence ccGTGGGAAACTGAGGCTTGGG (SEQ ID NO:58) is disclosed.

[0497] With reference to Figures 91 and 92, the off-target site sequence AGGTGGGAAAACTGAGGCTgAGG (SEQ ID NO:59) is disclosed.

[0498] With reference to Figures 93 and 94, the off-target site sequence AGGaGGGAAACTGAGGCTcAGG (SEQ ID NO: 60) is disclosed.

[0499] With reference to Figures 95 and 96, the off-target site sequence AaATaCATATGTATGTGTGTGG (SEQ ID NO: 61) is disclosed.

[0500] With reference to Figures 97 and 98, the sequence of the off-target site ACATGtATATGTATaTGTGTGG (SEQ ID NO: 62) is disclosed.

[0501] With reference to Figures 99 and 100, the off-target site sequence ACATatATATGTATGTGTGTGG (SEQ ID NO: 63) is disclosed.

[0502] With reference to Figures 101 and 102, the off-target site sequence GGGTGGGtGGAGTTTGCTaCTGG (SEQ ID NO: 64) is disclosed.

[0503] With reference to Figures 103 and 104, the off-target site sequence aGGTGGtGGGAGcTTGtTCCTGG (SEQ ID NO: 65) is disclosed.

[0504] With reference to Figures 105 and 106, the off-target site sequence GGtgGGGGtGgGTTTGCTCCTGG (SEQ ID NO:66) is disclosed.

[0505] With reference to Figures 107 and 108, the off-target site sequence GGGcaaGGGGAGgTTGCTCCTGG (SEQ ID NO: 67) is disclosed.

[0506] With reference to Figures 109 and 110, the sequence of the off-target site GGAtTgCCaTCcGCAGCACCTGG (SEQ ID NO: 68) is disclosed.

[0507] With reference to Figures 111 and 112, the off-target site sequence GGAgTCCCTcCTGCAGCACCTGA (SEQ ID NO:69) is disclosed.

[0508] With reference to Figures 113 and 114, the sequence of the off-target site aGAggCCCcTCTGCAGCACCAGG (SEQ ID NO: 70) is disclosed.

[0509] With reference to Figures 115 and 116, the off-target site sequence accATCCCTcCTGCAGCACCAGG (SEQ ID NO: 71) is disclosed.

[0510] Result 7. Further comparison of Extru-seq, GUIDE-seq, Digenome-seq, and DIG-seq Digenome-seq uses purified genomic DNA that has lost components such as chromatin proteins. To overcome this problem, previous studies have developed an improved version of Digenome-seq, named DIG-seq. DIG-seq, which uses cell-free chromatin DNA rather than histone-free DNA, predicted fewer false positives than Digenome-seq. The mild detergent used to dissolve cells in the DIG-seq approach may affect the chromatin state of cell DNA, which may affect the Cas9 cleavage mechanism, and therefore the inventors of this application expected that Extru-seq, which uses physical force for cell dissolution, would reflect more of the characteristics of cell-based methods compared to DIG-seq.

[0511] To compare Extru-seq with other in vitro methods, the inventors of the present application performed GUIDE-seq and Extru-seq using guide sequences targeting FANCF, VEGFA, and HBB in HeLa cells. Each of the guide sequences targeting FANCF, VEGFA, and HBB was used to compare DIG-seq and Digenome-seq in a previous study (see Kim, Daesik, and Jin-Soo Kim. "DIG-seq: a genome-wide CRISPR off-target profiling method using chromatin DNA." Genome research 28.12(2018):1894-1900.).

[0512] GUIDE-seq results in HeLa cells are shown in Figures 65 to 67 (using sgRNAs targeting FANCF, VEGFA, and HBB). For Figure 65, the target sequence (including PAM) GAATCCCTTCTGCAGCACCNGG (SEQ ID NO: 46) is shown. For Figure 66, the target sequence (including PAM) GGTGGGGGGAGTTTGCTCCNGG (SEQ ID NO: 47) is shown. For Figure 67, the target sequence (including PAM) TTGCCCCACAGGGCAGTAANGG (SEQ ID NO: 48) is shown. In particular, Figure 65 shows the results of predicting off-targets through GUIDE-seq using a FANCF targeting sgRNA (sequence read results). Figure 66 shows the results of predicting off-targets through GUIDE-seq using a VEGFA targeting sgRNA (sequence read results). Figure 67 shows the results of predicting off-targets through GUIDE-seq using a HBB targeting sgRNA (sequence read results).

[0513] Extru-seq results in HeLa cells are shown in Figures 68 to 73 (using sgRNAs targeting FANCF, VEGFA, and HBB). In particular, Figures 68 and 69 show the results of predicting off-targets through Extru-seq (Manhattan plot results) using FANCF-targeting sgRNA. Figures 70 and 71 show the results of predicting off-targets through Extru-seq (Manhattan plot results) using VEGFA-targeting sgRNA. Figures 72 and 73 show the results of predicting off-targets through Extru-seq (Manhattan plot results) using HBB-targeting sgRNA. The y-axis shows the DNA cleavage score.

[0514] Figures 6 and 7 show a comparison of the results for off-target candidates according to each method (Extru-seq, GUIDE-seq, DIG-seq, Digenome-seq) through Venn diagrams (using HeLa cells, human FANCF targeting sgRNA, human VEGFA targeting sgRNA, and human HBB targeting sgRNA). The analysis results through Venn diagrams show that Digenome-seq and DIG-seq predicted multiple different off-target loci. On the other hand, it shows that most off-target loci predicted by Extru-seq were identified by at least one of the different off-target prediction methods.

[0515] When tested to see if candidate loci could be validated, Extru-seq showed a higher validation rate than DIG-seq and Digenome-seq. The validation results are shown below in Table 5.

