Identification, characterization, and quantitation of crispr-introduced double-stranded DNA break repairs

Abstract

To provide a system and process for identifying and characterizing double-stranded DNA break repair sites that are based on biological information, and have improved accuracy.SOLUTION: A process is configured to: receive sample sequence data including a plurality of sequences; analyze and merge the sample sequence data; when a single-stranded or double-stranded DNA oligonucleotide donor is provided, output a target site sequence including a prediction result of a repair event; bine the merged sequence with the target site sequence or an arbitrary target prediction result; output a target-read realignment; realign a binned target-read realignment with a target site; generate a final alignment; analyze the final alignment; identify and quantify mutations within a pre-defined sequence distance window deriving from a canonical enzyme specific cut site; and output the final alignment, and data on results of the analysis and quantification as tables or graphics.SELECTED DRAWING: None

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Applications Nos. 62 / 870,426 and 67 / 870,471, both filed on July 3, 2019, and 62 / 952,603 ​​and 62 / 952,598, both filed on December 23, 2019, the contents of which are incorporated by reference herein in their entireties.

[0002] Described herein is a system and process for identifying and characterizing double-stranded DNA break repair sites with improved accuracy based on biological information.Also described is a sequence alignment process that uses biological data to inform the alignment matrix for position-specific alignment scoring, thereby resulting in accurate identification of non-standard target sites. [Background technology]

[0003] Genome editing has been transformed by the use of targeted nucleases such as CRISPR proteins. CRISPR enzymes form ribonucleoproteins (RNPs) when hybridized with either the two-part crRNA and tracrRNA or a single guide RNA (sgRNA). With either approach, a short protospacer sequence (guide RNA or "gRNA") targets a specific sequence in a complementary molecule. Upon finding a match, these enzymes introduce a break in one or both DNA (or RNA) strands. CRISPR enzymes targeting DNA (e.g., Cas9, Cas12a / Cpf1) introduce double-strand breaks (DSBs) at the predicted genomic location relative to the gRNA hybridization target. DNA DSBs are repaired by intracellular machinery, but the repair process often results in insertions and deletions (indels), substitutions, and other suboptimal allelic variants.

[0004] Because each cell in an affected population must repair itself, independent of neighboring cells, and a particular outcome may contain different resulting alleles, a population of cells may contain multiple alleles at the targeted location. In addition, the targeting capacity of these nucleases is often somewhat non-specific, which results in unwanted mutations at other off-target genomic locations.

[0005] Characterizing and quantifying multiple alleles at both on-target and off-target locations is highly desirable. Researchers often use DNA sequencing (e.g., Illumina's next-generation sequencing; NGS) to observe the resulting allelic diversity. Multiplexed polymerase chain reaction (PCR) can be performed to amplify and enrich all targeted locations. The resulting amplicons can be sequenced. Multiple alleles can be characterized and counted using specialized software.

[0006] A number of specialized software tools have been developed to characterize allelic variants resulting from DSBs. Earlier tools include CRISPResso [1], crispRvariants [2], and Amplican [3]. These tools generally work by aligning each sequence read to the expected amplicon target using Needleman-Wunsch, bwa, or bespoke alignment algorithms. The algorithms align possible reads: A list of target alignments is generated. Each alignment is scored based on the number of nucleotide matches, mismatches and deletions (gaps). The best scoring alignment is used for downstream data processing.

[0007] An alignment algorithm may create equally weighted query:target alignments, which is most likely to occur when the query contains an insertion or deletion. From equally weighted alternatives, the alignment method will either return all or select one alternative. When selecting, some methods make the selection randomly. Without a good predictive model or set of heuristic rules for making the selection, the selection of alignments is variable, which may result in incorrect indel annotations and lower accuracy results.

