A variety of experimental systems are widely being used to investigate the off-target effects of sgRNAs and they show different outcomes regarding the extent of off-targets. Many algorithms have been applied for the prediction of off-target effects and many Cas9 proteins with high specificity have successfully been produced for minimizing Cas9 promiscuity. However, sequencing-based approaches have been widely used to experimentally validate a variety of Cas9 off-targets that are not fully explained by such algorithms, raising an interesting question about mechanisms of binding of Cas9 complexes and cleavage of off-targets. Many tools have been developed and widely used on the basis of mismatch information, including the number and location of the mismatches, finding and evaluating the potential off-target sites. Several groups have used these algorithms for defining potential off-target sites.

Fig. 1 Mechanism of off-target editing.

PEM-seq (Primer-Extension-Mediated Sequencing)

PEM-seq has been developed and widely used to detect off-target effects and determination of specificity and editing efficiency of CRISPR/Cas9. LAM-HTGTS has been combined with targeted sequencing by PEM-seq for the effective analysis of CRISPR/Cas9 induced off-target mutations via translocation capture. PEM-seq can characterize off-target sites and other abnormal chromosomal structures, such as genome-wide translocations, small indels, and large deletions induced by Cas9. Furthermore, PEM-seq has been employed for testing a variety of extensively used methods which were developed for reducing Cas9 off-target activity, whereby PEM-seq can assess comprehensively the specificity and editing efficiency of CRISPR/Cas9. This could greatly help in choosing an appropriate GE strategy for given loci.

CRISPR-PLANT v2

CRISPR-PLANT is a platform for helping researchers in designing and constructing specific gRNAs for CRISPR/Cas9-mediated GE in plants. CRISPR-PLANT v2 detects every off-target, where other tools have failed in detecting a subset of hidden sequences. However, off-target sequences with gaps were not detected by other tools. This is important because of the considerable off-target activity by CRISPR/Cas9 at sites with one gap or one to three mismatches. Most of the tools use Bowtie, though Bowtie has been designed for mapping up to 1024 bases of sequence when only one hit is expected to exist in the target genome.

CCTop (CRISPR/Cas9 Target Online Predictor)

CCTop has widely been used to provide an intuitive user interface with easily adjustable default parameters by the user. It has the ability to identify and rank all the candidate sgRNA target sequences in accordance with their off-target score, and shows the entire output. CCTop has been widely applied in gene inactivation, HDR and NHEJ experiments. CCTop helps to rapidly and efficiently identify the high-quality target sites. It has numerous options for providing the list of top candidates to both a beginner and the expert, with complete documentation and flexible options. Therefore, the user will be well informed to select the suitable target sites.

CROP-IT (CRISPR/Cas9 Off-Target Prediction and Identification Tool)

CROP-IT is a web-based tool that has been widely used to perform improved site predictions of binding and cleavage of off-target editing. CROP-IT has the ability to integrate biological information in the genome wide scale from current Cas9 editing profile data base (binding and cleavage). CROP-IT has been shown to outperform current computational algorithms for predicting the Cas9 binding and cleavage sites. CROP-IT outputs scores and ranks the potential off-targets, enabling the improvements in the prediction of Cas9 editing profile and then accurate gRNA design.

CHOPCHOP

CHOPCHOP is a web-based tool that has been developed to accept numerous inputs (pasted sequences, genomic regions, gene identifiers) and provides a range of options to select any target. CHOPCHOP uses algorithms with efficient sequence alignment for minimizing search times, and prediction of off-target editing of TALENs and CRISPR/Cas9. Additionally, for each potential target site, the primer candidates and restriction sites are visualized to facilitate an efficient pipeline for generation and validation of mutations. CHOPCHOP has been considered as a valuable tool for genome engineering, allowing users for rapidly and easily select the optimum target sequences for CRISPR/Cas9 or TALEN in genes from a wide range of organisms.

CFD (Cutting Frequency Determination) Score

CFD score is a tool which has been widely used for scoring mismatch position and sequence-dependent off-targets with 240 fitting parameters. CFD score has been used to detect statistically significant increases in predicted off-target sites for lethal sgRNAs. In addition, it has been suggested that the sgRNAs with higher off-target risk are more easily improperly removed in the negative selection module. However, the avoidance of such uninhibited sgRNAs leads to improved library performance. CFD score predicts the probability of off-target cutting and allows the decrease of majority of off-target editing with high efficiency.

CT-Finder (CRISPR Target Finder)

CT-Finder is web tool developed by Zhu et al. with multiple functions.87 Besides the routine service for the off-target prediction for the CRISPR/Cas9 system, it also can be applied for the precise target prediction in two novel systems: RNA-guided FokI nuclease (RFNs) and Cas9 D10A nickases (Cas9 n). Like other web tools, CT-Finder can supports quite a few input setting including maximum numbers of gaps and numbers of mismatch as well as seed region length, with a very friendly and convenient visualization interface-JBrowse, which can visualize off-target and on-target sequences in a genomic context, providing users the incorporation of user inputs for various important features, such as PAM sequence and gRNA length.

CRISPOR

CRISPOR, as a web tool, was developed by Concordet and Haeussler and designed for the gRNAs selection in a wide range genomes type. More than 150 genomes have been added into the data of this tool in the past 3 years. Once the candidate sgRNA sequences loading, CRISPOR will rank them with different scores regarding possible off-target editing in the desired genome, and also to predict on-target activity. It offers an inclusive solution from cloning, expression, and selection of gRNA and provides primers required for analysis of guide activity and possible off-targets. CRISPOR ignores the candidate off-target mutation with the off-target score < 0.1 for the NGG PAM and those with a score less than 1.0 for NGA and NAG PAMs. CRISPOR currently supports 113 genomes.

CRISPR-GE

CRISPR-GE is an integrated web-based tool for expediting all experimental settings and analyses modifications by CRISPR/Cas/Cpf1 systems in various living organisms, including plants. It offers a robust toolkit for designing the target sgRNAs (target Design), primer design to construct the sgRNA expression cassettes, prediction of off-target sites (off Target) as well as the amplification of the target sites with genomic fragments (primer Design). CRISPR-GE provides a practical and comprehensive solution for GE in plants.

An Ensemble Learning Method

Through this method, off-target sites of a sgRNA from its thousands of candidates can be detected genome-wide. This method is based on the occurrence of considerable differences in GC count and the preferences in a mismatch between the positive on-target-off-target sequence pairs and those negative ones. This method enhances the efficiency of off-target site prediction as compared to other computational methods, and can identify more off-target sites consist with bona fide detections through high-throughput methods. According to two case studies, it is efficient in the selection of optimal sgRNAs for treatment of certain genetic disorders.

Cas-OFFinder

Cas-OFFinder is a versatile tool that detects possible off-target mutation sites in a user-defined sequence or a given genome. Compared with other tools for prediction of the off-target sites of RGEN, Cas-OFFinder has no limitations to the number of mismatches, allowing modifications in Cas9-recognized PAM sequences, the vital protein component in RGENs. In any sequenced genome, Cas-OFFinder allows rapid detection of potential off-target editing without restraining the PAM sequence or the number of mismatched bases.