[0516] In particular, Table 5(a) shows the validation results of the top 10 predicted off-target sites for sgRNAs targeting human FANCF. Table 5(b) shows the validation results of the top 10 predicted off-target sites for sgRNAs targeting human VEGFA. Table 5(c) shows the validation results of the top 10 predicted off-target sites for human HBB targeting sgRNAs. The off-target sites were validated through targeted deep sequencing. For validation, the indel frequency of the off-target sites must be higher than 0.1% and satisfy the formula '(indel frequency at off-target locus) / (indel frequency in control)>2'. In Table 5, * indicates that the target was manually validated. The results for manually validated targets are shown in Figures 101 to 116.

[0517] Table 5. Validation results of the top 10 off-target sites predicted by each off-target prediction method (a) [Table 9] (b) [Table 10] (c) [Table 11]

[0518] Figure 8 shows the validation results for off-target sites predicted by each method (DIG-seq, Digenome-seq, Extru-seq, and GUIDE-seq). In particular, the validation rates for the results related to Figures 6 and 7 and Table 6 are shown graphically (ns, not significant in two-tailed unpaired Mann-Whitney test).

[0519] Result 8. Comparison of rank distribution of off-target sites predicted by GUIDE-seq and Extru-seq The inventors of the present application compared the degree of agreement between the prediction results obtained by Extru-seq, cell-based, in vitro, and in silico methods. The top 10 candidate off-target loci for each prediction method were tabulated, and the ranking of these loci for each different method was tabulated. The ranking of off-target loci in each method was then compared. The results are shown in Tables 6 to 9. The loci predicted to be in the top 10 by pairs of methods were counted and the sharing ratio was calculated. For this purpose, the sharing ratio (%) of the top 10 (rank) was calculated. That is, the top 10 off-target candidates were extracted in method A, and the ranks occupied by the off-target candidates corresponding to the top 10 off-target candidates of A in another method (e.g., method B) were compared. If the corresponding off-target candidates were in the top 10 in another method, they were determined to contribute to the sharing ratio. As a result of the calculation, a low similarity was observed in most cases (overall average sharing ratio in the top 10 = 22%). The highest similarity was consistently found between GUIDE-seq and Extru-seq pairwise comparisons (average shared percentage in the top 10 for GUIDE-seq and Extru-seq pairs = 43%).

[0520] Table 6. Off-target prediction results for sgRNAs targeting human PCSK9 (in silico, GUIDE-seq, Digenome-seq, and Extru-seq) [Table 12-1] [Table 12-2] [Table 12-3] [Table 12-4] [Table 12-5]

[0521] Table 7. Off-target prediction results for sgRNA targeting human albumin (in silico, GUIDE-seq, Digenome-seq, and Extru-seq) [Table 13-1] [Table 13-2] [Table 13-3] [Table 13-4] [Table 13-5]

[0522] Table 8. Off-target prediction results for sgRNA targeting mouse PCSK9 (in silico, GUIDE-seq, Digenome-seq, and Extru-seq) [Table 14-1] [Table 14-2] [Table 14-3] [Table 14-4] [Table 14-5]

[0523] Table 9. Off-target prediction results for sgRNA targeting mouse albumin (in silico, GUIDE-seq, Digenome-seq, and Extru-seq) [Table 15-1] [Table 15-2] [Table 15-3] [Table 15-4] [Table 15-5]

[0524] A comparison of ranks with all off-target sites, including non-validated sites, can also be performed. Venn diagrams (Figures 3, 4, 6, and 7) show that there is a statistically significant number of candidate off-target sites in the intersection analyzed. To check the homogeneity of median scores of loci in the intersection of results from the two methods, scores / read counts were min-max normalized and Wilcoxon rank sum test was performed. The analysis results are shown in Figure 9. In Figure 9, the dotted line is p=0.05. Because a sample size of at least 16 was required to use the asymptotic nonparametric Wilcoxon rank test (see references [MUNDRY, ROGER, and JULIA FISCHER. "Use of statistical programs for nonparametric tests of small samples often leads to incorrect P values: examples from animal behavior." Animal behavior 56.1(1998):256-259.; and Dwivedi, Alok Kumar, Indika Mallawaarachchi, and Luis A. Alvarado. "Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method." Statistics in medicine 36.14(2017):2187-2205.]), intersections with fewer than 16 samples were not included in this analysis. See Table 10 below. In this test, low p-values ​​indicate that the locus scores are distributed differently from the intersection between the two populations. Except for the distribution of GUIDE-seq:Extru-seq and DIG-seq:Digenome-seq pairs, none of the pairs had a similar distribution and had high p-values ​​for N > 3. In Figure 9, p-values ​​were obtained from the normalized rank sum test for each off-target prediction method pair.Regarding sgRNAs, sgRNAs targeting FANCF, VEGFA, and HBB were used in HeLa cells, and sgRNAs targeting PCSK9 and albumin were used in human and mouse cells (n>16 were selected for analysis).

[0525] With reference to FIG. 9, Table 10 is shown below. Table 10 shows the number of samples found at the intersection of the Venn diagram. The table shows the number of samples for each sgRNA (sgRNAs targeting human PCSK9, human albumin, mouse PCSK9, mouse albumin, human FANCF, human VEGFA, and human HBB). Cases with n>=16 (16 is the minimum number of samples required for the asymptotic nonparametric Wilcoxon rank test) are underlined.