[0008] There is a need for biologically informed algorithms and processes for identifying and characterizing double-stranded DNA break repair sites with improved accuracy. Summary of the Invention

[0009] One embodiment described herein is a computer-implemented process for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising the steps of receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data, and outputting merged sequences; when single-stranded or double-stranded DNA oligonucleotide donors are provided, developing target site sequences that contain predicted results of repair events, and outputting target prediction results; using a mapper to bin the merged sequences by target site sequences or any target prediction results, and outputting target read alignment; realigning the binned target read alignment with the target site using an enzyme-specific position-specific scoring matrix derived from biological data that is applied based on the guide sequence and the position of the standard enzyme-specific cleavage site, and generating a final alignment; analyzing the final alignment, and identifying and quantifying the mutations within a predetermined sequence distance window from the standard enzyme-specific cleavage site; outputting the data of the final alignment, analysis and quantification results as a table or graphic.In one aspect, the sequence data comprises sequences from a population of cells or a subject. In another embodiment, the enzyme-specific cleavage site comprises one or more of Cas9, Cas12a or other Cas enzymes. In another embodiment, the predetermined sequence distance window is enzyme-specific and comprises 1 nt to about 15 nt. In another embodiment, the results show percent editing, percent insertion, percent deletion, or a combination thereof. In another embodiment, the accuracy of identifying variant target sites is improved by about 15 to about 20% compared to a comparable process.

[0010] Another embodiment described herein is a computer-implemented process for aligning biological sequences, comprising the steps of receiving sample sequence data comprising a plurality of sequences; aligning the sequence data with predicted target sequences using a matrix based on enzyme-specific position-specific scoring of specific nuclease target sites; and outputting the alignment results as a table or graphic, on a processor. In one aspect, the sequence data comprises sequences from a population of cells or a subject. In another aspect, the specific nuclease target sequences comprise target sites for one or more of Cas9, Cas12a or other Cas enzymes. In another aspect, the matrix uses position-specific gap opening and extension penalties.

[0011] Another embodiment described herein is a method for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising extracting genomic DNA from a population of cells or tissue from a subject; amplifying the genomic DNA using multiplex PCR to identify target site sequences. The method includes the steps of: generating enriched amplicons for a sequence; sequencing the amplicons to obtain sample sequence data; then receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data, and outputting merged sequences; developing target site sequences containing predicted results of repair events when single-stranded or double-stranded DNA oligonucleotide donors are provided, and outputting target prediction results; using a mapper to bin the merged sequences by target site sequences or any target prediction results, and outputting target read alignments; realigning the binned target read alignments with target sites using an enzyme-specific position-specific scoring matrix derived from biological data that is applied based on the guide sequence and the position of the standard enzyme-specific cleavage site, and generating a final alignment; analyzing the final alignment, and identifying and quantifying mutations within a predetermined sequence distance window from the standard enzyme-specific cleavage site; outputting the data of the final alignment, analysis and quantification results as a table or graphic.In one embodiment, the enzyme-specific cleavage site comprises one or more of Cas9, Cas12a or other Cas enzymes. In another embodiment, the predetermined sequence distance window is enzyme specific and comprises 1 nt to about 15 nt. In another embodiment, the results show percent editing, percent insertion, percent deletion, or a combination thereof. In another embodiment, the accuracy of identifying variant target sites is improved by about 15 to about 20% compared to a comparable process.

[0012] The patent or application contains at least one drawing executed in color. Copies of this application publication or patent with color drawing(s) will be provided by the Office upon request and payment of the necessary fee. [Brief description of the drawings]

