WHOLE GENOME SGRNA LIBRARY CONSTRUCTING SYSTEM AND APPLICATION THEREOF

1 . A system for constructing a genome-wide sgRNA library, comprising:

(1) an input module, which is configured to download genomic sequences and annotation files from a database, and extract a commonly deleted segment (CDS) sequence as an input target sequence;

(2) an sgRNA design module, which is configured to select candidate sgRNAs on a sense strand and an antisense strand of the target sequence according to a set parameter, perform a genome-wide sequence alignment according to a specified number of allowed mismatches, and evaluate off-target rates and grade sgRNAs according to off-target sites and a number of the off-target sites;

wherein 20 nt+NGG is selected as a candidate sgRNA on the sense strand and GGN+20 nt is selected as the candidate sgRNA on the antisense strand;

(3) an sgRNA filtering module, which is configured to screen evaluated and graded sgRNAs according to the following criteria: removing an sgRNA comprising 4 or more consecutive bases, ensuring that sgRNAs have no overlap, and ensuring that the sgRNAs are evenly distributed on a CDS as much as possible.

2 . The system of claim 1 , wherein a selection criterion of the target sequence in step (1) comprises that a CDS region is selected as the target sequence for a protein-encoding gene and an exon region is selected as the target sequence for a non-protein-encoding gene.

3 . The system of claim 1 , wherein the parameter in step (2) comprises a protospacer adjacent motif (PAM) sequence, a sequence length, guanine-cytosine (GC) content, a single/double-strand mode and a number of mismatches allowed in a genome alignment.

4 . The system of claim 1 , wherein the number of allowed mismatches in step (2) is 3 to 6, preferably 5; and

preferably, off-target rate evaluation criteria in step (2) comprise:

(a) filtering out an sgRNA capable of being accurately aligned to a plurality of sites in a genome;

(b) an sgRNA that is only aligned to a position corresponding to the sgRNA in the genome being Best; and

(c) for other sgRNAs, gradually decreasing a penalty point according to a mismatch position of 5′->3′, and comprehensively scoring the other sgRNAs in conjunction with a number of mismatches, wherein a larger penalty point corresponds to a higher risk.

5 . The system of claim 1 , wherein grading levels in step (2) comprise four levels: best, low-risk, moderate-risk and high-risk.

6 . The system of claim 1 , wherein screening criteria in step (3) further comprise any one or a combination of at least two of: selecting at most 6 sgRNAs for each target sequence, reserving only a best sgRNA and a low-risk sgRNA, ensuring that a selected sgRNA covers different transcripts of a gene as much as possible, a plurality of sgRNAs of each gene being targeted to different positions of the each gene as much as possible, and GC content of 20% to 80%, preferably, a combination of selecting at most 6 sgRNAs for each target sequence, reserving only the best sgRNA and the low-risk sgRNA, ensuring that the selected sgRNA covers the different transcripts of the gene as much as possible, the plurality of sgRNAs of each gene being targeted to the different positions of the each gene as much as possible, and the GC content of 20% to 80%.

7 . A method for constructing an sgRNA library by using the system of claim 1 , comprising:

(1) selecting a target sequence: downloading genomic sequences and annotation files from a database, and extracting a commonly deleted segment (CDS) sequence as an input target sequence;

(2) designing sgRNAs: selecting candidate sgRNAs on a sense strand and an antisense strand of the target sequence according to a set parameter, performing a genome-wide sequence alignment according to a specified number of allowed mismatches, and evaluating off-target rates and grading sgRNAs according to off-target sites and a number of the off-target sites;

wherein 20 nt+NGG is selected as a candidate sgRNA on the sense strand and GGN+20 nt is selected as the candidate sgRNA on the antisense strand;

(3) screening the sgRNAs: screening evaluated and graded sgRNAs according to the following criteria: removing an sgRNA comprising 4 or more consecutive bases, ensuring that sgRNAs have no overlap, and ensuring that the sgRNAs are evenly distributed on a CDS as much as possible.

8 . The method of claim 7 , wherein a selection criterion of the target sequence in step (1) comprises that a CDS region is selected as the target sequence for a protein-encoding gene and an exon region is selected as the target sequence for a non-protein-encoding gene;

preferably, the parameter in step (2) comprises a protospacer adjacent motif (PAM) sequence, a sequence length, guanine-cytosine (GC) content, a single/double-strand mode and a number of mismatches allowed in a genome alignment;

preferably, the number of allowed mismatches in step (2) is 3 to 6, preferably 5;

preferably, off-target rate evaluation criteria in step (2) comprise:

(a) filtering out an sgRNA capable of being accurately aligned to a plurality of sites in a genome;

(b) an sgRNA that is only aligned to a position corresponding to the sgRNA in the genome being Best; and

(c) for other sgRNAs, gradually decreasing a penalty point according to a mismatch position of 5′->3′, and comprehensively scoring the other sgRNAs in conjunction with a number of mismatches, wherein a larger penalty point corresponds to a higher risk;

preferably, levels for the grading in step (2) comprise four levels: best, low-risk, moderate-risk and high-risk.

9 . A method for constructing an sgRNA library by using the system of claim 1 , specifically comprising:

(1) selecting a target sequence: downloading genomic sequences and annotation files from a database, and extracting a commonly deleted segment (CDS) sequence as an input target sequence;

wherein a selection criterion of the target sequence in step (1) comprises that a CDS region is selected as the target sequence for a protein-encoding gene and an exon region is selected as the target sequence for a non-protein-encoding gene;

(2) designing sgRNAs: selecting candidate sgRNAs on a sense strand and an antisense strand of the target sequence according to a set parameter comprising a protospacer adjacent motif (PAM) sequence, a sequence length, guanine-cytosine (GC) content, a single/double-strand mode and a number of allowed mismatches, performing a genome-wide sequence alignment according to the number of allowed mismatches, and evaluating off-target rates and grading the sgRNAs as best, low-risk, moderate-risk and high-risk (off-target risk gradients) according to a number of mismatches and a mismatch position;

wherein 20 nt+NGG is selected as a candidate sgRNA on the sense strand and GGN+20 nt is selected as the candidate sgRNA on the antisense strand; and

off-target rate evaluation criteria comprise:

(a) filtering out an sgRNA capable of being accurately aligned to a plurality of sites in a genome;

(b) an sgRNA that is only aligned to a position corresponding to the sgRNA in the genome being Best; and

(c) for other sgRNAs, gradually decreasing a penalty point according to the mismatch position of 5′->3′, and comprehensively scoring the other sgRNAs in conjunction with the number of mismatches, wherein a larger penalty point corresponds to a higher risk;

(3) filtering the sgRNAs: screening evaluated and graded sgRNAs according to the following criteria: removing an sgRNA comprising 4 or more consecutive bases, ensuring that sgRNAs have no overlap, ensuring that the sgRNAs are evenly distributed on a CDS as much as possible, selecting at most 6 sgRNAs for each target sequence, reserving only a best sgRNA and a low-risk sgRNA, ensuring that a selected sgRNA covers different transcripts of a gene as much as possible, a plurality of sgRNAs of each gene being targeted to different positions of the each gene as much as possible, and GC content of 20% to 80%.

10 . A genome-wide sgRNA library constructed according to the method of claim 9 .