[0526] Table 10. Number of predicted off-target candidates found in the overlapping regions in Figures 3 and 4, and Figures 6 and 7 (comparison between each method) [Table 16] [Table 17]

[0527] The discrepancies between the results of GUIDE-seq or Extru-seq and the results of Digenome-seq or in silico predictions are speculated to be due to the low validation rate of Digenome-seq caused by a large number of false positives and the low validation rate of in silico predictions caused by the difference between machine learning-based prediction scores and real-world experimental values. Moreover, unlike the Digenome-seq:Extru-seq pair showing low p-values, the DIG-seq:Digenome-seq pair consistently showed high p-values, and the results obtained from DIG-seq were analyzed to be similar to the results obtained from in vitro prediction methods (denoted here as Digenome-seq). However, the results obtained from Extru-seq were analyzed to be more similar to the cell-based prediction methods (denoted here as GUIDE-seq) and different from the results obtained from in vitro prediction methods such as Digenome-seq. This is because, as mentioned above, the results for the GUIDE-seq:Extru-seq pair showed high p-values, and the results for the Extru-seq:Digenome-seq pair showed low p-values. In this regard, Extru-seq is distinguished from DIG-seq (which still shows similarity to Digenome-seq) in that it has lost similarity to in vitro Digenome-seq. All the experimental processes of Digenome-seq, DIG-seq and Extru-seq used in this experiment include WGS, and the analysis through GUIDE-seq is carried out based on PCR, so such results showing similarity between Extru-seq and GUIDE-seq are somewhat surprising. This indicates that the conditions for treating genomic DNA with Cas9 are more important than the analysis procedure.

[0528] Result 9. Comparison of the oversight rates of Extru-seq and GUIDE-seq It is known that cell-based methods, including GUIDE-seq, may miss true off-target candidates. The inventors of the present application calculated the miss rate (or the false negative rate calculated by the following formula: (number of false negatives) / (number of false negatives+number of true positives)) using a Venn diagram showing the overlap between the predicted and confirmed valid targets in samples analyzed by Extru-seq and GUIDE-seq and deep sequencing. As a result of the investigation, the average miss rate of Extru-seq was confirmed to be 2.3%, and the average miss rate of GUIDE-seq was confirmed to be 29% (see Figures 10 to 14 and Tables 11 to 17). The combined results for the miss rates of Extru-seq and GUIDE-seq are shown in the graph in Figure 14.

[0529] In particular, Figures 10 to 13 show Venn diagrams used to confirm the missed rate. In particular, the Venn diagrams shown in Figures 10 to 13 show the comparison results for off-target candidates predicted by Extru-seq and GUIDE-seq, and off-targets confirmed as valid (shown as validation). Off-target prediction and validation were performed using (a) sgRNA targeting human PCSK9 (Figure 10), (b) sgRNA targeting human albumin (Figure 10), (c) sgRNA targeting mouse PCSK9 (Figure 11), (d) sgRNA targeting mouse albumin (Figure 11), (e) sgRNA targeting human FANCF (Figure 12), (f) sgRNA targeting human VEGFA (Figure 12), and (g) sgRNA targeting human HBB (Figure 13). Validation represents targets (off-target and on-target) confirmed as valid by targeted deep sequencing. In Figures 10 to 13, * indicates the number of manually confirmed off-target sites.

[0530] Figure 14 shows the observed miss rates for Extru-seq and GUIDE-seq. Miss rates are disclosed for each method. Additionally, miss rates are disclosed for each sgRNA (sgRNA targeting human PCSK9, sgRNA targeting human albumin, sgRNA targeting mouse PCSK9, sgRNA targeting mouse albumin, sgRNA targeting human FANCF, sgRNA targeting human VEGFA, and sgRNA targeting human HBB) (*: P<0.05 in a two-tailed unpaired Mann-Whitney test).

[0531] FIG. 15 shows the distribution of the number of off-target mismatches missed in GUIDE-seq.

[0532] Tables 11 to 17 show the off-target analysis results predicted by each method (Extru-seq or GUIDE-seq) and the actual off-target analysis results confirmed through deep sequencing (related to Figures 10 to 14). That is, the results for off-target candidates predicted by Extru-seq or GUIDE-seq and the off-target sites confirmed as valid by deep sequencing were compared. (a) sgRNA targeting human PCSK9, (b) sgRNA targeting human albumin, (c) sgRNA targeting mouse PCSK9, (d) sgRNA targeting mouse albumin, (e) sgRNA targeting human FANCF, (f) sgRNA targeting human VEGFA, and (g) sgRNA targeting human HBB were each investigated. For validation, the indel frequency at off-target sites must be higher than 0.1% and must satisfy the formula '((indel frequency at off-target locus) / (indel frequency in control)>2)'.

[0533] In Tables 11 to 17, the target and position columns show information about targets that were validated through deep sequencing. + indicates that the target (validated on-target or off-target) was predicted by the indicated off-target prediction method. A blank cell indicates that the target (on-target or off-target) was not predicted by the indicated off-target prediction method (i.e., the indicated off-target prediction method missed a validated target). The missed rate was calculated by the formula '(number of blank cells) / (number of all columns)'. * indicates that the target was manually confirmed (see Figures 77 to 116).

[0534] Table 11. Confirmation of the off-target prediction method, Extru-seq, and GUIDE-seq miss-detection rates (results for sgRNA targeting human PCSK9) [Table 18-1] [Table 18-2]

[0535] Table 12. Results of confirming the off-target prediction method, Extru-seq, and GUIDE-seq miss rates (results for sgRNA targeting human albumin) [Table 19]

[0536] Table 13. Confirmation of off-target prediction methods, Extru-seq and GUIDE-seq miss rates (results for sgRNA targeting mouse PCSK9) [Table 20]

[0537] Table 14. Off-target prediction methods, Extru-seq and GUIDE-seq: Results of checking the missed-target rate (results for sgRNA targeting mouse albumin) [Table 21]

[0538] Table 15. Confirmation of off-target prediction methods, Extru-seq and GUIDE-seq miss rates (results for sgRNA targeting human FANCF) [Table 22]

[0539] Table 16. Confirmation of off-target prediction methods, Extru-seq and GUIDE-seq miss rates (results for sgRNA targeting human VEGFA) [Table 23]

[0540] Table 17. Off-target prediction methods, Extru-seq and GUIDE-seq: Results for sgRNA targeting human HBB [Table 24]

[0541] The results related to the miss rate in Figures 10 to 15 and Tables 11 to 17 indicate that the sensitivity of Extru-seq is much higher than that of the cell-based GUIDE-seq method and that Extru-seq rarely misses true off-target sites.