[0013] [Figure 1]Overall workflow for CRISPAltRations. Edited genomic DNA is extracted and amplified using targeted multiplex PCR to enrich for on-target and predicted off-target loci. Amplicons are sequenced on an Illumina MiSeq. Read pairs are merged into a single fragment (FLASH), mapped to the genome (minimap2), and binned by their alignment to predict amplicon locations. Reads in each bin are realigned with the predicted amplicon sequence after finding the cut site and generating a position-specific gap opening / extension bonus matrix to preferentially align indels closer to the cut site / predicted indel profile for each enzyme (CRISPAltRations code +psnw). Indels that intersected the window upstream or downstream of the cut site were annotated. Edit percentage is the sum of reads containing indels / total observed. [Diagram 2] Directed acyclic image of the CRISPRAltRations pipeline. The dashed boxes represent steps in the pipeline, each of which may include one or more software tools. The lines and arrows indicate the flow of information through the pipeline. Two important steps are that minimap2_orig_reads (orange) aligns the sequence reads to the reference genome using minimap2 [4], which is an optional step. minimap2 is then used to align the sequence reads to their predicted target regions. The CRISPy Python tool (in-house development) realigns the sequence reads to their target regions by calling a custom modified version of psnw, which performs cut site realignment to the target regions. CRISPy is also involved in characterizing the detected indels in the aligned reads. [Diagram 3]MinION sequencing read data aligned to the reference target demonstrates blunt insertions. Drops in the expected range (grey highlighting) indicate large insertions. Mismatches at the inner edge of the grey highlighting indicate mismatches between the observed read data and the expected reference. [Figure 4] Examples of schematic diagrams depicting the types of structural variants that can occur when using DNA oligos as templates for homology-directed repair (HDR). These examples are not exhaustive. In the schematic diagrams, blue represents the reference sequence, green represents the homology arms, and orange represents the desired insertion sequence. (A) and (B) are examples of double-stranded and single-stranded template oligos, respectively. (1) [Full repair] The region containing the DSB is repaired without any structural variants being introduced, even in the presence of single-stranded or double-stranded template oligos. (2) [HDR-mediated repair] The template oligo directs HDR, resulting in the desired insertion. Only the desired insertion sequence is observed in the repaired DNA. (3) [Non-homologous end joining (NHEJ) repair] The template oligo is inserted blunt-ended after the DSB. (4) [NHEJ repair with overlapping insertion] The template oligo is inserted blunt-ended multiple times after the DSB. Examples 3 and 4 most likely occur due to the double-stranded oligo being used as a repair template, where the homology arms present in the donor sequence are also inserted into the genome. [Diagram 5] Location of occurrence of editing events in Cas9 data. Location of indel occurrence from nuclease cleavage site using a set of 274 Cas9 to guide editing unique genomic targets in Jurkat cell lines. Both (A) deletion and (B) insertion location events were normalized by quantifying them as a total % of the respective indel events in the sample. Only sites with >50 reads and >5% indels were used for analysis to limit low confidence signals and remove noise. Indels were quantified within a 20 bp window from the cleavage site. Outliers were found to be primarily sites with non-reference indels present in Jurkat cell lines that were not likely caused by DSB activity. [Figure 6] Location of occurrence of editing events in Cas12a data. Location of indel occurrence from nuclease cleavage sites using a set of 199 Cas12a to guide editing unique genomic targets in Jurkat cell lines. Both (A) deletion and (B) insertion location events were normalized by quantifying them as a total % of the respective indel events in the sample. Only sites with >50 reads and >5% indels were used for analysis to limit the noise of low confidence signals. Indels were quantified within a 20 bp window from the cleavage site. Outliers were found to be primarily sites with non-reference indels present in the Jurkat cell line that were not likely caused by DSB activity. [Figure 7] Vector of gap opening / extension penalties near nuclease-induced DSBs. To positively weight indels near the cut site in a sequence alignment, we use a position-specific matrix of values ​​representing variable gap opening or extension penalties, where the length of the array is equal to the integer positions of each nucleic acid in the target (bold blue line). The vector values ​​(red circles and blue diamonds) are varied based on the proximity to the nuclease cut site (vertical dashed black line). Thus, the indels closest to the cut site have the smallest gap opening or extension penalty. [Figure 8] Selection of optimal window for observing true CRISPR-Cas9 editing events. (A) Editing window is the distance of nucleotides around the CRISPR cut site used to identify indels. True and false editing events were calculated as % indels from targeted sequencing of treated or untreated cells for (B) Cas9 (n=263 pairs of treated vs. control) or (C) Cas12a (n=384 pairs of treated vs. control), respectively. Most true editing events are collected in a 4nt window for Cas9 and a 7nt window for Cas12a, but extending the window further collects only additional false edits. [Figure 9]Screenshot of de-duplicated read alignment by observed frequency. Total region coverage is indicated by the height of the vertical grey bar (top) and reads are horizontal coloured bars. Lighter coloured reads indicate indels observed at higher frequencies. Thin horizontal lines indicate deletions, vertical purple "I" symbols indicate insertions and coloured bars within reads indicate mismatched bases. [Figure 10] CRISPAltRations reliably finds the correct indels. Each bar reports the percentage of correctly returned indels. Synthetically generated data was used representing 12,060 unique indel events, evenly distributed among each size bin (x-axis) and 603 unique amplicons. Error bars represent 95% confidence intervals across targets. [Figure 11] CRISPRAltRations more accurately reports the % total edits in 603 synthetic targets by simulated (A) Cas9 or (B) Cas12a editing. Dots represent the percent edits observed at each target site, and horizontal lines represent the percent of indels introduced at each target in the generated synthetic data. CRISPResso2Pooled and CRISPResso1Pooled were run using default pipeline parameters for each panel, with the minimum required read depth for quantification removed (to prevent spurious dropout of amplicons). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] One embodiment described herein is an analysis pipeline called CRISPAltRations. See Figures 1-2. Briefly, this pipeline takes in FASTQ files and uses FLASH to build a merged R1 / R2 consensus. At the same time, a target site reference is built that describes the sequence for all expected on-target locations. Optionally, a target is built that contains the expected outcome of a homology-directed repair (HDR) event. The merged sequence reads are then aligned with the target reference sequence by using minimap2 (originally developed for rapid alignment of long reads (e.g., those created by Oxford Nanopore Technologies MinION). The reads aligned with each target are then realigned by using a modified version of the Smith-Waterman aligner. The modified aligner can improve the detection of insertions and deletions resulting from DSB repair. All observed variants within a given distance of the DSB location are characterized and quantified. Finally, the results are summarized in tables and graphs. The various described programs, tools and file types (and those listed below) are well known and readily accessible to those skilled in the art. It should be understood that these programs, tools and file types are exemplary and are not intended to be limiting. Other tools and file types could be used to implement the described processes and analyses.