[0542] The results disclosed herein show that GUIDE-seq missed validated off-target sites containing 1 to 6 mismatches. Therefore, only GUIDE-seq-dependent IND studies (see Stadtmauer, Edward A., et al. "CRISPR-engineered T cells in patients with refractory cancer." Science 367.6481(2020):eaba7365.) are at risk of missing valid off-target candidates. For CTX001, in silico methods were used to complement GUIDE-seq (see Frangoul, Haydar, et al. "CRISPR-Cas9 gene editing for sickle cell disease and β-thalassemia." New England Journal of Medicine 384.3(2021):252-260.). However, only genomic sites with three or fewer mismatches, or two or fewer mismatches and a single DNA or RNA bulge are computationally identified, and valid off-target sites with three or more mismatches are still at risk of being overlooked.

[0543] Result 10. Receiver operating characteristic (ROC) curve for Extru-seq One of the powerful tools to evaluate a prediction model is the ROC curve. The ROC curve shows the sensitivity and specificity on the y and x axes, respectively. The ROC curves were plotted using sequence read counts (GUIDE-seq), DNA cleavage scores (Digenome-seq, DIG-seq, and Extru-seq), CDF scores (CDF), or CROP scores (CROP) depending on the metric to predict the validation outcome through binary classification. The ROC curves for different prediction methods are shown in Figures 16 to 18. In particular, Figures 16 to 18 show the ROC curves for GUIDE-seq, Digenome-seq, Extru-seq, CROP, and CFD prediction methods. Figure 16 (a) shows the results for the prediction method performed using sgRNA targeting human PCSK9. Figure 16 (b) shows the results for the prediction method performed using sgRNA targeting human albumin. FIG. 17(c) shows the results for a prediction method performed using a sgRNA targeting mouse PCSK9. FIG. 17(d) shows the results for a prediction method performed using a sgRNA targeting mouse albumin. FIG. 18(e) shows the results for a prediction method performed using a sgRNA targeting human FANCF. FIG. 18(f) shows the results for a prediction method performed using a sgRNA targeting human VEGFA. FIG. 18(g) shows the results for a prediction method performed using a sgRNA targeting human HBB.

[0544] Figure 19 shows the area under the curve of different methods calculated from the results (ROC curve results) disclosed in Figures 16 to 18. Extru-seq shows an area under the curve value of 0.83, GUIDE-seq shows an area under the curve value of 0.81, DIG-seq shows an area under the curve value of 0.80, Digenome-seq shows an area under the curve value of 0.72, CROP shows an area under the curve value of 0.69, and CFD shows an area under the curve value of 0.68. Error bars indicate standard deviation.

[0545] As mentioned above, the area under the ROC curve was calculated, and Extru-seq showed the highest area. In particular, Extru-seq showed an area under the ROC curve of 0.83, GUIDE-seq showed an area under the ROC curve of 0.81, DIG-seq showed an area under the ROC curve of 0.80, Digenome-seq showed an area under the ROC curve of 0.72, CROP showed an area under the ROC curve of 0.69, and CFD showed an area under the ROC curve of 0.68. The closer the area under the ROC curve is to 1, the better the model is at predicting the validation results. The highest area under the ROC curve for Extru-seq suggests a high performance in DNA cleavage score for Extru-seq. Furthermore, the use of different thresholds or cutoff values ​​may affect the predicted number of off-target sites. A high area under the ROC curve indicates a higher probability of finding a significant threshold for Extru-seq than other methods.

[0546] Result 11. Use of Extru-seq in primary cells The inventors of the present application confirmed that Extru-seq can be applied to primary cultured cells with less optimization. The GUIDE-seq method requires a high insertion rate of double-stranded oligodeoxynucleotide (dsODN) at double-strand break (DSB) sites, which may be difficult to experimentally achieve in some cell types and experimental conditions. For example, the inventors of the present application cannot obtain a high insertion rate of dsODN in primary mesenchymal stem cells (MSCs) derived from bone marrow. In contrast, Extru-seq does not require the insertion of dsODN. Considering these advantages of Extru-seq, the inventors of the present application carried out Extru-seq in MSCs using the above-mentioned promiscuous sgRNAs targeting human PCSK9 and albumin.

[0547] The Venn diagrams shown in Figures 74 and 75 show that differences exist between the Extru-seq results obtained in MSCs and HEK293T cells.

[0548] In particular, Figure 74 shows a comparison of predicted off-target sites for sgRNAs targeting human PCSK9, showing that only 1213 off-target sites predicted through Extru-seq performed on MSCs and those predicted through Extru-seq performed on HEK293T cells overlap (showing some differences depending on cell type).

[0549] Figure 75 shows a comparison of off-target candidates predicted using sgRNAs targeting human albumin, showing that only 26 off-target sites predicted by Extru-seq performed in MSCs and HEK293T cells overlap (showing some differences depending on cell type).

[0550] The test results for each prediction method in MSCs and HEK293T cells are further shown in Tables 18 and 19. The above-mentioned sgRNAs targeting PCSK9 and albumin were used as sgRNAs. In particular, Table 18 shows the top 10 candidates from off-target loci predicted by Extru-seq performed in MSCs for sgRNAs targeting human PCSK9. In addition, the ranking of the top 10 loci from Extru-seq predicted by other prediction methods is shown. The top 10 loci from Extru-seq were derived based on DNA cleavage scores.

[0551] Table 18 Comparison of predicted off-target candidates by cell type and method (sgRNAs targeting human PCSK9) [Table 25]

[0552] Table 19 shows the top 10 candidate off-target loci predicted by Extru-seq for sgRNA targeting human albumin in HEK293T cells. Additionally, Table 19 shows the ranking of the top 10 loci from Extru-seq predicted by other prediction methods. The top 10 loci from Extru-seq were derived based on DNA cleavage scores.