[0015] In this analysis pipeline, the following improvements over previous methods are described. First, the use of minimap2 [4] allows for the alignment of reads created from both short and long read sequences. Second, by constructing the expected outcome of homologous recombination repair events, the ability to fully characterize HDR events (i.e., occur correctly) is improved. Third, the use of a modified Needleman-Wunsch aligner that can accept Cas-specific bonus matrices allows for significantly improved characterization of indels and quantification of editing percentages (%) over previous methods. Fourth, the graphical visualization of introduced allelic variants is improved. Fifth, the predicted repair events described in the previous tool [5] can be compared against observed repairs, and the molecular pathways involved in the repair can be described.

[0016] In one embodiment, the process described herein has the following advantageous uses: Accurate characterization of indel profiles arising from DSBs. The percentage of reads that contain indels after the DSB has been repaired is used to calculate the percentage of editing. This metric (% edited) is used to determine the efficacy of gRNAs for use in CRISPR-Cas gene editing. Accurate characterization of the resulting indels similarly improves the ability to identify the percentage of chromosomes of cells in a population of cells that contain frameshift mutations that alter the protein encoded by the affected gene. Accurate characterization of the insertion sequence. Accurate characterization of multiple mutations resulting from delivery of multiple gRNA / Cas9 (i.e., ribonucleoprotein complexes) or dual guide region modifications. Analysis of indels sequenced on long-read platforms such as MinION. In addition, this allows for step-by-step characterization of both ends of large (>400 nt) insertions of deletion events that occur after DSB repair. Improved results visualization.

[0017] One embodiment described herein is a computer-implemented process for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising the steps of receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data and outputting the merged sequence; developing a target site sequence containing a predicted outcome of a repair event when a single-stranded or double-stranded DNA oligonucleotide donor is provided and outputting a target prediction result; binning the merged sequence by the target site sequence or any target prediction result using a mapper and outputting a target read alignment; realigning the binned target read alignment with the target site using an enzyme-specific position-specific scoring matrix derived from biological data applied based on the guide sequence and the position of the standard enzyme-specific cleavage site to generate a final alignment; analyzing the final alignment and identifying and quantifying mutations within a predetermined sequence distance window from the standard enzyme-specific cleavage site; and outputting the data resulting from the final alignment, analysis and quantification as a table or graphic.

[0018] In one embodiment, edited genomic DNA is extracted and amplified using targeted multiplex PCR to enrich for on-target loci and predicted off-target loci. Amplicons are sequenced on an Illumina MiSeq. Read pairs are merged into a single fragment (FLASH), mapped to the genome (minimap2), and binned by their alignment to predict the location of the amplicon. The reads in each bin are realigned with the predicted amplicon sequence after finding the cut site and creating a position-specific gap opening / extension bonus matrix to preferentially align indels closer to the cut site / predicted indel profile for each enzyme (CRISPAltRations code +psnw). Indels that intersected with the window upstream or downstream of the cut site were annotated. Editing percentage is the sum of reads containing indels / total observed.

[0019] In some embodiments, the process described herein uses minimap2 [4], which allows for the alignment of reads generated from both short and long read sequences. Previous tools typically only accept short read sequencing data, such as data generated by Illumina sequencers. Others have used long read sequencing data to examine large insertions or deletions [6-8], but no tools are believed to be publicly available on their own. Handling long read data is possible, in part, through the use of the minimap2 aligner. For example, alignment results can be visualized, which shows the identification of insertions of blunt-ended molecules in DNA following DSB repair. (Figure 3). Another embodiment sorts between real indels and noise-derived indels using a method similar to a previously published tool [7], in which small indels are ignored.

[0020] In another embodiment, the ability to characterize complete HDR events is improved by constructing expected outcomes of HDR events. The reference file contains each expected sequence target, and also the modified sequence target, in FASTA format. The first step aimed at constructing this file includes creating a reference sequence index that allows the reads to align with each expected structural variant. For example, when investigating the region targeted for DSB and the double-stranded DNA donor oligo that allows HDR, there are several different possible biological repair outcomes: complete repair (Figure 4-1), HDR-mediated repair (Figure 4-2), NHEJ repair (Figure 4-3), and NHEJ repair with overlapping insertion (Figure 4-4). Other outcomes such as template fragments or triple template insertions are also possible (not shown). Similar reference file construction approaches are used by other tools such as UDiTaS™ [9].

[0021] In another embodiment, a modified version of the Needleman-Wunsch algorithm is used to realign the reads to their predicted targets. The method described herein increases the accuracy of alignments containing indels (as annotations in the CIGAR string of the alignment). This significantly improves the characterization of indels and the quantification of edit % over previous methods. DNA sequencing aligners such as minimap2 and the Needleman-Wunsch approach consider alignment of indels using fixed penalties for opening and extending gaps. This method is improved by realigning reads to their targets using position-specific gap opening and extension penalties (possible in a tool called "psnw") so that alignments with indels favor those that overlap or are located close to the predicted DSB. This position-specific matrix is ​​a set that reflects the precisely characterized indel profile of the specific Cas enzyme used to edit (Figures 5-6). Therefore, indel-based alignment is most favorable at or near the predicted target cleavage site (variable scoring strategy; Figure 7). This method allows for accurate realignment of indels, especially those that occur in repetitive regions in the reference sequence. This approach improves the ability to identify the most biologically likely outcomes.

[0022] A recently developed tool (CRISPResso2

[11] ) uses a detailed alignment strategy of the cleavage site. However, the process described herein is adjusted by the actual editing data at the Cas9 / Cas12a site, and performs the detailed alignment method of the cleavage site using a full gap start / extension matrix during the alignment performed in C++. In contrast, CRISPResso2 allows only a single bonus at the cleavage site and uses a method performed in Python.