[0553] Table 19. Comparison of predicted off-target candidates by cell type and method (human albumin targeting sgRNA) [Table 26]

[0554] As shown in Tables 18 and 19, a comparison based on Extru-seq Top 10 in MSCs from the results associated with sgRNAs targeting human PCSK9 confirmed that Extru-seq Top 10 (MSC) and Extru-seq Top 10 (HEK293T) were 30% identical.

[0555] Comparison based on Extru-seq top 10 in MSCs from results associated with sgRNAs targeting human albumin confirmed that Extru-seq top 10 (MSC) and Extru-seq top 10 (HEK293T) were 70% identical. These results indicate that genome-wide off-target loci predicted by Extru-seq vary depending on the cell type.

[0556] Using a normalized rank sum test to analyze the intersection of the Venn diagrams, high p-values ​​were observed in tests involving sgRNAs targeting albumin, whereas low p-values ​​were observed in tests involving sgRNAs targeting PCSK9 (see Figure 76).

[0557] In particular, Figure 76 shows p-values ​​obtained from pairwise normalized rank sum tests of the off-target prediction methods for promiscuous sgRNAs targeting PCSK9 and albumin in MSCs and HEK293T cells. These results indicate that the ranking of off-targets may or may not change depending on the cell type.

[0558] It is known that cell-based methods such as GUIDE-seq miss more valid off-target candidates than in vitro and in silico methods. The inventors of this application confirmed that the miss rate of Extru-seq (2.33%) is 12.6 times lower than that of GUIDE-seq (29.5%). Furthermore, it was confirmed that Extru-seq, like other in vitro methods, can be universally applied to various cell types of different origins. This is because Extru-seq does not require dsODN insertion, unlike GUIDE-seq, which requires the insertion of dsODN into DSB sites. Extru-seq overcomes the major limitations of cell-based (high miss rate and need for optimization for different cell types) and in vitro methods (low validation rate and loss of cell type-specific information). Furthermore, the focused implementation of Extru-seq as a binary classifier of validation results is supported by the area under the ROC curve for Extru-seq. Therefore, Extru-seq is expected to be a strong candidate as a balanced method to obtain a comprehensive list of off-target sites in various cell type and patient-specific clinical safety studies.

[0559] Most cell-based methods use "surrogate" cell lines to predict genome-wide off-target sites for human clinical samples. However, as can be seen by comparing the results of Extru-seq for HEK293T cells and MSCs, there may be differences in chromatin and epigenetic states between dividing in vitro cell lines and most non-dividing in vivo cells. Therefore, it is preferable that off-target prediction is performed by Extru-seq in more clinically relevant cells rather than in surrogate cell lines. Recently, two cell-based methods, such as DISCOVER-seq (see Wienert, Beeke, et al. "Unbiased detection of CRISPR off-targets in vivo using DISCOVER-Seq." Science 364.6437(2019):286-289.) and GUIDE-tag (see Liang, Shun-Qing, et al. "Genome-wide detection of CRISPR editing in vivo using GUIDE-tag." Nature communications 13.1(2022):1-14.), have been directly performed in vivo in mouse models. However, for preclinical studies on human treatment, it is nearly impossible to directly perform these methods in human organs. Extru-seq has the advantage of being performed in primary human cells isolated from a specific patient or organ. In the experimental examples of this application, whole genome sequencing (WGS) was used as a genome analysis method after extrusion of Extru-seq.The present application is not limited to the methods disclosed in the experimental examples of the present application, and it is strongly anticipated that different methods such as PCR-based amplification protocols (e.g., PCR-based amplification protocols used in SITE-Seq, see Cameron, Peter, et al. "Mapping the genomic landscape of CRISPR-Cas9 cleavage." Nature methods 14.6 (2017): 600-606.) may also be used for post-extrusion genomic analysis (e.g., analysis at DNA cleavage sites). Furthermore, it is strongly anticipated that optimization of the algorithm used (e.g., optimization of the algorithm to increase analytical sensitivity), optimization of the extruder (e.g., optimization of the size, cost, and throughput of the extruder, etc.) may be performed for Extru-seq optimization. Inventions that follow the inventive concept of Extru-seq disclosed in the present application and are developed after the filing date of the present application are included in the scope of the present application. Furthermore, Extru-seq can be used in combination with recently developed tools such as CAST-seq (see Turchiano G, Andrieux G, Klermund J, Blattner G, Pennucci V, El Gaz M, Monaco G, Poddar S, Mussolino C, Cornu TI et al: Quantitative evaluation of chromosomal rearrangements in gene-edited human stem cells by CAST-Seq. Cell Stem Cell 2021, 28(6):1136-1147 e1135.) to detect Cas9-mediated large-scale deletions, chromosomal losses and translocations.

[0560] Below, we disclose some of the references referred to in this specification, which may or may not be mentioned in the paragraph related to the corresponding reference.