[0023] In another embodiment, the process described herein collects indels near the nuclease cleavage site, as well as tag indels that cross the cleavage site or within a fixed distance. Some published accounts suggest a fixed distance of 1-2 nt, but data supporting these choices are limited. In the development of the embodiments described herein, the optimal distance (i.e., window size) around the cleavage site was studied using a set of treated Cas9-RNPs and paired untreated control samples. It was observed that a window of 4 nt for Cas9, or a window of 7 nt for Cas12a, provided the highest sensitivity and acceptable specificity (Figure 4). The requirement for a larger window for Cas12a may be due to the mechanism of action, with Ca S12a carries out double-stranded cleavage by generating two single-stranded breaks 5 bp apart (leaving "sticky" ends)

[12] . Thus, the process described here can be extended to other nucleases (e.g., CasX)

[13] that have biological data informing their target window size and enzymatic mechanism of action.

[0024] In another embodiment, the graphical visualization of allelic variants is improved. Downstream of the alignment step, several other analyses specific to the described method are performed. To create improved visualization, reads are de-duplicated based on the identification of indel sequences identified within the CRISPR editing window after alignment. The de-duplicated reads are written back to a BAM file, and the frequency of each de-duplicated read within the original population of reads is written to the associated BAM tag. After the file is indexed, the indels and their associated frequency in the de-duplicated reads can be visualized using the publicly available IGV tool

[10] (Figure 9).

[0025] In another embodiment, predicted repair events as described in previous tools [5] can be compared against observed repair and used to determine the molecular pathways involved in repair. The system described herein also adds the ability to compare observed indel profiles against expected indel profiles, allowing rapid identification of whether an experimental treatment has altered the intracellular machinery of DNA repair.

[0026] The utility of the systems and methods described herein is demonstrated by generating a synthetic set of 603 gRNA:amplicon pairs. For each target, 4000 read pairs (2 × 150 bp) were generated using a simulated Illumina MiSeq sequence analysis. v3 platform error profiles. In half of the reads, random indels are introduced based on models created from observed editing profiles for Cas9 and Cas12a (Figures 4-5). The synthetic data is analyzed using the CRISPRAltRations system described herein, as well as the previously published CRISPResso1 and CRISPResso2 tools

[11] . By implementing the method described herein, the ability to correctly characterize indels is improved by approximately 15-20% (Figure 10). The algorithm described herein has increased accuracy because it provides a biologically informed selection of the best alignment in targets where multiple equivalent scored alignments are possible. In addition, the method described herein more accurately calculates the percentage of modified DNA molecules (Figure 11). The process and strategy described herein are important enhancements to the characterization and quantification of indels introduced after DSB repair.

[0027] Another embodiment described herein is a computer-implemented process for aligning biological sequences, comprising the steps of receiving sample sequence data comprising a plurality of sequences; aligning the sequence data with predicted target sequences using a matrix based on enzyme-specific position-specific scoring of specific nuclease target sites; and outputting the alignment results as a table or graphic, on a processor. In one aspect, the sequence data comprises sequences from a population of cells or a subject. In another aspect, the specific nuclease target sequences comprise target sites for one or more of Cas9, Cas12a or other Cas enzymes. In another aspect, the matrix uses position-specific gap opening and extension penalties.

[0028] Another embodiment described herein is a method for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising: extracting genomic DNA from a population of cells or tissue from a subject; amplifying the genomic DNA using multiplex PCR to generate amplicons enriched for target site sequences; sequencing the amplicons and analyzing the sample sequences. obtaining sequence data; then receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data and outputting the merged sequences; when a single-stranded or double-stranded DNA oligonucleotide donor is provided, developing target site sequences containing predicted outcomes of repair events and outputting target prediction results; using a mapper to bin the merged sequences by the target site sequences or any target prediction results and outputting target read alignments; re-aligning the binned target read alignments with the target sites using an enzyme-specific position-specific scoring matrix derived from biological data that is applied based on the guide sequence and the position of the standard enzyme-specific cleavage site to generate a final alignment; analyzing the final alignment and identifying and quantifying mutations within a predetermined sequence distance window from the standard enzyme-specific cleavage site; and outputting the data resulting from the final alignment, analysis and quantification as a table or graphic.

[0029] Many different arrangements of the various components and processes described herein, as well as components or processes not shown, are possible without departing from the spirit and scope of the present disclosure. It should be understood that an embodiment or aspect may include or otherwise be performed by a combination of various hardware, software, or electronic components. For example, various microprocessors and application specific integrated circuits ("ASICs") may be utilized, as well as software in various languages. Also, servers and various computing devices may be used, including one or more processing units, one or more computer readable media, one or more input / output interfaces, and various connections (e.g., a system bus) connecting the components.

[0030] It will be apparent to those skilled in the art that suitable modifications and adaptations to the compositions, formulations, methods, processes and applications described herein can be made without departing from the scope of any of its embodiments or aspects. The compositions and methods provided are exemplary and are not intended to limit the scope of any particular embodiment. All the various embodiments, aspects and options disclosed herein can be combined with any variations or iterations. The scope of the methods and processes described herein includes all actual or possible combinations of the embodiments, aspects, options, examples and preferred ones described herein. The methods described herein may exclude any component or step, substitute any component or step disclosed herein, or include any component or step disclosed anywhere in the present specification. If the meaning of any term in any of the patents or publications incorporated by reference conflicts with the meaning of the term used in this disclosure, the meaning of the term or expression in this disclosure shall control. Furthermore, this specification discloses and describes merely exemplary embodiments. All patents and publications cited herein are incorporated by reference herein for their specific teachings.

[0031] References 1.Pinello, L. et al., “Analyzing CRISPR genome-editing experiments with CRISPResso.” Nat Biotechnol. 34(7): 695-697 (2016). 2. Lindsay, H. et al., “CrispRVariants: precisely charting the mutation spectrum in genome engineering experiments,” Nat. Biotechnol. 34(7): 701-703 (2015). 3.Labun, K. et al., “Accurate analysis of genuine CRISPR editing events with ampliCan Kornel,” bioRxiv 249474 (2018); now published in Genome Research 29: 843-847 (2019) 4.Li, H., “Minimap2: Pairwise alignment for nucleotide sequences,” Bioin formatics 34(18): 3094-3100 (2018). 5.Shen, M. W. et al., “Predictable and precise template-free CRISPR editing of pathogenic variants,” Nature 563 (7733): 646-651 (2018). 6.Hendel, A. et al., “Quantifying genome-editing outcomes at endogenous loci with SMRT sequencing.” Cell Rep. 7(1): 293-305 (2014). 7.Iyer, S. et al., “Precise therapeutic gene correction by a simple nuclease-induced double-stranded break,” Nature 568 (7753): 561-565 (2019). 8.Vu, G. T. H. et al., “Endogenous sequence patterns predispose the repair modes of CRISPR / Cas9-induced DNA double-stranded breaks in Arabidopsis thaliana,” Plant J. 92(1): 57-67 (2017). 9.Giannoukos, G. et al., “UDiTaSTM , a genome editing detection method for indels and genome rearrangements,” BMC Genomics 19: 212 (2018). 10. Robinson, J., “Integrated genomics viewer,” Nat. Biotechnol. 29(1), 24-26 (2012). 11. Clement, K. et al., “Analysis and comparison of genome editing using CRISPResso2,” bioRxiv 1-20 (2018). Now published in Nat. Biotechnol. 37(3): 224-226 (2019) 12. Zetsche, B. et al., “Cpf1 Is a Single RNA-Guided Endonuclease of a Class 2 CRISPR-Cas System,” Cell 163(3): 759-771 (2015). 13. Liu, JJ et al., “CasX enzymes comprise a distinct family of RNA-guided genome editors,” Nature 566(7743): 218-223 (2019).