[0561] References 1. Mullard A:Gene-editing pipeline takes off. Nat Rev Drug Discov 2020、19(6):367-372. 2. Tsai SQ、Zheng Z、Nguyen NT、Liebers M、Topkar VV、Thapar V、Wyvekens N、Khayter C、Iafrate AJ、Le LP et al:GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 2015、33(2):187-197. 3. Liang SQ、Liu P、Smith JL、Mintzer E、Maitland S、Dong X、Yang Q、Lee J、Haynes CM、Zhu LJ et al:Genome-wide detection of CRISPR editing in vivo using GUIDE-tag. Nat Commun 2022、13(1):437. 4. Wienert B、Wyman SK、Richardson CD、Yeh CD、Akcakaya P、Porritt MJ、Morlock M、Vu JT、Kazane KR、Watry HL et al:Unbiased detection of CRISPR off-targets in vivo using DISCOVER-Seq. Science 2019、364(6437):286-289. 5. Yan WX、Mirzazadeh R、Garnerone S、Scott D、Schneider MW、Kallas T、Custodio J、Wernersson E、Li Y、Gao L et al:BLISS is a versatile and quantitative method for genome-wide profiling of DNA double-strand breaks. Nat Commun 2017、8:15058. 6. Crosetto N、Mitra A、Silva MJ、Bienko M、Dojer N、Wang Q、Karaca E、Chiarle R、Skrzypczak M、Ginalski K et al:Nucleotide-resolution DNA double-strand break mapping by next-generation sequencing. Nat Methods 2013、10(4):361-365. 7. Wang X、Wang Y、Wu X、Wang J、Qiu Z、Chang T、Huang H、Lin RJ、Yee JK:Unbiased detection of off-target cleavage by CRISPR-Cas9 and TALENs using integrase-defective lentiviral vectors. Nat Biotechnol 2015、33(2):175-178. 8. Chiarle R、Zhang Y、Frock RL、Lewis SM、Molinie B、Ho YJ、Myers DR、Choi VW、Compagno M、Malkin DJ et al:Genome-wide translocation sequencing reveals mechanisms of chromosome breaks and rearrangements in B cells. Cell 2011、147(1):107-119. 9. Petri K、Kim DY、Sasaki KE、Canver MC、Wang X、Shah H、Lee H、Horng JE、Clement K、Iyer S et al:Global-scale CRISPR gene editor specificity profiling by ONE-seq identifies population-specific、variant off-target effects. bioRxiv 2021:2021.2004.2005.438458. 10. Kim HS、Hwang GH、Lee HK、Bae T、Park SH、Kim YJ、Lee S、Park JH、Bae S、Hur JK:CReVIS-Seq:A highly accurate and multiplexable method for genome-wide mapping of lentiviral integration sites. Mol Ther Methods Clin Dev 2021、20:792-800. 11. Breton C、Clark PM、Wang L、Greig JA、Wilson JM:ITR-Seq、a next-generation sequencing assay、identifies genome-wide DNA editing sites in vivo following adeno-associated viral vector-mediated genome editing. BMC Genomics 2020、21(1):239. 12. Huang H、Hu Y、Huang G、Ma S、Feng J、Wang D、Lin Y、Zhou J、Rong Z:Tag-seq:a convenient and scalable method for genome-wide specificity assessment of CRISPR / Cas nucleases. Commun Biol 2021、4(1):830. 13. Kim D、Bae S、Park J、Kim E、Kim S、Yu HR、Hwang J、Kim JI、Kim JS:Digenome-seq:genome-wide profiling of CRISPR-Cas9 off-target effects in human cells. Nat Methods 2015、12(3):237-243、231 p following 243. 14. Kim D、Kim JS:DIG-seq:a genome-wide CRISPR off-target profiling method using chromatin DNA. Genome Res 2018、28(12):1894-1900. 15. Cameron P、Fuller CK、Donohoue PD、Jones BN、Thompson MS、Carter MM、Gradia S、Vidal B、Garner E、Slorach EM et al:Mapping the genomic landscape of CRISPR-Cas9 cleavage. Nat Methods 2017、14(6):600-606. 16. Tsai SQ、Nguyen NT、Malagon-Lopez J、Topkar VV、Aryee MJ、Joung JK:CIRCLE-seq:a highly sensitive in vitro screen for genome-wide CRISPR-Cas9 nuclease off-targets. Nat Methods 2017、14(6):607-614.17. Lazzarotto CR、Malinin NL、Li Y、Zhang R、Yang Y、Lee G、Cowley E、He Y、Lan X、Jividen K et al:CHANGE-seq reveals genetic and epigenetic effects on CRISPR-Cas9 genome-wide activity. Nat Biotechnol 2020、38(11):1317-1327. 18. Bae S、Park J、Kim JS:Cas-OFFinder:a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics 2014、30(10):1473-1475. 19. Montague TG、Cruz JM、Gagnon JA、Church GM、Valen E:CHOPCHOP:a CRISPR / Cas9 and TALEN web tool for genome editing. Nucleic Acids Res 2014、42(Web Server issue):W401-407. 20. Concordet JP、Haeussler M:CRISPOR:intuitive guide selection for CRISPR / Cas9 genome editing experiments and screens. Nucleic Acids Res 2018、46(W1):W242-W245. 21. Shapiro J、Iancu O、Jacobi AM、McNeill MS、Turk R、Rettig GR、Amit I、Tovin-Recht A、Yakhini Z、Behlke MA et al:Increasing CRISPR Efficiency and Measuring Its Specificity in HSPCs Using a Clinically Relevant System. Mol Ther Methods Clin Dev 2020、17:1097-1107. 22. Gillmore JD、Gane E、Taubel J、Kao J、Fontana M、Maitland ML、Seitzer J、O'Connell D、Walsh KR、Wood K et al:CRISPR-Cas9 In Vivo Gene Editing for Transthyretin Amyloidosis. N Engl J Med 2021、385(6):493-502. 23. Maeder ML、Stefanidakis M、Wilson CJ、Baral R、Barrera LA、Bounoutas GS、Bumcrot D、Chao H、Ciulla DM、DaSilva JA et al:Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10. Nat Med 2019、25(2):229-233. 24. Poirot L、Philip B、Schiffer-Mannioui C、Le Clerre D、Chion-Sotinel I、Derniame S、Potrel P、Bas C、Lemaire L、Galetto R et al:Multiplex Genome-Edited T-cell Manufacturing Platform for "Off-the-Shelf"Adoptive T-cell Immunotherapies. Cancer Res 2015、75(18):3853-3864. 25. MacLeod DT、Antony J、Martin AJ、Moser RJ、Hekele A、Wetzel KJ、Brown AE、Triggiano MA、Hux JA、Pham CD et al:Integration of a CD19 CAR into the TCR Alpha Chain Locus Streamlines Production of Allogeneic Gene-Edited CAR T Cells. Mol Ther 2017、25(4):949-961. 26. Stadtmauer EA、Fraietta JA、Davis MM、Cohen AD、Weber KL、Lancaster E、Mangan PA、Kulikovskaya I、Gupta M、Chen F et al:CRISPR-engineered T cells in patients with refractory cancer. Science 2020、367(6481). 27. Goh WJ、Zou S、Ong WY、Torta F、Alexandra AF、Schiffelers RM、Storm G、Wang JW、Czarny B、Pastorin G:Bioinspired Cell-Derived Nanovesicles versus Exosomes as Drug Delivery Systems:a Cost-Effective Alternative. Sci Rep 2017、7(1):14322. 28. Kim D、Kim S、Park J、Kim JS:Genome-wide target specificities of CRISPR-Cas9 nucleases revealed by multiplex Digenome-seq. Genome Res 2016、26(3):406-415. 29. Chu VT、Weber T、Wefers B、Wurst W、Sander S、Rajewsky K、Kuhn R:Increasing the efficiency of homology-directed repair for CRISPR-Cas9-induced precise gene editing in mammalian cells. Nat Biotechnol 2015、33(5):543-548. 30. Akcakaya P、Bobbin ML、Guo JA、Malagon-Lopez J、Clement K、Garcia SP、Fellows MD、Porritt MJ、Firth MA、Carreras A et al:In vivo CRISPR editing with no detectable genome-wide off-target mutations. Nature 2018、561(7723):416-419. 31. Liu Q、Cheng X、Liu G、Li B、Liu X:Deep learning improves the ability of sgRNA off-target propensity prediction. BMC Bioinformatics 2020、21(1):51. 32. Doench JG、Fusi N、Sullender M、Hegde M、Vaimberg EW、Donovan KF、Smith I、Tothova Z、Wilen C、Orchard R et al:Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat Biotechnol 2016、34(2):184-191. 33. Mundry R、Fischer J:Use of statistical programs for nonparametric tests of small samples often leads to incorrect P values:examples from animal behaviour. . Animal Behaviour 1998、56:256-259. 34. Dwivedi AK、Mallawaarachchi I、Alvarado LA:Analysis of small sample size studies using nonparametric bootstrap test with pooled resampling method. Stat Med 2017、36(14):2187-2205. 35. Frangoul H、Altshuler D、Cappellini MD、Chen YS、Domm J、Eustace BK、Foell J、de la Fuente J、Grupp S、Handgretinger R et al:CRISPR-Cas9 Gene Editing for Sickle Cell Disease and beta-Thalassemia. N Engl J Med 2020、384(3):252-260. 36. Turchiano G、Andrieux G、Klermund J、Blattner G、Pennucci V、El Gaz M、Monaco G、Poddar S、Mussolino C、Cornu TI et al:Quantitative evaluation of chromosomal rearrangements in gene-edited human stem cells by CAST-Seq. Cell Stem Cell 2021、28(6):1136-1147 e1135. 37. Park J、Bae S、Kim JS:Cas-Designer:a web-based tool for choice of CRISPR-Cas9 target sites. Bioinformatics 2015、31(24):4014-4016. 38. Cho SW、Kim S、Kim JM、Kim JS:Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease. Nat Biotechnol 2013、31(3):230-232. 39. Kim E、Koo T、Park SW、Kim D、Kim K、Cho HY、Song DW、Lee KJ、Jung MH、Kim S et al:In vivo genome editing with a small Cas9 orthologue derived from Campylobacter jejuni. Nat Commun 2017、8:14500. 40. Kim D、Kang BC、Kim JS:Identifying genome-wide off-target sites of CRISPR RNA-guided nucleases and deaminases with Digenome-seq. Nat Protoc 2021、16(2):1170-1192. 41. Park J、Childs L、Kim D、Hwang GH、Kim S、Kim ST、Kim JS、Bae S:Digenome-seq web tool for profiling CRISPR specificity. Nat Methods 2017、14(6):548-549. 42. DiGiusto DL、Cannon PM、Holmes MC、Li L、Rao A、Wang J、Lee G、Gregory PD、Kim KA、Hayward SB et al:Preclinical development and qualification of ZFN-mediated CCR5 disruption in human hematopoietic stem / progenitor cells. Mol Ther Methods Clin Dev 2016、3:16067. 43. A Safety and Efficacy Study Evaluating CTX110 in Subjects With Relapsed or Refractory B-Cell Malignancies(CARBON). https: / / clinicaltrials.gov / ct2 / show / NCT04035434. Accessed 15 Dec 2022. 44. Safety、Tolerability、and PK of LBP-EC01 in Patients With Lower Urinary Tract Colonization Caused by E. Coli. https: / / clinicaltrials.gov / ct2 / show / NCT04191148. Accessed 15 Dec 2022. 45. Miller JC、Paschon D、Rebar EJ:METHODS AND COMPOSITIONS FOR TREATING HEMOPHILIA. World Intellectual Property Organization 2015、WO:2015 / 089046. 46. Kwon J、Kim M、Lee J:Extru-seq:A method for predicting genome-wide off-target sites with high sensitivity. NCBI Bioproject 2022、PRJNA796642. https: / / www.ncbi.nlm.nih.gov / bioproject / ?term=PRJNA796642