[0032] Computer Code Example code used to generate a 1D scoring matrix for gap opening bonuses during alignment using psnw.

number

number

number

[0033] Example code used to generate a 1D scoring matrix for gap extension bonuses during alignment using psnw.

number

Claims

1. 1. A computer-implemented process for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising: receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data and outputting the merged sequences; developing a target site sequence containing the predicted outcome of a repair event when a single-stranded or double-stranded DNA oligonucleotide donor is provided, and outputting the target prediction outcome; binning the merged sequences by target site sequences or any target prediction results using a mapper, and outputting the target read alignments; realigning the binned target read alignments with the target sites using an enzyme-specific, position-specific scoring matrix derived from the biological data that is applied based on the guide sequence and the position of the standard enzyme-specific cleavage site to generate a final alignment; analyzing the final alignment to identify and quantify mutations within a predefined sequence distance window from a standard enzyme-specific cleavage site; Outputting the final alignment, analysis and quantification results data as tables or graphics. on a processor.

2. The process of claim 1 , wherein the sequence data comprises sequences from a population of cells or a subject.

3. 2. The process of claim 1, wherein the enzyme-specific cleavage site comprises one or more of Cas9, Cas12a or other Cas enzymes.

4. 2. The process of claim 1, wherein the predetermined sequence distance window is enzyme-specific and comprises from 1 nt to about 15 nt.

5. 2. The process of claim 1, wherein the results indicate a percentage of editing, a percentage of insertions, a percentage of deletions, or a combination thereof.

6. 10. The process of claim 1, wherein the accuracy of identifying variant target sites is improved by about 15 to about 20% compared to a comparable process.

7. 1. A computer-implemented process for aligning biological sequences, comprising: receiving sample sequence data comprising a plurality of sequences; aligning the sequence data with predicted target sequences using a matrix based on enzyme-specific position-specific scoring of specific nuclease target sites; Steps for outputting alignment results as tables or graphics on a processor.

8. The process of claim 7 , wherein the sequence data comprises sequences from a population of cells or a subject.

9. 8. The process of claim 7, wherein the specific nuclease target sequence comprises a target site for one or more of Cas9, Cas12a or other Cas enzymes.

10. The process of claim 7 , wherein the matrix uses position-specific gap opening and extension penalties.

11. 1. A method for identifying and characterizing double-stranded DNA break repair sites with improved accuracy, comprising: Extracting genomic DNA from a population of cells or tissue from the subject; amplifying genomic DNA using multiplex PCR to generate amplicons enriched for target site sequences; sequencing the amplicons to obtain sample sequence data; after that, receiving sample sequence data comprising a plurality of sequences; analyzing and merging the sample sequence data and outputting the merged sequences; developing a target site sequence containing the predicted outcome of a repair event when a single-stranded or double-stranded DNA oligonucleotide donor is provided, and outputting the target prediction outcome; binning the merged sequences by target site sequences or any target prediction results using a mapper, and outputting the target read alignments; realigning the binned target read alignments with the target sites using an enzyme-specific, position-specific scoring matrix derived from the biological data that is applied based on the guide sequence and the position of the standard enzyme-specific cleavage site to generate a final alignment; analyzing the final alignment to identify and quantify mutations within a predefined sequence distance window from a standard enzyme-specific cleavage site; Outputting the final alignment, analysis and quantification results data as tables or graphics. to run on the processor A method comprising:

12. 2. The process of claim 1, wherein the enzyme-specific cleavage site comprises one or more of Cas9, Cas12a or other Cas enzymes.

13. 2. The process of claim 1, wherein the predetermined sequence distance window is enzyme-specific and comprises from 1 nt to about 15 nt.

14. 2. The process of claim 1, wherein the results indicate a percentage of editing, a percentage of insertions, a percentage of deletions, or a combination thereof.

15. 10. The process of claim 1, wherein the accuracy of identifying variant target sites is improved by about 15 to about 20% compared to a comparable process.

Images