Claims

1. A method for identifying information about off-targets generated in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) preparing a starting composition comprising a Cas protein, a guide RNA, and a cell; (ii) obtaining a test composition through physical disruption of the cells; wherein, through said physical disruption of the cells, genomic DNA is contacted with a Cas / gRNA complex formed by said Cas protein and said guide RNA, thereby cleaving said genomic DNA at one or more cleavage sites; and (iii) analyzing the test composition to obtain information about the one or more cleavage sites. A method comprising:

2. 2. The method of claim 1, wherein the physically disrupting the cells comprises passing the cells through a filter having pores, the average diameter of the pores of the filter being smaller than the size of the cells.

3. The method of claim 2 , wherein the force that causes the cells to pass through the filter is pressure.

4. The method of claim 2 , wherein the average diameter of the pores of the filter is between 5 and 15 μm.

5. 10. The method of claim 1, wherein the physically disrupting the cells is accomplished through the use of an extruder equipped with a filter with holes.

6. The method of claim 5 , wherein the average diameter of the pores of the filter contained in the extruder is smaller than the size of the cells.

7. The method of claim 5, wherein the average diameter of the pores of the filter is between 5 and 15 μm.

8. The information about the cleavage site may include genomic DNA location of each cleavage site relative to said one or more cleavage sites; a cleavage score for each cleavage site for the one or more cleavage sites; and Number of cleavage sites The method of claim 1 , comprising one or more of:

9. The method comprises: (iv) identifying information about potential off-targets from the information about the cleavage sites obtained from (iii); The method of claim 1 further comprising:

10. The information about the off-target candidates may include: genomic DNA location of each off-target candidate relative to one or more off-target candidates; an off-target prediction score for each off-target candidate for one or more off-target candidates; and Number of predicted off-target candidates The method of claim 9 , comprising one or more of:

11. 10. The method of claim 1, wherein the step of analyzing the test composition comprises analyzing the cleaved genomic DNA contained in the test composition through sequencing.

12. 10. The method of claim 1, wherein the step of analyzing the test composition comprises analyzing cleaved genomic DNA contained in the test composition via a PCR-based method.

13. 2. The method of claim 1, wherein said physically disrupting the cells disrupts membranous structures including cell membranes, thereby creating an environment in which the Cas / gRNA complex can contact the genomic DNA from the cells.

14. 2. The method of claim 1, wherein said physically disrupting the cells disrupts membrane structures including the nuclear membrane of the cells, thereby creating an environment in which the Cas / gRNA complex can contact the genomic DNA from the cells.

15. The method comprises: Identifying a predetermined CRISPR / Cas genome editing system, wherein the step of identifying the predetermined CRISPR / Cas genome editing system is performed prior to (i). The method of claim 1 further comprising:

16. The predetermined CRISPR / Cas genome editing system comprises the use of a predetermined guide RNA having a predetermined guide sequence, The predetermined guide sequence and the guide sequence of the guide RNA are the same. The method of claim 15.

17. The predetermined CRISPR / Cas genome editing system includes the use of a predetermined cell, and the predetermined cell and the cell are the same. The method of claim 15.

18. The test composition comprises cleaved genomic DNA from physically disrupted cells, the genomic DNA being cleaved by the Cas / gRNA complex. The method of claim 1.

19. The concentration of the Cas protein contained in the starting composition is 4000 nM to 6000 nM. The method of claim 1.

20. The concentration of the guide RNA contained in the starting composition is 4000 nM to 6000 nM. The method of claim 1.

21. The concentration of the Cas / gRNA complex contained in the starting composition is 4000 nM to 6000 nM. The method of claim 1.

22. The concentration of the cells contained in the starting composition is 1×10 ７ The method of claim 1, wherein the concentration is cells / mL.

23. The step of obtaining the test composition further comprises a step of incubating the composition obtained through the disruption of the cells.

23. The method of any one of claims 1 to 22.

24. The step of obtaining the test composition comprises: removing RNA from the composition obtained through said disrupting said cells. Further having 23. The method of any one of claims 1 to 22.

25. The step of obtaining the test composition comprises: purifying DNA from the composition obtained through said disrupting said cells. Further having 23. The method of any one of claims 1 to 22.

26. A method for identifying information about off-targets generated in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) charging a starting composition comprising a Cas protein, a guide RNA, and cells into a first vessel of an extruder; (ii) using the extruder, applying pressure to the first vessel to move the components of the starting composition from the first vessel of the extruder to the second vessel of the extruder; wherein the components of the starting composition are forced by the applied pressure through a filter with holes located between the first container and the second container, thereby filling the second container with the mixture; Here, the cells, which are components larger in size than the diameter of the pores of the filter, are broken by the applied pressure and pass through the pores of the filter, wherein an environment is created in which genomic DNA is accessible to the Cas protein and the guide RNA through physical disruption of the cells; This brings the genomic DNA into contact with the Cas / gRNA complex, This causes the genomic DNA to be cleaved at one or more cleavage sites. obtaining the test composition; and (iii) analyzing the test composition to obtain information on the cleavage site. A method comprising:

27. 27. The method of claim 26, wherein the pressure applied to the first container is generated through a process of pushing a piston designed to apply the pressure to the first container in a direction toward the first container and the filter.

28. A method for identifying information about off-targets generated in a genome editing process using a CRISPR / Cas genome editing system, comprising: (i) charging a starting composition comprising a Cas protein, a guide RNA, and cells into a first vessel of an extruder; (ii) using the extruder, (a) applying pressure to the first vessel to move components of the starting composition from a first vessel of the extruder to a second vessel of the extruder; wherein the components of the starting composition are forced through a filter with holes located between the first container and the second container by the applied pressure, thereby filling the second container with the mixture. (b) applying pressure to the second vessel to transfer components of the mixture from the second vessel to the first vessel. wherein the components of the mixture are passed through the filter with holes located between the first container and the second container by the applied pressure, whereby the mixture displaced from the second container by passing through the filter is filled into the first container; and (c) repeating the processes of (a) and (b) a predetermined number of times; wherein the predetermined number of times is counted in increments of 0.5, where 0.5 represents execution of a single process of (a) or (b); Here, the cells, which are components larger in size than the diameter of the pores of the filter, are broken by the applied pressure and pass through the pores of the filter, wherein an environment is created in which genomic DNA is accessible to the Cas protein and the guide RNA through physical disruption of the cells; This brings the genomic DNA into contact with the Cas / gRNA complex, This causes the genomic DNA to be cleaved at one or more cleavage sites. obtaining the test composition; and (iii) analyzing the test composition to obtain information on the cleavage site. A method comprising:

29. the pressure applied to the first container is generated through a process of pushing a piston designed to apply the pressure to the first container in a direction toward the first container and the filter; The pressure applied to the second container is generated through the process of pushing a piston designed to apply pressure to the second container in a direction toward the second container and the filter.

29. The method of claim 28.

Images